Geochemistry of Oilfield Waters

Developments in Petroleum Science, 1
GEOCHEMISTRY OF OILFIELD WATERS

This Page Intentionally Left Blank
Developments in Petroleum Science, 1
GEOCHEMISTRY OF OILFIELD
WATERS
A. GENE COLLINS
Bartlesville Energy Research Center
Bureau o f Mines
United States Department of the Interior
Bartlesville, Oklahoma, U.S.A.
ELSEVIER SCIENTIFIC PUBLISHING COMPANY

Amsterdam - Oxford - New York 1975
ELSEVIER SCIENTIFIC PUBLISHING COMPANY
335 Jan van Galenstraat
P.O. Box 211,Amsterdam, The Netherlands
AMERICAN ELSEVIER PUBLISHING COMPANY, INC.

52 Vanderbilt Avenue
New York, New York 10017
Library of Congress Card Number: 73-89149
ISBN 0-444-41183-6
With 132 illustrations and 87 tables
Copyright 0 1975 by Elsevier Scientific Publishing Company, Amsterdam

All rights reserved. No part of this publication may be reproduced, stored in a retrieva
system, or transmitted, in any form or by any means, electronic, mechanical, photo
copying, or otherwise without the prior written permission of the publisher, Elseviei
Scientific Publishing Company, Jan van Galenstraat 335,Amsterdam
Printed in The Netherlands

PREFACE
The purpose of this book is to provide information relevant to the analyt-

ical chemistry and geochemistry of oilfield waters. The book attempts to
recognize the importance of subsurface oilfield waters as they are related t o
origin, migration, accumulation, and maturation of oil and gas and thus their
relationship t o exploration for and production of oil and gas. One chapter
presents a simplistic introduction t o the origin of oilfield waters. Because oil-
field waters can constitute an environmental pollution hazard, the book de-
scribes and comments on methods of their disposal or of recovering valuable
constituents from them.
The numerous references indicate that the book relies heavily upon the
work of others. The reader will vastly expand his knowledge of the subject
by consulting these references.
The writer appreciates the understanding and thoughtfulness of his Wife,
Barbara, and children, Sandy and Mike, during the preparation of part of this
book at our home. He acknowledges With appreciation the criticisms, opin-
ions, and suggestions of various portions of the book by O.C. Baptist, W.H.
Caraway, P.H. Dickey, G.L. Gates, R.V. Huff, P.H. Jones, and C.C. Linville.
M.E. Crocker and Ms. C.A. Pearson, did an invaluable service of proof-reading
and index preparation. He extends appreciation t o Ms. D.J. Forbes, Ms. M.G.
Goff, and Ms. J. Haimson for typing the manuscript; t o D.W. Anderson, Ms.
E.S. Baldwin, J.A. Chidester, G.E. Fletcher, R.M. Horn, and W.A. McClung
for preparing the figures; and to authors, book publishers, companies, and
technical journals who granted permission t o use various illustrations.
Permission t o publish this manuscript was granted by the Director of the
United States Bureau of Mines. Bureau of Mines officials who generously
helped obtain this permission were: J.S. Ball, R.T. Johansen, and J.W.Watkins.
Finally inasmuch as it is the writer’s belief that this book is not perfect, he
takes this opportunity to solicit constructive criticism from its readers.
A. GENE COLLINS
Bartlesville Energy Research Center
U.S.Bureau of Mines
Bartlesville, Oklahoma
CONTENTS
Preface . . . . . . . . . . . . . . . . . . . . . . . . . . V
Chapter 1. Introduction . . . . . . . . . . . . . . . . . . . . 1
References . . . . . . . . . . . . . . . . . . . . . . . . . 5
Chapter 2 . Sampling subsurface oilfield waters . . . . . . . . . . . . . 7

Drill-stem test . . . . . . . . . . . . . . . . . . . . . . . 8
Sample containing dissolved gases . . . . . . . . . . . . . . . . . 12
Sampling at the flow line . . . . . . . . . . . . . . . . . . . . 13
Sampling at the wellhead . . . . . . . . . . . . . . . . . . . . 14
Sample for determining unstable properties or species . . . . . . . . . . . 14
Sample for stable-isotope analysis . . . . . . . . . . . . . . . . . 15
Sample containers . . . . . . . . . . . . . . . . . . . . . . 16
Tabulation of sample description . . . . . . . . . . . . . . . . . 17
References . . . . . . . . . . . . . . . . . . . . . . . . . 17
Chapter 3. Analysis of oilfield waters for some physical properties and inorganic
chemical constituents . . . . . . . . . . . . . . . . . 19
Quality control . . . . . . . . . . . . . . . . . . . . . . . 19
Preliminary sample treatment . . . . . . . . . . . . . . . . . . . 22
Reporting the analytical results . . . . . . . . . . . . . . . . . . 25
Synthetic brine . . . . . . . . . . . . . . . . . . . . . . . 27
Determination of pH . . . . . . . . . . . . . . . . . . . . . 27
Determination of Eh . . . . . . . . . . . . . . . . . . . . . 29
Suspended solids . . . . . . . . . . . . . . . . . . . . . . . 31
Resistivity . . . . . . . . . . . . . . . . . . . . . . . . . 32
Specific gravity . . . . . . . . . . . . . . . . . . . . . . . 35
TITRIMETRIC METHODS . . . . . . . . . . . . . . . . . . . 37
Acidity. alkalinity. and borate boron . . . . . . . . . . . . . . . . 37
Calcium and magnesium . . . . . . . . . . . . . . . . . . . . 40
Ammonium nitrogen . . . . . . . . . . . . . . . . . . . . . 43
Chloride . . . . . . . . . . . . . . . . . . . . . . . . . . 44
Bromide and iodide . . . . . . . . . . . . . . . . . . . . . . 45
Oxygen . . . . . . . . . . . . . . . . . . . . . . . . . . 47
Carbon dioxide . . . . . . . . . . . . . . . . . . . . . . . 50
Sulfide . . . . . . . . . . . . . . . . . . . . . . . . . . 51
Sulfur compounds . . . . . . . . . . . . . . . . . . . . . . 52
FLAME SPECTROPHOTOMETRIC .METHODS . . . . . . . . . . . . 53
Lithium . . . . . . . . . . . . . . . . . . . . . . . . . . 54
Sodium . . . . . . . . . . . . . . . . . . . . . . . . . . 57
Potassium . . . . . . . . . . . . . . . . . . . . . . . . . 58
Rubidium and cesium . . . . . . . . . . . . . . . . . . . . . 59
Manganese . . . . . . . . . . . . . . . . . . . . . . . . . 61
Strontium . . . . . . . . . . . . . . . . . . . . . . . . . 62
Barium . . . . . . . . . . . . . . . . . . . . . . . . . . 63
VIII CONTENTS
ATOMIC ABSORPTION METHODS . . . . . . . . . . . . . . . . 65

Interferences . . . . . . . . . . . . . . . . . . . . . . . . 66
Burners and solvents . . . . . . . . . . . . . . . . . . . . . . 66
Lithium . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
Sodium . . . . . . . . . . . . . . . . . . . . . . . . . . 68
Potassium . . . . . . . . . . . . . . . . . . . . . . . . . 70
Magnesium (1) . . . . . . . . . . . . . . . . . . . . . . . . 71
Calcium (1) . . . . . . . . . . . . . . . . . . . . . . . . . . 72
Magnesium (2) . . . . . . . . . . . . . . . . . . . . . . . . 74
Calcium (2) . . . . . . . . . . . . . . . . . . . . . . . . . 75
Strontium . . . . . . . . . . . . . . . . . . . . . . . . . 76
Barium . . . . . . . . . . . . . . . . . . . . . . . . . . 77
Manganese . . . . . . . . . . . . . . . . . . . . . . . . . 78
Iron . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
Copper . . . . . . . . . . . . . . . . . . . . . . . . . . 80
Zinc . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
Lead(1) . . . . . . . . . . . . . . . . . . . . . . . . . . 81
Lead(2) . . . . . . . . . . . . . . . . . . . . . . . . . : 82
EMISSION SPECTROSCOPY . . . . . . . . . . . . . . . . . . . 83
Barium, boron, iron, manganese, and strontium . . . . . . . . . . . . . 83
Beryllium . . . . . . . . . . . . . . . . . . . . . . . . . 89
Aluminum . . . . . . . . . . . . . . . . . . . . . . . . . 90
MASS SPECTROMETRIC METHODS FOR STABLE ISOTOPES . . . . . . . 91
Deuterium . . . . . . . . . . . . . . . . . . . . . . . . . 91
Oxygen-18 . . . . . . . . . . . . . . . . . . . . . . . . . 91
COLORIMETRIC METHODS . . . . . . . . . . . . . . . . . . 92
Interferences . . . . . . . . . . . . . . . . . . . . . . . . 93
Iron . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
Copper . . . . . . . . . . . . . . . . . . . . . . . . . . 96
Nickel . . . . . . . . . . . . . . . . . . . . . . . . . . 98
Lead . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
Zinc . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
Cadmium . . . . . . . . . . . . . . . . . . . . . . . . . 103
Phosphate . . . . . . . . . . . . . . . . . . . . . . . . . 105
Silica . . . . . . . . . . . . . . . . . . . . . . . . . . . 107
Nitrate nitrogen . . . . . . . . . . . . . . . . . . . . . . . 107
Arsenic . . . . . . . . . . . . . . . . . . . . . . . . . . 108
Fluoride . . . . . . . . . . . . . . . . . . . . . . . . . . 109
Iodide . . . . . . . . . . . . . . . . . . . . . . . . . . 110
Selenium . . . . . . . . . . . . . . . . . . . . . . . . . 111
Barium . . . . . . . . . . . . . . . . . . . . . . . . . . 114
GRAVIMETRIC METHODS . . . . . . . . . . . . . . . . . . . 114
Sulfate . . . . . . . . . . . . . . . . . . . . . . . . . . 114
Barium . . . . . . . . . . . . . . . . . . . . . . . . . . 115
OTHER METHODS . . . . . . . . . . . . . . . . . . . . . . 116
Sodium . . . . . . . . . . . . . . . . . . . . . . . . . . 116
Dissolved solids . . . . . . . . . . . . . . . . . . . . . . . 117
Spent acid . . . . . . . . . . . . . . . . . . . . . . . . . 118
Acetic acid solutions . . . . . . . . . . . . . . . . . . . . . . 120
References . . . . . . . . . . . . . . . . . . . . . . . . . 121
Chapter 4 . Interpretation of chemical analysis of oilfield waters . . . . . . . 125

Calculating probable compounds . . . . . . . . . . . . . . . . . 125
CONTENTS IX
Determining a sought compound . . . . . . . . . . . . . . . . . 127

Graphic plots . . . . . . . . . . . . . . . . . . . . . . . . 128
References . . . . . . . . . . . . . . . . . . . . . . . . . 132
Chapter 5 . Significance of some inorganic constituents and physical properties of oil-

field waters . . . . . . . . . . . . . . . . . . . . . 133
Lithium . . . . . . . . . . . . . . . . . . . . . . . . . . 133
Sodium . . . . . . . . . . . . . . . . . . . . . . . . . . 136
Potassium . . . . . . . . . . . . . . . . . . . . . . . . . 138
Rubidium . . . . . . . . . . . . . . . . . . . . . . . . . 140
Cesium . . . . . . . . . . . . . . . . . . . . . . . . . . 141
Beryllium . . . . . . . . . . . . . . . . . . . . . . . . . 141
Magnesium . . . . . . . . . . . . . . . . . . . . . . . . . 142
Calcium . . . . . . . . . . . . . . . . . . . . . . . . . . 143
Strontium . . . . . . . . . . . . . . . . . . . . . . . . . 145
Barium . . . . . . . . . . . . . . . . . . . . . . . . . . 147
Manganese . . . . . . . . . . . . . . . . . . . . . . . . . 149
Iron . . . . . . . . . . . . . . . . . . . . . . . . . . . 149
Copper . . . . . . . . . . . . . . . . . . . . . . . . . . 150
Zinc . . . . . . . . . . . . . . . . . . . . . . . . . . . 151
Mercury . . . . . . . . . . . . . . . . . . . . . . . . . . 151
Lead . . . . . . . . . . . . . . . . . . . . . . . . . . . 152
Cadmium . . . . . . . . . . . . . . . . . . . . . . . . . 152
Boron . . . . . . . . . . . . . . . . . . . . . . . . . . 153
Aluminum . . . . . . . . . . . . . . . . . . . . . . . . . 155
Silica . . . . . . . . . . . . . . . . . . . . . . . . . . . 156
Ammonium nitrogen . . . . . . . . . . . . . . . . . . . . . 157
Phosphorus . . . . . . . . . . . . . . . . . . . . . . . . . 158
Arsenic . . . . . . . . . . . . . . . . . . . . . . . . . . 158
Oxygen . . . . . . . . . . . . . . . . . . . . . . . . . . 158
Sulfur . . . . . . . . . . . . . . . . . . . . . . . . . . 159
Selenium . . . . . . . . . . . . . . . . . . . . . . . . . 160
Fluorine . . . . . . . . . . . . . . . . . . . . . . . . . . 161
Chlorine . . . . . . . . . . . . . . . . . . . . . . . . . . 161
Bromine . . . . . . . . . . . . . . . . . . . . . . . . . . 162
Iodine . . . . . . . . . . . . . . . . . . . . . . . . . . 164
Significance of some physical properties . . . . . . . . . . . . . . . 166
References . . . . . . . . . . . . . . . . . . . . . . . . . 174
Chapter 6. Organic constituents in saline waters .

. . . . . . . . . . . . 177
Nitrogen-free organic compounds . . . . . . . . . . . . . . . . . 178
Hydrocarbons containing nitrogen . . . . . . . . . . . . . . . . . 182
Fatty acids . . . . . . . . . . . . . . . . . . . . . . . . . 183
Naphthenic and humic acids . . . . . . . . . . . . . . . . . . . 185
Determination of oil in water . . . . . . . . . . . . . . . . . . . 186
Organic acids in oilfield brines . . . . . . . . . . . . . . . . . . 188
References . . . . . . . . . . . . . . . . . . . . . . . . . 188
Chapter 7. Origin of oilfield waters . . . . . . . . . . . . . . . . . 193

Definitions of some water terms . . . . . . . . . . . . . . . . . . 194
Sedimentary rocks . . . . . . . . . . . . . . . . . . . . . . 195
Composition of oilfield waters . . . . . . . . . . . . . . . . . . 213
Research studies related t o the originof oilfield brines . . . . . . . . . . 219
X CONTENTS
Conclusions . . . . . . . . . . . . . . . . . . . . . . . . 245
References . . . . . . . . . . . . . . . . . . . . . . . . . 246
Chapter 8 . Classification of oilfield waters . . . . . . . . . . . . . . 253

Palmer’s classification . . . . . . . . . . . . . . . . . . . . . 254
Sulin’s classification . . . . . . . . . . . . . . . . . . . . . . 257
Modification of Sulin’s system by Bojarski . . . . . . . . . . . . . . 260
Chebotarev’s classification . . . . . . . . . . . . . . . . . . . . 262
Schoeller’s system . . . . . . . . . . . . . . . . . . . . . . 267
Oilfield brine analyses . . . . . . . . . . . . . . . . . . . . . 272
Application of the classification systems . . . . . . . . . . . . . . . 274
Discussion . . . . . . . . . . . . . . . . . . . . . . . . . 289
Conclusions . . . . . . . . . . . . . . . . . . . . . . . . 291
References . . . . . . . . . . . . . . . . . . . . . . . . . 291
Chapter 9. Some effects of water upon the generation. migration. accumulation. and
alteration of petroleum . . . . . . . . . . . . . . . . . 293
Compaction . . . . . . . . . . . . . . . . . . . . . . . . 294
Generation and migration . . . . . . . . . . . . . . . . . . . . 295
Accumulation . . . . . . . . . . . . . . . . . . . . . . . . 298
Alteration . . . . . . . . . . . . . . . . . . . . . . . . . 299
Conclusions . . . . . . . . . . . . . . . . . . . . . . . . 304
References . . . . . . . . . . . . . . . . . . . . . . . . . 304
Chapter 10. Geochemical methods of exploration for petroleum and natural gas . . 307
Introduction . . . . . . . . . . . . . . . . . . . . . . . . 307
Hydrogeochemical research and methods . . . . . . . . . . . . . . . 313
Reviews . . . . . . . . . . . . . . . . . . . . . . . . . . 322
Case history of the Delaware sand (Bell Canyon formation). Texas. by Visher (1961) 322
Formation water maps of others areas . . . . . . . . . . . . . . . . 330
Concluding remarks . . . . . . . . . . . . . . . . . . . . . . 335
References . . . . . . . . . . . . . . . . . . . . . . . . . 337
Chapter 11. Geopressured reservoirs . . . . . . . . . . . . . . . . 343

Geopressure . . . . . . . . . . . . . . . . . . . . . . . . 343
Origin of abnormal pressures . . . . . . . . . . . . .
. . . . . . 344
Abnormal pressures in the Gulf Coast area . . . . . . . .
. . . . . . 346
Detection of abnormal pressures . . . . . . . . . . . .
. . . . . . 362
References . . . . . . . . . . . . . . . . . . .
. . . . . . 364
Chapter 12. Compatibility of oilfield waters . . . . . . . .

. . . . . 367 .
Wellbore and formation damage . . . . . . . . . . . .
. . . . . 368 .
Solubility of calciumcompounds invarioussaltsolutions . . .
. . . . . 370.
Solubilities of the sulfates of barium and strontium in saline solutions . . . . . 372
Experimental determination of some solubilities of the sulfates of barium and
strontium . . . . . . . . . . . . . . . . . . . . . . . . . 372
Resultsand discussion of the experimental investigation . . . . . . . . . . 373
Brine stabilization . . . . . . . . . . . . . . . . . . . . . . 380
Mixing of subsurface waters . . . . . . . . . . . . . . . . . . . 382
References . . . . . . . . . . . . . . . . . . . . . . . . . 386
.
Chapter 13 Valuable minerals in oilfield waters . . . . . . . . . . . . . 389
Recovery of iodine and bromine from oilfield brines . . . . . . . . . . . 390
CONTENTS XI
Minerals recovered from saline waters . . . . . . . . . . . . . . . . . 392

Fresh-water production . . . . . . . . . . . . . . . . . . . . . 401
Preliminary economic evaluation . . . . . . . . . . . . . . . . . 402
Disposal brines . . . . . . . . . . . . . . . . . . . . . . . 411
Worth and value estimates . . . . . . . . . . . . . . . . . . . . 411
Conclusions . . . . . . . . . . . . . . . . . . . . . . . . 417
References . . . . . . . . . . . . . . . . . . . . . . . . . 417
Chapter 1 4.Subsurface disposal . . . . . . . . . . . . . . . . . . 419

History of brine disposal operations . . . . . . . . . . . . . . . . 419
Subsurface injection . . . . . . . . . . . . . . . . . . . . . . 420
Present-day technology in subsurface disposal . . . . . . . . . . . . . 421
Economics and oilfield brine disposal . . . . . . . . . . . . . . . . 422
Injection well versus disposal well . . . . . . . . . . . . . . . . . 424
Acceptable geologic areas . . . . . . . . . . . . . . . . . . . . 425
Suitable disposal zones . . . . . . . . . . . . . . . . . . . . . 426
Evaluation of the disposal zone . . . . . . . . . . . . . . . . . . 427
Considerations during drilling and well completion . . . . . . . . . . . . 432
Fluid travel . . . . . . . . . . . . . . . . . . . . . . . . . 433
Hazards of underground waste disposal . . . . . . . . . . . . . . . 434
State regulations and tax incentives . . . . . . . . . . . . . . . . . 434
Costs of disposal systems . . . . . . . . . . . . . . . . . . . . 437
Conclusions . . . . . . . . . . . . . . . . . . . . . . . . 438
References . . . . . . . . . . . . . . . . . . . . . . . . . 438
.
Chapter 1 5 Solubilities of some silicate minerals in saline waters . . . . . . . 441
Composition and structure of minerals studied . . . . . . . . . . . . . 441
Silicate solubilities a t 25°C and 1 atm . . . . . . . . . . . . . . . . 443
Experimental equipment . . . . . . . . . . . . . . . . . . . . 447
Experimental method . . . . . . . . . . . . . . . . . . . . . 449
Fundamental equations . . . . . . . . . . . . . . . . . . . . . 449
Experimental dataandempirical equations . . . . . . . . . . . . . . 453
Summary and conclusions . . . . . . . . . . . . . . . . . . . . 457
References . . . . . . . . . . . . . . . . . . . . . . . . . 459
.
Chapter 16 Environmental impact of oil- and gas-well drilling. production and
associated waste disposal practices . . . . . . . . . . . . . 461
Drilling . . . . . . . . . . . . . . . . . . . . . . . . . . 461
Production . . . . . . . . . . . . . . . . . . . . . . . . . 467
Disposal . . . . . . . . . . . . . . . . . . . . . . . . . . 471
References . . . . . . . . . . . . . . . . . . . . . . . . . 474
Reference Index . . . . . . . . . . . . . . . . . . . . . . . 477

Subject Index . . . . . . . . . . . . . . . . . . . . . . . * 485
Chapter 1. INTRODUCTION
Petroleum, known to exist long before an oil well was drilled, first found
limited use as a medicine, lubricant, and waterproofing agent. The American
Indians knew of several oil and gas springs and gave this information t o the
early American settlers. Early settlements were commonly located close to
salt licks which supplied salt to the population. Often these salt springs were
contaminated with petroleum, and many of the early efforts to acquire more
salt by digging wells were rewarded by finding unwanted increased amounts
of oil and gas associated with the saline waters. In the Appalachians: many
saline water springs occurred along the crests of anticlines (Rogers and
Rogers, 1843).
In 1855 it was found that distillation of petroleum produced a light oil
similar to coal oil, which was better than whale oil as an illuminant (Howell,
1934, p.2). This knowledge spurred the search for saline waters which con-
tained oil. Colonel Edward Drake, utilizing the methods of the salt pro-
ducers, drilled a well on Oil Creek, near Titusville, Pennsylvania, in 1859. He
struck oil at a depth of 21 m, and this first oil well produced about 35 barrels
of oil per day (Dickey, 1959).
The early oil producers did not realize the significance of the oil and saline
waters occurring together. In fact, it was not until 1938 that the existence of
interstitial water in oil reservoirs was generally recognized (Schilthuis, 1938).
Torrey (1966) was convinced as early as 1928 that dispersed interstitial
water existed in oil reservoirs, but his belief was rejected by his colleagues
because most of the producing oil wells did not produce any water upon
completion. Occurrences of mixtures of oil and gas with water were rec-
ognized by Griswold and Munn (1907), but they believed that there was a
definite separation of the oil and water, and that oil, gas, and water mixtures
did not occur in the sand before a well tapped the reservoir.
It was not until 1928 that the first commercial laboratory for the analysis
of rock cores was established (Torrey, 1966); the first core tested was from
the Bradford Third Sand (from the Bradford field, McKean County, Pennsyl-
vania). The percent saturation and percent porosity of this core were plotted
versus depth t o construct a graphic representation of the oil and water
saturation. The soluble mineral salts that were extracted from the core led
Torrey to suspect that water was indigenous t o the oil productive sand.
Shortly thereafter a test well was drilled near Custer City, Pennsylvania,
which encountered higher than average oil saturation in the lower part of the
Bradford Sand. This high oil saturation resulted from the action of an un-
2 INTRODUCTION
suspected flood, the existence of which was not known when the location
for the test well had been selected. The upper part of the sand was not
cored. Toward the end of the cutting of the first core with a Baker cable tool
core barrel, oil began t o come into the hole so fast that it was not necessary
t o add water for the cutting of the second section of the sand. The lower 1 m
of the Bradford Sand therefore was cut with oil in a hole free from water.
Two samples from this section were preserved in sealed containers for satura-
tion tests, and both of them, when analyzed, had a water content of ap-
proximately 20% of pore volume. This well made about 10 barrels of oil per
day and no water after being shot with nitroglycerine. Thus, the evidence
developed by the core analysis and the productivity test after completion
provided a satisfactory indication of the existence of immobile water, in-
digenous t o the Bradford Sand oil reservoir, which was held in its pore
system and which could not be produced by conventional pumping methods
(Torrey, 1966).
Fettke (1938) was the first t o report the presence of water in an oil-
producing sand. However, he thought that it might have been introduced by
the drilling process.
It was recognized by Munn (1920)that moving underground water might
be the primary cause of migration and accumulation of oil and gas. However,
this theory had little experimental data t o back it until Mills (1920)con-
ducted several laboratory experiments on the effect of moving water and gas
on water-oil-as-sand and water-oil-sand systems. Mills concluded that
“the up-dip migration of oil and gas under the propulsive force of their
buoyancy in water, as well as the migration of oil, either up or down dip,
caused by hydraulic currents, are among the primary factors influencing
both the accumulation and the recovery of oil and gas.” This theory was
seriously questioned and completely rejected by many of his con-
temporaries.
Rich (1923)postulated that “hydraulic currents, rather than buoyancy,
are effective in causing accumulation of oil or its retention.” He did not
believe that the hydraulic accumulation and flushing of oil required a rapid
movement of water, but rather that the oil was an integral constituent of the
rock fluids and that it could be carried along with them whether the move-
ment was very slow or relatively rapid.
The effect of water displacing oil during production was not recognized in
the early days of the petroleum industry in Pennsylvania. Laws were passed, a
however, to prevent operators from injecting water into the oil reservoir
sands through unplugged wells. In spite of these laws, some operators at
Bradford surreptitiously opened the well casing opposite shallow ground-
water sands in order to start a waterflood in the oil sands. Effects of artificial
waterfloods were noted in the Bradford field, McKean County, Pennsylvania,
in 1907, and became evident about 5 years later in the nearby oilfields of
New York (Torrey, 1950). Volumetric calculations of the oil-reservoir
volume which were made for engineering studies of these waterflood opera-
INTRODUCTION 3
tions proved that interstitial water was generally present in the oil sands.
Publications by Garrison (1935) and Schilthuis (1938) give detailed informa-
tion concerning the distribution of water and oil in porous rocks, and of the
origin and occurrence of “connate” water with information concerning the
relationship of water saturation to formation permeability.
The word “connate” was first used by Lane and Gordon (1908) to mean
interstitial water that was deposited with the sediments. The processes of
rock compaction and mineral diagenesis result in the expulsion of large
amounts of water from sediments and movement out of the deposit through
the more permeable rocks. It is therefore highly unlikely that the water
now in any pore is the same as that which was there when the particles that
surround it were deposited. White (1957) redefined connate water as “fossil”
water; it has been out of contact with the atmosphere for an appreciable part
of a geologic time period. Connate water is thus distinguished from meteoric
water which has entered the rocks in geologically recent times, and from
juvenile water which has come from deep in the earth’s crust and has never
been in contact with the atmosphere.
Meanwhile petroleum engineers and geologists had learned that waters
associated with petroleum could be identified with regard to the reservoir in
which they occurred by a knowledge of their chemical characteristics. Com-
monly the waters from different strata differ considerably in their dissolved
chemical constituents, making the identification of a water from a particular
strata easy. However, in some areas the concentrations of dissolved con-
stituents in waters from different strata do not differ significantly, and the
identification of such waters is difficult or impossible.
The amount of water produced with the oil often increases as the amount
of oil produced decreases. If this is edge water, nothing can be done about it.
If it is bottom water, the well can be plugged back. However, it often is
intrusive water from a shallow sand gaining access t o the well from a leaky
casing or faulty completion and this can be repaired.
Enormous quantities of water are produced with the oil in some fields,
and it is necessary to separate the oil from the water. Most of the oil can be
removed by settling. Often, however, an oil-in-water emulsion forms which is
very difficult t o break. In such cases, the oil is heated and various surface-
active chemicals are added to induce separation.
In the early days, the water was dumped on the ground where it seeped
below the land surface. Until about 1930, the oilfield waters were disposed
into local drainage, frequently killing fish and even surface vegetation. After
1930, it became common practice t o evaporate the water in earthen pits or
to inject it into the producing sand or another deep aquifer. The primary
concern in such disposal practice is to remove all oil and basic sediment from
the waters before pumping them into injection wells, to prevent clogging of
the pore spaces in the formation receiving the waste water. Chemical com-
patibility of waste water and host aquifer water must also be assured.
Waters produced with petroleum are growing in importance. In years past,
4 INTRODUCTION
these waters were considered waste and had t o be disposed of in some

manner. Injection of these waters into the petroleum reservoir rock serves
three purposes: (1) it produces additional petroleum (secondary recovery);
(2) it utilizes a potential pollutant; and (3) in some areas it controls land
subsidence.
The volume of water produced with petroleum in the United States is very
large. In 1970, daily production was about 3.78 trillion liters of water with
about 1.51 trillion liters of oil. In older fields, the production is frequently
95%water and 5%petroleum.
. Waterflooding in petroleum production is expanding rapidly, and in 1970
one-third to one-half of the production in the United States came from fields
into which water was injected. The volume of injected water has grown each
year. In many fields the volume of petroleum produced by secondary re-
covery by waterflooding is equal to the volume recovered by primary
met hods.
To inject these waters into reservoir rocks, suspended solids and oil must
be removed from the waters to prevent plugging of the porous formations.
Water injection systems require separators, filters, and, in some areas,
deoxygenating equipment utilizing chemical and physical control methods to
minimize corrosion and plugging in the injection system.
In waterflooding most petroleum reservoirs, the volume of produced
water is not sufficient t o efficiently recover the additional petroleum. There.
fore, supplemental water must be added t o the petroleum reservoir. The use
of waters from other sources requires that the blending of the produced
water with supplemental water must yield a chemically stable mixture so
that plugging solids will not be formed. For example, a produced water
containing considerable calcium should not be mixed with a water con-
taining considerable carbonate because calcium carbonate may precipitate
and prevent injection of the flood water. The design and successful operation
of a secondary recovery waterflood requires a thorough knowledge of the
composition of the waters used.
Chemical analyses of waters produced with oil are useful in oil production
problems, such as identifying the source of intrusive water, planning water-
flood and salt-water disposal projects, and treating t o prevent corrosion
problems in primary and secondary recovery. Electrical well-log interpreta
tion requires a knowledge of the dissolved solids concentration and composi
tion of the interstitial water. Such information also is useful in correlationof
stratigraphic units and of the aquifers within these units, and in studiesof
the movement of subsurface waters. It is impossible to understandthe
processes that accumulate petroleum or other minerals without insight in to
the nature of these waters.
REFERENCES 5
References
Dickey, P.A., 1959. The first oil well. J. Pet. Technol., 11:14-26.
Fettke, C.R., 1938. Bradford oil field, Pennsylvania, and New York. Pa. Geol. Surv.,
Fourth Ser., Bull., M21:l-454.
Garrison, A.D., 1935. Selective wetting of reservoir rock and its relation to oil produc-
tion. In: Drilling and Production Practice. American Petroleum Institute, New York,
N.Y., pp.130-140.
Griswold, W.T. and Munn, M.J., 1907. Geology of oil and gas fields in Steubenville,
Burgettstown and Claysville Quadrangles, Ohio, West Virginia and Pennsylvania. U.S.
Geol. Sum. Bull., No.318, 196 pp.
Howell, J.V., 1934. Historical development of the structural theory of accumulation of
oil and gas. In: W.E. Wrather and F.H. Lahee (Editors), Problems of Petroleum
Geology. American Association of Petroleum Geologists, Tulsa, Okla., pp.1-23.
Lane, A.C. and Gordon, W.C., 1908. Mine waters and their field assay. Bull. Geol. SOC.
A m . , 19:501-512.
Mills, R. van A., 1920. Experimental studies of subsurface relationships in oil and gas
fields. Econ. Geol., 15:398-421.
Munn, M.J., 1920. The anticlinal and hydraulic theories of oil and gas accumulation.
Econ. Geol., 4:509-529.
Rich, J.L., 1923. Further notes on the hydraulic theory of oil migration and accumula-
tion. Bull. Am. Assoc. Pet. Geol., 7:213-225.
Rogers, W.B. and Rogers H.D., 1843. On the connection of thermal springs in Virginia
with anticlinal axes and faults. A m . Geol. Rep., 1313.323-347.
Schilthuis, R.J., 1938. Connate water in oil and gas sands. In: Petroleum Development
and Technology, AIME, pp.199-214.
Torrey, P.D., 1950. A review of secondary recovery of oil in the United States. In:
Secondary Recovery of Oil in the United States. American Petroleum Institute, New
York, N.Y., pp.3-29.
Torrey, P.D., 1966. The discovery of interstitial water. Prod. Monthly, 30:8-12.
White, D.E., 1957. Magmatic, connate, and metamorphic water. Bull. Geol. SOC.A m . ,
68:1659-1682.
Chapter 2. SAMPLING SUBSURFACE OILFIELD WATERS
Subsurface waters associated with petroleum are subjected to forces that

promote mixing and homogeneity, but the assumption cannot be made that
they are so well mixed that no attention to sampling technique is required.
Localized conditions within an aquifer are commonly such that a given
subsurface body of water may not be of uniform composition. The com-
position of subsurface water commonly changes with depth, and also later-
ally in the same aquifer. Changes may be brought about by the intrusion of
other waters, and by discharge from and recharge to the aquifer. It is thus
difficult to obtain a representative sample of a given subsurface body of
water because any one sample is a very small part of the total mass, which
may vary widely in composition. To develop a comprehensive picture of the
composition characteristics of the total mass, it is generally necessary t o
obtain and analyze many samples. Also, the samples may change with time
as gases come out of solution and supersaturated solutions approach satura-
tion.
The sampling sites should be selected, if possible, t o fit into a comprehen-
sive network t o cover an oil-productive geologic basin. Considerations in
selecting sampling sites are as follows:
(1)Which sites will better fit into an overall plan to evaluate the chemistry
of the waters on a broad basis?
(2) Which sites will yield the better information for correlation with data
obtained from other sites?
(3) Which sites are more representative of the total chemistry of brines
from a given area?
The value of the sample is directly proportional to the facts known about
its source; therefore, sites should be selected for which the greater source
knowledge is available.
For surveillance purposes, samples can be collected from the same site at
sufficiently frequent intervals that no important change in quality will occur
between sampling times. Change in composition may result from changes in
rate of water movement, pumpage rates, or infiltration of other water.
Changes that can occur in petroleum-associated water are illustrated in Table
2.1.
Well 1 shows the sort of change that commonly occurs. The water from
well 2 did not change between 1947 and 1957, within the accuracy of the
analytical determination. Water from well 3 changed drastically, suggesting
the intrusion of water from a different source.
8 SAMPLING SUBSURFACE OILFIELD WATERS
TABLE 2.1
Changes in the composition of petroleum-associated waters (mg/l)
Constituent Well 1 Well 2 Well 3
1947 1957 1947 1957 1956 1959
Sodium and
potassium 29,062 25,000 46,038 45,924 1,491 856
Magnesium 1,100 1,200 2,011 2,200 30 2
Calcium 5,900 5,500 14,200 14,400 60 10
Bicarbonate 34 12 24 12 600 1,800
Sulfate 14 50 3 52 200 0
Chloride 58,500 51,800 102,100 102,800 2,000 300
Total dissolved
solids 94,610 83,562 164,376 165,388 4,381 2,968
~- _.-
There is a tendency for some petroleum-associated waters t o become more

dilute as the oil reservoir is produced. Such dilution may result from the
movement of dilute water from adjacent compacting clay beds into the
petroleum reservoir as pressure declines with the continued removal of oil
and brine (Wallace, 1969).
The composition of petroleum-associated water varies with the position
within the geologic structure from which it is obtained. For example, if the
water table is tilted, the more dilute water probably will be on the structural-
ly high side. In some cases the salinity will increase upstructure t o a
maximum at the point of oil-water contact. Usually this is caused by in-
filtrating meteoric waters.
Few of the samples collected by drill-stem test are truly representative
formation-water samples. During drilling, the pressure in the well bore is
intentionally maintained higher than that in the formations. Filtrate from
the drilling mud seeps into the permeable strata, and this filtrate is the first
liquid to enter the test tool.
The most truly representative formation-water sample usually is obtained
after the oil well has produced for a period of time and all extraneous fluids
adjacent t o the wellbore have been flushed out. Samples taken immediately
after the well is completed may be contaminated with drilling muds, with
drilling fluids, and/or with well completion fluids, such as filtrate from
cement, tracing fluids, and acids, which contain many different chemicals.
Drill-stem test
The drill-stem test, if properly made, can provide a reliable formation

water sample. Mud filtrate will be the first fluid to enter the drill-stem test
tool, and it will be found at the top of the fluid column immediately below
DRILL-STEM TEST 9
tester
Multiple closed i
pressure s a m p l e
Flow stream
pressure recorde
V e r t i c a l and rot
Locked down
Blanked o f f
pressure r e c o r d
RUNNING
IN HOLE
Fig. 2.1. Multiple closed-in-pressure subsurface sampler. (Courtesy of the Halliburton

Company.)
Droin
YOIV4 Oroln
Floottnq p l s t o n
' h
SAMPLE UNIT,'
for l o w p e r m o b i l i t i e s
Rubber doughnut Sample
Mud -
Dump chamber v o l v e ( l o c k s open)
L " - J
MECHANICAL U N I T F I M - A SAMPLE U N I T
RECORDED TESTER
m E
SP POSITION
Tester positioning depth
SURFACE CONTROL
I N 0 ICATIDNS S A M P L I N G PRESSURE
Pod set
?
action
Tool open
I n i t i a l shut-in
prss1ure
Sampling
pressure
RECORDED LOG
3 F i n a l shul-in
pressure
Hydroltotic
head
Fig. 2.2. Formation interval tester. (Courtesy of Schlumberger Well Services Company.)
DRILL-STEM TEST 11
the oil. At some point down the column a representative formation-water

sample can be found. The point is variable and will be influenced by rock
characteristics, mud pressure, type of mud, and duration of the test. It is
best to sample the water after each stand of pipe is removed. Normally, the
total dissolved solids content will increase downwards and become constant
when pure formation water is obtained, if the concentration continues t o
increase all the way t o the bottom, no representative sample can be ob-
tained. A test that flows water will give even higher assurance of an un-
contaminated sample. If only one drill-stem test water sample is taken for
analysis, it should be taken just above the tool, as this is the last water to
enter the tool and is least likely t o show contamination.
Fig. 2.1 and 2.2 illustrate two drill-stem test tools with their various
components. Fig. 2.1 illfistrates a Halliburton Company tool; Fig. 2.2
illustrates a Schlumberger Well Services Company tool. Other companies
supply equally adequate tools, and reference t o specific brands throughout
this test is made for identification only and does not imply endorsement by
the US. Bureau of Mines. The drill-stem test can provide pressure head and
head decline and buildup data useful in permeability calculation (Brede-
hoeft, 1965) and other information for the determination of additional
reservoir conditions, such as the gas/oil ratio and reservoir depletion
(McAlister et al., 1965). A stratigraphic interval of interest is isolated in the
drilled hole by use of packers attached t o the drill string. Opening the tester
valve in the test string allows the formation fluid to enter the drill pipe.
Pressures are recorded by gages in the bottom of the test tool.
To insure that a representative sample is obtained, the pH, resistivity, and
chloride content of samples taken at intervals down the drill pipe can be
determined. Usually a transition zone will be found below which apparently
uncontaminated formation water will be located. The pH, resistivity, and
chloride content will vary above the transition zone, and they will become
constant below it. The sample taken for analysis in the laboratory can yield
positive evidence of contamination, if present. The two most indicative tests
are pH and the color of a filtered sample. If the filtered sample remains tan
or brown and the color cannot be removed even with pressure filtration, it
probably is contaminated with drilling-mud filtrate. A sample can be placed
on a white-spot plate for color evaluation. For positive identification of the
presence of mud filtrate, a sample of the drilling mud used in drilling the
well can be obtained and allowed t o react with distilled water, the reacted
water .is analyzed to determine the mud-contributed ions, and the suspected
contaminated sample is analyzed t o determine if it contains these ions.
Analyses of water obtained from a drill-stem test of Smackover Limestone
water in Rains County, Texas, show how errors can be caused by improper
sampling of drill-stem test water. Analyses of top, middle, and bottom
samples taken from a 15-m fluid recovery are shown in Table 2.11. These data
show an increase in salinity with depth in the drill pipe, indicating that the
first water was contaminated by mud filtrate (Noad, 1962). The middle
TABLE 2.11
Drill-stem test recovery of Smackover Limestone water
Constituent Concentration (mg/l)
top middle bottom

-
Sodium 29,600 43,500 71,800
Calcium 8,100 13,100 22,400
Magnesium 600 900 1,400
Bicarbonate 500 500 400
Sulfate 2,000 1,300 500
Chloride 59,900 91,800 154,000
Total dissolved
solids 101,000 151,000 251,000
sample is approximately half mud filtrate and half formation water. The
bottom sample is the most representative of Smackover water.
No single procedure is universally applicable for obtaining a sample of
oilfield water. For example, information may be desired concerning the
dissolved gas or hydrocarbons in the water, or the reduced species present -
such as ferrous or manganous compounds. Sampling procedures applicable to
the desired information must be used.
Sample containing dissolved gases
Knowledge of certain dissolved hydrocarbon gases is used in exploration.

Methane is quite soluble in water, but samples must be collected in a sampler
that keeps the subsurface pressure on the sample until it is opened in the
laboratory. The testing tool is kept open until the head of water in the drill
pipe is equalized with the formation pressure or until water flows at the
surface. The pressure equalization may require 4 or more hours. However, a
surface recording subsurface pressure gage can be lowered into the drill pipe
to determine when the pressure has equalized. After equalization of pressure,
formation-water samples can be obtained by lowering a subsurface sampler
into the drill pipe (Buckley et al.,1958). Zarrella et al. (1967) determined
the content of dissolved benzene. For this it is not necessary t o use a sub-
surface sampler; the samples are caught in buckets on opening the pipe
string, and immediately transferred from the buckets t o new narrow-necked
glass or metal containers.
A preferred method of obtaining a sample for subsequent gas analysis is t o
catch the aqueous sample in a metal container of about one-quart capacity.
This sample is immediately transferred to another metal sample container.
The second container should be filled completely t o the top, then the sides
of the can are lightly squeezed t o allow for fluid expansion, and the lid is
sealed tightly. A foil-lined (not plastic) lid should be used. If possible, the
SAMPLING AT THE FLOW LINE 13
sample should be analyzed immediately. If this is not possible, cool or freeze

the sample.
Sampling at the flow line
Another method of obtaining a sample for analyses for dissolved gases is

to place a sampling device in a flow line. Fig. 2.3 illustrates such a device.
-I
container
Valve
P i p e line
f
Valve
Rubber
tube
-
Fig. 2.3. Flow-line sampler.
The device is connected to the flow line, and water is allowed t o flow into
and through the container, which is held above the flow line, until 10 or
more volumes of water have flowed through. The lower valve on the sample
container is closed and the container removed. If any bubbles are present in
the sample, the sample is discarded and a new one is obtained.
Sampling at the wellhead
It is common practice in the oil industry to obtain a sample of formation

water from a sampling valve at the wellhead. A plastic or rubber tube can be
used to transfer the sample from the sample valve into the container. The
source and sample container should be flushed t o remove any foreign
material before a sample is taken. After flushing the system, the end of the
tube is placed in the bottom of the container, and several volumes of fluid
are displaced before the tube is slowly removed from the container and the
container is sealed. Fig. 2.4 illustrates a method of obtaining a sample at the
wellhead. An extension of this method is to place the sample container in a
larger container, insert the tube to the bottom of the sample container, allow
the brine to overflow both containers, withdraw the tube, and cap the
sample under the fluid.
At pumping wellheads the brine will surge out in heads and will be mixed
with oil. In such situations a larger container equipped with a valve at the
bottom can be used as a surge tank or an oil-water separator, or both. To
use this device, place the sample tube in the bottom of the large container,
open the wellhead valve, rinse the large container with the well fluid, allow
the large container t o fill, and withdraw a sample through the valve at the
bottom of the large container. This method will serve to obtain samples that
are relatively oil-free.
We1 l h e a d
O i l and water
Fig. 2.4. Schematic of method of obtaining a sample at the wellhead.
Sample for determining unstable properties or species
The pH, Eh, and various species of elements are unstable and will change
with changes in pressure and temperature, and when the formation water is
exposed to the atmosphere. The pH of the sample will change because of the
oxidation of reduced species, because of release of dissolved gases, and be-
cause of hydrolysis reactions such as:
H+ H+
c03-* + HCO,-+ H,CO,
SAMPLE FOR STABLE-ISOTOPE ANALYSIS 15
Because the pH of the formation water sample will change, the pH should
be determined using a flowing sample. A pH/Eh flow sampling chamber
(Collins, 1964) is shown in Fig. 2.5. The Eh determination is difficult and
for corroboration it should be checked using a knowledge of the dissolved
Fe+*/Fe+3ratio of the water.
Ther mocompensator
Fig. 2.5. Flow chamber for use in determining pH and Eh at the wellhead.
Ferrous iron will oxidize to ferric and should be determined immediately

after collecting a fresh sample. Some of the other dissolved constituents that
should be determined immediately after securing a fresh sample are oxygen,
hydrogen sulfide, thiosulfate, and manganous manganese.
Sample for stable-isotope analysis
A sample that is t o be analyzed for stable isotopes should be collected

with care. If possible, such a sample should be taken at reservoir tempera-
tures and pressures to minimize any isotope fractionation. However, because
this usually is impossible, caution should be exercised t o insure that a
representative sample is collected at the prevailing wellhead temperature and
pressure.
The sample should be collected at the wellhead. If this proves impossible,
it may be feasible t o collect the sample from a nonheated separator or

heater; samples are not to be taken of water that has been heated or treated
with any chemicals. Glass sample bottles (about 100 ml usually is sufficient)
should be used, and the sample should overflow the bottle. The bottle
should be closed with a cap equipped with a plastic insert, and the top
should be sealed with wax to minimize exchange reactions with air.
Sample containers
Various factors influence the type of sample container that is selected.

Containers that are used include polyethylene, other plastics, hard rubber,
metal cans, and borosilicate glass. Glass will absorb various ions such as iron
and manganese, and may contribute boron or silica to the aqueous sample.
Plastic and hard rubber containers are not suitable if the sample is to be
analyzed to determine its organic content. A metal container is used by some
laboratories if the sample is to be analyzed for dissolved hydrocarbons such
as benzene.
TABLE 2.111
Description needed for each petroleum-associated water sample

__ --
Sample number Field
Farm or lease Well No. in the
of Section Township Range __
County State - Operator __
Operator’s address (main office) - -___
Sample obtained by -_____ Date ~
Address ~ Representing ~
Sample obtained from (lead line, separatory flow tank, etc.)

Completion date of well Elevation of well ___
Name of productive zone from which sample is produced -
Sand -Shale Lime~ Other
Name of productive Names of formations
formation well passes through
Depths: Top of formation Bottom of formation
Top of producing zone Bottom of producing zone
Total depth drilled Present depth
Bottom hole pressure and date of pressure
Bottom hole temperature
Date of last workover b e any chemicals If yes,
added to treat well? -what?
Well production Initial Present Casing service record:
Oil, barrelslday -
Water, barrels/day
Gas, cubic feetlday _______
Method of production (primary or secondary)
Remarks: (such as casing leaks, communication, or other pays in same well, lease, or field)
TABULATION OF SAMPLE DESCRIPTION 17
The type of container selected is dependent upon the planned use of the
analytical'data. Probably the more satisfactory container, if the sample is to
be stored for some time before analysis, is the polyethylene bottle. All
polyethylenes are not satisfactory because some contain relatively high
amounts of metals contributed by catalysts in their manufacture. The
approximate metal content of the plastic can be determined using a qualita-
tive emission spectrographic technique. If the sample is transported during
freezing temperatures, the plastic container is less likely to break than glass.
The practice of obtaining two samples and acidifying one sample so that
the heavy metals will stay in solution works better if the plastic container is
used. Some of the heavy metals are adsorbed by glass even if the sample is
acidified.
Tabulation of sample description
The sample is of little value if detailed information concerning it is not

available. Information such as that in Table 2.111 should be obtained for each
sample of petroleum-associated water, and for certain types of studies, addi-
tional information may be needed.
References
Bredehoeft, J.D., 1965. The drill-stem test: the petroleum industry's deep-well pumping
test. Ground Water, 3:15-23.
Buckley, S.E., Hocott, C.R. and Taggart, Jr., M.S., 1958.Distribution of dissolved hydro-
carbons in subsurface waters. In: L.C. Weeks (Editor), Habitat of Oil. American
Association Petroleum Geologists, Tulsa, Okla., pp.850-882.
Collins, A.G., 1964. Eh and pH of oilfield waters. Prod. Monthly, 28:ll-12.
McAlister, J.A., Nutter, B.P. and Lebourg, M., 1965. A new system of tools for better
control and interpretation of drill-stem tests. J. Pet. Technol., 17 :207-214.
Noad, D.F., 1962. Water analysis data, interpretation and applications. J. Can. Pet.
TechnoL , 1 :82-89.
Wallace, W.E., 1969. Water production from abnormally pressured gas reservoirs in South
Louisiana, J. Pet. Technol., 21 :969-982.
Zarrella, W.M., Mousseau, R.J., Coggeshall, N.E., Norris, M.S. and Schrayer, G.T., 1967.
Analysis and significance of hydrocarbons in subsurface brines. Geochim. Cosmochim.
Acta, 31 :1155-1166.
Chapter 3. ANALYSIS OF OILFIELD WATERS FOR SOME PHYSICAL
PROPERTIES AND INORGANIC CHEMICAL CONSTITU-
ENTS
Water analyses are used by the petroleum industry in studies related to

subsurface formation identification, pollution problems, water compatibili-
ties, corrosion, water-quality control, waterflooding, and exploration.
Efforts to standardize methods applicable to analyzing oilfield waters have
been made by the American Petroleum Institute (1968),and currently
similar efforts are being made by the American Society for Testing and
Materials.
The methods discussed in this chapter include wet chemical procedures
for calcium, magnesium, barium, carbon dioxide, sulfide, sulfur compounds,
selenium, oxygen, spent acid, fluoride, chloride, bromide, and iodide. In-
strumental methods are described for pH, Eh, specific gravity, resistivity,
suspended solids, acidity, alkalinity, oxygen isotopes, ammonium nitrogen,
phosphate, boron, arsenic, copper, nickel, lead, manganese, zinc, cadmium,
and silica. Also described are emission and atomic absorption methods for
lithium, sodium, potassium, rubidium, cesium, magnesium, calcium, barium,
manganese, zinc, copper, iron, and lead; and emission spectroscopic methods
for aluminum, beryllium, boron, iron, manganese, and strontium. ._
The methods used to analyze oilfield waters should be capable of producing
precise and accurate results. Methods applicable to analyzing fresh waters
may or may not be directly applicable to a petroleum-associated water, but in
general such a method will need modification or complete redevelopment
because the petroleum-associated water contains a more complex and con-
centrated array of dissolved salts than the fresh water.
Quality control
Data provided by the analytical laboratory are used in decision-making,

and the data must describe precisely and accurately the characteristics or
concentrations of the constituents in the sample. Usually an approximate or
incorrect result is less valuable than no result because it leads t o faulty
interpretations.
The analyst needs t o be aware of his responsibility to provide results that
reliably describe the sample. Further, he should know that the procedures
that he uses, his professional competence, and his reported values may be
used or challenged. To meet any challenge his results must be adequately
20 ANALYSIS OF OILFIELD WATERS
documented. The value of research investigations which use oilfield brine

analyses depends upon the validity of the laboratory results.
A program to insure the reliability of analytical data is mandatory because
of the importance of the laboratory results and the actions that they
produce. An established routine control program applied to analytical tests is
necessary to assure the precision and accuracy of the final results. The use of
spiked samples can measure quality, while the use of analytical-grade
reagents is a control measure. Quality control varies with the type of as*-
sis. For example, the frequent standardization of the titrant used in a titra-
tion is an element of quality control, while instrument calibration in an
instrumental method is also a quality control function.
The specific methodology employed should be carefully documented
regardless of the method used; thus the data user or reviewer can apply the
associated precision and accuracy when interpreting the laboratory data.
Choosing an analytical method
Widespread use of an analytical method indicates that it probably is

reliable and will produce valid results. Use of a little-known procedure forces
the data user t o accept the judgment of the analyst.
The following criteria are useful in selecting analytical methods:
(1)The desired constituent should be measured with sufficient precision
and accuracy in the presence of the interferences normally found in oilfield
waters.
(2) The method must utilize the skills and equipment available in the
oilfield water laboratory.
(3) The method should be sufficiently tested and used by several labora-
tories to establish its validity.
Precision
Precision is the reproducibility among replicate observations, and in

quality control it is determined on actual water samples containing inter-
fering constituents. Several methods to determine precision are available and
the following is representative:
(1) Study four separate concentration levels, including a low concentra-
tion near the sensitivity level of the method, two intermediate concentra-
tions, and a concentration near the upper limit of application of the method.
(2) Make seven replicate determinations at each of the concentrations
tested.
(3) To allow for changes in conditions, the precision study should use at
least 2 hours of normal laboratory operation.
(4) To permit the maximum interferences in sequential operation, the
samples should be run in the following order: high, low, intermediate, inter-
mediate. Repeat this series seven times to obtain the desired replication.
QTJALITY CONTROL 21
TABLE 3.1
Precision data on oilfield brine samples for boron

___ ~~
Sample Concentrations of boron found (mg/l)
Taylor Eagle Ford Paluxy Douglas

-- . - _ _ _ _- -
10.1 15.2 20.1 30.3

10.1 15.3 20.1 30.2
10.2 15.1 20.3 30.1
10.3 15.2 20.2 30.1
10.1 15.3 20.3 30.3
10.2 15.2 20.3 30.2
10.2 15.1 20.1 30.1
Average 10.2 15.2 20.2 30.2

Standard deviation 0.1 0.1 0.1 0.2
(5) The precision statement includes a range of standard deviations over

the t A e d range of concentrations. Thus, four standard deviations are ob-
tained over a range of four concentrations, but the statement contains only
the extremes of standard deviations and concentrations studied.
An example of data generated from such an approach is shown in Table 3.1.
Using the data of Table 3.1 the precision statement would read: “In a
single ldboratory, using oilfield water samples containing concentrations of
10.2 and 30.2 mg B/1, the standard deviation was kO.1.”
Accuracy
The degree of difference between observed and actual values is accuracy.

The accuracy of a method can be determined as follows:
(1) Add known amounts of the constituent t o be determined to actual
samples at concentration levels where the precision of the method is adequ-
ate. The added amount should double the concentration of the low-
concentration sample and bring the concentration of an intermediate sample
to about 75% of the upper limit of application of the method.
(2) Make seven replicate determinations at each concentration.
(3) Report the accuracy as the percent recovery found in the spiked
sample, where the percent at each concentration is the mean of the seven
replicate tests.
Table 3.11 illustrates the application of this approach, where two of the
samples used in the precision study, Table 3.1, were used. An appropriate
accuracy statement is: “In a single laboratory, using oilfield water samples
containing concentrations of 20.2 and 35.3 mg B/1, recoveries were 100.0%
and 100.3%,respectively.”
TABLE 3.11
Accuracy data on oilfield brine samples for boron
Sample Concentrations of boron found (mg/l)
Taylor Paluxy
(added 10 mg/l boron) (added 15 mg/l boron)
20.2 35.5
20.2 35.4
20.1 35.2
20.1 35.2
20.3 35.3
20.3 35.2
20.4 35.1
Average 20.2 35.3

20-2 35.3
Percent recovery ] x 100 = 100.0 ] x 100 = 100.3
[ 10.2 + 10 [ 20.2 + 15
The precision and accuracy data are valuable in determining that the
analyst and the method are capable of generating valid data. Once this is
proven, the data can be used to evaluate systematic performance. This can be
done by using spiked samples about 10% of the time to determine that the
accuracy is favorable, and evaluating replicate samples to determine that the
precision is favorable.
Preliminary sample treatment
The following determinations should be made in the field immediately

after sampling:
(1) temperature (in "C),(2) pH, (3) dissolved oxygen, (4) resistivity, ( 5 )
acidity, (6) alkalinity, (7) sulfide, and (8) carbon dioxide.
If possible, the oilfield water sample should be filtered immediately after
sampling in the field. A preferred method-is to use pressure filtration
through a 0.45-micrometer (pm) membrane filter. A liter of filtrate usually is
sufficient and the following determinations can be made on aliquots: (1)
iodide, (2) bromide, (3) chloride, (4) selenium, ( 5 ) sulfate, (6) nitrogen, (7)
phosphate, (8) silica, (9) boron, (10) potassium, (11) sodium, and (12)
lithium.
If a field-filtered sample cannot be provided, a laboratory-filtered sample
may be substituted with slightly less confidence in the reported data.
PRELIMINARY SAMPLE TREATMENT 23
Standard solutions
Examples of standardization procedures are given for some of the

methods. The concentrations of standard solutions are indicated as the
weight of a given element equivalent to, or contained in, 1 ml of solution.
The strength of acids and bases are given in terms of molarities or normali-
ties.
Accuracy of measurements
In the instructions for making the analysis and preparing the solutions,
significant figures are utilized to define the accuracy of weights and
measures.
Required accuracy for measurement of volume in the analysis and prepa-
ration of reagents is also shown. Standard solutions are prepared in and
measured from volumetric glassware.
Reagent chemicals and solutions
All of the chemicals used in the analytical procedures should conform to

the specifications of the Committee on Analytical Reagents of the American
Chemical Society. Chemicals not listed by this organization can be tested
according to procedures given by Rosin (1955). Primary standard chemicals
can be obtained from the National Bureau of Standards or from companies
marketing chemicals of the same purity.
Water used to dilute samples or t o prepare chemical solutions should first
be demineralized by passage through mixed cation-anion exchange resins or
by distillation. Its specific conductance a t 25°C must not exceed 1.5 pmho/
cm, and it should be stored in polyethylene bottles.
Carbon-dioxide-free water may be prepared by boiling and cooling
demineralized water immediately before use. Its pH should be between 6.2
and 7.2.
Ammonia-free water should be prepared by passing distilled water through
a mixed-bed ion-exchange resin.
Sampling
A field-filtered acidified sample also should be taken. It is pressured

filtered using a 0.45-pm membrane filter and then the filtrate is immediately
acidified to a pH of 3.0 or less with reagent-grade HCl. The acidified sample
is used for the following determinations: (1) aluminum, (2) arsenic, (3)
barium, (4) cadmium, (5) calcium, (6) copper, (7) iron, (8) lead, (9) magne-
sium, (10) manganese (11)nickel, (12) strontium, and (13) zinc.
TABLE 3.111
Units in which water analyses may be reported
milligrams per liter = mg/l 1 grain per U.S. gallon = 17.12 mg/l
part per million = ppm 1 grain per Imperial gallon = 14.3 mg/l
milligrams per liter 1 ppm = 0.012 milligram atom per liter
ppm = specific gravity of the water
T o convert compounds expressed as parts per million t o ions expressed,as parts per
million (where compound is A, Bm):
,(atomic weight A)
ppm ion A = ppm compound A,
Bm molecular weight A, B,
,(atomic weight B)
ppm ion B = ppm compound A,
Bm molecular weight A, ,
B
To convert parts per million t o equivalents per million (epm):
Example: sample contains 28.3 ppm Ca” , what is the concentration of calcium in epm?
Solution; atomic weight Ca = 40.08; valence = 2; equivalent weight =
40.08
2
= 20.04; then:
28 3
28.3 ppm Ca+’ = -
20.04
= 1.41 epm Ca+’
Titrime tric analysis
milliequivalent weight 106

(ml x N of standard solution) x
of determined ion
= mg/l of determined ion
ml of sample used
Gravimetric analysis
(grams of preninitnta\
.,.F.’U”’, ,. atomic
Y
weight of determined element
molecular
..._. - weig,ht of precipitate
106
ml of sample used
.._.- determined
me/l
=. ~
element
%O = parts per thousand or g/kg
Chlorinity (CZ) = mass in grams of silver required to precipitate the halogens in

328.5233 g of sea water
Salinity (S) = total amount of solid material, in grams contained in 1 kg of

sea water when all of the bromide and iodide have been replaced
by the equivalent amount of chloride, when all of the carbonate
is converted t o oxide and when all the organic matter is completely
oxidized
%o S = 1.805 x %o Cl + 0.03
REPORTING THE ANALYTICAL RESULTS 25
Reporting the analytical results
A study conducted by the American Petroleum Institute (1968) indicated

that some laboratories reported the results of oilfield water analysis as parts
per million (ppm) or as milligrams per liter (mg/l) without regard to the
specific gravity of the sample. For example, a sample with a specific gravity
of 1.200 containing 12,000 mg/l of calcium does not contain 12,000 ppm of
calcium but contains 12,000/1.200 = 10,000 pprn of calcium. Such an error
obviously is more serious in reporting the analytical results for a brine than
in reporting the results for a fresh water. The unit ppm means parts per
million by weight, while the unit mg/l means milligrams per liter or weight
per unit volume; therefore, they are not interchangeable until the volume is
changed to a unit weight. Table 3.111 indicates the relation between various
units of measurement.
Because the American Petroleum Institute now recommends that oilfield-
water analysis be reported in units of mg/l, other associations will no doubt
recommend the same uniform practice. Such standardization implements
studies concerned with the chemistry and geochemistry of waters.
Sign i f ican t figures
The term significant figure (Ballinger et al., 1972) is used rather loosely to
describe some judgment of the number of reportable digits in a result. Often
the judgment is not soundly based and meaningful digits are lost or meaning-
less digits are accepted.
Proper use of significant figures gives an indication of the reliability of the
analytical method used. The following definitions and rules are suggested for
retention of significant figures.
A number is an expression of quantity. A figure or digit is any of the
characters 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, which, alone or in combination, serves
to express a number. A significant figure is a digit that denotes the amount
of the quantity in the place in which it stands.
Reported values should contain only significant figures. A value is made
up of significant figures when it contains all digits known to be true and one
last digit in doubt. For example, if a value is reported as 18.8 mg/l, the “18”
must be a firm value while the “0.8” is somewhat uncertain and may be
“0.7” or “0.9”.
The number zero m a y or may not be a significant figure:
(a) Final zeros after a decimal point are always significant figures. For
example, 9.8 g t o the nearest milligram is reported as 9.800 g.
(b) Zeros before a decimal point with other preceding digits are signifi-
cant. With no other preceding digit, a zero before the decimal point is not
significant.
(c) If there are no digits preceding a decimal point, the zeros after the
decimal point but preceding other digits are not significant. These zeros only
indicate the pbsition of the decimal point.
(d) Final zeros in a whole number may or may not be significant.
A good measure of the significance of one or more zeros before or after
another digit is to determine whether the zeros can be dropped by expressing
the number in exponential form. If they can, the zeros are not significant.
For example, no zeros can be dropped when expressing a weight of 100.08 g
in exponential form; therefore, the zeros are significant. However, a weight
of 0.0008 g can be expressed in exponential form as 8 x g, and the
zeros are not significant. Significant figures reflect the limits of the particular
method of analysis. It must be decided beforehand whether this number of
significant digits is sufficient for interpretation purposes. If not, there is little
that can be done within the limits of normal laboratory operations t o im-
prove these values. If more significant figures are needed, a further improve-
ment in method or selection of another method will be required t o produce
an increase in significant figures.
Once the number of significant figures is established for a type of analysis,
data resulting from such analyses are reduced according t o set rules for
rounding off.
R o unding-of f numbers
Rounding off of numbers is a necessary operation in all analytical areas. It

is automatically applied by the limits of measurement of every instrument
and all glassware. However, it is often applied in chemical calculations in-
correctly by blind rule or prematurely, and in these instances can seriously
affect the final results. Rounding off should normally be applied only as
follows.
Round ing-of f rules

(a) If the figure following those to be retained is less than 5 , the figure is
dropped, and the retained figures are kept unchanged. As an example,
11.443 is rounded off to 11.44.
(b) If the figure following those to be retained is greater than 5, the figure
is dropped, and the last retained figure is raised by 1.As an example, 11.446
is rounded off to 11.45.
(c) When the figure following those t o be retained is 5 , and there are no
figures other than zeros beyond the 5, the figure is dropped, and the last
place figure retained is increased by 1 if it is an odd number, or it is kept
unchanged if an even number. As an example, 11.435 is rounded off t o
11.44,while 11.425 is rounded off to 11.42.
Rounding-off single arithmetic operations

(a) Addition: when adding a series of numbers, the sum should be
rounded off to the same number of decimal places as the addend with the
SYNTHETIC BRINE 27
smallest number of places. However, the operation is completed with all

decimal places intact, and rounding off is done afterward. As an example,
11.1+ 11.12 + 11.13 = 33.35, and the sum is rounded off t o 33.4.
(b) Subtraction : when subtracting one number from another, rounding off
should be completed before the subtraction operation, to avoid invalidation
of the whole operation.
(c) Multiplication: when two numbers of unequal digits are t o be
multiplied, all digits are carried through the operation; then the product is
rounded off to the number of significant digits of the less accurate number.
(d) Division: when two numbers of unequal digits are t o be divided, the
division is carried out on the two numbers using all digits. Then the quotient
is rounded off t o the number of digits of the less accurate of the divisor or
dividend.
(e) Powers and roots: when a number contains n significant digits, its root
can be relied on-for n digits, but its power can rarely be relied on for n digits.
Synthetic brine
Synthetic brine solutions are used in many of the analytical procedures

for analyzing oilfield waters (American Petroleum Institute, 1968). Such
solutions are a necessity in the development of analytical methods to study
the effects of possible interfering ions. Often these synthetic solutions are
used as an integral part of the analytical technique (Collins, 1967). Prepara-
tion of a fairly stable synthetic brine involves saturating distilled water with
carbon dioxide by bubbling carbon dioxide through it, followed by adding
the bicarbonate and sulfate compounds to one portion of the C 0 2-saturated
water, adding the alkali chlorides to one portion, and adding the alkaline
earth chlorides to one portion. The alkali chloride solution is mixed with the
bicarbonate-sulfate solution, and t o this mixture the alkaline earth chloride
solution is added. Carbon dioxide is bubbled through the synthetic brine to
mix it, and the synthetic brine container is sealed immediately after re-
moving the carbon dioxide source.
Determination of pH
The pH of the water can be determined with a pH meter which utilizes the
principle of measuring the electrical potential between an indicator electrode
and a reference electrode (Potter, 1956, p.56). pH meters measure the elec-
trical potential between two suitable electrodes immersed in the solution to
be tested. The reference electrode assumes a constant potential, and the
indicating electrode assumes a potential dependent on the pH of the solu-
tion. Electrode potential is the difference in poteptial between the electrode
and the solution in which it is immersed. The calomel electrode, which is a
widely used reference electrode in water analysis, consists of a mercury-
calomel rod immersed in a saturated solution of potassium chloride; this
electrode has a potential of +0.246 V. Electrical connection with the sample

is provided through porous fibers sealed into the immersion end. A
hydrogen-ion-selective glass electrode is normally used as an indicating elec-
trode. The glass electrode has several features that recommend it for pH
measurements. Among the most important are that it is not affected by
oxidizing or reducing substances in the sample and that it can be used to
measure the pH of turbid samples and/or colloidal suspensions. The basic
design is a silver-silver chloride or mercury-mercurous chloride electrode
immersed in a solution of known pH and the whole completely sealed in
glass.
The mechanism by which the glass membrane responds to hydrogen-ion
activity involves absorption of hydrogen ions on both sides of the membrane
proportionally to the activity of the hydrogen ions in solution. The cell for
measuring the pH of a solution is of the following type:
Ago :AgC1 1I solution of

known pH;
glass
membrane;
solution of
unknown pH
glass electrode I 1 Hgo :HgC1
The voltage of the glass electrode is a logarithmic function of the differ-
ence in hydrogen-ion activity of the solutions on either side of the glass
membrane. To measure this voltage an electron-tube voltmeter is used be-
cause the resistance of the glass membrane is so great.
The pH should be determined at the time of sampling. A device similar to
that shown in Fig. 2.5, can be used, or the electrodes can be placed in a
container and then a stream of the sample allowed to flow from the oil-
water separator (Fig. 2.4.) into the container while the pH is measured. If
accurate results are desired, at least two pH buffer solutions should be used
to calibrate the pH meter and electrodes before determining the pH. Because
TABLE 3.IV
pH buffer solutions (pH values of NBS standards from 0-30°C)
Temperature 0.5M Potassium acid 0.05M 0.025M 0.01M

("C) potassium tartrate (sat. at potassium acid potassium dihydrogen sodium
tetroxalate 25OC) phthalate phosphate + 0.025M tetra-
sodium dihydrogen borate
phosphate
0 1.67 - 4.01 6.98 9.46

10 1.67 - 4.00 6.92 9.33
15 1.67 - 4.00 6.90 9.27
20 1.68 - 4.00 6.88 9.22
25 1.68 3.56 4.01 6.86 9.18
30 1.69 3.55 4.01 6.85 9.14
DETERMINATION O F Eh 29
TABLE 3.V
Performance characteristics of typical pH meters
Normal scale Expanded scale

~~~
Range 0-14 pH (+ 1,400 mV) 1 pH (* 100 mV)

Smallest scale division 0.1 pH (10 mV) 0.005 pH (0.5 mV)
Accuracy f 0.05 pH (5 5 mV) f 0.002 pH (+ 2%of reading)
Reproducibility f 0.02 pH (+ 2 mV) f 0.002 pH (+ 0.2 mV)
Temperature
compensation O - I O O ~ C(manual or
automatic)
Input impedance
-
> 1014 > 1013
the pH probably will fall between 5 and 7, the standard pH buffer solutions
used could be for pH 5 and pH 7.
Standard buffer solutions, covering a range of pH, may be purchased from
almost any chemical supply house and are satisfactory for routine use. Table
3.IV gives a list of NBS buffers (easily made in the laboratory) and the
resulting pH at several different temperatures.
An idea of the effect of temperature on pH may be obtained by observing
temperature versus pH of various buffers shown in Table 3.IV. Theoretically,
the potential response of the electrode system changes 0.20 mV per pH unit
per degree Celsius. Since all pH meters measure potential but read out in pH,
a variable compensation is used. A rough rule of thumb is that temperature
compensation is about 0.05 pH unit per 5' increase in temperature. Perfor-
mance data of a conventional and an expanded scale pH meter are shown in
Table 3.V.
Determination of Eh
The Eh, called the oxidation-reduction potential or the redox potential,

is a measure of the relative intensity of oxidizing or reducing conditions in a
chemical system. It is expressed in volts, and at equilibrium it is related to
the proportions of oxidized and reduced species present. Standard equations
of chemical thermodynamics express the relationships (Collins, 1964).
Eo is the standard potential of a redox system when unit activities of
participating substances are present under standard conditions. Eo is related
to standard free energy change in a reaction by the equation:
A P = -nfEo
where n is the number of unit negative charges (electrons) shown in the

redox reaction and f is the Faraday constant in units that give a potential in
volts (94,484absolute coulombs). Standard free energy values are given in
texts such as that of Latimer (1952).
When the system is not under standard conditions, the redox potential is
expressed by the Nernst equation:
R T log oxidized species)
Eh=Eo+ -
nf (reduced species)
where R is the gas constant (1.987 calories per degree mole) and T is the
temperature in degrees Kelvin. Geochemical literature and biochemical litera-
ture such as that of Pourbaix (1949) use increasing positive potential values
to represent increasing oxidizing systems, and decreasing potential values to
represent reducing systems. The sign of Eh used in this manner is opposite to
standard American practice in electrochemistry.
Reagents. An Eh standard' which can be used is a solution of M/300

K3Fe(CN)6 and M/300 KqFe(CN), in M/10 KC1 (Zobell, 1946). The Eh of
this mixture is 0.430 V at 25'C.
Equipment. A pH meter equipped with a thermometer, a glass electrode, a

calomel electrode, a platinum electrode and a thermocompensating elec-
trode.
Eh flowchamber, a design similar to Fig. 2.5 can be used.
Procedure. Buff the platinum electrode lightly with a fine abrasive cloth
and wipe it carefully with a dry soft tissue. Install the glass electrode, the
calomel electrode, the platinum electrode, the thermocompensator, and the
thermometer in the flowchamber. Standardize the instrument using the Eh
standard.
Connect a line t o the wellhead or waterline and install an oil-water
separator if oil and water both are present. Connect the flowchamber to the
waterline, allowing the water to flow into the bottom and out the top. Make
certain that all air bubbles are excluded at the top.
Take at least three readings of the Eh (in mV), and the temperature at
10-minute intervals. These readings should agree; if they do not, continue
making readings until three successive readings do agree. Make certain the
water is continually flowing, that there are no air bubbles in the flow-
chamber, and that the solution is being stirred. It may be necessary t o
remove and rebuff the platinum electrode.
Calculation. Because a thermocompensator is used in determining the pH, a

temperature correction need not be made. However, if a thermocompensator
is not used, a temperature correction should be made.
The Eh value is obtained by algebraically adding the measured voltage E
and the voltage of the constant voltage reference electrode, which in this
case is the saturated calomel electrode. The potential of the saturated
calomel electrode at 25OC is 0.242 V. Therefore, if the millivolt reading of
the sample is +300:
SUSPENDED SOLIDS 31
Eh = E + voltage of reference electrode

Eh= +300+242=+542
However, if the millivolt reading is -300, then:
E = +0.242 - 7.6x ( t - 24)
Note: the calculations for Eh are correct only if the temperature of the
brine is 25OC at the time of measurement. If the temperature is not 25"C, a
correction should be made. For example, the potential of the saturated
calomel electrode is 0.246 V at 20°C and 0.238 V at 30°C. The following
formula can be used to obtain the correct potential:
E = +0.242 - 7.6 x ( t - 24)
where t is in degrees Celsius.

Suspended solids
Various inorganic and organic materials are found in petroleum-associated

water. Knowledge of the composition of such material is useful in deter-
mining the source of the material and what treatments can be used t o
remove it or prevent it from recurring. Such material may be particles of
oxides of the metals from well casings, pumps, or precipitates caused by
oxidation of the formerly reduced species, such as iron or manganese. Other
suspended solids may be silt, sand, and clay.
An estimation of the amount of material in suspension can be ac-
complished by using a turbidimeter (Rainwater and Thatcher, 1960).This is
done by comparing the intensity of light passing through the solution with
the Tyndall effect produced by lateral illumination of the solution with the
same source of light.
Turbidimeter
Instruments for the measurement of turbidity employ principles of design

related to transmission or reflectance of light. The lack of a primary standard
for turbidity, however, has resulted in a complete absence of uniformity
among the available instruments. Further, the Jackson candle turbidimeter,
which does not depend upon the use of a primary standard, is a primitive
instrument, subject to many interferences, and the measurements generally
are not reproducible.
Recently developed turbidimeters often use for calibration a suspension of
formazin permanently embedded in a cylinder of Lucite. These instruments
produce reproducible readings up t o 40 Jackson candle units (JCU), and
samples containing turbidities in excess of 40 JCU should be diluted to
32 ANALYSIS O F OILFIELD WATERS
values below this level and the results multiplied by the correct dilution
factor.
To obtain maximum accuracy and precision the following precautions
should be observed:
(a) Protect the Lucite standard from scratches, nicks, and fingerprints.
(b) While calibrating the instrument, use a constant orientation of the
Lucite standard.
(c) Use a homogeneous sample in the sample cuvette; do not take readings
until finely dispersed bubbles have disappeared.
(d) Dilute samples containing. excess tubidity to some value below 40
JCU; take reading, and multiply results by correct dilution factor.
Suspended solids analysis
To determine the composition of the suspended solids they can be

removed by filtration using a 0.45-pm membrane or less porous filter. The
filtered solids can then be subjected to chemical analysis. To determine the
exact composition of the solids may require the filtration of a large sample
in order to procure enough solid material. The heavy-metal content can be
determined by subjecting a portion of the sample t o an emission spectro-
metric analysis; X-ray diffraction can be used to determine which, if any,
clays are present; extraction with organic solvents followed by infrared mass
spectrometric, chromatographic, and gas chromatographic analysis will give
an indication of organic compounds present; thermogravimetric analysis will
provide clues; wet chemical analysis can be used t o determine many of the
anions; and X-ray fluorescence can be used to determine some of the anions.
Resistivity
The resistivity of petroleum-associated waters is used in electric log inter-

pretations (Wyllie, 1963), and for such use the values must be adjusted t o
the formation temperature. This can be done by referring t o curves such as
those shown in Fig.3.1, which gives resistivity values for sodium chloride
solutions. The resistivity of a formation water will not be exactly the same as
that of a pure sodium chloride solution of equal dissolved solid (DS) con-
tent, but for practical purposes the assumption that the resistivities are
approximately equal is satisfactory.
It is possible to calculate the resistivity from water-mineral analysis by
using methods such as those developed by Dunlap and Hawthorne (1951) or
Jones (1944). The calculated values are less accurate and usually lower than
the directly measured resistivities. The direct-measurement method is essen-
tially the electrical resistance of a cube of oilfield water. In well-logging
practice, the edge of the cube considered is 1 m in length. Therefore, resis-
tivity of an oilfield water is expressed in ohm-meters ( a m ) .
Temperature has a profound effect on resistivity; therefore, all resistivities
RESISTIVITY 33
R E S I S T I V I T Y . .hm-n.ttrl
Fig. 3.1. Plots of resistivity of aqueous solutions containing various concentrations of

sodium chloride.
should be determined at a known constant temperature. The sample should

be freshly filtered and free of oil. Nonionized silica and other materials in
suspension in an oilfield water can affect the resistivity determination, but in
general such interferences can be ignored. Cell polarization can be trouble-
some with highly mineralized waters and will vary directly with the current
that flows between the electrodes and inversely with the frequency of the
current. High input voltage t o the bridge or low cell resistance (highly
mineralized waters) increases the likelihood of polarization. Cell resistance
can be increased by increasing the cell constant.
Reagents. The necessary reagents are standard potassium chloride solutions

of l.OOON, O.lOOON, and 0.01OON (use only certified reagent-grade KC1 that
has been oven-dried t o constant weight at 110OC);chromic-sulfuric acid
cleaning solution; platinizing solution (dissolve 3 g of chloroplatinic acid and
0.02 g of lead acetate in 100 ml of water); and a 10%aqueous sulfuric acid
solution.
Equipment. The necessary resistivity measurement equipment includes a

Wheatstone bridge; resistivity cells, either dip or pipet type, with platinum
electrodes; water bath, complete with stirrer, thermostat, and thermometer,

with 0.loC graduations; source of alternating current, 25- t o 60-cycle a.c.
galvanometer, and an appropriate isolating transformer.
Selection of the cell constant is limited by the accuracy and sensitivity of
the bridge when measuring very high and very low resistivities. Also, current
frequency should not be excessively high since a.c. resistance is a complex
function of frequency; e.g., at frequencies necessary to avoid polarization,
the differences between a.c. resistance and d.c. resistance may be appreciable
unless the cell has been carefully designed t o minimize this difference. In
essence, the ideal single apparatus for measurement of resistivity throughout
a wide range necessarily incorporates compromises between low input
voitage, high cell constant, high current frequency, and accuracy and sensi-
tivity of the bridge.
Cell preparation
To prepare the cell, clean it with chromic-sulfuric acid solution and rinse
thoroughly with water. Immerse the cell or fill it, depending upon whether a
dip or pipet cell is used, in the platinizing solution. Connect the electrodes of
the cell to three dry cells (1-1/2 V each) in parallel through a limiting
resistance of approximately 1,000 52. Reverse the direction of the current
once a minute for 6 minutes or until the shiny platinum surface is covered
with a dense black coating. Repeat the electrolytic process using 10%
sulfuric acid solution to remove chlorine. Remove the electrodes, rinse with
distilled water, and store in distilled water.
Note: new cells should be cleaned and platinized before use. They should
be cleaned and replatinized whenever the readings become erratic or when
the platinum black flakes off.
Cell resistance
To determine the cell resistance using the standard potassium chloride

solutions, adjust the temperature of each potassium chloride solution to
exactly 25OC and obtain a reading in ohms for each solution with the cell.
Calculate the cell constant using the following formula:
C = R K C l x specific conductance of standard KC1 solution
where R K C l = reading obtained in ohms for standard KC1 solution.

Note: the specific conductivities of the standard KC1 solutions a t 25°C are
as follows (Hodgman et al., 1962, p. 2690):
1.OON KCl = 0.11173 mho/cm

0.1ON KC1 = 0.012886 mho/cm
0.01N KCl = 0.0014114 mho/cm
SPECIFIC GRAVITY 35
Method of determination
Procedure. To determine the resistivity of the petroleum-associated water,

filter the sample to remove oil and transfer the sample t o the cell or cell
container and place it in a water bath. Allow sample sufficient time to adjust
to bat!i temperature, and measure resistance of sample and record the
temperature to nearest 0.1Oc.
Calculation. The resistivity calculation is dependent upon the type of cell

and bridge used, but in general the following formula will apply:
R, =-
D2 V
4LXT
where R, = resistivity of water, a m ; V = difference in potential between
potential-measuring electrodes, V; I = current flowing through the cell, A;
D = inside diameter of potential-measuring electrodes, m; and L = distance
between potential-measuring electrodes, m.
Because D and L are constant for any one cell and I is held constant for
most waters, these values can be combined into a single constant, K , and the
following simplified equation used :
R, = KV
Calculated resistivity
The resistivity of petroleum-associated waters often is calculated using the

laboratory analysis (Dunlap and Hawthorne, 1951). The concentrations of
the ionic constituents are used in the calculation method.
Dunlap and Hawthorne (1951) caution users of their calculation method
that the sulfate factor 0.50 may give unreliable results if the water contains
appreciable concentrations. of sulfate. If the sulfate concentration exceeds
2,500 mg/l, a factor of 0.40 will give a better calculated resistivity value.
Specific gravity
Specific gravity is the ratio of the weight of a given volume of material to

the weight of an equal volume of some other material used as a standard
(Mellon, 1956, p.306), and pure water is the usual standard for liquids and
solids. Depending upon the accuracy desired, the specific gravity of a
petroleum-associated water can be determined with a pycnometer, specific
gravity balance, or hydrometer. Because any oil in or on the sample will
interfere with the specific gravity determination, the sample should be
filtered.
TABLE 3.W
Approximate relation of specific gravity (Sp. gr.) to mg/l of dissolved solids (DS)
-
Sp. gr. DS Sp.gr. DS Sp. gr. DS Sp.gr. DS Sp. gr. DS Sp.gr. DS
1.ooo 0 1.038 '50,800 1.076 97,400 1.114 141,200 1.152 182,100 1.190 221,000
1.001 1,400 1.039 52,000 1.077 98,700 1.115 142,300 1.153 183,200 1.191 222,000
1.002 2,800 1.040 53,300 1.078 99,800 1.116 143,400 1.154 184,200 1.192 223,?00
1.003 4,200 1.041 54,600 1.079 101,000 1.117 144,500 1.155 185,300 1.193 224,000
1.004 5,600 1.042 55,900 1.080 102,200 1.118 145,600 1.156 186,300 1.194 225,000
1.005 7,000 1.043 57,100 1.081 103,400 1.119 146,700 1.157 187,400 1.195 226,000
1.006 8,300 1.044 58,300 1.082 104,600 1.120 147,900 1.158 188,400 1.196 227,000
1.007 9,700 1.045 59,600 1.083 105,800 1.121 148,900 1.159 189,500 1.197 228,000
229,000
1.008 11,100 1.046 60,900 1.084 106,900 1.122 150.00 0 1.160 190,500 1.198
1.009 12,400 1.047 62,100 1.085 108,100 1.123 151,100 1.161 191,600 1.199 230,000
1.010 13,700 1.048 63,400 1.086 109,300 1.124 152,100 1.162 192,600 1.200 230,800
1.011 15,200 1.049 64,600 1.087 110,400 1.125 153,200 1.163 193,600 1.201 231,800
1.012 16,600 1.050 65,900 1.088 111,600 1.126 154,400 1.164 194,700 1.202 232,800
1.013 17,800 1.051 67,100 1.089 112,800 1.127 155,500 1.165 195,700 1.203 233,700
1.014 19,100 1.052 68,400 1.090 114,000 1.128 156,600 1.166 196,700 1.204 234,700
1.015 20,500 1.053 69,600 1.091 115,100 1.129 157,700 1.167 197,800 1.205 235,700
1.016 21,900 1.054 70,900 1.092 116,200 1.130 158,800 1.168 198.800 1.206 236,700
1.017 23,200 1.055 72,000 1.093 117,400 1.131 159,900 1.169 199,800 1.207 237,600
1.018 24,500 1.056 73,300 1.094 118,600 1.132 161,000 1.170 200,900 1.208 238,600
1.019 25,900 1.057 73,600 1.095 119,600 1.133 162,000 1.171 201,900 1.209 239,500
1.020 27,300 1.058 75,800 1.096 120,800 1.134 163,100 1.172 202,900 1.210 240,500
1.021 28,500 1.059 77,100 1.097 122,000 1.135 164,100 1.173 203,900 1.211 241,500
1.022 29,800 1.060 78,200 1.098 123,100 1.136 165,200 1.174 204,900 1.212 242,400
1.023 31,000 1.061 79,400 1.099 124,400 1.137 166,200 1.175 206,000 1.213 243,400
1.024 32,400 1.062 80,600 1.100 125,500 1.138 167,300 1.176 207,000 1.214 244,300
1.025 33,900 1.063 81.800 1.101 126,700 1.139 168,400 1.177 208,000 1.215 245,300
1.026 35,100 1.064 83,100 1.102 127,800 1.140 169,400 1.178 209,000 1.216 246,200
1.027 36,400 1.065 84,300 1.103 128.800 1.141 170,400 1.179 210,000 1.217 247,700
1.028 37,700 1.066 85,600 1.104 130,000 1.142 171,500 1.180 211,000 1.218 248,100
1.029 39,100 1.067 86,700 1.105 131,100 1.183 172,500 1.181 212,000 1.219 249,100
1.030 40,400 1.068 87,800 1.I06 132,300 1.144 173,600 1.182 213,000 1.220 250,000
1.031 41,700 1.069 89,100 1.107 133,400 1.145 174,700 1.183 214,000 1.221 250,900
1.032 43,000 1.070 90,300 1.108 134,500 1.146 175.7 00 1.184 215,000 1.222 251,900
1.033 44,300 1.07 1 91,500 1.109 135,600 1.147 176,800 1.185 216,000 1.223 252,800
1.034 45,600 1.072 92,700 1.110 136,800 1.148 177,900 1.186 217,000 1.224 253,800 8
1.035
1.036
46,900
48,300
1.073
1.074
93,900
95,100
1.111
1.112
137,900
139,100
1.149
1.150
178,900
180,000
1.187
1.188
218,000
219.000
1.225 254,700 3
m
1.037 49,500 1.075 96,200 1.113 140.1 00 1.151 181,100 1.189 22o;ooo
TITRIMETRIC METHODS 37
Knowledge of the specific gravity of the sample is necessary to convert the

analytical data determined for the sample from milligrams per liter to parts
per million. In addition, the specific gravity will give an indication of the
amount of dissolved solids present in the sample, as indicated in Table 3.VI.
TITRIMETRIC METHODS
Acidity, alkalinity, and borate boron
If the pH of the water is less than 4.5, the water possesses what is called
“mineral-acid acidity”. The acidity of a petroleum-associated water may
indicate a contaminant because of acid treatment of the well or it could
indicate the presence of various dissolved gases and salts. Most petroleum-
associated waters contain little or no acidity. If a water contains acidity, it
does not contain alkalinity.
The acidity of a water is determined by adding a standard base such as
0.02N sodium hydroxide to the water until the pH of the water is 4.5
(Collins et al., 1961) as monitored with a pH meter. To obtain a value close
to natural conditions, the acidity should be determined at the sampling
point.
The alkalinity of a water is determined by adding a standard acid such as
0.05N hydrochloric acid t o the water and recording the volume used to
neutralize it to pH 8.1 and pH 4.5. The amounts of hydroxide, carbonate,
and/or bicarbonate can then be calculated using the relationships shown in
Table 3.VII. Because the alkalinity will change when the sample is exposed
to the atmosphere, the alkalinity should be determined as rapidly as possible
after sampling.
TABLE 3.VII
Relationships for determining alkalinity after neutralization with a standard acid
Volume of standard acid used
OH co3 HC03
P=O 0 0 T
<
P = 1/2T 0 2P T - 2P
P = 1/2T 0 2P 0
P > 1/2T 2P-T 2(T-P) 0
P=T T 0 0
P = volume used t o titrate t o pH 8 . 1 ; T = volume used t o titrate t o pH 4.5 plus

volume used to titrate to pH 8.1.
Reagents. The necessary reagents are standard hydrochloric acid, standard

sodium hydroxide, pH buffer solutions (preferably for pH 4, 7, and lo),
mannitol, and nitrogen gas.
Equipment The necessary equipment are a pH meter, 10-ml microburets,

boron-free glassware, and boron-free reflux condensers.
Standardization of 0.02N sodium hydroxide
The 0.02N sodium hydroxide solution should be standardized t o deter-

mine its exact normality. One of the better methods is to standardize it with
potassium acid phthalate. Obtain a potassium acid phthalate sample of
known purity, such as a National Bureau of Standards standard sample, and
dry the salt at 105°C. Weigh 0.1 g of salt, dissolve it in 50 ml of distilled
water, and titrate with sodium hydroxide to a pH of 7.0.
Reaction:
HOOC , ,’COONa
NaOH +
‘, ChH, - -‘ChH,’ + H,O
/
KOOC ’ ‘COOK
Normality calculation:
weight KHCs H, 0,
= 0.20422 x ml NaOH
Standardization of 0.05N hydrochloric acid
If constant boiling-point hydrochloric acid is not used in preparing 0.05N

hydrochloric acid, the normality should be checked. One method is t o use a
potassium iodate sample of known purity; for example, a National Bureau of
Standards standard sample. Dry the salt at 180’C for 2 hours, weigh 0.1 g,
dissolve it in 50 ml of distilled water, add 2 g of potassium iodide and 2 g of
sodium thiosulfate, and titrate with hydrochloric acid t o a pH of 7.0.
Reaction:
KI03 + 5KI = 6HC1 + 6KC1+ 312 + 3H20
I2 + 2Na2S203 + Na2S406+ 2NaI
Normality calculation :
weight KI03
N=
0.03567 x ml HC1
TITR IMETR IC METHODS 39
Procedure. Calibrate the pH meter to pH 4 and 7 with appropriate buffer

solutions .and recheck the calibration often. Transfer an undiluted 50- or
100-ml sample t o a beaker and determine the pH and record it. If the pH is
above 8.1, titrate it to 8.1 with 0.05N hydrochloric acid and record the titer
for the carbonate calculation. Continue the titration to pH 4.5 and record
the titer for the bicarbonate calculation. If salts of organic acids are in the
water sample, special precautions must be taken to separate the bicarbonate
titer from that required for the organic salts. This may be done by extracting
the acids, usually naphthenic, with a neutral organic solvent such as
petroleum ether. If the initial pH is below 4.5,titrate to pH 7 with 0.02N
sodium hydroxide and record the titer for the acidity calculation.
Next reduce the pH to 3.5 with 0.05N hydrochloric acid, and reflux the
sample 5 minutes. Remove the sample and immediately cool in an ice-water
bath. Carefully adjust the pH of the cooled brine to 7 with sodium hy-
droxide while nitrogen is aspirated gently over the top. Add 10 g of mannitol
and titrate the sample back to pH 7 with 0.02N sodium hydroxide. Record
this titer for the borate boron calculation. If more than 1 mg of boron is
present in the titrated sample, the results may be low.
Calculations. If the initial pH was more than 8.1, the titer for carbonate and
bicarbonate is determined:
m1 HC1 30y000= mg carbonate per liter
ml sample
To convert carbonate t o bicarbonate, multiply the carbonate value by
2.03.
If the initial pH is less than 8.1 but more than 4.5,only bicarbonate is
present :
ml HC1 x N x 61,000
= mg bicarbonate per liter
ml sample
If the initial pH is below 4.5,the brine is acid:
ml NaOH x N x 50,000
= acidity as mg CaCO, per liter
ml sample
If the total titration is equal to the titer found to pH 8.1, only hydroxide
is present:
ml HC1 x N x 50,000
= acidity as mg CaC0, per liter
mg sample
The borate boron is calculated by using the titer for borate boron:
ml NaOH x N x 10,820
= mg borate boron per liter
mg sample
Calcium and magnesium
Probably the most common method currently used to determine calcium

and magnesium in waters is the complexometric method (American Petro-
leum Institute, 1968)which utilizes a salt such as disodium ethylenediamine-
tetraacetic acid (EDTA) or disodium 1,2-cyclohexanediaminetetraacetic
acid (CDTA) to chelate calcium or magnesium. At a pH of 10,both calcium
and magnesium are chelated, while at a pH of 12, only calcium is chelated
because magnesium will precipitate as the hydroxide.
Disodium ethylenediaminetetraacetate has the following structure
(Welcher, 1957,p.128):
0 -CO - CHI CHI - C O - 0

Na’
/
‘\
, Na’
H N - (CH2) - N H
\ 0 - CO - CH, / “CH, - CO - 0
/
’
Its molecular weight is 372.254, and it forms 1:l complexes with most
cations according t o the following equations:
Me+’ + H2Y-’ * MeY-’ + 2H+
Me+3 + H2Y-’ * MeY- + 2H+
Me+4 + H2Y-’ * MeY + 2H+
where Me = the cation, H2Y = EDTA, and MeY = the complex.
Therefore, 1 gram-ion of EDTA reacts with 1 gram-ion of the metal,
regardless of its valence. The resulting complexes have the same composition,
differing only in the charge they carry.
A metal indicator in an EDTA titration can be represented by the follow-
ing expression :
M-In + EDTA * M-EDTA + In
where M-In = the metal indicator complex, M-EDTA = the metal-EDTA com-
plex, and In = the indicator. The metal indicator complex must be weaker
than the metal-EDTA complex. The color change occurs because the metal-
indicator complex ionizes, and the free metal is completely complexed by
the EDTA, leaving a free indicator.
Sample size
Because many petroleum-associated waters contain high concentrations of

dissolved solids including calcium and magnesium, it usually is necessary t o
dilute them or to use a micropipet t o obtain a small sample before
performing a complexometric titration. The dilution and subsequent aliquot
TABLE 3.VIII
Aliquot size for total hardness, calcium, and magnesium determination
Specific gravity Dilution Aliquot (ml)
1.ooo--1.010 none 50
1.010-1.025 none 25
1.025-1 .O50 dilute 25 ml to 100 ml, 12.5
take 50 ml
1.050-1.090 dilute 25 ml to 100 ml, 6.25
take 25 ml
1.090-1.1 20 dilute 25 ml to 500 ml, 1.25
take 25 ml
1.120-1.150 dilute 25 ml to 1,000 ml, 0.625
take 25 ml
TABLE 3.IX
Comparison of errors in direct reading of sample size using a micropipet versus the
dilution technique
Direct-reading using micropipet Dilution technique
sample size error ml taken after error

(ml) diluting 10 ml (ml)
to 100 ml
~~
200 h f 0.004 2 f 0.01

500 h f 0.01 5 f 0.025
1,000 h * 0.002 10 f 0.005
1ml f 0.006 10 * 0.005
2 ml f 0.006 20 f 0.009
5 ml f 0.01 50 * 0.02
size usually can be determined by using data such as that illustrated in Table
3.VII1, which is applicable to most oilfield brines.
A more rapid method of obtaining a fraction of a milliliter of a liquid
sample is direct measurement using a micropipet. A micropipet in the hands
of a competent analyst can also yield a more accurate sample size than the
dilution technique illustrated in Table 3.VIII; e.g., two reading errors are
omitted because only one meniscus reading is necessary with the direct
measurement as compared to three using the dilution. Table 3.IX illustrates a
comparison of errors in sample sizes of the direct reading method versus the
dilution method.
Reagents: CDTA (disodium 1,2-cyclohexanediaminetet,raaceticacid) standard

solution, approximately 0.025M:dissolve 10.66 g CDTA in water and dilute
to 1liter. (To standardize, dissolve 2.4971 g calcium carbonate in the smallest

amount of hydrochloric acid possible and dilute to 1 liter with water (0.025M
Ca solution). Pipet triplicate 10-ml aliquots into flasks and dilute to approxi-
mately 50 ml with water. Begin at paragraph 2 of the procedure, and carry
the standard solutions through all the steps of that paragraph. Calculate the
molarity using the following equation:
10.0 x 0.025 0.25
M= or -
V V
where V = volume of CDTA required to titrate 1 0 ml of 0.025M calcium
solution.
Sodium hydroxide: dissolve 320 g sodium hydroxide pellets in water and
dilute to 1liter.
Ammonium chlorideammonium hydroxide buffer solution : dissolve
67.5 g ammonium chloride in approximately 200 ml of water. Add 570 ml
concentrated ammonium hydroxide and dilute to 1 liter with water.
Eriochrome Black T indicator (sodium 1-( l-hydroxy-2-naphthylaz0)-6-
nitro-2-naphtol-4-sulfonate): dissolve 0.5 g of the indicator and 4.5 g of
hydroxylamine hydrochloride in 100 ml of water.
Calcon: dissolve 0.4 g sodium 1-(2-hydroxy-l-naphthylazo)-2-naphthol-4-
sulfonate in 100 ml of methanol.
Triethanolamine: dilute 30 ml triethanolamine to 100 ml with water.
Procedure. Filter the sample to remove undissolved solids and traces of oil
from the water. Transfer, by means of “Lambda” pipet or volumetric trans-
fer pipet, an aliquot of sample containing not more than 10 mg of calcium
into an Erlenmeyer flask. Dilute t o approximately 50 ml with water. (No
more than 10 ml of standard CDTA are t o be used in a titration for either
calcium or magnesium.)
Add two to three drops of triethanolamine solution and approximately 4
ml of sodium hydroxide solution. The pH of the solution at this point
should be 12.0-12.5. Add six drops of calcon indicator and titrate with
standard CDTA solution until the indicator blue endpoint is reached. Record
the volume of CDTA titrant used t o titrate calcium.
Using the same pipet, pipet another aliquot into another Erlenmeyer flask
and dilute t o 50 ml with water. Add two to three drops of triethanolamine
solution, 3-5 ml of the ammonium chloride ammonium hydroxide buffer
solution, and three to four drops of Eriochrome Black T indicator. Refill the
buret with the same standard CDTA titrant and titrate the sample until the
color changes from wine to pure blue. This endpoint is sometimes delayed,
so proceed cautiously with the titration near the endpoint. Record the
volume of CDTA used t o titrate calcium and magnesium, or hardness (Ca +
Mg).
TITRIM ETRIC METHODS 43
Calcula t ions :
B x M x 100,100
= mg total hardness as CaC03 per liter
sample volume
A x M x 40,000
= mg Ca per liter
sample volume
(B-A) x M x 24,300
= mg Mg per liter
sample volume
where A = calcium titer (blank); B = hardness titer (blank); and M = molarity
of CDTA.
The complexometric determination of calcium and magnesium will give a
precision of about 2% of the amounts present. The accuracy is dependent
upon the interferences present, and the major interferences are strontium and
barium, both of which will be complexed along with calcium, thus producing
high results. In the absence of strontium and barium the accuracy of the
method is about 4% of the amount of calcium and magnesium present.
Ammonium nitrogen
Organic compounds containing nitrogen decompose in a reducing environ-

ment and form ammonia and the ammonium ion. A reducing environment is
characteristic of a petroleum genetic environment (Collins et al., 1969).
Bogomolov et al. (1970) call it an indicator of petroleum.
The ammonium ion is too weak an acid t o be successfully titrated;
however, when treated with formaldehyde, hexamethylenetetramine and a
strong acid are produced. This strong acid can be titrated with a base using
indicators or a potentiometer to determine the endpoint.
Reagents. The necessary reagents are hydrochloric acid, 12N; sodium

hydroxide standard, OJN, and 0.02N;and formaldehyde.
Equipment. The necessary equipment includes an expanded-scale pH meter,

a hotplate, microburets, flasks, and an ice-water bath.
Procedure. The method should not be used if less than 5 mg/l of ammonium
nitrogen is present. Acidify the brine or water when sampling t o a pH of
about 1.5 with 12N HC1. The acid will stabilize the sample by changing to
the ammonium ion any ammonium hydroxide which could volatilize as
ammonia and be lost. Transfer a 100-ml aliquot of the acidified sample to a
250-ml Erlenmeyer flask and boil the solution for 5 minutes on a hotplate.
Cool the solution as quickly as possible to about 25°C. An ice bath will
facilitate .rapid adjustment to this temperature.
Using an expanded-scale pH meter, adjust the pH of the sample to 7.0

with 0.1N NaOH, add 5 ml of 37 wt.% formaldehyde solution, and heat the
mixture t o 4OoC. (Do not exceed this temperature.) Cool immediately t o
ambient temperature using an ice water bath, and titrate the sample with
0.02N NaOH to pH 8.6 using an expanded-scale pH meter t o detect the
endpoint .
The weak hydroxide titrant must be protected from atmospheric carbon
dioxide, and a reagent blank must be determined because formaldehyde
contains formic acid.
The reactions are :
6HCHO + 4NH4 C1 (CH2 )6 N4 + 4HC1+ 6H2 0
--f
HC1+ NaOH + NaCl + H20
Calculation:
(ml NaOH x N used for sample) -

(ml NaOH x N used for reagent blank) x 14,007
ml sample = mg/l NH4N
Chloride
A modification of the Mohr method (Furman, 1962) is satisfactory for

the determination of chloride in petroleum-associated waters. Common
interferences are bromide, iodide, sulfide, and iron. Sulfide can be removed
by acidifying the sample with nitric acid and boiling. Iron can be removed by
ion exchange or precipitation with sodium hydroxide or sodium peroxide
followed by filtration.
Because most petroleum-associated waters contain high concentrations of
chloride, it usually is necessary to dilute the sample before titrating with
silver nitrate, because the voluminous precipitate masks the endpoint. About
50 mg of chloride is maximum for a satisfactory titration. The indicator
usually is potassium chromate or sodium chromate, and at the endpoint the
chromate ion combines with excess silver to form the slightly soluble red
silver chromate:
Ag+ + C1- + AgCl

ZAg+ + --f Ag2Cr04
The specific gravity of the sample can be used t o estimate the correct
aliquot size. Table 3.X indicates aliquot sizes that will contain less than 50
mg of chloride. The micropipet can be used as demonstrated in the
calcium-magnesium procedure and Table 3.IX.
TABLE 3.X
Aliquots that contain less than 50 mg of chloride as estimated from the specific gravity
Specific gravity Dilution Aliquot (ml)
1.000-1.002 none 100

1.003-1.004 none 50
1.005-1.012 dilute 10 ml to 100 ml, take 50 ml 5.0
1.01 3-1.019 dilute 10 ml to 100 ml, take 20 ml 2.0
1.020-1.032 dilute 10 ml to 100 ml, take 1 0 ml 1.0
> 1.163 dilute 10 ml to 1,000 ml, take 10 ml
-
0.1
Reagents. The necessary reagents include silver nitrate, standard solution,

0.05N;potassium or sodium chromate, neutral 5% aqueous solution; and
nitric acid, 0.1N (nitrous free); and sodium bicarbonate.
Equipment. The necessary equipment includes a hotplate, a 10-ml micro-

buret, flasks, and pipets.
Procedure. After removal of interferences and selection of correct aliquot

size, dilute the sample to 20 ml or more, adjust the pH to 8.3 with sodium
bicarbonate or 0.1N nitric acid, add 1 ml of a 5% aqueous potassium chro-
mate solution, and titrate with an 0.05N silver nitrate solution until the red
endpoint just persists.
Calculation:
ml AgN03 x N x 35,500 = mg,l cl-
ml sample
The precision and accuracy of the method are about 1%and 2%, respec-
tively, of the amount present.
Bromide and iodide
Bromide and iodide are present in almost all petroleum-associated waters.

In the following procedure, iodide is selectively oxidized t'o iodate with
bromine water; excess bromine is reacted with sodium formate. The iodate
reacts with added iodide t o produce iodine which is titrated with thiosulfate.
Hypochlorite is added to another sample to oxidize both bromide and iodide
to bromate and iodate, respectively. Excess hypochlorite is reacted with
sodium formate, and the bromate and iodate are reacted with iodide t o
liberate iodine for titration with thiosulfate.
Iron, manganese, and organic matter can interfere but are removed in the
procedure. Fluoride is added to mask interference from any remaining traces
of iron.
Reagents. The necessary reagents include a 2% ammonium molybdate solu-

tion; glacial acetic acid; calcium hydroxide; calcium carbonate; 0.05N
hydrochloric acid; 6N hydrochloric acid; potassium iodide; sodium fluoride;
starch indicator solution; 0 . O l N sodium thiosulfate (standardize prior t o
use); 3.8M sodium formate (prepare fresh daily); saturated bromine water;
and methyl red indicator solution.
Equipment. The necessary equipment includes a mechanical shaker, 200-ml

bottles, a hot-water bath, flasks, pipets, and microburets.
Procedure. To remove iron, manganese, and organic matter from the sample,
add exactly 100 ml of sample to a stoppered bottle. Add 1 g of calcium
hydroxide, and place the mixture in a shaker for 1 hour. Allow the mixture
to stand overnight and filter through a dry folded filter, discarding the first
20 ml that comes through. Brines with specific gravities of less than 1.009
may be filtered without standing overnight. Prepare a blank in the same
manner.
Transfer an aliquot of the filtrate containing 1-2 mg of iodide t o a
250-ml Erlenmeyer flask. Add sufficient water t o make the total volume 75
ml, and three drops of methyl red indicator. Add 0.05N hydrochloric acid
until the mixture is just slightly acid, add 10 ml of sodium acetate solution,
1 ml of glacial acetic acid, and 4 ml of bromine water, and allow to stand for
5 minutes. Next add 2 ml of sodium formate solution, blow out any bromine
vapor from the neck of the flask, and wash down the sides with water.
When the solution is completely colorless, add 0.2 g of sodium fluoride
and 0.5 g of potassium iodide. Mix until dissolved and add 15 ml of 6N
hydrochloric acid. Titrate with 0.01N sodium thiosulfate using starch indica-
tor. Disregard any return of blue color after the endpoint. Record this titra-
tion volume for the iodide calculation.
Transfer another aliquot of the filtrate containing 1-2 mg of bromide t o a
250-ml Erlenmeyer flask and add sufficient water to make the total volume
75 ml. Add 10 ml sodium hypochlorite solution and approximately 0.4 g of
calcium carbonate (or enough so that approximately 0.1 g will remain after
the next step). Adjust the pH of the solution with 3N hydrochloric acid to
between 5.5 and 6.0 and heat in a water bath t o 90°C for 10 minutes. (A
small amount of undissolved calcium carbonate should remain at this point.)
Remove the flask and cautiously add 10 ml of sodium formate solution,
return the flask to the water bath, and keep the contents hot for 5 minutes
more and observe the timing very closely. Rinse down the inside of the flask
with a few milliliters of distilled water and allow the solution to cool t o
room temperature. (Do not use a cold water bath.) To the ambient solution
TITRlMETRIC METHODS 47
add three drops of ammonium molybdate solution, 0.5 g sodium fluoride (if
iron is present), and 0.5 g potassium iodide, mix until dissolved, and acidify
with 15 ml of 6N hydrochloric acid. Titrate with 0 . O W sodium thiosulfate
using starch indicator. Disregard any return of blue color after the endpoint.
Record this titration for the bromide calculation.
Calculations. Iodide: ml of Na, S2O 3 for sample - ml of Na, S2O3 for blank =
corrected ml of Na, S2O 3 :
(ml x N) N a 2 S 2 0 3x 21,150-
- mg/l I-
ml sample
Bromide: ml of Na2S2O 3 for sample - ml of Na2 S2O3 for blank = corrected
ml of Na2S203:
(ml x N) Na, S 2 0 3 x 13.320
ml sample
- mg/l I- x 0.63 = mg/l Br-
The precision and accuracy of the method are about 3%and 676, respec-
tively, of the amounts of bromide and iodide present.
Oxygen
The solubility of a gas varies directly with pressure and inversely with
temperature and usually is reduced by the presence of dissolved minerals.
Most petroleum-associated waters contain little or no dissolved oxygen in
situ at depth. Knowledge of the dissolved oxygen content of waters that are
to be reinjected for waterflooding or disposal is needed to determine treat-
ment required t o prevent corrosion. Instrumental and wet chemical methods
(American Petroleum Institute, 1968) are available for the determination of
dissolved oxygen. Instrumental methods usually are modifications of the
rotating platinum electrode method (Marsh, 1951), but with them the residual
current (when no oxygen is present) is difficult to determine. The modified
Winkler method probably is the most accurate wet chemical method available
(Watkins, 1954).
In the Winkler method for quantitatively determining dissolved oxygen in
water, a glass-stoppered bottle is completely filled with the water to be
tested. Manganous sulfate (MnS04) and potassium hydroxide (KOH) are
added, forming a precipitate of manganous hydroxide (Mn(OH), ) in accor-
dance with the following reaction:
MnS04 + 2KOH + Mn (OH), + K2S04
The manganous hydroxide combines with the oxygen dissolved in the
water to form a higher oxide of uncertain composition, assumed t o be man-
ganese hydroxide (MnO(OH), ), as follows:
2Mn (OH), + 0, + MnO (OH),
On acidification in the presence of an iodide, the higher oxide of man-

ganese liberates a quantity of iodine stoichiometrically equivalent t o the
quantity of dissolved oxygen present in the sample in the following manner:
MnO (OH), + 2H2 SO4 +. Mn(S04 )Z + 3H20

Mn(S04), + 2KI +. MnS04 + K2S04 + I2
The quantity of iodine liberated is determined by titrating an aliquot

portion of the sample with a standard solution of sodium thiosulfate
( N a 2 S 2 0 3 )using starch solution as an indicator, as shown by the equation:
2Na2S, O3 + I, +. Na, S40, + 2NaI
The iodine modification of the Winkler method depends upon the conver-
sion of any hydrogen sulfide t o hydrogen iodide and free sulfur by reducing
the iodine added t o the brine. This reaction proceeds as follows:
H,S+I2 + 2 H I + S
Tests have shown that interfering substances other than hydrogen sulfide
that might be present in oilfield brines also are counteracted by the iodine
added.
Reagents. It is important to use sterile glassware or polyethylene bottles in

preparing and storing reagents for this test to prevent contamination and t o
make longer storage of reagents possible without appreciable changes in their
normality.
Iodine solutions, 0.5N and 0.W.
Hydrogen sulfide water: saturate distilled water (which has been boiled
and cooled recently t o drive off dissolved oxygen) with hydrogen sulfide gas.
Starch solution.
Manganous sulfate solution: dissolve 480 g of manganous sulfate
(MnS04 *4H20) or 400 g of manganous sulfate (MnS04 *2H20) in distilled
water, filter, and dilute to 1 liter.
Alkaline iodide solution: dissolve 700 g of potassium hydroxide (KOH) or
500 g of sodium hydroxide (NaOH) and 150 g of potassium iodide (KI), or
135 g of sodium iodide (NaI).in distilled water, and dilute the solution t o 1
liter. If a white carbonate precipitate is formed, separate the precipitate by
settling and then siphoning off the supernatant liquid. The solution should
give no color with starch indicator when diluted and acidified, which indi-
cates the absence of nitrates, iodates, and ferric salts.
Sulfuric acid, concentrated.
Sodium thiosulfate solution, 0.1N.
Standard sodium thiosulfate solution, 0.025N.
Equipment. The necessary equipment includes glass-stoppered bottles,

pipets, flasks, and microburets.
Sa mp 1ing
Care must be taken t o obtain uncontaminated samples of water for deter-

mining dissolved gases. Glass-stoppered bottles should be used for sample
containers. To determine dissolved oxygen in water, 300-ml bottles with
pointed, ground-glass stoppers and overflow lips of the type used for bio-
chemical oxygen-demand tests are particularly suitable. These bottles are so
designed that samples may be obtained without contamination by atmo-
spheric oxygen and so the necessary chemical reagents may be introduced
during the analysis without excessive overflow from the lip of the bottle.
Before a sample is taken, rinse the bottle three times with the water to be
sampled and fill through a rubber tube extending to the bottom of the bottle.
A quantity of water equal t o at least three times the capacity of the bottle
should be allowed to overflow the bottle, and the rubber tube should be
withdrawn slowly so that the space in the bottle occupied by the tube is
filled simultaneously with water. The glass stopper, when placed in the
mouth of the bottle, will displace all excess water. If any bubbles are seen,
the sample is immediately analyzed. If the temperature of the water taken
for analysis of dissolved gases is above 2OoC, a cooling coil should be used to
cool the sample before the water enters the bottle. It is important that the
samples contain no included atmospheric oxygen or carbon dioxide, as errors
may be introduced into many of the analyses if extraneous oxygen or carbon
dioxide is present in the water.
Procedure. All reagents in the following steps 1 through 8 should be added

slowly and carefully under the surface of the water near the bottom of the
bottle, using pipets, permitting the displaced water t o overflow the top of
the bottle. The quantities of reagents added should be recorded for use in
the final calculation. After each reagent is added, the stopper should be
carefully replaced and the bottle inverted gently several times so as not to
introduce air into the bottle while adding and mixing reagents.
Collect the sample as described previously. Add excess 0.5N iodine solu-
tion to give the sample a yellow color and let stand 5 minutes. Add saturated
hydrogen sulfide water until the sample is a very light straw-yellow, and 1ml
of starch solution as an indicator. Add dilute hydrogen sulfide water until
the blue color just disappears and then add, drop by drop, 0.1N iodine
solution until a faint blue color persists. Add 1 ml of manganous sulfate
solution, 1 ml of alkaline iodide solution, and 1ml of concentrated sulfuric
acid, letting it run down the neck of the bottle.
Transfer 200 ml of the solution by pipet from the sample bottle t o a
500-ml Erlenmeyer flask. Titrate the 200-ml sample in the Erlenmeyer flask
with 0.025N sodium thiosulfate solution. The starch indicator should be
added when the yellow color of free iodine has been almost eliminated by
the sodium thiosulfate titration, and the titration should be continued until
one drop changes the solution from a light blue to colorless. (Subsequent
blue recoloration should be disregarded.)
If no hydrogen sulfide or other interfering substances are present, the first
six steps of the determination may be eliminated, using only the part of the
procedure starting with the addition of the alkaline iodide solution.
Calculation. The dissolved oxygen content of the water is determined by the

following equations:
v = 200 x-(Y-1)
u=-200 w
V
where U = dissolved oxygen content, ppm; V = volume of sample titrated,
ml; W = volume of 0.025N sodium thiosulfate required, ml; X = volume of
sample bottle, ml; Y = total volume of all reagents added, ml; and 1 = the 1
ml of acid added, which does not change the effective oxygen-tested volume
of the sample because it is added after all the oxygen has been absorbed.
The factor used to take into account the volume of reagents added may
involve a slight error, because it is based on the assumption that the reagents
contain no dissolved oxygen.
Carbon dioxide
Petroleum-associated waters containing carbon dioxide and bicarbonate or

carbonate will contain a weak acid H2C03 or its salt, which buffers the
solution. This combination controls the pH of waters in the range of about
pH 4.5-8.0. Such buffering is caused by the presence of slightly dissociated
acids or bases, and when H+ or OH- ions are added they first convert the
undissociated acid or base to its salt or vice versa.
Loss of carbon dioxide will disturb the carbon dioxide-bicarbonate-
carbonate buffer systems. For example, the pH probably will change and
precipitation of calcium carbonate or other compounds may occur. An in-
crease in carbon dioxide will shift t h e . carbon dioxide-carbonate-
bicarbonate equilibria, allowing more material such as calcite t o go into
solution.
Bacterial reduction of sulfate can cause the amount of dissolved carbon
dioxide and hydrogen sulfide in petroleum-associated waters t o be quite
high. Several hundred milligrams per liter of C 0 2 can be present in such
waters. Knowledge of the amount of carbon dioxide in solution is useful in
carbonate equilibria studies (Garrels and Christ, 1965) and in water com-
patibility studies (Watkins, 1954).
TITR IMETR IC METHODS 51
Reagents. The necessary reagents are 0.05N sodium carbonate solution and
phenolphthalein indicator solution.
Procedure. Collect the water sample in the same manner used in taking the
sample t o be analyzed for dissolved oxygen. Pipet 100 ml of the water into a
flask and add five drops of phenolphthalein indicator. If the sample turns
red, no free carbon dioxide is present; if it remains colorless, titrate the
sample with the standard sodium carbonate solution to a red endpoint.
Calcula tion :
ml Na2CO, x N x 22,000
= mg/l C02
ml sample
Sulfide
As mentioned above, the bacterial reduction of sulfate causes some

petroleum-associated waters t o contain appreciable concentrations of hydro-
gen sulfide. Knowledge of the amount of dissolved sulfide present is neces-
sary for corrosion and water compatibility studies (Watkins, 1954).
The following method depends upon the reduction of iodine by the hy-
drogen sulfide in the brine, as shown by the following equation:
H2S+I2 + 2 H I + S
Because of the unstable nature of the hydrogen sulfide in solution in waters

and brines, the sulfide is not titrated directly. To prevent the loss of hydro-
gen sulfide t o the air, an excess of iodine solution is added, and the sample is
back-titrated with standard sodium thiosulfate solution, in accordance with
the following equation:
2Na2 S2 0, + I2 +. Na2S4 O6 + 2NaI

Experiments conducted by the US. Bureau of Mines indicate that residual
reducing agents that cannot be removed by aeration or boiling are present in
some oilfield brines. Brine from the Arbuckle (siliceous) Limestone forma-
tion originally containing 96 mg/l hydrogen sulfide showed such residual
reducing agents t o equal 9 mg/l of hydrogen sulfide after air has been
bubbled through the brine for 28 hours. This dropped to 4 mg/l after
standing another 24 hours. Further tests in which the hydrogen sulfide was
driven off by boiling indicated the presence of 5 mg/l residual reducing
agents. When the brine was neutralized with hydrochloric acid (using methyl
orange indicator) before boiling, residual reducing agents equal t o 2 mg/l
hydrogen sulfide remained.
Reagents. The necessary reagents are potassium iodide, standard sodium

thiosulfate, 0.1N and 0.01N; standard iodine solutions, 0.1N and 0.01N; and
starch indicator solution.
Procedure. Collect the sample in a glass-stoppered bottle (approximately

200-mi capacity) in the manner previously described for dissolved oxygen.
Analysis should be made as soon as possible after sampling.
Pipet 5 ml of 0.1N or 0.01N standard iodine solution, depending upon the
hydrogen sulfide concentration expected, into each of two Erlenmeyer
flasks. It may be necessary to use a larger quantity of 0.01N solution if the
hydrogen-sulfide content of the sample is high.
Add approximately 1 g of potassium iodide crystals t o each flask. (This
step usually may be omitted in determinations on brine samples because of
the high mineral content of the water.) Add 50 ml of distilled water t o the
flask to be used for a blank determination, and pipet 50 ml of the water
sample into the other flask. Titrate both the distilled water blank and the
water sample with standard sodium thiosulfate solution of the same normal-
ity as the iodine solution used, adding 1 ml of starch indicator near the end
of the titration. Record the milliliters of thiosulfate used in each titration.
Calculation. Subtract the milliliters of thiosulfate solution used for the

sample from the milliliters used for the blank and use the difference in the
following formula:
(ml x N ) I2 - (ml x N) N a 2 S 2 0 3x 17,000
= mg/l H 2 S
ml sample
Sulfur compounds
The redox potential of petroleum-associated waters indicates that sulfur

compounds other than sulfate and sulfide may exist in solution. When the
water is brought t o the land surface, the change in pressure and temperature
will affect the redox potential and, if the sample is allowed t o come into
contact with the atmosphere, the equilibria of the sample will start t o change
immediately. Better methods are needed t o determine the composition of a
water in situ. The following method can be used to gain a semiqualitative
estimation of the amomts of thiosulfate, sulfite, and sulfide in a water.
Reagents. Zinc carbonate suspension: add zinc acetate to a solution of

sodium carbonate, filter and wash the precipitate with several volumes of
cold water. Prepare the zinc carbonate suspension by vigorously shaking the
precipitate with water.
The other reagents are glycerol iodine, 0.01N; sodium thiosulfate, 0.01N;
starch indicator solution; glacial acetic acid; and formaldehyde.
FLAME SPECTROPHOTOMETRIC METHODS 53
Determination o f thiosulfate, sulfate, and sulfide
Procedure. Collect a water sample as described in the dissolved oxygen

procedure. Pipet 100 ml of the sample into a 300-ml flask, and add 20 ml of
glycerol, 100 ml of an aqueous suspension of zinc carbonate, and 70 ml of
distilled water. Shake the mixture vigorously for 1 minute, filter, and discard
the precipitate.
Pipet 50 ml of the filtrate into a flask and add 5 ml of formaldehyde, and
3 ml of acetic acid, add starch indicator and titrate to the blue endpoint with
0.01N iodine. Record the amount of iodine used to calculate thiosulfate (A).
Pipet another 50-ml aliquot of the filtrate into another flask; add 0.01N
iodine until the solution remains yellow. Add starch indicator and titrate to
a colorless endpoint with 0.Ol.N sodium thiosulfate. Record the amount of
iodine used for thiosulfate plus sulfite (B).
Pipet 25 ml of water that was not treated with the zinc carbonate into a
flask and add an excess of 0.Ol.N iodine, 3 ml of acetic acid, add starch
indicator and titrate t o the colorless endpoint with 0.01N sodium thiosul-
fate, sulfite, and sulfide (C).
Calculations. Milliliters iodine used in A = X ml

X ml x N x 112,000
= mg/l S2 03-2
ml sample
Milliliters iodine used in A - milliliters iodine used in B = Y ml
Y ml x N x 40,000
= mg/l SO,-2
ml sample
Milliliters iodine used in C - milliliters iodine used in B = 2 ml
2 ml x N x 16,000
= mg/l S-’
ml sample
FLAME SPECTROPHOTOMETRIC METHODS
When a metal salt in solution is sprayed into a flame, the solvent evapo-
rates and the salt decomposes and vaporizes, producing atoms. Some of these
atoms can be raised t o an excited state by the thermal energy of the flame,
although a major portion of the atoms present in the flame remain at the
grourid state. The return of the excited atoms to the ground state results in
the emission of radiant energy characteristic of the element atomized. The
quantitative measurement of this radiation is the basis of emission flame
spectrophotometry, and the essential difference between this form of
analysis and classical arc-emission spectrography is the temperature of the
source used to excite the atoms. Because the g a s a i r and gas-oxygen flames
are much cooler than the spark and arc sources used in spectrography, analy-
sis by emission flame spectrophotometry is usually limited t o the more easily
excitable elements - lithium, sodium, and potassium.
Instrumentation requirements include:
(1) A method of introducing the sample into the flame for vaporization.
(2) A method of detecting and recording the radiation intensity emitted.
(3) A method of selecting the correct wavelength, ordinarily a variable
monochromator.
A more complete discussion of the theory and instrumentation can be
found in books by Burriel-Marti and Ramirez-Munoz (1957) and Dean
(1960), as well as in publications of commercial instrument manufacturers.
Lithium
Lithium usually is calculated as a part of the sodium content in reporting

the results of oilfield water analyses rather than being determined and
reported separately. One of the more accurate methods t o determine lithium
in petroleum-associated waters is the flame spectrophotometric method
(Collins, 1962).
Reagents. The reagents are lithium, standard solutions, 0.1 mg/ml and 0.01
mg/ml; and n-propanol.
Equipment. The necessary equipment includes a flame spectrophotometer,

10-ml'microburets, and volumetric flasks.
Preliminary calibration curves. Preliminary calibration curves are useful in

determining approximately how much lithium is in the sample and in deter-
mining the optimum amount of standard lithium solution t o use in the
analysis. Because n-propanol is easier t o work with, it usually is used; how-
ever, if additional sensitivity is needed, the acetone-n-amyl alcohol mixture
can be used (Collins, 1965).
To prepare the preliminary calibration curves, transfer t o 50-ml volumet-
ric flasks aliquots of diluted standard lithium solution containing the follow-
ing amounts of lithium: 0.01 mg, 0.05 mg, 0.1 mg, 0.15 mg, and 0.2 mg.
Add 20 ml of n-propanol to each flask and dilute to volume with distilled
water. Aspirate, burn, and record the emission intensity of each of these five
standards at 670.8 mp and their background at about 665 mp. Record
several peaks for each standard at various sensitivity levels and slit widths.
Plot the results on linear graph paper by plotting milligrams of lithium
versus intensity. Prepare a curve for each sensitivity level and slit width used,
as illustrated in Fig. 3.2. The sensitivity of the instrument will determine the
optimum concentrations of lithium and this will require some experimenta-
tion.
The analyst may find it convenient t o scan all the emission lines of
26 -
- 1 'I I I I
2 4 ~ mm r l i t
0.01
22- 1,620 volta to I T 1 F W 6836
- tOppri 02
-x 20-
0
0 18-
- 5 p s i C2H2
12.5 mm burner height
L -
-
-
-
-
-
5 10-
-
-
L
-
-
-
-
mg L i / m l 5 0 % n - P R O P A N O L
Fig. 3.2. Preliminary calibration curves for use in selecting optimum standard additions:
Instrument: 0.01-mm slit, 1,620 V to ITT FW 6836, 1 0 psi 02,5 psi C2HZ,and 12.5-mm
burner height.
interest; e.g., lithium, sodium, potassium, rubidium, cesium, and perhaps

others. This will give information concerning what elements are present.
Procedure. To determine the amount of lithium in the petroleum-associated

water, transfer an aliquot of about 10 ml of brine to a 50-ml volumetric
flask, add 20 ml of n-propanol, and dilute to volume with distilled water.
(The size of the aliquot will vary with the sample. The specific gravity can be
used to help decide the aliquot size. For a brine with a specific gravity of
1.1, an aliquot of 10 ml or less probably will be sufficient.) Aspirate the
sample into the flame and read or record the emission intensity of the
background at 665 mp and lithium line at 670.8 mp. With these readings and
the preliminary calibration curves, calculate approximately how much
lithium is in the sample.
Determine an aliquot size that will contain about 0.05 mg of lithium.
Transfer equal aliquots to three 50-ml volumetric flasks. Add no lithium
standard to the first flask, 0.05 mg to the second flask, and 0.1 mg to the
third flask. Add 20 ml of n-propanol t o each flask and dilute to volume with
distilled water.
Aspirate and record the background at 665 mp and the emission intensity
of each sample at 670.8 mp. Optimum accuracy is attained by this method
when the two standard additions are respectively equal to and twice the
amount of lithium in the sample. Care should be taken that too much
lithium is not present in the final samples, because self-absorption will cause
errors.
v)
13
r
2 3
W
a
c
a
.x2
L
I
u
1 2s
0
I I I I I
I 3 4 5 6 7
COI -
tENTRATION OF STANDARD ADDITIONS
Fig. 3.3. Standard-addition calculation graph. In this ideal case the unknown would con-
tain 2 x the dilution factor ( 2 could be 2 mg or 2 pg or whatever unit the analyst used).
Calculation. A graph can be used in the calculation, as illustrated in Fig. 3.3.

Plot the concentrations in milligrams of the standard-addition samples on the
horizontal axis of linear graph paper and the emission intensities on the
vertical axis. Plot the emission intensity of the sample to which no standard
lithium soiution was added at 0 concentration. The plot should produce a
straight line as shown in Fig. 3.3.
Multiply the chart reading at 0 concentration by 2, place this value on the
y-axis, and draw a line parallel to the x-axis until it intersects the line
plotted. From this point, draw a line parallel to the y-axis until it intersects
the x-axis. The vrlue obtained in milligrams can be converted to milligrams
per liter by the following formula:
mg Li x 1,000
= mg/l Li+
ml sample
The formula, shown in Table 3.X1, can be used to calculate the amount of
lithium in the sample, using the flame spectrophotometric readings in lieu of
the graph method. Optimum accuracy is attained with this method using
either type of calculation when the two standard additions respectively are
equal to and twice the amount of lithium that is present in the sample. The
addition of alcohols t o the aqueous phase before aspiration into the flame
increases the sensitivity of the flame method, allowing the use of more dilute
solutions and consequently less dissolved solids, which reduces burner plug-
ging. The average precision and accuracy of the lithium method are about 2%
and 4%, respectively, of the amount present.
TABLE 3.XI
C
Formula for standard-addition calculation C, = (rx - r b ) -
where the following are true*: r - r,
. .-
Solution Concentration Reading
Unknown c, r,
Mixture . c, = c, + c r
*C is a standard addition.
Sodium
The flame spectrophotometer offers an excellent instrumental technique

for determining sodium in a petroleum-associated water. The flames con-
taining alkali metals give strong resonance lines of these metals plus some
additional continyous radiation. The strongest line for sodium results from a
transition between the lowest excited level and the ground state. The yellow
doublet of sodium at 589.0-589.6 mp results from such a transition.
Reagents. The necessary reagents are sodium standard solutions, 1 mglml

and 0.01 mg/ml; and n-propanol.
Preliminary Calibration curves. Preliminary calibration curves similar to those

shown in Fig. 3.2 should be used to determine the approximate amount of
sodium in the sample. These curves are prepared in the same manner as the
lithium curves, except that standard sodium solutions are used; the emission
intensity of the sodium at 589 mp is determined, minus a background at
about 582 mp.
Procedure. To analyze the petroleum-associated water, transfer an aliquot of

water t o a 50-ml volumetric flask, add 20 ml of n-propanol, and dilute to
volume with distilled water. (The size of the aliquot will vary with the
sample. The specific gravity can be used to help decide the aliquot size. For
a water with a specific gravity of 1.1,an aliquot of 1ml or less probably will
be sufficient.) Aspirate the sample into the flame and record the emission in-
tensity of the background at 582 mp and sodium line at 589 mp. With these
readings, calculate approximately how much sodium is in the sample by using
the preliminary calibration curves.
Determine the aliquot size that will contain about 0.05 mg of sodium.
Transfer equal aliquots t o three 50-ml volumetric flasks. Add no sodium
standard to the first flask, 0.05 mg to the second flask, and 0.1 mg t o the
third flask. Add 20 ml of n-propanol to each flask and dilute to volume with
water. Aspirate and record the emission intensity of each sample at 589 mp
and its background at 582 mp.
Calculation. Use the graph or formula illustrated in the lithium method. The
value obtained in milligrams can be converted to milligrams per liter by the
following formula:
mg Na x 1,000
= mg/l Na'
ml sample
The precision and accuracy of the method are approximately 3%and 6%,
respectively, of the amount of sodium present. Some elements, when present
in the solution being analyzed, will cause a change in the emission intensity
of the sodium. The use of a standard addition technique largely compensates
for these interferences.
Potassium
Potassium usually is included with sodium without any differentiation in

reporting the results of brine analyses, although potassium is known to be
present in many oilfield brines. Potassium compounds often are dissolved
before sodium compounds; however, they do not remain dissolved as readily
because they are readily adsorbed and enriched in clays. In sea water and
oilfield brines, only a small part of the originally dissolved potassium remains
in solution. The fact that many oilfield brines are low in potassium with
respect to sodium, whereas surface waters and young volcanic waters are
enriched in potassium with respect to sodium, is an important criterion in
identifying the sources of brines.
The flame spectrophotometer provides a sensitive method for the determi-
nation of potassium. The strongest lines for potassium detection in a flame
are the doublet at 766.5 and 769.9 mp.
Reagents. The necessary reagents are potassium standard solution, 0.1

mg/ml; and n-propanol.
Preliminary calibration curves. Preliminary calibration curves are useful in

determining the approximate amount of potassium in the sample, so that the
optimum sample size for standard addition can be selected for the analysis.
These curves can be prepared in the same manner used in the preparation of
the lithium preliminary calibration curves (Fig.3.2) except that standard
potassium solutions are used. The emission intensity of the potassium line at
766.5 mp minus the background at about 750 mp can be used in preparing
the curves.
Procedure. To determine the amount of potassium in the sample, transfer an

aliquot of sample to a 50-ml volumetric flask, add 20 ml of n-propanol, and
dilute to volume with distilled water. The specific gravity can be used to help
decide the aliquot size. For a brine with a specific gravity of 1.1,an aliquot
of 5 ml or less probably will be sufficient. Aspirate the sample into the flame
and record the emission intensity of the background at 750 mp and potas-
sium line at 766.5 mp. With this reading, use the preliminary calibration
curves and calculate approximately how much potassium is in the sample.
Determine an aliquot size that will contain about 0.05 mg of potassium.
Transfer equal aliquots t o three 50-ml volumetric flasks. Add no potassium
standard to the first flask, 0.05 mg to the second flask, and 0.1 mg t o the
third flask. Add 20 ml of n-propanol to each flask and dilute t o volume with
distilled water. Aspirate and record the emission intensity of each sample at
766.5 mp and the background at 750 mp.
Optimum accuracy is attained by this method when the two standard
additions are respectively equal to and twice the amount of potassium in the
sample. Care should be taken that too much potassium is not present in the
final samples, because self-absorption will cause errors.
Calculation. The graph or formula illustrated in the lithium method can be

used. The value obtained in milligrams can be converted to milligrams per
liter by the following formula:
mg K x 1,000
= mg/l K+
ml sample
The precision and accuracy of the method are approximately 2% and 4%
of the amount present. Several elements can interfere in the flame analysis of
potassium. Elements which ionize easily will lower the degree of ionization
of potassium, and elements which are difficult to ionize or have high ioniza-
tion energies will give the opposite effect. By using the Saha equation
(Herrmann and Alkemade, 1963), it is possible t o estimate such interfer-
ences. Generally, the use of a standard addition compensates for inter-
ferences.
Rubidium and cesium
The flame spectrophotometer provides one of the most sensitive methods

available for determining rubidium and cesium. Cesium has a pair of emission
lines at 852.1 and 894.4 mp. Both lines are of about equal intensity, but
water produces a molecular band system at 900 mp which can interfere at
894.4 mp. Rubidium also has two strong lines in the red region at 780.0 and
794.8 mp.
It is necessary t o use a photomultiplier with an S-1 response t o detect
cesium and rubidium at the levels found in many waters. Examples of such
tubes are ITT type 6836/FW118, RCA types 1P22 and 7102, and DuMont
type 6911. Such tubes also are useful for lithium and potassium deter-
minations.
Several elements can interfere in the determination of cesium and
rubidium. However, because a solvent extraction or standard-addition tech-
nique is used most interferences are either removed or compensated (Collins,
1965).
Reagents. The necessary reagents are cesium standard solution, 0.01 mg/ml;
rubidium standard solution, 0.01 mg/ml; buffer solution, pH 6.6 (adjust the
pH of a 1M sodium citrate solution to 6.6 with 0.5M nitric acid); sodium
tetraphenylboron, 0.05M (dissolve 0.855 g of sodium tetraphenylboron in
distilled water and dilute t o 50 ml - prepare a fresh solution daily);
nitroethane; hydrochloric acid, 0.1N; sodium hydroxide, 0.W;synthetic
brine solution.
Procedure. To determine the amount of rubidium and cesium in the

petroleum-associated water, transfer an aliquot of brine containing 0.005 to
0.05 mg of cesium and rubidium to a 100-ml beaker and add 25 ml of the
citrate buffer solution. Transfer the solution to a 125-ml Teflon-stoppered
separatory funnel and adjust to 100-ml volume. Add 2 ml of 0.05M sodium
tetraphenylboron aqueous solution and 1 0 ml of nitroethane, and shake the
mixture vigorously for 2 minutes. Allow the phases t o separate for 30
minutes, after which time withdraw the aqueous phase. Centrifuge the
nitroethane phase. Determine the cesium and rubidium emission intensities
by burning the nitroethane phase in the flame spectrophotometer and
automatically scanning the 780.0 mp, 794.8 mp, and 894.4 mp lines.
Calibration curves. Prepare calibration curves by using appropriate portions

of the standard cesium and rubidium solutions. Add 5 ml of synthetic brine
solution t o each standard sample before buffering and extraction. Plot the
resultant emission intensities versus milligrams of cesium or rubidium or
linear graph paper.
Calculation. Determine the milligrams of cesium or rubidium in the sample

by referring t o the calibration curves. The milligrams can be converted to
mg/l by the following formula:
mgx 1,000
= mg/l Cs+ or Rb'
ml sample
Fig.3.4 illustrates the relative emission intensities obtained with cesium
and rubidium in nitrobenzene, nitroethane, 1-nitropropane, and 2-nitro-
propane. 15 ml of each of these solvents.are used t o extract 0.1 mg each of
cesium and rubidium tetraphenylboron from aqueous solutions. The organic
phases then are aspirated directly into the flame, and the peaks scanned
automatically. Good resolution is obtained with a 0.01 mm slit width. Amy1
alcohol gives poorer results than nitrobenzene.
NITROBENZENE NITROETHANE I- NITROPROPANE 2 - NITROPROPANE
1
Fig. 3.4. Relative intensities obtained by burning organic solvents containing tetraphenyl-
boron salts of cesium and rubidium.
Standard-addition technique to determine rubidium
Some waters contain sufficient rubidium to enable use of the standard-

addition technique. To analyze such waters, preliminary calibration curves
similar to those used to determine lithium (Fig.3.2) are recommended, to aid
in selecting the optimum amount of standard rubidium solution to use.
Manganese
The amounts of sodium, potassium , calcium, and strontium in most

petroleum-associated waters are too high t o permit determination of man-
ganese with the flame spectrophotometer without preliminary separations.
These interferences can be obviated by extracting the manganese into a
chloroform 8-hydroxyquinoline solution. The chloroform is removed by
evaporation, and the manganese hydroxyquinoline is dissolved in n-propanol.
This solution is burned in the flame spectrophotometer, and the emission
intensity of its resonance triplet at 403.2 mp is recorded (Collins, 1962).
Reagents. The necessary reagents are standard manganese solution (dissolve

0.583 g of manganese dioxide in 10 ml of hydrochloric acid and dilute to 1
liter with distilled water, transfer a 100-ml aliquot of this solution t o another
1-liter flask, add 10 ml of hydrochloric acid, and dilute to volume with

distilled water; ( 1 ml of this solution contains 10 pg of manganese); chloro-
form solution of 8-hydroxyquinoline (dissolve 1.O g of 8-hydroxyquinoline
in 100 ml of chloroform); hydrogen peroxide (3% solution); ammonium
hydroxide ( 3 N ) ; sodium potassium tartrate (10% solution); ammonium
fluoride (5%solution); n-propanol; and chloroform.
Procedure. Transfer an aliquot of brine containing up to 150 pg of manga-

nese to a 100-ml beaker; add 1ml of hydrogen peroxide, 5 ml of ammonium
fluoride, and 10 ml of sodium potassium tartrate; and adjust the pH of the
mixture to 9.0 with ammonium hydroxide. Transfer the solution to a 125-ml
Teflon-stoppered separatory funnel, add 10 ml of 8-hydroxyquinoline
chloroform solution, and bring the mixture t o equilibrium by shaking it
vigorously for 1minute.
Draw the chloroform phase off into a 100-ml beaker and strip the aqueous
phase by an additional extraction with chloroform. Evaporate the combined
chloroform extracts to dryness over a hotplate, taking care t o prevent the
residue from charring. Dissolve the residue in n-propanol and make to 50 ml
volume with n-propanol. Aspirate the n-propanol solution directly into the
flame and determine the net emission by subtracting the background emission
at 400 mp.
Calculate the amount of manganese in the sample from a calibration curve
prepared by adding known amounts of manganese t o a synthetic brine solu-
tion. The calibration curve should be linear for up t o 150 pg of manganese
when the emission intensity is plotted versus micrograms of manganese on
linear graph paper.
Calculation :
pg Mn (from curve)
= mg/l Mn +*
ml sample
The intensity of the emission of manganese in a flame spectrophotometer
is enhanced by a factor of 16 by using n-propyl alcohol rather than water as
the solvent. With this increased intensity, the sensitivity of the method is
about 1 mg/l, although additional sensitivity is attainable by concentrating
the brine by evaporation. The precision of the method is about 3%,and the
accuracy is about 6% of the amount present.
Strontium
Several flame photometric methods are available for determining stron-

tium in oilfield brines; a standard curie may be unreliable if there are
instrument changes, such as a slightly plugged burner, change of resistance in
the amplifying circuit, or other variables. Chemical precipitation of stron-
tium as the sulfate does not satisfactorily separate strontium from barium
and calcium without several preliminary separations. Precipitations as the

carbonate or oxalate have the same disadvantages, and precipitation as the
nitrate and subsequent solvent extraction of calcium with butylcellosolve
still leaves barium in the precipitate. The use of a standard addition flame
photometric method gives reproducible results without the necessity of
several separations.
Reagents. The necessary reagents are standard strontium solution, 1mg/ml;

and n-propanol.
Preliminary calibration curves. To determine approximately how much

strontium is present in the samples, it is advantageous t o prepare preliminary
calibration curves. A procedure similar to that used in the lithium method
can be used, except that the strontium emission should be determined at 680
mp with a background reading at 690 mp. The data are plotted in a manner
similar to Fig. 3.3.
Procedure. To determine the amount of strontium, transfer an aliquot of

brine to a 50-ml volumetric flask, add 20 ml of n-propanol, and dilute t o
volume with distilled water. Aspirate the sample into the flame and read;
record the emission intensity of the background at 690 mp and the stron-
tium line at 680 mp. With these readings and the preliminary calibration
curves, calculate approximately how much strontium is in the sample.
Determine an aliquot size that will contain about 1.0 mg of strontium.
Transfer equal aliquots t o three 50-ml volumetric flasks. Add no strontium
standard t o the first flask, 1.0 mg t o the second flask, and 2.0 mg t o the
third flask. Add 20 ml of n-propanol to each flask and dilute t o volume with
distilled water. Aspirate and record the background at 690 mp and the
emission intensity of each sample at 680 mp.
Calculation. A graph can be used in the calculation as illustrated in Fig.3.3.

The value obtained in milligrams can be converted t o milligrams per liter by
the following formula:
mg Sr x 1,000
= mg/l Sr+*
ml sample
The formula, shown in Table 3.X1, can be used to calculate the amount of
strontium in the sample using the flame spectrophotometric readings in lieu
of the graph method.
Barium
A flame spectrophotometric method was developed which utilizes the

chromate precipation followed by dissolution in nitric acid, mixing with an
alcohol, and burning in the flame (Collins, 1962). The flame method is
64 ANALYSIS OF OILFJELD WATERS
subject to few interferences except from calcium, but by using the chromate
precipitation, calcium is eliminated and barium is concentrated.
Reagents. The necessary reagents are barium standard solution, 1 mg/ml;

ammonium chromate solution (dissolve 10 g of ammonium chromate in
distilled water and dilute to 100 ml); 10%ammonium acetate aqueous solu-
tion; nitric acid ( 4 N ) ;n-propanol; acetic acid; and synthetic brine solution
(use carbon dioxide-saturated distilled water and dissolve the following
amounts of constituents in 1 liter of water: sodium bicarbonate, 0.4 g;
sodium chloride, 61 g; potassium. chloride, 5 g; calcium chloride, 19 g; mag-
nesium chloride, 1 2 g; and strontium chloride, 5 g).
Procedure. Transfer an aliquot of the sample containing 0.5-15 mg of

barium t o a 100-ml beaker, add 1 ml of the ammonium acetate solution, 10
ml of the ammonium chromate solution, and adjust the pH t o 4.6 using
acetic acid. Cover the beaker with a watchglass; heat the solution t o near
boiling (90°C), remove from the hotplate, and allow to stand for 1 hour.
Filter the solution through a 0.45-pm membrane filter using vacuum. Take
care to transfer all of the precipitate from the beaker to the filter funnel. Use
ammonium chromate solution rather than distilled water to aid in this trans-
fer.
Wash the precipitate with 50 ml of ammonium chromate or until stron-
tium and calcium are absent. Wash the precipitate with 50 ml of hot water to
remove excess chromate.
Add 5 ml of 4N nitric acid t o the filter and swirl the solution on the filter
gently to dissolve the precipitate. A clean test tube should be placed below
the filter to catch the dissolved precipitate. When all of the precipitate is
dissolved, turn on the vacuum and catch the solution in the test tube. Repeat
this procedure using an additional 5 ml of 4N nitric acid.
Transfer the solution from the test tube t o a 50-ml volumetric flask.
Carefully wash the test tube with two 5-ml portions of water. Add 25 ml of
n-propanol, dilute to 50 ml volume with water, and mix the solution
thoroughly. Burn the sample in the flame spectrophotometer and record the
emission intensity at 873 mp and the background at 900 mp.
Prepare calibration curves by adding up t o 25 mg of barium to 10 ml
portions of the synthetic brine followed by analysis according t o the fore-
going procedure, and use in the calculation.
Calculation:
mg Ba x 1,000
= mg/l Ba+’
ml sample
ATOMIC ABSORPTION METHODS 65
ATOMIC ABSORPTION METHODS
Atomic absorption is complementary to flame spectroscopy. The spectra

emitted are analyzed by absorption of resonance lines by free atoms of a
constituent in the vapor phase. The unexcited or ground-state atoms pro-
duced in the flame can absorb radiant energy when supplied by a suitable
external radiation source at a frequency coinciding with that of the emission
frequencies of the element atomized. The measurement of this radiation
absorbed forms the basis of absorption flame spectrophotometry - or
atomic absorption spectrophotometry, as it is usually called.
At temperatures up t o 2,7OO0C, ground-state atoms usually account for
more than 90%of the atoms in the vapor phase. Hollow cathode discharge
tubes generally are used as a light source. The sensitivity of detection does
not depend upon the spectral response of the light receiver, since the absorp-
tion coefficient is a measure of the relative intensity of the light which passes
through an absorption cell versus that which does not. Additional theory can
be found in a book by Robinson (1966).
Atomic absorption is useful in water and brine analysis, and there are
several publications on the subject. Publications oriented to oilfield and sea
water analysis are Fabricand et al. (1966), and Angino and Billings (1967).
Table 3.XII illustrates the sensitivities that can be obtained using atomic
absorption t o determine some metals in aqueous solutions. The sensitivities
listed are obtainable if no interferences are present. Interference usually
TABLE 3.XII
Approximate sensitivities for some metals to atomic absorption
Metal Wavelength Sensitivity Fuel and oxidant

(A) (mg/l)
-- .-
Aluminum 3093 1.o nitrous oxide-acetylene

Barium 5536 0.2 nitrous oxide-acetylene
Beryllium 2348 0.1 nitrous oxide-acetylene
Cadmium 2288 0.04 air-acetylene
Calcium 4226 0.08 air-acetylene
Chromium 3579 0.15 air-acetylene
Copper 3247 0.2 air-acetylene
Iron 2483 0.3 air-acetylene
Lead 2833 0.5 air-acetylene
Magnesium 2852 0.02 air-acetylene
Manganese 2794 0.15 air-acetylene
Mercury 2536 0.01 air-acetylene
Nickel 2320 0.15 air-acetylene
Silver 3281 0.1 air-acetylene
Sodium 5890 0.03 air-acetylene
Zinc 21 38 0.04 air-acetylene
results from lack of absorption of atoms bound in molecular combination in

the flame and can occur when the flame is not hot enough to dissociate the
molecule. It also occurs when a dissociated atom immediately oxidizes to a
compound that cannot dissociate further at the temperature of the flame.
Interferences
Ionization
When a significant number of the atoms of the element being determined

are ionized in the flame, an error in the analysis can result. This ionization is
because of excessive flame temperature, which, however, can be changed to
control this interference. Another type of interference can be caused by the
presence in the sample of other, more easily ionizable elements than the one
sought. The resulting increase can be controlled by the addition of a suffi-
cient amount of the interfering element t o both sample and standards t o
produce a “plateau” in the absorbance above which no further increase
occurs.
Che m ica 1
A chemical interference is caused by the formation, in the flame, of salts

of the element sought which are difficult to decompose, thus reducing the
amount of the element available for absorption. The formation of such
compounds may often be precluded by the addition of another element,
such as lanthanum, which forms a less-soluble salt with the interfering anion
than does the element desired. The interfering anion is thus removed from
the flame, and the interference is eliminated.
Phosphate combines with calcium and magnesium and produces an inter-
ference; however, the addition of lanthanum largely overcomes this inter-
ference. Addition of an excess of a cation having a similar or lower ioniza-
tion potential usually reduces interference problems.
Matrix
Matrix interference is caused by unequal amounts of dissolved solids in

the standards and samples. This can cause error because of differences in
aspiration rates through the atomizer. Often this can be controlled by
matching the specific gravities of the standards and samples or by adding
salts t o the standards.
Burners and solvents
Various types of burners are used with atomic absorption spectrophoto-

meters. For example, a Boling burner usually is used for aqueous solutions,
while a premix burner is used for organic solutions. A nitrous oxide burner
head with. a 2-inch slot is used for determining aluminium, barium, and
beryllium because overheating is often encountered wit,h a 3-inch slot
burner.
The use of concentration steps, such as solvent extraction of a chelated
compound, enables sensitivities lower than those shown in Table 3.XII to be
achieved. For example, aluminium and beryllium can be complexed with
8-quinolinol and extracted with chloroform; cadmium and lead can be com-
plexed with ammonium pyrrolidine dithiocarbamate and extracted with
methyl isobutyl ketone. When burning the organic solvents, it usually is
necessary t o reduce the fuel air ratio because the burning organic solvent
contributes to the fuel supply producing an undesirable luminescent flame
and may also lift the flame off the burner. An optimum fuel/air ratio can be
found by noting the characteristics of the flame before burning the organic
solvent and then reducing the fuel flow, while burning the organic solvent
until the flame characteristics are similar t o those noted before the organic
solvent was burned. Ramirez-Munoz (1968) provides additional information.
Burner height is very important and adjustment often is necessary when
changing from one element t o another. Some instruments have a Vernier
adjustment for reproducing burner-height settings and some do not. Fig. 3.5
illustrates a device which can be used for reproducing exact burner height
(Ballinger et al., 1972).
0-m from hollow cathode lamp
Fig. 3.5. Device for reproducing burner height for emission and atomic absorption spec-
trometers.
Lithium
Lithium is determined at the 6707.8 A wavelength with an air-acetylene

flame.
Interferences. Ionization interference is suppressed by adding 1,000 pg/ml of

potassium.
Reagents. The necessary reagents are:

(1)Potassium solution: see reagents preparation under “Sodium”.
(2) Standard lithium solution: obtain commercially or dissolve 5.324 g of
lithium carbonate, Liz CO, , in a minimum volume of one part Hz0 to one
part of HC1 (1+ 1).Dilute to 1liter with water. 1ml of this solution contains
1,000 pg of lithium.
Preliminary calibration. Prepare standard lithium solutions containing 1-5

pg/l of lithium using the standard lithium solution and 50-ml volumetric
flasks. Add to each of these and to a blank, 0.5 ml of the potassium stock
solution. Aspirate these standards and the blank as recommended in the
calcium method and determine the absorbance at a wavelength of 6707.8 A.
Procedure. Transfer an aliquot of brine to a 50-ml volumetric flask. The

specific gravity of the brine can be used as a guide in estimating the size of
an aliquot containing about 0.05 mg of lithium. Add 0.5 ml of the potassium
stock solution, dilute t o volume with water, and aspirate. Calculate the
approximate sample concentration from the preliminary calibration readings
and determine the aliquot size that will contain about 0.05 mg of lithium.
Transfer equal aliquots containing about 0.05 mg of lithium t o three
50-ml volumetric flasks. Add no lithium standard t o the first flask, 0.05 mg
of the lithium standard t o the second flask, and 0.10 mg t o the third. Add
0.5 ml of the potassium stock solution t o each of the three flasks and dilute
to volume. Aspirate and record the absorbance readings for each sample.
Calculations. See calculations under “Lithium” in the flame spectrophoto-

metric section, Fig.3.3, or Table 3.XI:
mgLix 1,000
= mg/l Li+
ml sample
Precision. In a single laboratory using oilfield water samples containing con-
centrations of 90 and 190 mg Li+/l, the standard deviations were k 3 and +5,
respectively. The recoveries were 100.6% and 92.996, respectively.
Sodium
Two wavelengths are used: the 5890-5896 A doublet for the 1-mg/l
aliquots and the 3302-3303 a doublet for the 100-mg/l aliquots. Because of
the wide .range of sodium concentrations found in brines, the higher wave-
length can be used for the lower gravity brines and the lower wavelength for
the higher gravity brines, thus avoiding making two dilutions with some of
the heavier brines. It is usually necessary t o make a preliminary determina-
tion so that the correct aliquot can be used with the standard additions.
Interferences. Ionization interference is usually overcome by adding potas-

sium.

(1) Potassium solution: dissolve 190.70 g of potassium chloride, KC1, in
water and dilute to 1 liter. 1 ml of this solution contains 100 mg of potas-
sium.
(2) Standard sodium solution: obtain commercially or dissolve 25.420 g
of sodium chloride in 1 liter of water. 1 ml of this solution contains 10 mg
of sodium. Dilute 1 0 ml of this solution t o a liter. 1 ml of this solution
contains 100 pg of sodium.
Preliminary calibration. Prepare standard solutions containing 1.O-5.0 and

100-500 pg/ml of sodium using the standard sodium solutions and 50-ml
volumetric flasks. Add to each of these, and to a blank, 0.5 ml of the potas-
sium stock solution. Aspirate these standards and blank as recommended in
the calcium method and determine the absorbance at 5890-5896 a for the
1.0-5.0 pg/ml Na solutions and at 3302-3303 a for the 100-500 pg/ml Na
solutions.
Procedure. Transfer an aliquot of brine t o a 50-ml volumetric flask. The

an aliquot containing either about 0.05 mg or about 5 mg of sodium. Add
0.5 ml of the potassium stock solution, dilute t o volume, and aspirate.
Calculate the approximate sample concentration from the preliminary cali-
bration readings. Determine the aliquot size that will contain either about
0.05 mg or 5 mg of sodium, depending on the wavelength to be used.
Transfer equal aliquots t o three 50-ml volumetric flasks. For the 0.05-mg
aliquots, add no sodium standard t o the first flask, 0.05 mg of sodium
standard to the second flask, and 0.10 mg to the third. For the 5-mg
aliquots, add no sodium standard to the first flask, 5 mg t o the second, and
10 mg t o the third. Add 0.5 ml of the potassium stock solution t o each flask
and dilute to volume. Aspirate and record the absorbance readings for each
sample.

metric section, Fig. 3.3, or Table 3.XI:
mg Na x 1,000
= mg/l Na+
ml sample

centrations of 22,700 and 43,200 mg Na+/l, the standard deviations were
+485 and ?1,890, respectively. The recoveries were 100.8% and 100.9%,
respectively.
Potassium
Potassium is determined at the 7664.9 A wavelength with an air-acetylene

flame.

sodium.

(1) Sodium solution: dissolve 254.20 g of sodium chloride in 1 liter of
water. 1 ml of this solution contains 100 mg of sodium.
(2) Standard potassium solution: obtain commercially or dissolve 1.907 g
of potassium chloride, KCl, in 1liter of water. 1 ml of this solution contains
1,000 pg of potassium.
Preliminary calibration. Prepare standard solutions containing 1-5 pg/l of

potassium using the standard potassium solution and 50-ml volumetric
flasks. Add 0.5 ml of the sodium stock solution t o each of these and to a
blank. Aspirate these standards and the blank as recommended in the cal-
cium method and determine the absorbance at a wavelength of 7664.9 A.

an aliquot containing about 0.05 mg of potassium. Add 0.5 ml of the sodium
approximate potassium concentration from the preliminary calibration
readings and determine the aliquot size that will contain about 0.05 mg of
potassium.
Transfer equal aliquots containing about 0.05 mg of potassium t o three
50-ml volumetric flasks. Add no potassium standard to the first flask, 0.05
mg of the potassium standard t o the second flask, and 0.10 mg to the third.
Add 0.5 ml of the sodium stock solution to each flask and dilute t o volume.
Aspirate and record the absorbance readings for each sample.

mg K x 1,000
= mg/l K+
ml sample
Precision. I n a single laboratory using oilfield water samples containing con-

centrations of 456 and 5,680 mg K+/1, the standard deviations were *25 and
+325, respectively. The recoveries were 93.7% and 97.8%, respectively.
Magnesium (1)
Magnesium is determined at the 2852.1 A wavelength with an air-

acetylene flame.
Interferences. The silicon and aluminum suppression of the magnesium

absorption is generally removed by the addition of lanthanum or by the use
of a nitrous oxide-acetylene flame.
Reagents. The reagents are:

(1)Lanthanum solution (same as used in the calcium procedure).
(2) Standard magnesium solution: obtain commercially or dissolve 1.OOO g
of magnesium ribbon in a minimum of (1 + 1) HC1, and dilute to 1 liter
with 1%(v/v) HC1. 1ml of this solution contains 1,000 pg of magnesium per
ml and should be made up daily t o use for the standard additions.
Preliminary calibration. Prepare standard solutions containing 0.1-0.5 pg/l

of magnesium using the standard magnesium solution and 50-ml volumetric
flasks. Add t o each of these and t o a blank 5 ml of the stock lanthanum
solution. Aspirate as suggested in the calcium method and determine the
absorbance at 2852.1 A.

specific gravity of the brine often can be used as a guide in estimating the
size of an aliquot containing about 0.005 mg of magnesium. Add 5 ml of the
lanthanum stock solution, dilute t o volume with water, and aspirate. Calcu-
late the approximate sample concentration from the preliminary calibration
readings, and determine the aliquot size that will contain about 0.005 mg
magnesium.
Fig. 3.6 illustrates a plot of the concentration of magnesium found in
some oilfield brines compared to their specific gravity. This figure cannot
necessarily be applied to all oilfield brines, however, because some will
contain more and some less. The concentrations of magnesium in brines
from the same formation at about the same depth often are similar.
Transfer equal aliquots containing about 0.005 mg magnesium to three
50-ml volumetric flasks. Add no magnesium standard t o the first flask, 0.005
mg t o the second flask, and 0.010 mg t o the third. Add 5 ml of the
lanthanum stock solution t o each of the three flasks and dilute t o volume.
-L
4.000
1.00 1.05 1.10 1.15 1.20 I. 5

SPECIFIC G R A V I T Y
Fig. 3.6. Relationship of the concentration of magnesium to specific gravity for some
oilfield brines.

mgMgx 1,000
= mg/l Mg+2
rnl sample
centrations of 1,470 and 2,000 mg Mg+2/1, the standard deviations were
k36 and +128, respectively. The recoveries were 97.3%and 103.2% respec-
tively.
Calcium (1)
Calcium is determined at the 4226.7 A wavelength with an air-acetylene

flame.
Interferences. The chemical suppressions caused by silicon, aluminium, and

phosphate are controlled by adding lanthanum. The lanthanum also controls
a slight ionization interference. A pH above 7 causes low calcium values, so
dilute HC1 is added t o standards and samples. For samples containing large
amounts of silica, it often is preferable t o use the nitrous oxide-acetylene
flame. The analysis appears to be free from chemical suppressions, but a
large amount of alkali salt should be added' t o control ionization inter-
ferences.
Reagents. The reagents are:

(1) Lanthanum solution: wet 58.65 g of L a 2 0 3 with water, add 250 ml
concentrated HC1 very slowly until the material is dissolved and dilute t o 1
liter. This provides a 5% lanthanum solution in 25% (v/v) HC1.
(2) Standard calcium solution: obtain commercially or prepare by adding
50 ml of water t o 0.2497 g of primary standard calcium carbonate, CaC03.
Add dropwise a minimum volume of HC1 to dissolve all of the CaCO, and
dilute to 1 liter. 1ml of solution contains 100 pg of calcium.
Preliminary calibration. Use the standard calcium solution (1ml-100 pg Ca)

and transfer the following amounts t o six 50-ml volumetric flasks. To the
first flask add 0.5 ml, t o the second 1.0 ml, t o the third 1.5 ml, to the fourth
2.0 ml, and to the fifth 2.5 ml; and the sixth flask should have 0.0 ml. To
each flask add 5 ml of the lanthanum solution and sufficient distilled water
to adjust the volume t o 50.0 ml. The first flask now contains 1.0 pg/ml Ca,
the fifth contains 5.0 pg/ml Ca, and the sixth is a blank. Aspirate these five
standards and the blank into an air-acetylene flame and determine the ab-
sorbance at 4226.7 A. If the atomic absorption instrument has curvature
correction controls, make the necessary adjustments to obtain a linear
relationship between absorbance and the actual concentration of the
standards. If the instrument does not have these controls, plot the results on
linear graph paper as illustrated in Fig. 3.2 by substituting absorbance for
intensity.

an aliquot containing about 0.05 mg of calcium. Add 5 ml of the lanthanum
stock solution, dilute t o volume with water, aspirate the sample into an air-
acetylene flame, and determine the absorbance of 4226.7 A. Calculate the
approximate sample concentration from the preliminary calibration readings,
and determine the aliquot size that will contain 0.05 mg of calcium.
Transfer equal aliquots containing 0.05 mg Ca+2 t o three 50-ml volu-
metric flasks. Add no calcium standard t o the first flask, 0.5 mg to the
second flask, and 0.10 mg t o the third.
Add 5 ml of the lanthanum stock solution to each of the three flasks and
dilute to volume. Aspirate and record the absorbance readings for each
sample.

metric section, Fig.3.3, or Table 3.XI:
mg Ca x 1,000
= mg/l Ca+’
ml sample
centrations of 17,400 and 32,500 mg Cat’ /1, the standard deviations were
k430 and +1,090, respectively. The recoveries were 103.5% and 100.3’36,
respectively.
Magnesium (2)
The following method for the determination of magnesium in an oilfield

water was supplied through courtesy of the Halliburton Company (1970),
and can be used t o determine all concentrations of the magnesium ion in a
brine.
Reagents. The necessary reagents are magnesium standard solution, 1 mg/ml;

lanthanum solution, 1g/ml; and hydrochloric acid.
Magnesium standard working so 1ut ions
Pipet 1.0 ml of the magnesium standard stock solution into a 1liter flask,
add 11.0 ml t o a second 1-liter flask, and add 21.0 ml to a third 1-liter flask.
To each flask add 50 ml of concentrated hydrochloric acid, 10 ml of the
lanthanum stock solution, and dilute each to an overall 1,000 ml volume
with water. This yields standards of 1.0, 11.0, and 21.0 mg/l of magnesium
in the first, second, and third flasks, respectively.
Procedure. Filter the sample with the micropore filter apparatus t o remove
solids and traces of hydrocarbons from the water. Transfer, by means of
“Lambda” pipet or volumetric transfer pipet, an aliquot of sample t o con-
tain not more than 1.0 mg magnesium into a 100-ml volumetric flask. Add
5.0 ml hydrochloric acid, 1.0 ml lanthanum stock solution, and sufficient
water to dilute to exactly the 100-ml mark. Mix thoroughly. Aspirate the
5-mg/l standard through the burner, positioning the burner angle as neces-
sary until the recorder indicates a stable reading of about 25% absorption
using a wavelength setting of 2852 a. Record the reading and aspirate distil-
led water through the burner until the recorder returns t o the original base-
line. Next, aspirate the sample through the burner until a maximum stable
reading is obtained on the recorder. Record the reading and if the sample
reading on the recorder is greater than the 5-mg/l standard, aspirate the
9-mg/l standard through the burner until a maximum stable reading is ob-
tained. Record the reading and if the sample reading on the recorder is less
than the 5-mg/l standard, aspirate the 1-mg/l standard through the burner
until a maximum, stable reading is obtained, and record the reading.
Calculations:
where %A = percent absorption of high standard; %I2 = percent absorption

of low standard; %A, = percent absorption of sample; mg/ll = mg Mgt2 /1 of
high standard; mg/12 = mg Mg+2/1 of low standard; mg/l, = mg Mg+2/1 of
sample; and DF= dilution factor of sample (100/ml sample).
Derivation of above equation:
%A 1-%A 2 - %A s-%A 2
mg/l1-mg/l2 mg/l,-mg/l2
or :
when mg/l, = 11,mg/12 = 1 ; A mg/l(l-2) = 1 0

when mg/ll = 21, mg/12 = 11;A mg/l( - 2 = 1 0
mg/ll -mg/12 = 10, when standards of 21 mg/l and 11 mg/l or 11 mg/l and
1mg/l are used.
Calcium (2)
The same apparatus used in determining magnesium by atomic absorption

can be used t o determine calcium.
Reagents. The necessary reagents are calcium standard solution, 1 mg/ml;

lanthanum solution, 1g/ml; and hydrochloric acid.
Calcium standard working solutions
Pipet 1.0 ml of the calcium standard stock solution into a 1-liter flask, add
11.0 ml t o a second 1-liter flask, and 21.0 ml to a third 1-liter flask. To each
flask add 50 ml of concentrated hydrochloric acid, and 10 ml of the
lanthanum stock solution, and dilute each to an overall 1,000 ml volume
with water. This yields standards of 1.0, 11.0, and 21.0 mg/l of calcium in
the first, second, and third flasks, respectively.
Procedure. Filter the sample through the micropore filter apparatus to remove
solids and traces of hydrocarbons from the water. Transfer, by means of
micropipet or volumetric transfer pipet, an aliquot of sample containing not
more than 2.0 mg calcium into a 100-ml volumetric flask. Add 5.0 ml hy-
drochloric acid, 1.0 ml lanthanum stock solution, and sufficient water t o di-
lute to exactly the 100-ml mark and mix thoroughly. Aspirate the ll mg/l
standard through the burner, positioning the burner angle as necessary until
the recorder reaches a maximum stable reading of about 22% absorption using
a wavelength setting of 4227 A. Record the reading and aspirate distilled water
through the burner until the recorder returns t o the original baseline. Re-
move and aspirate the sample through the burner until a maximum stable
reading is obtained on the recorder. Record the reading and aspirate distilled
water through the burner until the recorder returns t o the original baseline.
If the sample reading on the recorder is greater than the 11 mg/l standard,
aspirate the 21 mg/l standard through the burner until a maximum stable
reading is obtained. Record the reading and if the sample reading on the
recorder is less than the 11 mg/l standard, aspirate the 1 mg/l standard
through the burner until a maximum, stable reading is obtained. Record the
reading.
Calculations:
(%A,--%A 2 )
10 + mg/12 x DF = mg/l Ca+2
%A1 -76 2
where %A = percent absorption of high standard; %A2 = percent absorption
of low standard; %A, = percent absorption of sample; mg/ll = mg Ca+2/lof
high standard; mg/12 = mg Ca+2/l of low standard; mg/l, = mg Ca+?/l of
sample; and DF = dilution factor of sample (100/ml sample).
Strontium
Strontium is determined at the 4607 A wavelength with an air-acetylene

flame.
Interferences. The chemical suppression caused by silicon, aluminum, and

phosphate is controlled by adding lanthanum. The lanthanum also controls
ionization interference. The nitrous oxide-acetylene flame can be used t o
control chemical interferences, but a large excess of alkali salt should be
added t o control ionization.

(1) Lanthanum solution (same as used in the calcium standard-addition
procedure).
(2) Standard strontium solution: obtain commercially or dissolve 2.415 g
of strontium nitrate, Sr(N03)2,in 1 liter of 1% (v/v) HNO,. 1 ml of the
solution contains 1,000 pg of strontium.
Preliminary calibration. Prepare standard strontium solutions containing

1-10 pg/ml of strontium using the standard strontium stock solution and 50
ml of volumetric flasks. Add to each of these and t o a blank, 5 ml of the
lanthanum stock solution. Aspirate these standards and the blank as
suggested in the calcium method and determine the absorbance of strontium
at 4607 A.

an aliquot containing about 0.1 mg of strontium (see Fig. 3.6). Add 5 ml of
the lanthanum stock solution, dilute t o volume, and aspirate. Calculate the
approximate sample concentration from the preliminary calibration readings,
and determine the aliquot size that will contain about 0.1 mg strontium.
Transfer equal aliquots containing about 0.1 mg of strontium t o three
volumetric flasks. Add no strontium standard to the first flask, 0.1 mg to the
second, and 0.2 t o the third. Add 5 ml of the lanthanum stock solution to
each of the three flasks and dilute to volume. Aspirate and record the absor-
bance readings for each sample.

mg Sr x 1,000
= mg/l Sr+2
ml sample
centrations of 840 and 2,250 mg Sr+2/1,the standard deviations were +48
and +110, respectively. The recoveries were 106.8%and 103.1%, respec-
tively.
Barium
Barium is determined at the 5336 A wavelength with an acetylene

nitrous-oxide flame.

sodium.

(1) Sodium solution: see reagents preparation under “Potassium”.
(2) Standard barium solution: obtain commercially or dissolve 1.5161 g of
BaClz in 1 liter of water. 1 ml of this solution contains 1,000 pg of barium.
Preliminary calibmtion. Prepare standard barium solutions containing 2-1 0

pg/ml of barium using the standard barium solution and 50-ml volumetric
flasks. Add t o each of these and to a blank, 0.5 ml of the sodium stock
solution. Aspirate these standards and the blank as recommended in the
calcium method and determine the absorbance at a wavelength of 5336 8.

an aliquot containing about 0.1 mg of barium. Add 0.5 ml of the sodium
approximate sample concentration from the preliminary calibration readings
and determine the aliquot size that will contain about 0.1 mg of barium.
Transfer equal aliquots containing about 0.1 mg of barium to three 50-ml
volumetric flasks. Add no barium standard t o the first flask, 0.1 mg of the
barium standard t o the second flask, and 0.2 mg to the third. Add 0.5 ml of
the sodium stock solution t o each of the three flasks and dilute to volume.

metric section, Fig. 3.3 or Table 3.XI:
mg Ba x 1,000
= mg/l Ba+’
ml sample
centrations of 7 and 8 mg Ba+’/l, the standard deviations were k0.5 and
kO.9, respectively. The recoveries were 108.2% and 97.3% respectively.
Manganese
Manganese is determined at the 2794.8 8 wavelength with an air-acetylene

flame.
Reagents. The necessary reagent is a standard manganese solution: obtain

commercially or dissolve 1.000 g of manganese in a minimum volume of (1
+1)nitric acid. Dilute t o 1 liter with 1%(v/v) HC1.l ml of solution contains
1mg of manganese.
Preliminary calibration. Prepare standard manganese solutions containing

1-5 pg/ml using the standard manganese solution and 50-ml volumetric
flasks. Aspirate these standards arid a blank as recommended in the calcium
method, and determine the absorbance at a wavelength of 2794.8 8.
Procedure. Transfer an aliquot containing about 0.05 mg of manganese to a

50-ml volumetric flask. Dilute t o volume and aspirate. Calculate the approxi-
mate sample concentration from the preliminary calibration readings and
determine the aliquot size that will contain 0.05 mg of manganese.
’
Transfer equal aliquots containing about 0.05 mg of manganese to three

50-ml volumetric flasks. Add no manganese standard t o the first flask, 0.05
mg of the manganese standard to the second flask, and 0.10 mg t o the third.
Dilute t o volume. Aspirate and record the absorbance readings for each
sample.

mg Mn x 1,000
= mg/l Mn+’
ml sample

centrations of 20 and 97 mg Mn+2/1,the standard deviations were k 1 and
+3, respectively. The recoveries were 102.2% and 105.4% respectively.
Iron
Iron is determined at the 2483.2 A wavelength with an air-acetylene

flame.
Interferences. The sensitivity is reduced if nitric acid and nickel are present.
This effect can be controlled by using a very lean (hot) flame.
Reagents. The necessary reagent is a standard solution: obtain commercially

or dissolve 1.000 g of iron wire in 50 ml of (1+ 1)nitric acid and dilute to 1
liter with water. 1 ml of solution contains 1mg of iron.
Preliminary calibration. Prepare standard iron solutions containing 1-5 Mg/ml

using standard iron solution and 50-ml volumetric flasks. Aspirate these
standards and a blank as recommended in the calcium method and determine
the absorbance at a wavelength of 2483.2 A.
Procedure. Transfer an aliquot containing about 0.05 mg of iron to a 50-ml

volumetric flask. Dilute t o volume and aspirate. Calculate the approximate
sample concentration from the preliminary calibration readings and deter-
mine the aliquot size that will contain about 0.05 mg of iron.
Add no iron standard t o the first flask, 0.05 mg of the iron standard to
the second flask, and 0.10 mg to the third. Dilute to volume. Aspirate and
record the absorbance readings for each sample.

mg Fe x 1,000
= mg/l Fe+’
ml sample
centrations of 6.3 and 6.8 mg Fe+2/1, the standard deviations were k0.5 and
k0.3, respectively. The recoveries were 115.6%and 97%,respectively.
copper
Copper is determined at the 3247.5 8 wavelength with an air-acetylene
flame.
Reagents. The necessary reagent is a standard copper solution: obtain com-

mercially or dissolve 1.000 g of copper metal in a minimum volume of (1 +
1) nitric acid. Dilute 1 liter with 1% (v/v)’ nitric acid. 1 ml of solution
contains 1 mg of copper.
Preliminary calibration. Prepare standard copper solutions containing 1-5

pg/ml using the standard copper solution and 50-ml volumetric flasks.
Aspirate these standards and a blank as recommended in the calcium method
and determine the absorbance at a wavelength of 3247.5 8.
Procedure. Transfer an aliquot containing about 0.05 mg of copper t o a

50-ml volumetric flask. Dilute t o volume and aspirate. Calculate the approxi-
mate sample concentration from the preliminary calibration readings and
determine the aliquot size that will contain 0.05 mg of copper.
Transfer equal aliquots containing about 0.05 mg of copper to three 50-ml
volumetric flasks. Add no copper t o the first flask, 0.05 mg of the copper
standard to the second flask, and 0.10 mg t o the third. Dilute t o volume.

mgCux 1,000
= mg/l CU+’
ml sample
Precision. In a single laboratory using an oilfield water sample containing a
concentration of 3 mg Cu+’ /1, the standard deviation was k0.2. The recovery
was 100.5%.
zinc
Zinc 1- determined at the 2138.6 A wavelength with an air-acetylene

flame.
Reagents. The necessary reagent is a standard zinc solution: obtain commer-

cially or dissolve 0.500 g of zinc metal in a minimum volume of ( 1 +1) HC1
and dilute to 1 liter with 1%(v/v) HCl. 1 ml of solution contains 500 pg of
zinc.
Preliminary calibration. Prepare standard zinc solutions containing 0.2-1.0

pg/ml using the standard zinc solution and 50-ml volumetric flasks. Aspirate
these standards and a blank as recommended in the calcium method and
determine the absorbance at a wavelength of 2138.6 a.
Procedure. Transfer an aliquot containing about 10 pg of zinc to a 50-ml
volumetric flask. Dilute t o volume and aspirate. Calculate the approximate
mine the aliquot size that will contain 10 pg of zinc.
Transfer equal aliquots containing about 10 pg of zinc t o three 50-ml
volumetric flasks. Add no zinc standard to the first flask, 10 pg of the zinc
standard t o the second flask, and 20 pg t o the third. Dilute t o volume.


centrations of 27 and 120 mg Zn+2/1, the standard deviations were +1. The
recoveries were 103.5%and 102.3%,respectively.
Lead (1)
Lead is determined at the 2833.1 A wavelength with an air-acetylene

flame.
Reagents. The necessary reagent is a standard lead solution: obtain commer-

cially or dissolve 1.598 g of lead nitrate, Pb(N03)2,in 1 liter of 1%(v/v)
HN03. 1 ml of solution contains 1,000 pg of lead.
Preliminary calibration. Prepare standard lead solutions containing 2-10

pg/ml using the standard lead solution and 50-ml volumetric flasks. Aspirate
these standards and a blank as recommended in the calcium method and
determine the absorbance at a wavelength of 2833.1 8.
Procedure. Transfer an aliquot containing 100 pg of lead t o a 50-ml volu-

metric flask. Dilute to volume and aspirate. Calculate the approximate
mine the aliquot size that will contain 100 pg of lead.
Transfer equal aliquots containing about 100 pg of lead t o the three 50-ml
volumetric flasks. Add no lead standard t o the first flask, 100 pg of lead
standard to the second flask, and 200 pg t o the third. Dilute t o volume.

metric section, and Fig. 3.3, or Table 3.XI:
mg Pb x 1,000
= mg/l Pb+’
ml sample
Precision. In a single laboratory using an oilfield water sample containing a
concentration of 16 mg Pb+*/l,the standard deviation was k2.6. The recovery
was 74.8%.
Lead (2)
Lead is determined by chelating with ammonium pyrollidine dithiocar-

bamate (APDC) and extracting with methyl isobutyl ketone (MIBK) (Brooks
et al., 1967). The organic extract is analyzed by means of atomic-absorption
spectrophotometry. Interferences have not been observed in the air-
acetylene flame.
Reagents. The necessary reagents are methyl isobutyl ketone (MIBK); 0.3M
hydrochloric acid; ammonium pyrollidine dithiocarbamate (APDC) (dissolve
1.0 g of APDC in 100 ml of distilled water); bromphenol blue indicator
solution (dissolve 0.1 g bromphenol blue in 100 ml of 50% ethanol); 2.5M
sodium hydroxide; and lead standard solution. The latter can be bought
commercially or made from lead nitrate. The presence of 0.5% nitric acid in
the lead standards of low concentrations retards the plating of the lead on
the sides of the container.
Procedure. Pipet the sample into a 200-ml volumetric flask and adjust the
volume to approximately 100 ml with distilled water. Add two drops of the
bromphenol blue indicator solution. Adjust the pH by adding 2.5M NaOH
by drops until a blue color persists. Add 0.3M HC1 until the blue color
disappears. Add 2.0 ml of HC1 in excess. The pH should be 2.4. Add 2.5 ml
of the APDC solution and mix. Add 10 ml of MIBK and shake vigorously for
1 minute. Allow the layer to separate and add distilled water until the
ketone layer is in the neck of the flask. Aspirate the ketone layer for lead
content. Prepare a calibration curve by adding known amounts of lead t o a
synthetic brine solution.
Calculations:
mg Pb (from curve) -
ml sample
- mg/l Pb’ *
EMISSION SPECTROMETRY a3
EMISSION SPECTROSCOPY
The basic requirements for all spectroscopic measurements are a source, a

dispersion element, and a detector. The source may be an emitter whose
emission is to be measured, or it may be a continuum that emits all wave-
lengths, within a certain range, so that absorption by material in the light
path may be measured.
In general, emission spectra are concerned with transitions from upper
state to lower state electronic levels in atoms and in simple molecular
species. Some flames are hot enough to excite upper electronic levels in
neutral atoms (un-ionized) and in molecules. Electric discharges produce
more vigorous excitation, and a high-voltage spark tends t o increase the
ionization of the emitters.
In spectrographic analysis the light source first vaporizes and dissociates
the sample and second excites the atoms causing them to radiate charac-
teristic spectra. The intensities of the spectral lines of elements excited in a
light source are proportional to the concentration of the elements in the
sample, thus providing a basis for quantitative analysis. Excitation is mainly
thermal in the sources, flames, arcs, and sparks.
Temperature is very important in spectrographic analysis because some
elements are not easily excited in a thermal source while others are. The
ionization potential of the element determines the ease of exciting its
spectra. The alkali elements with ionization potentials of 4-5 V are excited
in low energy sources while the rare gases with ionization potentials up to 25
V require high temperatures t o be excited. A Bunsen flame gives a tempera-
ture of about 1,700'C; an oxyacetylene flame, about 2,700'C; an electric
arc, 3,700'C-6,700'C; and an electric spark, about 9,700"C.In the follow-
ing procedures a plasma arc source was used, capable of temperatures up t o
7,700' C .
The plasma arc was adapted to analytical spectrography by Scribner and
Margoshes (1961).The temperature of a direct current arc is increased by
thermal-pinch effect.
The internal standard method is used in the following procedures, and
with this method the intensity of a line of the element present in unknown
concentration is measured relative to that of an invariant line of a reference
element. With this method the intensity ratio must be highly reproducible.
Barium, boron, iron, manganese, and strontium
The emission characteristics of barium, boron, iron, manganese, stron-

tium, and lanthanum in 10 solvent systems have been studied (Collins,
1967).The greatest emission enhancement was found in a mixture consisting
of 30 ml of water plus 20 ml of 35% n-amyl-alcohol and 65% acetone, as
illustrated by Fig. 3.7.Because n-propanol is easier t o work with, it w a s used
in the following procedure; however, if additional sensitivity is needed, the
n-amyl-alcohol-acetone mixture can be used.
"'i
w
2 50
90
35 percent N-Amy1
65 percent acetone
Proplonlc Add,
Acetone
N-Proponoi
a
I-
430 ~ t h y l ~ l f ~ ~ i d ~
2
I I I I I I I I I I
0 I 2 3 4 5 6 7 8 9 101I
CARBON, grams per SOml
Fig. 3.7. Relative intensity of lanthanum versus grams of carbon in the solvent aspirated
into a plasma arc.

Helium.
Eastman Kodak D-19 developer.
Eastman Kodak rapid fixer and hardener.
Stop-bath solution, e.g. 5%acetic acid.
Standard spectroscopic stock solutions containing 1 mglml of the follow-
ing metals (use spectroscopic grade reagents): (1) barium: dissolve 1.4368 g
of barium carbonate in a minimum amount of hydrochloric acid and dilute
t o 1 liter with distilled water; (2) strontium: dissolve 1.6848 g of strontium
carbonate in a minimum amount of hydrochloric acid and dilute to 1 liter
with distilled water; (3) boron: dissolve 5.7153 g of boric acid in distilled
water and dilute to 1 liter with distilled water; (4) manganese: dissolve
1.5823 g of manganese dioxide in hydrochloric acid and dilute to 1 liter
with 6N hydrochloric acid; (5) iron: dissolve 1.00 g of iron wire in aqua regia
and dilute t o 1 liter with 6N hydrochloric acid.
Internal standard solution: dissolve 2.3455 g of lanthanum oxide in a
minium amount of hydrochloric acid and dilute t o 1 liter with distilled
water. 1 ml contains 2 mg of lanthanum.
Standard solution: prepare a composite standard containing 0.025 mg/ml
of manganese, 0.075 mg/ml of iron, and 0.03 mg/ml of strontium by trans-
ferring appropriate quantities of the standard spectroscopy stock solutions
to a 1-liter volumetric flask. Dilute the resultant mixture t o volume with 6N
hydrochloric acid.
Synthetic brine solution: prepare a solution containing the following ions,
in mg/l: sodium, 32,000; calcium, 4,000; and magnesium, 2,500. Dissolve
73.766 g of sodium carbonate in hydrochloric acid, 9.989 g of calcium
carbonate in hydrochloric acid and 2,500 g of magnesium metal in
EMISSION SPECTROMETRY 85
hydrochloric acid. Evaporate these acid solutions to dryness, dissolve the

residues in distilled water, combine, and dilute t o 1liter with distilled water.
Hydrochloric acid, concentrated.
n-Propanol.
Equipment. The necessary equipment includes a spectrograph; a d.c. arc

source, 18-A minimum; a plasma arc assembly; a plate-developing machine; a
microphotometer; 50-ml volumetric flasks; 10-ml microburets; pipets; and
spectrographic plates, Eastman Kodak Type 1-N.
Spectrochemical excitation conditions. The conditions which are used to

determine barium, boron, iron, manganese, and strontium are as follows:
Source, d.c. arc.
Current, 18-25 A (keep constant).
Voltage, 220 d.c.
Pre-exposure time, 5 seconds.
Exposure time, 15 seconds.
Spectral region, 3200-5200 8,first order and second order.
Dispersion, reciprocal linear 8.21 8 / m m first order, 4.02 8 / m m second
order or better.
Plasma arc assembly.
Helium lift gas, 7 liter per minute.
Helium tangential gas, 60 liter per minute.
Atomizer, Beckman Model 4030 with medium-bore capillary.
Arc length above orifice, 7 mm.
Full arc length, anode t o cathode, 18 mm.
Portion of arc viewed, 2 mm above orifice.
Orifice electrodes; lower anode, Ultra Carbon 106 drilled to 3.97 mm in
center hole, tapered to 9.53 mm at bottom; center ring, neutral Ultra Carbon
861 drilled to 5.95 mm center hole, tapered to 9.53 mm at bottom.
Cathode electrode, vertical position, Ultra Carbon 6.35 mm graphite rod
with pointed tip.
Slit, 20-p.
Filter, 3-step.
Spectrographic plate development conditions: 5 minutes in Eastman Kodak

D-19 at 2OoC with constant agitation; 30 seconds in stop bath at 3OoC with
agitation; 5 minutes in Eastman Kodak rapid fixer and hardener solution at
2OoC with constant agitation; 30 minutes in water rinse with constant fresh
supply of water; 30 seconds rinse with distilled water; and 30 minutes in
constant air bath to dry.
Microphotometer criteria. Slit, 1 2 p wide and 0.5 mm high.

Read the background and intensity of the following lines: Ba, 11, 4554.03
8;B, I, 4995.46 8;Fe, 11, 5198.80 8;Mn, 11, 5152.20 8;Sr, 11,4215.52 8;
La, 11, 4086.62 8;and La, 11, 4429.90 8.
The background and intensity of the following lines can be read if some of
those above are too intense or if more than one line for a given element is
wanted: Ba, 11, 4934.09 8; B, I, 4993.56 8;Fe, 11, 4196.74 8; Mn, 11,
5187.46 8; La, 11, 4077.35 8; La, 11, 4123.23 8;Sr, 11, 4077.71 8;and Sr,
I, 4607.33 8.
Calibration. A preliminary curve, gamma curve, and calibration curves are

needed unless a direct-reading instrument is used. To make a preliminary
curve, record an iron spectrum using d.c. arc current excitation at about 4 A.
Read the percent transmittance (% 2') of several iron lines at 100%unfiltered
portion. (Any filter can be used as long as the 5% T is known.) Plot the 100%
unfiltered lines versus the 63.10%filtered lines. The % T of these lines should
vary from about 10%T t o 90%T to give a good preliminary curve, shown in
Fig. 3.8.
After the preliminary curve is plotted, the gamma or emulsion calibration
curve is made, as shown in Fig.3.9. There are several methods of establishing
a gamma curve.
The following example is given: 98 on x-axis set t o equal 0.2, and 96 on
the y-axis intersects curve at the same point on the curve that 98 does on the
x-axis. The filter factor is now used. In this case, it is 100%/63.10%= 1.585.
%?' Relative intensity
98 = 0.2 arbitrarily set at 0.2
96 = 0.2 x 1.585 0.317
Owwit). rotio of filter i s 1.585
100
0
FILTERED, percent
Fig. 3.8. Preliminary curve for emission spectrometry.

-
20-
z
- 40-
P
v)
-
z -
I I I 1 1 1 ,
0
R E L A T I V E INTENSITY
Fig. 3.9. Gamma or emulsion calibration curve for emission spectrometry.
A t 96 on the x-axis, find the curve intersection point on the y-axis; in this
case, it is 91.
%T Relative intensity
91 = 0.317 x 1.585 0.502
Repeat above procedure to obtain the following data:
81 = 0.502 x 1.585 0.796
63 = 0.796 x 1.585 1.262
38 = 1.262 x 1.585 2.000
19.5 = 2.000 x 1.585 3.17
9.5 = 3.17 x 1.585 5.024
4.6 = 5.024 x 1.585 7.963
2 = 7.963 x 1.585 12.621
Plot the gamma curve using the above values and plot the values on
3-cycle semilogarithmic paper. Place the 7% T values on the linear portion,
usually the x-axis, and place the relative-intensity values on the log portion.
The resultant curve should be an inverted S if the linear portion or % T is the
x-axis. (Theoretically, only one gamma curve need be plotted for all plates
with the same emulsion number.)
After the gamma curve is plotted, a calibration curve for each element
desired can be plotted, as shown in Fig. 3.10. To do this, spectra are re-
corded for various concentrations of the element in question. The % T of
each of the desired lines is determined, and these % T are referred to the
gamma curve to obtain their relative intensities. Ordinarily, internal stan-
dards are used t o permit a ratio of the relative intensity of the internal
standard line to the relative intensity of the element line to be calculated for
each concentration of the element. These ratios are plotted versus the
element concentration on 2 x 2-cycle logarithmic paper.
I I I I 1 I I
I ' 0.2 0.4 0.6 0.8 1.0 2.0 4.0 t 0
INTENSITY RATIO
Fig. 3.10. Calibration curve for emission spectrometry.
To obtain data for calibration curves for barium, boron, iron, manganese,
and strontium, use size 50-ml volumetric flasks. To one flask add no stan-
dard solution; add 1.0 ml t o the second flask; and add 2.5 ml, 5.0 ml, 7.5 ml,
and 10.0 ml of standard solution to the third, fourth, fifth, and sixth flasks,
respectively. (These aliquots will vary with the sensitivity of your instru-
ment.) Add 2 ml of concentrated hydrochloric acid, 2 ml of internal stan-
dard solution, 5 d of synthetic brine solution, 20 ml of n-propanol, and
sufficient distilled water t o adjust the final volume t o 50 ml at ambient
temperature. For optimum accuracy, prepare duplicate or triplicate samples.
Aspirate and burn the samples using the excitation conditions, the
development conditions, and the microphotometer conditions described
above; plot the curves using the above procedure.
The water sample should be adjusted t o a pH of about 1.5 at the time of
sampling t o prevent precipitation and adsorption. The sample should be
contained in a good quality plastic bottle that has been rinsed first with
dilute nitric acid and then with distilled water.
Transfer to a 50-ml volumetric flask an aliquot of the sample of sufficient
size to provide absolute quantities of the elements which will fall within the
calibration curves. The optimum aliquot size will vary from brine to brine;
however, equal-size aliquots often can be used for waters with similar
specific gravities from the same geologic formation. Add 2 ml of concen-
trated hydrochloric acid, 2 ml of internal standard solution, 5 ml of
synthetic brine solution (or try to approximate the ionic composition of the
standard), 20 ml of n-propanol, and sufficient distilled water to adjust the

volume to 50 ml at ambient temperature. For optimum accuracy, prepare
duplicate or triplicate samples.
Aspirate and excite the sample, develop the plate, and read the plate as
suggested above. Determine the relative intensity ratios for the following: Ba
4554.03/La 4429.90; B 4995.46/La 4429.90; Mn 5152.20/La 4086.72; Sr
4215.52/La 4086.72; and Fe 5198.80/La 4086.72.
Calculations. Refer the calculated ratio to the appropriate calibration curve

to determine milligrams of tested ion in the sample. Convert this value to
milligrams per liter by use of the following equation:
mg from curve x 1,000
= mg/l
ml sample
The relative intensity ratios for other line pairs can be calculated and used if
desired. The precision and accuracy of the method are approximately 2-3%
and 4-696, respectively, for strontium and barium; and 5 4 %and 10-1196,
respectively, for boron, iron, and manganese.
Beryllium
Beryllium forms a complex with acetylacetone which can be extracted

into chloroform from an aqueous solution. The chloroform extracted is
aspirated into a plasma arc, and the beryllium I1 line at 3131.07 A is read.
An apparent carbon line at 3036.3 A is used for an internal standard.
Reagents. Spectrographic plates, Eastman Kodak Type SA No. 1.

Standard beryllium stock solution: dissolve 1.00 g of fused metallic beryl-
lium (spectroscopic grade) in a small amount of 6N hydrochloric acid and
dilute t o 1 liter with 1%hydrochloric acid. 1 ml contains 1 mg of beryllium.
Standard beryllium solution: prepare a standard by transferring a suitable
aliquot of the standard stock solution t o a 1-liter volumetric flask and
diluting t o volume with 1%hydrochloric acid. The standard prepared will
depend upon the resolution and dispersion of the spectrograph. However, for
many instruments, a 0.01 pg/ml solution should be adequate.
EDTA solution: dissolve 10 g of disodium ethylenediaminetetraacetic acid
and 2 g of sodium hydroxide in water and dilute to 100 ml.
Synthetic brine solution: dissolve 80 g of sodium chloride, 30 g of calcium
chloride, 10 g of magnesium chloride, 5 g of strontium chloride, and 3 g of
potassium chloride in distilled water that is saturated with carbon dioxide
and dilute t o 1liter.
Sodium hydroxide, 0.5N.
Chloroform.
Acetylacetone.
Procedure. The spectrochemical excitations used are the same as those

shown in the procedure t o determine barium, etc., with the exception that
the spectral region is 2300-3300 8, first order and the slit is 10 p. The plate
development conditions are the same as those shown in the procedure t o
determine barium, etc., and the microphotometer conditions are the same
except that the background and the intensity of only the following lines are
read: Be, 11, 3131.07 a, internal standard line, 3036.3 8;or if the 3131.07-8
line is too intense, the Be, 11, 3130.42-a line can be used.
To prevent precipitation and adsorption, immediately acidify the clean,
oil-free sample to a pH of approximately 1.5 with concentrated hydrochloric
acid. Store the sample for transportation t o the laboratory, in a good quality
plastic bottle which previously was washed with dilute nitric acid, rinsed with
distilled water, and dried.
Transfer an aliquot of the sample estimated to contain 0.01-0.05 pg of
beryllium t o a 100-ml beaker, adjust the pH to 0.5 with concentrated
hydrochloric acid, adjust the volume to about 30-50 ml with distilled water,
boil gently for 5 minutes, and then cool. Add 2 ml of the EDTA solution
and adjust the pH of the mixture to 7.0 with 0.5N sodium hydroxide. Add 2
ml of acetylacetone, readjust the pH t o 7.0, mix thoroughly, and allow the
solution t o stand for 15 minutes. Transfer the sample to a 125-ml Teflon-
stoppered, separatory funnel and adjust the volume t o 75 ml with distilled
water, add 10 ml of chloroform, and shake the mixture vigorously for 2
minutes. After the phases separate, extract the chloroform phase and cen-
trifuge it. Aspirate the centrifuged extract into the plasma arc using the
above excitation conditions. For optimum accuracy, prepare duplicate
samples.
Develop the plates, make background corrections, and determine the rela-
tive intensity ratios for the following lines:
Be 3131.07 Be 3130.42
and
3036.3 3036.2
Determine the concentration of beryllium using a calibration curve pre-
pared by using 0.01-0.05 pg of beryllium standard. This concentration in
micrograms can be converted to milligrams per liter by this formula:
pg Be (from curve)
= mg/l Be+2
ml sample
Less than 1 ppb of beryllium can be detected with this method, the
precision and accuracy of the method are about 2%and 496, respectively, of
the amount present.
Aluminum,
Petroleum-associated water containing more than 5 mg/l of aluminum can

be analyzed using the same procedure and internal standard that are
MASS SPECTROMETRIC METHODS 91
described above for barium, boron, iron, manganese, and strontium; the
aluminum emission lines at 3082.5 acan be used. However, if the alumi-
num concentration is less than 5 mg/l, the aluminum should be separated
and concentrated from the aqueous phase. This can be done by adjusting the
pH of a sample containing up t o 100 pg of aluminum to pH 0.4 with
hydrochloric acid, adding 10 ml of a 6% aqueous solution of cupferron,
adjusting the pH t o 4.8 with sodium acetate, and extracting the aluminum
complex into chloroform. The chloroform phase then is aspirated into the
plasma arc using the same conditions and internal standard line that is
described above for beryllium.
MASS SPECTROMETRIC METHODS FOR STABLE ISOTOPES
The ratios of the stable isotopes of deuterium and hydrogen and of

oxygen-18 and oxygen-16 differ in water taken from various sources. These
differences are useful in studying the origin of a water, and of studying
paleoenvironments if the water is geologically old. The isotopic ratios are
measured on a mass spectrometer and are always compared to the ratios
found in a standard material because such a comparison proyides greater
precision than direct analysis of absolute ratios.
Deuterium
Friedman and Woodcock (1957) developed a method whereby deuterium

is converted t o hydrogen gas by reacting a 0.01-ml sample with hot uranium
metal. A mass spectrometer (Friedman, 1953) is used t o compare the
deuterium/hydrogen ratio in the emitted gas to the ratio in a standard gas.
Replicates agreeing within k0.176 usually are considered satisfactory. The
results usually are expressed as deuterium enrichments (+6 values) or deple-
tion (-6 values) relative t o SMOW (standard mean ocean water, with a D/H
ratio of 158 x (Craig, 1961b). The standard deviation is about 0.2%,
and a sample with a 6 value of -5 has 5% less deuterium than SMOW.
Oxygen-18
Epstein and Mayeda (1953) developed a method t o analyze water samples

for l 8 0 . A 10-ml sample of water is equilibrated with carbon dioxide at
25OC and an aliquot of the COz is analyzed using a mass spectrometer for
l 8 0 . The isotope ratios in the sample are compared to those in a standard
material, using the mass spectrometer, which gives a greater precision than
direct analysis of the absolute ratios. The standard generally used in SMOW
(standard mean ocean water) which is distributed by the National Bureau of
Standards (Craig, 1961a). Delta units express the isotopic data as:
where R is the isotope ratio such as 180/160 or D/H, and the delta values are
expressed in per mil like salinity, and &MOW = O%,.
COLORIMETRIC METHODS
The instrumental measurement of the absorption of radiant energy at a

certain wavelength involves spectrophotometry. The essential components of
a spectrophotometer include:
(1)Radiant energy source such as a tungsten-filament incandescent lamp
for the visible region, while hydrogen or deuterium discharge lamps usually
are used for the ultraviolet region.
(2) A monochromator, which is a device that isolates a narrow band of the
radiant energy.
(3) Containers, cells, or cuvettes usually made of glass to hold the solution
being analyzed.
(4) A detector, which is a device (usually a phototube) that measures the
radiant energy passed through the solution.
In the application of spectrophotometric analysis the two terms “trans-
mittance” and “absorbance” are important. Transmittance is:
T =-I 2
I1
where T = transmittance; II = radiant energy incident upon the first surface

of the sample; and I2 = radiant energy leaving the sample.
The term absorbance is defined as:
1
A = -1ogIJ” = lOg1,T
or the negative logarithm of the transmittance.
In the preparation of spectrophotometric curves of light-intensity ratio
plotted against concentration, it is preferable, for convenience, t o use ab-
sorbance as the basis of the plot. Under these conditions a system that
conforms t o Beer’s law gives a straight-line plot, and the commonly used
colorimetric systems that do not conform will usually show only a moderate
curvature (Willard et al., 1965). Extreme curvature, when the curve is
plotted on the basis of absorbance data, is sometimes a sign that the system
is not sufficiently stable for analytical purposes. Semicolloidal suspensions of
colored substances often give extreme curvatures. When transmittance data
are used for plotting, a curve is always obtained unless semilogarithmic
coordinates are used. The modern. spectrophotometers have an absorbance
calibration as well as the conventional “percent transmittance”, and it is
common practice t o use the absorbance scale. The relations between trans-
COLORIMETRIC METHODS 93
mittance and absorbance plots for potassium permanganate solutions at

three wavelengths are illustrated by Mellon (1950, p.95).
Several other terms for light absorption are given in the literature and are
still found on the printed scales of some photometers. “Optical density” is
often used; it is the same as absorbance.
Interferences
In spectrophotometric determinations, interferences often result from the

presence in the sample of dissolved or suspended foreign material that either
absorbs radiant energy or reacts with the color reagent to form a complex
that absorbs radiant energy. In either case, the absorbance of the sample is
decreased. Where the interference results from the formation of an absorbing
complex by ions in solution, dilution of the sample can eliminate the inter-
ference if the sensitivity of the color reagent for the element sought is
sufficiently greater than for the interfering ions. If this is not the case, other
methods must be found t o increase the selectivity of the method. Among
such methods are:
(1)pH adjustment: if pH is an important factor in complex ion formation,
its adjustment can favor the formation of the complex of the element
desired instead of the interfering ions.
(2) Masking: compounds such as EDTA (ethylenediaminetetraacetic acid)
are added t o the sample t o form a stable complex with interfering ions, thus
preventing their reaction with the color reagent.
(3) Solvent extraction: preferential solubility of some ions in organic
solvents permits the removal of interfering ions.
Another common source of interference in spectrophotometry is the use
of color reagents that absorb at the wavelength at which the complex of the
element desired is measured. Such interference usually can be reduced or
eliminated by the use of a reagent blank.
In some samples a significant source of interference results from the
presence of natural color. The natural color in water samples often gives
appreciable absorbance and requires either compensation or elimination. In
some cases it is possible t o select a spectrophotometric reagent of sufficient
sensitivity that the absorbance of the constituent sought will exceed the
absorbance of the natural color by a large factor. If this factor is 50 or higher
the error caused by the natural color is 2% or less. Knowledge of the relative
sensitivity of the constituent to be determined relative to the natural color in
the sample is necessary before such a factor can be used. If the relative
sensitivity is unknown the natural color of the sample should be com-
pensated for or removed. This can be done by determining the absorbance of
the test sample versus the blank specified for the procedure. Determine the
absorbance of the naturally colored sample versus distilled water. The differ-
ence is the corrected absorbance and is used to determine concentration
values.
Iron
The spectrochemical procedure will give values only for total iron and will
not differentiate ferrous iron from ferric iron. The following procedure can
be used t o determine F-+* and Fe+3 in a freshly sampled water (Collins et
al., 1961).
Reagents and apparatus. Standard iron solution: dissolve 1.00 g of

hydrogen-reduced iron in a minimum of hydrochloric acid and dilute to 1
liter with distilled water. This solution contains 1mg/ml of iron. Transfer 1 0
ml of this solution to a l-liter flask and dilute to volume with distilled water.
1ml of this solution contains 0.01 mg of iron.
Hydroquinone solution: dissolve 1 g of hydroquinone in 100 ml of distil-
led water.
IRON, m i l l i g r a m
Fig. 3.11. Plot of the optical density at 522 m p of the ferrous iron complex with 2,2'-
bipyridine.
O-phenanthroline or 2,2‘-bipyridine (either reagent can be used, how ever,

2,2’-bipyridine is subject t o less interferences): dissolve 0.5 g of either re-
agent in 100 ml of distilled water. The solution can be warmed to 60°C t o
effect more rapid dissolution.
Sulfuric acid, approximately 9N (441.36 g per liter): cautiously pour 270
ml of pure concentrated sulfuric acid into 650 ml of distilled water. Care-
fully mix the solution, cool, and dilute to 1liter with distilled water.
Spectrophotometer capable of measurements at 508 mp or 522 mp, glass-
electrode pH meter, 100-ml volumetric flasks, 10-ml microburet, and pipets.
Procedure. Prepare a calibration curve by transferring aliquots of the stan-

dard iron solution, containing from 0.02 mg t o 0.20 mg of iron, t o 100-ml
volumetric flasks. To separate aliquots, add 5 ml of the sodium citrate solu-
tion and determine how much sulfuric acid is necessary to adjust the pH t o
3.5. Add this amount to the aliquots in the volumetric flasks. Add reagents
in the following order: 5 ml of hydroquinone solution, 5 ml of
2,2’-bipyridine or O-phenanthroline solution, and 5 ml of sodium citrate.
The citrate must always be added last. Convert t o volume with distilled
water, mix well, and let stand for 1hour. Prepare a reagent blank in the same
manner.
Determine the absorbance at 522 mp if 2,2’-bipyridine is used or 508 mp
if O-phenanthroline is used. Plot the absorption versus iron concentration on
coordinate graph paper. The resulting curve should be linear, as shown in
Fig. 3.11.
Obtain a clean sample of brine, free of oil. Determine ferrous iron, by
following the above procedure, but omit the addition of hydroquinone. To
determine dissolved iron, filter the sample and follow the above procedure.
To determine total iron, do not filter the sample. The amount of ferric iron
can be calculated from the difference.
Calculations:
1,000 x mg iron from curve
= mg/l Fe+2 or Fe+’
sample volume
Concentrating copper, iron, lead, and nickel by ion exchange
To determine accurately, using colorimetric methods, copper, nickel, lead,

zinc, and cadmium in oilfield brines, they should be separated from inter-
fering ions. Many oilfield brines contain metals in such minute amounts that
they must be concentrated before analyses can be made. Concentration
methods investigated were ion exchange, electro-deposition, solvent extrac-
tion, and evaporation. An ion-exchange method proved t o be the most
practical for concentrating copper, nickel, and lead, because it is less time
consuming and requires less expensive equipment than any of the other
methods studied.
Acidifying the samples to pH 3.5 with acetic or hydrochloric acid

minimizes precipitation and adsorption. If acetic acid is used, 2 ml of formal-
dehyde per liter of sample should be added to retard mold growth. These
precautions will aid in obtaining representative heavy metal analyses; how-
ever, to obtain optimum results, the samples should be analyzed as quickly
after sampling as possible. If it is necessary to store the samples, they should
be stored in a cool, dark place and should not be moved frequently. Light
accelerates photochemical reactions, and high temperatures and moving
accelerate chemical reactions. Once the seal of the cap of the sample bottle
has been broken, the sample should be analyzed immediately.
A chelating ion-exchange resin such as Dowex A-1 can be used to separate
copper, iron, nickel, and lead from an aqueous solution. Slurry the resin into
a plastic column about 36 cm long and 1.7 cm in diameter. Convert the resin
to the sodium form by washing with 2 volumes of distilled water, 1 volume
being equal to the amount of resin used, followed by 2 volumes of 1 N
sodium hydroxide, and then with 10 volumes of distilled water. Because the
resin expands more than 100%when changing from the hydrogen form to
the sodium form, the column must be backwashed frequently t o reduce
compaction of the resin and to prevent shattering of the column. Pass the
brine which has been neutralized to pH 7.0 with sodium hydroxide through
the column. 2 liters or more probably will be necessary, depending upon the
amount of heavy metals present in the brine. Elute the chelated metals with
2 volumes of 2N hydrochloric acid and water effluents t o a small volume;
cool and adjust t o a predetermined volume (for example, 200 ml) with
water. Use aliquots of this solution for determining copper, iron, nickel, and
lead.
The resin must be changed back t o the sodium form as soon as the metals
have been eluted, because the resin tends to lose its chelating capacity if left
in the water-rinsed hydrogen form for longer than a few hours. If this
happens, the resin can be regenerated by heating it at 6OoC in a 30-50%
sodium hydroxide solution for 24 hours.
Once the metals are separated from the brine and concentrated, they can
be analyzed using various methods such as atomic absorption spectrometry,
flame spectrometry, emission spectrometry, or colorimetry (Collins et al.,
1962).
The compound 2,9-dimethyl-1,lo-phenanthroline, assigned the name neo-

cuproine (Diehl and Smith, 1958, p.23), has the following structure:
This reagent is used t o determine copper because of its nearly specific reac-
tion with cuprous copper. The combining ratio is 2 moles of neocuproine to
1 mole of copper. The increased selectivity of neocuproine for copper is
caused by a steric hindrance effect. The cuprous neocuproine compound is
formed over a pH range of 3-10 and is bright orange. The compound can be
extracted with n-amyl alcohol, isoamyl alcohol, n-hexyl alcohol, or chloro-
form. The maximum absorption of the compound in isoamyl alcohol occurs
at a wavelength of 454 mp.
Hydroxylamine hydrochloride can be used t o reduce the cupric ion t o
cuprous. Citrate will hold any iron present in solution when the pH is
adjusted to between 5 and 6. The chromate ion can cause low results; how-
ever, this effect does not occur when iron is present, which is almost always
the case with an oilfield brine. The anions such as sulfide, cyanide, periodate,
nitrate, t hiocyanate, and ferricyanide can interfere by reacting with
hydroxylamine; however, they are eliminated in the ion exchange separation.
Reagents. Neocuproine solution: dissolve 1 g of 2,9-dimethyl-l,1 O-phenan-

throline in 1liter of ethyl alcohol.
Hydroxylamine hydrochloride solution : dissolve 10 g of hydroxylamine
hydrochloride in 100 ml of water.
Isoamyl alcohol, analytical reagent grade.
Sodium citrate solution: dissolve 300 g of sodium citrate in 1 liter of
water, add 2 ml of the hydroxylamine hydrochloride solution, add 1 ml of
neocuproine solution, and extract with 10-ml portions of chloroform until a
colorless chloroform extract is obtained.
Standard copper solution: dissolve 0.100 g of copper in 5 ml of nitric acid
and 5 ml of water by heating gently to dissolve the copper. Add 5 ml of
perchloric acid and evaporate to fumes of perchloric acid. Cool, dilute with
water, transfer t o a l-liter volumetric flask, and dilute t o volume. Pipet a
100-ml aliquot of this solution t o another l-liter volumetric flask. Dilute to
volume with water. This solution contains 10 mg/ml of copper.
Sodium acetate.
Procedure, Add 5 ml of 10% hydroxylamine hydrochloride solution and 20

ml of 30% sodium citrate solution to a sample of effluent from the ion
exchange column containing 4-150 pg of copper, and adjust the pH of the
mixture to between 5 and 6 with 1 g or more of sodium acetate. Extract
with a 10-ml portion of isoamyl alcohol. Separate the liquids and discard the
alcohol layer. Add 10 ml of 0.1%neocuproine solution and 10 ml of isoamyl
alcohol, and shake the mixture vigorously for 1 minute. Let the phases
separate and transfer the alcohol layer to a 50-ml volumetric flask. Make
additional extractions until the alcohol layer remains colorless. Dilute the
combined alcohol extracts t o 50 ml with isoamyl alcohol, mix, and measure
the absorbance at 454 mp in a l-cm cell with a spectrophotometer.
30
-.# ANALYSIS OF OILFIELD WATERS
Calculations. Estimate the amount of copper present by using a calibration

curve prepared by using about 10-200 pg of copper:
pg Cu (from curve)
= mg/l C U + ~
ml sample
Nickel
Nickel forms a wine-red or brown compound with dimethylglyoxime

(Sandell, 1959, p.555). The structure of the chelate on the basis of available
evidence is:
H3C - C-C = CH3
//
/I /I
H3C - C C = CH,
Dimethylglyoxime gives a nearly specific reaction with nickel that has been
oxidized t o its higher valences with an oxidizing agent such as bromine. The
wine-red compound is somewhat unstable; therefore, the absorbance mea-
surements should be made within 10 minutes after formation of the nickel
dimethylglyoximate. Cobalt and copper also give colored compounds with
dimethylglyoxime, but they can be removed by washing the chloroform
extract of nickel dimethylglyoximate with dilute ammonium hydroxide.
Iron interference is removed by extracting the nickel dimethylglyoximate
with chloroform from a solution containing citrate. Palladium, platinum, and
gold also give colored compounds when nickel dimethylglyoximate is
extracted with chloroform; however, they are removed, if present, by the
ion-exchange separation.
Reagents. Dimethylglyoxime solution: dissolve 1 g of dimethylglyoxime in

100 ml of ethyl alcohol.
Saturated bromine water.
Ammonium hydroxide solution, approximately 4N: add 200 ml of con-
centrated ammonium hydroxide to 800 ml of water.
Standard nickel solution: dissolve 0.100 g of nickel in dilute nitric acid by
heating gently. Cool, dilute with water, transfer to a 1-liter volumetric flask,
and dilute to volume. Pipet a 100-ml aliquot of this stock solution into
another 1-liter volumetric flask and dilute to volume. This solution contains
10 pg/ml of nickel.
Hydrochloric acid, approximately 6N: cautiously add 500 ml of concen-
trated hydrochloric acid to 500 ml of water.
Chloroform.
Ammonium citrate solution: dissolve 1 0 g of ammonium citrate in water

and dilute to 100 ml.
Hydroxylamine hydrochloride solution: dissolve 1 0 g of hydroxylamine
hydrochloride in water and dilute to 50 ml.
Procedure. Add 10 ml of ammonium citrate solution, and 5 ml of hydroxy-

lamine hydrochloride solution to a sample of effluent containing up to 100
pg of nickel, and adjust the pH t o 8 with ammonium hydroxide. Transfer the
mixture t o a 125-ml separatory funnel, add 10 ml of dimethylglyoxime
solution and 10 ml of chloroform, and shake the mixture vigorously for 1
minute. Let the phases separate and extract the chloroform phase into
another 125-ml separatory funnel. Make additional extractions of the sample
with 10-ml portions of chloroform until a colorless chloroform extract is
obtained. Add 10 ml of 4N ammonium hydroxide solution to the combined
chloroform extracts in the 125-ml separatory funnel, and shake the mixture
vigorously for 1 minute. Let the phases separate and discard the ammonium
hydroxide phase.
Acidify the chloroform phase with 1 ml of 6N hydrochloric acid, shake
the mixture vigorously for 2 minutes, let the phases separate, and discard the
chloroform phase. Add 10 ml of chloroform t o the acid phase, shake the
mixture vigorously for 1 minute, and discard the chloroform phase. Adjust
the pH of the acid phase to 6.9, transfer it t o a 100-ml volumetric flask, add
bromine water until a yellow color persists, swirl the mixture, and allow it to
stand for 10 minutes. Add 10 ml of 4N ammonium hydroxide and 10 ml of
dimethylglyoxime solution. Swirl to mix, cool to room temperature in an ice
water bath, and adjust t o 100-ml volume with water. After 5 minutes deter-
mine the absorbance at 445 mp using a 1-cm cell and a spectrophotometer.
Calculations. Calculate the nickel concentration in the water by using a

calibration curve prepared by using about 10-100 pg of nickel:
pg Ni (from curve)
ml sample
= mg/l Ni+’
Lead
Dithizone (Sandell, 1959, p. 665) is an excellent reagent for the determi-

nation of traces of lead. Lead dithizonate probably has a formula similar to
the following:
r
C6HS
I
Lead can be extracted from a basic solution with dithizone in chloroform or

carbon tetrachloride in the presence of citrate or tartrate, which prevent the
precipitation of several metal hydroxides. The optimum pH range for extrac-
tion of the lead dithizonate with chloroform is 8.5-11. Cyanide will com-
plex all interfering metals except bismuth, thallium, and stannous tin.
Because these metals are separated by ion,exchange, their interference is
eliminated. Ferric iron can form a ferricyanide that will oxidize dithizone;
however, this reaction can be prevented by adding a reducing agent such as
hydroxylamine hydrochloride.
Excess of calcium, magnesium, and phosphorus retards the lead dithizo-
nate extraction, but thz ion exchange separation excludes phosphorus as well
as much of the calcium and magnesium. The lead dithizonate in chloroform
absorbs at 510 mp. The amount of lead in the chloroform phase should not
be much greater than 2.5 mg/l for optimum results.
Reagents. Hydroxylamine hydrochloride solution : dissolve 10 g of hydroxy-

lamine hydrochloride in water and dilute to 50 ml.
Standard lead solution: dissolve 0.100 g of lead in 10-15 ml of nitric
acid. Dilute to 1liter volume with water. Pipet a 100-ml aliquot of this stock
solution into another 1-liter volumetric flask, add 1 0 ml of nitric acid, and
dilute t o 1 liter volume with water. This solution contains 10 pg/ml of lead.
Ammonia-cyanide-sulfite solution: add 350 ml of concentrated am-
monium hydroxide, 30 ml of a 10%potassium cyanide solution, and 1.5 g of
sodium sulfite, t o a 1-liter volumetric flask and dilute to volume with water.
Dithizone solution: dissolve 0.01 g of dithizone in 200 ml of chloroform.
Chloroform.
Procedure. Transfer a sample of the ion exchange effluent containing up t o

80 pg of lead to a 125-ml separatory funnel, and add 5 ml of hydroxylamine
hydrochloride solution, 75 ml of ammonia-cyanidesulfite solution, and 10
ml of chloroform. Shake the mixture vigorously for 1minute, let the phases
separate, and discard the chloroform phase. Add 1 ml of 0.005% dithizone-
chloroform solution, shake the mixture vigorously for 1 minute, let the
phases separate, and extract the dithizone-chloroform phase into a 25-ml
volumetric flask. If the dithizone-chloroform phase is green or some color
other than cinnabar red, three possibilities exist: (1)there is no lead present;
(2) there is an oxidizing agent present; or (3) an excess of dithizone has been
used. In any event, if the dithizone-chloroform phase is not red, acidify it
with 15 ml of 1 : l O O nitric acid, shake the mixture for 1 minute t o transfer
the lead to the nitric acid phase, and discard the chloroform. Treat the nitric
acid phase with hydroxylamine hydrochloride solution, ammonia-cyanide-
sulfite solution, and make another dithizone-chloroform extraction using
0.5 ml or less of the dithizone-chloroform solution. If the dithizone-
chloroform phase still does not turn red, take a larger sample of the effluent.
However, if the original dithizone-chloroform extraction did turn red, make
additional extractions until the dithizone-chloroform phase remains green.

Dilute the combined red dithizone-chloroform phases in the 25-ml volu-
metric flask t o volume with chloroform, mix well, and determine the absor-
bance with a spectrophotometer at 510 mp.
Calculations. Prepare a calibration curve by using aliquots of the standard

lead solution containing 10-80 pg of lead:
pg Pb (from curve)
= mg/l Pb+2
ml sample
zinc
Extraction of zinc with dithizone from a weakly ammoniacal solution

containing a reducing agent and citrate prevents the precipitation of iron.
Extraction of zinc at a pH of 4.75 in the presence of sodium thiosulfate
largely eliminates interference from copper, mercury, lead, and cadmium.
The zinc dithizonate complex can be broken in 0.02N hydrochloric acid,
whereas cupric dithizonate cannot. Lead and cadmium dithizonates will dis-
sociate in 0.02N hydrochloric acid, but only traces of them should be
present after the preliminary extractions.
More accurate results are obtained by applying a zincon (Platte and
Marcy, 1959) method t o the zinc which has been isolated by the dithizone
extractions than by making another dithizone extraction of the isolated zinc
and using it for absorption measurements. Therefore, the following method
is a combination of the dithizone and zincon methods. Traces of any re-
maining interferences can be complexed.
Zinc reacts with dithizone to form a compound similar to:
Zinc reacts with zincon:

OH
1
CsHS
to form a 1:lblue complex that absorbs at a wavelength of 620 mp.

Reagents. Standard zinc solution: dissolve 1.00 g of zinc metal in hydrochloric

acid and dilute to 1 liter with water. Dilute 1 0 ml of the stock solution to
1liter to prepare a standard containing 1 0 pg/ml of zinc.
Sodium citrate solution: dissolve 1 0 g of sodium citrate in water and
dilute to 100 ml.
Hydroxylamine hydrochloride solution : dissolve 1 0 g of hydroxylamine
hydrochloride in water and dilute to 100 ml.
Buffer solution, pH 4.75: dissolve 130 g of sodium acetate and 57 ml of
glacial acetic acid in water and dilute to 1 liter.
Dithizone solution: dissolve 0.1 g of dithizone in a liter of alcohol-free
carbon tetrachloride. Extract any alcohol from the carbon tetrachloride by
shaking it with distilled water. Keep a water blanket on the extracted carbon
tetrachloride when storing it.
Potassium cyanide solution: dissolve 1.0 g of potassium cyanide in water
Buffer solution, pH 9.0: dilute 213 ml of lN sodium hydroxide to 600 ml
with water. Dissolve 37.3 g of potassium chloride and 31.0 g of boric acid in
water, mix with the sodium hydroxide, and dilute to 1liter.
Zincon solution: dissolve 0.13 g of zincon in 2 ml of 1N sodium
hydroxide and dilute to 100 ml with water.
Chloral hydrate solution: dissolve 10 g of chloral hydrate in water and
dilute to 100 ml.
Hydrochloric acid, 0.02N:add 1.7 ml of concentrated hydrochloric acid
to water and dilute to 1liter.
Ammonium hydroxide.
Sodium ascorbate.
Sodium thiosulfate solution: dissolve 25 g of sodium thiosulfate in water
Procedure. Add 1 0 ml of the hydroxylamine hydrochloride solution to an

aliquot of brine containing up to 200 pg of zinc, mix, add 1 0 ml of sodium
citrate solution, and adjust the pH to 8.3 with ammonium hydroxide.
Transfer the sample to a separatory funnel, add 3 ml of 0.01% dithizone
solution, and shake the mixture vigorously for 1 minute. Let the phases
separate and note the color of the dithizone phase. If any zinc is present, the
dithizone phase will be red or violet, but not green. If the dithizone phase is
green, take a larger aliquot of brine. If the dithizone phase is red or violet,
extract it into another separatory funnel containing 1ml of sodium thiosul-
fate solution and 1 0 ml of pH 4.75 buffer solution. Make additional
extractions of the brine solution with dithizone solution until the dithizone
remains green, which indicates that all the zinc has been extracted. This is
important because the final dithizone phase must be green, not violet.
Discard the brine solution and wash the combined dithizone extracts by
mixing them vigorously for 1minute with the buffer solution. Let the phases
separate, extract the dithizone phase into another separatory funnel con-
taining 10 ml of 0.02N hydrochloric acid, and shake this mixture vigorously

for 2 minutes. Let the phases separate, and extract and discard the carbon
tetrachloride phase. Wash the acid phase twice with carbon tetrachloride,
extract, and discard the carbon tetrachloride. Transfer the acid phase to a
50-ml volumetric flask and make t o volume with water.
Pipet 10-ml aliquots from the 50-ml volumetric flask to two 50-ml Erlen-
meyer flasks. To both flasks add 0.5 g of sodium ascorbate followed by, in
this order and with mixing, 1 ml of potassium cyanide solution, 5 ml of pH
9.0 buffer solution, and 3 ml of zincon solution. To one sample add 3 ml of
chloral hydrate solution, and t o the other (which is the reference solution)
add 3 ml of water. Within 2-5 minutes after adding the last reagent, measure
the absorbance of the sample versus the reference solution at 620 mp in
1-cm cells with a spectrophotometer.
Calculations. Prepare a calibration using aliquots of the standard zinc solu-

tion containing 10-80 pg of zinc, and use the curve t o calculate the amount
of zinc in the sample:
pg Zn (from curve)
= mg/l Zn+*
ml sample
Cadmium
Cadmium can be extracted from aqueous solutions as cadmium dithizo-

nate into carbon tetrachloride or chloroform. Cadmium dithizonate is
extracted more readily into carbon tetrachloride than is zinc dithizonate, but
zinc dithizonate is extracted more readily into chloroform than the cadmium
compounds. Therefore, because many oilfield brines contain more zinc than
cadmium, the cadmium extraction should be made with carbon tetrachloride
to insure the best possible separation from zinc.
Although citrate and tartrate do not hinder the cadmium dithizonate
extraction, they do impede the extraction of lead and zinc. Cadmium
dithizonate can be extracted from an alkaline solution containing cyanide
and tartrate; the dithizonates of nickel, copper, silver, and tin are not extrac-
ted. Most of the interference from iron can be eliminated by oxidizing it
with peroxide and filtering.
Cadmium reacts with dithizone to form a compound of the type:
104 ANALYSIS O F OILFIELD WATERS
Cadmium dithizonate in carbon tetrachloride absorbs strongly at a wave-

length of 620 mp.
Reagents. Standard cadmium solution: dissolve 0.100 g of cadmium metal in

hydrochloric acid and dilute t o 1 liter with water. Pipet a 10-ml aliquot of
this stock solution into another l-liter volumetric flask and dilute to volume.
This solution contains 1pg/ml of cadmium.
Ammonium chloride, 1N: dissolve 53.5 g of ammonium chloride in water
and dilute to 1liter.
-
Rochelle salt solution: dissolve 100 g of rochelle salt (KNaC4H406 4H20)
in water and dilute t o 1liter.
Sodium citrate solution: dissolve 100 g of sodium citrate in water and
dilute to 1 liter.
Hydrogen peroxide, reagent grade 30%hydrogen peroxide.
Tartaric acid solution: dissolve 20 g of tartaric acid in 1 liter of water.
Store in a refrigerator and discard if any mold is present.
No.1 dithizone reagent: dissolve 0.12 g of dithizone in 1 liter of carbon
tetrachloride. Store in a refrigerator in a dark bottle.
No.2 dithizone reagent: dilute 5 ml of No.1 reagent t o 100 ml with
carbon tetrachloride. Store in the refrigerator.
Hydroxylamine hydrochloride solution : dissolve 10 g of hydroxylamine
hydrochloride in 50 ml of water. Prepare fresh weekly.
Sodium hydroxide (35%)-potassium cyanide (1%) solution: dissolve 175 g
of sodium hydroxide and 0.5 g of potassium cyanide in water and dilute to 1
liter.
Ammonium hydroxide 5M: dilute 16.0 ml of concentrated ammonium
hydroxide (14.8M) to 50 ml.
Sodium hydroxide, 5% solution: dissolve 5 g of sodium hydroxide in
water and dilute to 100 ml.
Procedure. Filter the brine through Whatman No.4 filter paper (double
thickness). Transfer 900 ml or less of the filtered brine to a 2-liter beaker,
add 5 ml of 30% hydrogen peroxide, and heat until complete decomposition
of the excess hydrogen peroxide is attained. Cool the solution and filter if
any precipitate is present. Add 100 ml of ammonium chloride solution, 10
ml of rochelle salt solution, 25 ml of sodium citrate solution, and adjust the
pH to between 8 and 8.5 with 5M ammonium hydroxide. Transfer the
solution to a liter separatory funnel, add 15 ml of the No.1 dithizone solu-
tion, and shake the mixture vigorously for 5 minutes. Let the phases separate
and extract the dithizone phase into a 50-ml separatory funnel. Reextract
the brine with another 15 ml of No.1 dithizone solution. Separate the
dithizone phase into the 50-ml separatory funnel and discard the brine
phase.
Add 10 ml of tartaric acid solution to the combined dithizone extractions
in the 50-ml separatory funnel and shake the mixture vigorously for 2
minutes. Discard the carbon tetrachloride phase and wash the tartaric acid
phase twice with a 3-ml portion of carbon tetrachloride. To the tartaric acid
phase add 1 ml of hydroxylamine hydrochloride solution, 5 ml of the 35%
sodium hydroxide-1% potassium cyanide solution, and 10 ml of the No. 2
dithizone solution Shake the mixture vigorously for 1minute. Let the phases
separate and extract the dithizone phase into another 50-ml separatory
funnel. Reextract the aqueous phase with 10 ml of No.2 dithizone solution,
and add the dithizone extraction of the previous separation. Wash the
aqueous phase with 5 ml of carbon tetrachloride, extract the carbon
tetrachloride, and combine it with the two dithizone extractions. Discard the
aqueous phase. Add 15 ml of 5% sodium hydroxide solution to the com-
bined dithizone extractions, shake the mixture vigorously for 1 minute,
extract the carbon tetrachloride phase, and determine its absorbance at 620
mp in a l-cm cell with a spectrophotometer.
Calculations Run a blank and make appropriate corrections, using a cali-

bration curve prepared by using aliquots of the standard cadmium solution
containing 1-7 pg of cadmium:
pg Cd (from curve)
= mg/l Cd+2
ml sample
Phosphate
Only orthophosphate will respond to the test. Polyphosphates must be

reverted to orthophosphates by boiling with acid (American Petroleum
Institute, 1968).
Interferences. Color development in the test is inhibited when the dissolved

solids content of the sample is greater than 8% (a specific gravity greater
than 1.06) or when the total iron is greater than 50 mg/l. In such cases the
sample taken for analysis must be diluted with distilled water so that these
limits are not exceeded. Sulfide interferes by giving high results, and should
be destroyed by adding potassium permanganate solution t o the acidified
sample.
Reagents. Hydrochloric acid, concentrated.

Reagent No. 1: dissolve 46 g of ammonium molybdate [ (NH), )a Mo, 02,
4H201 in 700 ml of distilled water. The ammonium molybdate used should
consist of white crystals without a bluish-green tinge. Add 2.5 ml of concen-
trated ammonium hydroxide to the solution and dilute t o 1liter with distil-
led water.
Amino solution: dissolve 10 g (about 1 level tablespoon) of amino powder
mixture in 100 ml of distilled water. If solution remains turbid, filter. Store
solution in a well-stoppered, brown glass bottle and prepare fresh at least
every 2 weeks.
The amino powder mixture is made up by adding 5 g of sodium sulfite

and 1.0 g of l-amino-2-naphthol-4-sulfonic acid to a dry mortar. Grind the
materials to a fine powder. Transfer the powder t o a large wide-mouthed
bottle containing 66.5 g of sodium bisulfite (meta, powder, Naz Sz 0, ) and
35 g of sodium sulfite. Mix well by shaking. If the mix is not uniform, it
should be passed through a 20-mesh screen and again shaken in the large
bottle. Store mixture in a well-stoppered, wide-mouthed brown bottle.
Standard phosphate solution: dissolve in distilled water 0.1335 g of potas-
sium dihydrogen phosphate (KHz PO4 ) which has been dried in an oven at
105°C. Dilute t o 1 liter. 1ml of this solution is equivalent to 0.1 mg sodium
metaphosphate (NaP03).
Procedure. Thoroughly shake a freshly drawn sample to disperse the solids

and pipet 100-ml aliquot into each of two 250-ml beakers. If the expected
concentration of sodium metaphosphate is greater than 10 mg/l, take smaller
aliquots diluted t o 100 ml with distilled water.
Note: phosphate-free glassware must be used in this determination. The
glassware should be soaked in dilute hydrochloric acid, followed by rinsing
with distilled water.
Add 7 ml of concentrated hydrochloric acid to one of the samples. If it is
suspected t o contain sulfide, stir the solution vigorously for a minute to
remove as much of the sulfide as possible, then add potassium permanganate
solution (8%) dropwise until the solution just turns pink. Boil solution
vigorously for 30 minutes while maintaining the volume between 75 and 100
ml by adding distilled water. Cool sample t o a temperature between 70" and
95°F and dilute t o 107 ml with distilled water in a graduated cylinder
bearing a mark at the 107-ml level.
Add 7 ml of concentrated hydrochloric acid t o the unboiled sample and
treat with permanganate as above if sulfide is suspected. Filter both boiled
and unboiled samples if turbid.
Add 5 ml of reagent No.1 to both samples and mix well. Add 5 ml of
amino solution t o both and again mix well.
Ten minutes after the amino solution addition, measure the color with a
spectrophotometer at a wavelength of 690 mp, after adjusting the meter to
100%transmittance with a proper blank.
Calibration curve. Prepare a calibration curve by using aliquots of the stan-

dard phosphate solution containing up t o 10 mg/l of sodium metaphosphate.
Calculations. Refer the spectrophotometer readings t o the calibration curve

(expressed as milligrams of NaP03 versus photometer reading) t o obtain the
sodium metaphosphate concentration. The results on the heated sample cor-
respond to total phosphate, whereas, those on the unheated sample cor-
respond to orthophosphate, the difference being polyphosphate, usually ex-
pressed as sodium metaphosphate (NaP03):
where A = mg NaPO, (heated sample), and B = mg NaPO, (unheated sample).
Precision. The precision is about k3% of the amount present.
Silica
Silicon is the second most abundant element in the earth’s crust and is
found in most rocks as the oxide Si02 or as a silicate such as Mg,Si2O5
(OH), . The solubilities of silicate minerals in saline waters are a function of
temperature, pressure, pH, Eh, dissolved gases, and other ions in solution. A
limited amount of research has been done concerning silicate solubilities
(Collins, 1969) in saline solutions. Some investigators believe that most silica
exists in solution as H4Si04 (White et al., 1956);others that it exists both in
colloidal form and as H4Si04 (Krauskopf, 1956).Hydration of silica gives
the following reaction:
Si02 + 2H20 + Si(OH), or H4 Si04
A method developed by Schrink (1965)was used to study silicate solu-

bilities in saline waters (Collins, 1969)and it gave satisfactory results. It also
has been used to analyze some petroleum-associated waters. The method
involves adding 1 ml of a 4% ammonium molybdate solution in 0.75 molar
sulfuric acid solution to an appropriate aliquot of the water sample; add 15
ml of 4.5N sulfuric acid; extract for 1 minute with ethyl acetate; and deter-
mine the absorbance of the ester extract with a spectrophotometer at a
wavelength of 335 mp.
Nitrate nitrogen
Nitrate is the most highly oxidized form of nitrogen and is the most stable
form in an oxidizing environment. Many fertilizers contain nitrate, and
waters will leach the nitrate from soil or rock. Most rocks do not contain
much nitrate; therefore, it is unlikely that petroleum-associated waters con-
tain appreciable quantities of nitrate. The nitrate in deep waters also may be
depleted through anion exchange (George and Hastings, 1951).
Chloride is a serious interference in many of the methods used to deter-
mine nitrate nitrogen. Oxidizing or reducing agents such as ferric or ferrous
iron also interfere. The Brucine method (Fisher et al., 1958)can be applied
to a petroleum-associated water. To determine the nitrate concentration,
transfer an aliquot of the sample containing up to 15 pg of nitrate into a
50-ml Erlenmeyer flask, add 15 ml of water, 1 ml of Brucine reagent (2%
aqueous solution of Brucine hydrochloride), acid, place in a dark area and

allow to cool to 30°C. Determine the absorbance of the sample with a
spectrophotometer at a wavelength of 410 mp.
Arsenic
The determination of arsenic in brines has received little attention despite

its toxic relationship t o fish and animals. The arsenic content of sea water
was first investigated by Gautier (1903),who found inconsistent variations.
He attributed the higher amounts found at great depth to volcanic in-
fluences, and the higher amounts found at the surface t o evaporation and
disturbances caused by marine animals.
Rakestraw and Lutz (1933)and Gorgy et al. (1948)also studied arsenic in
sea water. They conclude that 50-60% of the arsenic is in the arsenite form,
with 8--10% each of arsenate, dissolved organic arsenic, and arsenic sus-
pended in particulate matter. Smales and Pate (1952)used an activation
analysis method t o determine submicrogram quantities of arsenic in sea
water. They found an average of 2.6 pg of arsenic per liter, with a range of
1.6-5.0 pg/l. The water analyzed is believed representative for Atlantic
Ocean water.
The Gutzeit method can be used to analyze a petroleum-associated water
for arsenic (Collins et al., 1961). Arsenic is reduced t o arsine with zinc in
acid solution. A yellow t o brown stain is produced when AsH3 passes through
paper impregnated with mercuric chloride or mercuric bromide. The color-
ation is produced by A s H ( H ~ B ~-) ~yellow, A s ( H ~ B ~-) ~ brown, and
As2Hg3 - black. By comparing unknowns with a series of standard papers
prepared with known amounts of arsenic, a quantitative estimation can be
made. Papers prepared from mercuric bromide can be preserved for several
months in a dark, dry atmosphere.
Arsenic silver diethyldithiocarbamate method
Arsine gas is liberated from arsenic compounds upon the addition of zinc
in an acid medium (Stratton and Whitehead, 1962). The arsine gas is passed
through a lead acetate scrubber and into an absorbing tube containing silver
diethyldithiocarbamate solution. The arsine and the silver diethyldithiocar-
bamate solution react forming a red color that can be measured spectro-
photometrically.
Apparatus. Arsine generator, scrubber, and absorber.

Spectrophotometer set at the following operating conditions: wavelength
- 535 mp; cells - 10 mm; phototube -blue sensitive; and slit width - 0.02
mm.
Reagents. Standard arsenious oxide solution: dissolve 1.320 g of As203 in

10 ml of 1OM NaOH and dilute to 1 liter with distilled water. 1 ml of this

solution contains 1.00 mg of A% 0 3 .Dilute this stock standard solution as
required.
Hydrochloric acid, concentrated, analytical-grade.
Lead acetate solution: dissolve 1 0 g of Pb(C2H302)2 * 3 H 2 0 in distilled
water and dilute to 100 ml.
Potassium iodide solution: dissolve 1 5 g of KI in distilled water and dilute
to 100 ml. Store in an amber colored bottle.
Silver diethyldithiocarbamate solution: dissolve 1 g of AgS[SN(C2H5 )2 ]
in 200 ml of pyridine. Store in an amber colored bottle.
Stannous chloride solution: dissolve 40 g of arsenic-free SnC12*2H20 in
1:3 HC1 and dilute to 100 ml with the same acid.
Zinc, 20 mesh, arsenic-free.
Procedure. Place a 25-ml sample, or suitable aliquot, containing less than 20

pg of arsenic in a Gutzeit generator.
Add to the flask successively, 5 ml of concentrated HCl, 2 ml of KI
solution, and eight drops of SnC12 solution. Thoroughly mix after each
addition. Allow 15 minutes for reduction of the arsenic to the tervalent
state.
Insert a plug of glass wool that has been impregnated with the lead acetate
solution into the scrubber. Assemble the generating apparatus and add 4 mi
of the silver diethyldithiocarbamate solution to the absorber. Glass beads
should be added to the absorber until the liquid just covers them.
Add 3 g of zinc to the generator and reconnect immediately. Allow 30
minutes for complete evolution of the arsine. Warm the generating flask
gently to assure complete evolution of the arsine and then pour the solution
from the absorber directly into the spectrophotometer cells. Make the deter-
minations within 30 minutes as the color developed is not permanent.
CuZcuZutions. The quantity of arsenic in the sample is determined from a plot

of absorbances of the standards:
pg As (from curve)
= mg/l As
ml sample
Fluoride
Because of interferences from large amounts of chloride present in

petroleum-associated waters, a standard addition method was developed
which is accurate in the presence of large amounts of chloride and sulfate
and is more rapid than methods requiring distillation (Collins et al., 1961).
Up to 0.01 mg of phosphate in the aliquots taken for analysis can be toler-
ated. Larger amounts of phosphate than this decolorize the zirconium
cyanine R complex completely, and distillation is necessary to remove the

phosphate.
Reagents. Eriochrome Cyanine R stock solution: dissolve 1.80 g of

Eriochrome Cyanine R in 200 ml of distilled water.
Zirconyl nitrate stock solution: dissolve 0.40 g of zirconyl nitrate dihy-
drate in 100 ml of concentrated hydrochloric acid and dilute to 200 ml.
Fluoride indicator solution: add 20.0 ml of the Eriochrome Cyanine R
solution to about 500 ml of water, stir and add 10.0 ml of the zirconyl
nitrate solution, 75 ml of concentrated hydrochloric acid, and 4 g of barium
chloride. This mixture is stable for 4 - 6 months.
Thiosemicarbazide, powdered solid.
Procedure. Measure equal amounts of brine containing less than 0.03 mg of

fluoride into each of three 50-ml volumetric flasks. Add lOpg of fluoride to
one of the flasks and add 20 pg to another. Add a few milligrams of solid
thiosemicarbazide and 25 ml of fluoride indicator solution to each 50-ml
volumetric flask. If sulfate is present, it will precipitate as barium sulfate and
must be centrifuged out of suspension. Arbitrarily adjust the transmission of
the blank (25 ml of fluoride indicator solution made to 50-ml volume with
distilled water) at 540 mp to 32% and measure the transmission of the three
solutions.
Calculations. Using coordinate graph paper, plot the transmission of the

standard-addition samples on the y-axis and their concentrations in milli-
grams of fluoride per liter on the x-axis. Multiply the sample reading at 0
concentration by 2, and from this point on the y-axis, draw a line parallel to
the x-axis until it intersects the line plotted. From this point of intersection,
draw a line parallel to the y-axis until it intersects the x-axis. This value from
the x-axis multiplied by the dilution factor equals the amount of fluoride in
milligrams per liter. Fig. 3.3 illustrates this procedure.
Iodide
A rapid, accurate method for the determination of iodide suitable for field
work utilizes the principle whereby iodide is oxidized to iodine with nitrous
acid and extracted into carbon tetrachloride. Hydrogen sulfide will interfere,
but it can be removed by acidifying the sample and boiling (Collins et al.,
1961).
Reagents. Bromphenol blue: dissolve 0.1 g of bromphenol blue in 100 ml of

distilled water.
Carbon tetrachloride.
Iodide standard solution: dissolve 1.3081 g of potassium iodide in distilled
water and dilute to 1,000ml. 1 ml contains 1 mg of iodide.
Potassium nitrite solution: dissolve 10 g of potassium nitrite in 100 ml of

distilled water.
Sulfuric acid, 9N.
Procedure. Pipet a sample containing less than 3 mg of iodide into a separa-

tory funnel, and add three drops of the bromphenol blue solution and a few
drops of 9N sulfuric acid until the indicator turns yellow. Add 10 ml of car-
bon tetrachloride and 1 ml of a 10% aqueous potassium nitrite solution, and
vigorously mix the combined phases. Extract the carbon tetrachloride phase
into a glass-stoppered cylinder. A violet color in the carbon tetrachloride
indicates iodine. Repeat the extractions with 5-ml portions of carbon
tetrachloride until all of the iodine is extracted. Dilute the combined
extracts to 25 ml with carbon tetrachloride and measure the absorbance
using a spectrophotometer at a wavelength of 517 mp. Use a calibration
curve prepared with standard iodide solutions t o determine the milligrams of
iodide in the sample.
Calculation:
mg I (from curve) x 1,000
ml sample
= mg/l r
Selenium
Selenium can be reduced t o the elemental form with sulfur dioxids

(Collins et al., 1964), hydrazine, hydroxylamine hydrochloride, hypo-
phosphorous acid, ascorbic acid, and stannous chloride. From hydrochloric
acid solutions exceeding 8N,selenium is precipitated free of tellurium when
the reducing agent is sulfur dioxide. Both selenium and tellurium are precipi-
tated by sulfur dioxide from 3 t o 5N hydrochloric acid solutions. Traces of
nitric acid should be removed before sulfur dioxide reduction. When precipi-
tating selenium, it is important that the temperature of the solution be kept
below 30°C because the volatile selenium monochloride easily can form and
be lost. A large excess of reducing agent helps to prevent loss of the mono-
chloride.
Selenium can be determined semiquantitatively by comparing the color of
the red amorphous form, or it can be adjusted to the quadrivalent form,
reacted with 3,3’-diaminobenzidine to form the monopiazselenol, and quan-
titatively determined spectrophotometrically. If sufficient selenium is
present, it also can be determined gravimetrically .
Selenate (VI) can be reduced to selenite (IV) by heating in concentrated
hydrochloric acid. Selenite is the only form that reacts with
3,3’-diaminobenzidine; the reaction is :
% N w - NH2 + H, SeO, + N= NH2 + 3Ha0
H2 N NH2 SeN
i NH2
Selenium adsorption on glassware can introduce a significant error. Much

of this adsorption can be eliminated by treating the glassware with a solution
of chlorosilane.
Reagents. Hydrobromic acid, 48%.

Selenium, stock solution: dry some selenium dioxide by placing it in a
desiccator over phosphorous pentoxide for 24 hours. Dissolve 0.141 g of the
dry selenium dioxide in water, add 80 ml of 48% hydrobromic acid, and
dilute to 1 liter with water. 1 ml of this solution contains 0.1 mg of
selenium. (Note: particles of red selenium may appear in this stock solution
after long standing as a result of reduction. When this happens, a new stock
solution must be prepared.)
Selenium solution: pipet 100 ml of the selenium stock solution into a
1-liter volumetric flask, add 80 ml of 48% hydrobromic acid, and dilute with
water. 1ml of this solution contains 0.01 mg, or 10 pg of selenium.
Sulfur dioxide selenium free.
Sulfuric acid, concentrated.
3,3'-diaminobenzidine hydrochloride: dissolve 0.25 g of 3,3'-diaminoben-
zidine hydrochloride in 50 ml of water. Prepare a fresh solution each day.
Formic acid, 2.5M: dissolve 11.5 g of formic acid in water, and dilute to
100 ml with water.
Toluene, spectro-grade.
Ammonium hydroxide: dilute 10 ml of concentrated ammonium hy-
droxide to 100 ml with water.
Barium chloride solution: dissolve 5 g of barium chloride in 100 ml of
water.
EDTA solution, 0.1M: dissolve 37.225 g of disodium ethylenediamine-
tetraacetate in water and dilute t o 1liter.
Procedure. Pipet an aliquot of brine (50 ml or less) into a 100-ml volumetric

flask and dilute to volume with concentrated hydrochloric acid. If desired,
the detection limit can be increased by first concentrating the brine by
careful evaporation after adjusting the pH t o 2 with hydrochloric acid. Mix
the solution and allow it t o stand until most of the sodium chloride precipi-
tates.
Carefully withdraw 50 ml of the supernatant clear liquor into a 150-ml
beaker and add 10 ml of concentrated hydrochloric acid. Heat the mixture
to near boiling for 10 minutes. Place the beaker in an ice-water bath beneath
an exhaust hood, let the mixture cool t o the temperature of the ice water,
and then bubble sulfur dioxide gas rapidly into the solution for about 8
minutes. If a heavy turbidity develops, filter the solution through a micro-
pore filter. Wash the precipitate with 20 ml of cold water if a 30-ml crucible
is used, or with 5 ml if a 1.5-ml crucible is used. Take care that no air is
pulled through the precipitate until the entire filtration and washing process
is complete.
Transfer the filter and precipitate back t o the 150-ml beaker and add 5 ml
of a 1:l mixture of hydrochloric acid and nitric acid. Heat the mixture to
near boiling for a few minutes, taking care not t o let the mixture boil
violently or to dryness. Examine the mixture carefully t o make sure that all
of the selenium has dissolved. Place the beaker containing the mixture in a
vacuum desiccator over anhydrous magnesium perchlorate and sodium
hydroxide and let the mixture evaporate t o dryness. Dissolve the residue in 5
ml of concentrated hydrochloric acid and heat the mixture to near boiling
for a few minutes.
Cool the mixture, add 20 ml of water, and filter it through Whatman No.4
filter paper into a 100-ml volumetric flask. Adjust the volume to 100 ml and
pipet an aliquot containing 1-100 pg of selenium (IV) from the 100-ml flask
into a 100-ml beaker. Add 5 ml of 0.1M EDTA and 2 ml of 2.5M formic
acid and adjust the pH to 1.5 with hydrochloric acid. Adjust the volume to
about 50 ml with water, add 2 ml of 3,3'-diaminobenzidine solution, mix,
and let stand for 30 minutes. Adjust the pH to 8 and transfer the solution to
a 1 2 5 4 Teflon-stoppered separatory funnel containing 10.0 ml of toluene.
Shake this mixture vigorously for 2 minutes and let the phases separate.
Extract the toluene phase, which now contains the monopiazselenol, into a
centrifuge tube. Centrifuge briefly t o clear the toluene of water droplets. If a
centrifuge is not available, the organic phase can be filtered through a dry
filter paper to which has been added 100 mg of anhydrous sodium sulfate.
Determine the absorbance of the toluene phase at 420 mp versus a reagent
blank. 1-cm cells are used; however, longer path length cells will increase
sensitivity.
Calculation. Prepare a calibration curve by plotting log I,JI, which is the

extinction or optical density of the solution versus the concentration, using
solutions containing known amounts of selenium and treated as previously
described. Estimate the amount of selenium from this curve, and calculate as
follows:
g' Se = mg/l Se-*

ml sample
Semiquantitative determination of selenium
Pipet a 20- to 50-ml aliquot of brine into a 100-ml volumetric flask and
dilute t o volume with concentrated hydrochloric acid. (To increase the
detection limit, the brine can first be concentrated by careful evaporation
after acidifying it to pH 2 with hydrochloric acid.) Mix the solution and
allow it to stand until most of the sodium chloride precipitates.
Withdraw an aliquot of the supernatant clear liquor into a small beaker,
add additional concentrated hydrochloric acid if necessary, and bubble

sulfur dioxide gas into the solution for 3 minutes. If the solution remains
clear, there is less than 25 pg of selenium present. Filter the solution through
a 5-ml micropore filter.
Compare the color in the crucible with a series of color standards com-
prising 3-20 pg of selenium. These cofar standards are prepared with known
amounts of selenium and will give the fobwing colorations (in pg of
selenium): 3 - very pale yellow; 6 - very pale orange; 10 - pale orange; 15
- orange; and 20 - red orange.
Barium
Qualitative test
This test can be used to detect barium and strontium in an oilfield brine.
It is possible to detect barium and strontium individually by using chromate
to precipitate the barium.
Transfer an aliquot of brine to a test tube, add a few millimeters of 0.5%
aqueous sodium rhodizonate solution, stopper the tube, and shake the
mixture vigorously. Barium and/or strontium is present if a bright red, a
brownish-red, or a yellow-red precipitate forms. The deeper brown indicates
barium, while the lighter yellow may indicate strontium. In any event, if a
precipitate forms, barium and/or strontium is present. A series of standards
can be prepared to help in determining the approximate amounts present.
To differentiate between barium and strontium, a few milliliters of a 10%
aqueous solution of ammonium chromate can be added to a sample brine
30-60 minutes before the sodium rhodizonate solution is added. The more
soluble strontium chromate will react with the rhodizonate while the less
soluble barium chromate will not.
GRAVIMETRIC METHODS
Gravimetric methods involve isolating a compound and determining its

weight. Their use can involve considerable time because preliminary separa-
tions often are necessary to remove interfering elements; e.g., to determine
barium as the sulfate, all strontium should be removed before the final
precipitation of the barium sulfate. One constituent present in most oilfield
waters that has resisted development of a good instrumental method of
analysis is sulfate, and perhaps the most accurate method to determine
sulfate in oilfield waters is still the gravimetric method.
Sulfate
Sulfate is precipitated as barium sulfate from an acid solution. The preci-

pitate is baked, cooled, and weighed.
GRAVIMETRIC METHODS 115
Reagents. Hydrochloric acid, concentrated.

Barium chloride solution, 10%aqueous.
Procedure. Use an aliquot that will produce no more than 100 mg of precipi-
tated barium sulfate. Dilute the aliquot t o 250 ml with distilled water and
add 1 ml of hydrochloric acid. If the sample volume itself is larger than 250
ml, add 1 ml of hydrochloric acid per 250 ml of volume. Heat t o boiling and
add an excess of hot, 10% barium chloride solution while stirring. Cover the
solution and allow it t o stand for about 4 hours at a temperature of about
85OC. Filter through a very retentive filter paper such as Munktells No.OOH
or Whatman No.42, and wash with hot water until the filtrate is chloride
free. Place the filter plus precipitate in a tared crucible, char slowly without
igniting, and bake at 800°C for 1 hour. Place the crucible in a desiccator t o
cool and then weigh.
Calculation:
Barium
Interest in the knowledge of the barium concentration in most

petroleum-associated waters has spurred the development of several types of
methods to determine barium. Perhaps the most rapid but least accurate
method is the turbidimetric method, which measures the turbidity of the
sample caused by precipitated barium sulfate after addition of excess sulfate.
The gravimetric method also measures precipitated barium sulfate or more
preferably barium chromate. It is a more time-consuming method than the
turbidimetric method, but will yield more accurate results. The double
precipitation as chromate reduces the interference from calcium and stron-
tium.
Reagents. Ammonium chromate solution: dissolve 10 g of ammonium

chromate in water, dilute t o 100 ml, and filter.
Ammonium chromate wash solution: dissolve 5 g of ammonium chromate
in water and dilute to 1 liter. Adjust the pH of this solution to 4.6 with
ammonium acetate or acetic and filter.
Ammonium acetate solution: dissolve 30 g of ammonium acetate in water,
dilute to 100 ml, and filter.
Nitric acid, 4N: cautiously add 30 ml of concentrated nitric acid t o 90 ml
of water.
Ammonium hydroxide, concentrated.
Hydrogen peroxide, 30%solution.
Procedure. Because iron will interfere if present, it should be removed by

adding a few drops of hydrogen peroxide to the sample before heating to

near boiling, adding ammonium hydroxide dropwise, and stirring until the
odor of ammonia is faint but distinct. Heat to boiling to remove excess
peroxide and flocculate any precipitate, and then filter out the iron
hydroxide.
To an aliquot of iron-free filtered water (the water should be filtered even
if iron is not specifically removed) containing less than 500 mg of barium
and strontium, add acetic acid until the pH is 4.6. Then add 1 0 ml of
ammonium chromate and 1 ml of ammonium acetate. Readjust the pH to
4.6. The final volume should be about 200 ml. Boil the mixture for 5
minutes, stirring occasionally. Remove the mixture from the heat, cool it
fairly rapidly to room temperature, and allow it to stand at room tempera-
ture for 1hour.
Filter the solution through a fine porosity filter. Wash the precipitate
from the beaker into the filter with the ammonium chromate wash solution.
Since a second precipitation is made, it is not necessary to police the beaker.
Wash the precipitate on the filter with 50 ml or more of ammonium
chromate wash solution or until calcium and strontium are absent.
Dissolve the precipitate in 3 or 4 ml of 4N nitric acid. Transfer the
dissolved precipitate back to the beaker and repeat the precipitation. The
same filter can be used, but make sure that it is acid-free.
Dry the second precipitate for 1 hour at llO°C or until it reaches a
constant weight.
Calcuhtions. Weigh the barium chromate and calculate barium as follows:

mgBaCr04 x 542
ml sample
= mg/l Ba+ *
Precision and accuracy. The precision and accuracy of this method with
optimum conditions are 1% and 2% respectively, of the amount of barium in
the sample.
OTHER METHODS
The approximate concentration of sodium in an oilfield water can be

calculated by using a knowledge of the amounts of other major cations and
anions in the sample. Likewise, the dissolved solids concentration in an
oilfield water can be calculated.
Sodium
The practice of determining sodium by calculation does not give an accu-

rate sodium value. For example, this value is calculated after determining all
of the major common anions plus two or more cations, usually calcium and
OTHER METHODS 117
magnesium. The excess of equivalents per million of anions over cations is

assumed to be sodium plus potassium, and this practice includes all the
errors of the analysis plus the undetermined ions in the combined sodium
plus potassium value. The ions are converted t o milliequivalents per liter
(me/l) by dividing each ion concentration (mg/l) by its milliequivalent
weight (mglme) to give the milliequivalents per liter for each ion determined.
After adding the milliequivalents per liter for both the anions and the
cations, the difference is multiplied by the milliequivalent weight of sodium
to give the calculated milligrams per liter of sodium.
Procedure. The calculation method is demonstrated as follows:
Anions: chloride (50,000 mg/1)/(35.5 mg/me) = 1,410 me/l

sulfate (1,290 mg/1)/(48.0 mg/me) = 27 me/l
bicarbonate (204 mg/1)/(61.0 mg/me) = 3 me/l
Total anions 1,440 me/l
Cations: calcium (5,900 mg/1)/(20.0 mg/me) = 295 me/l

magnesium (2,000 mg/1)/(12.1 mg/me) = 164 me/l
Total cations 459 me/l
Determination of sodium:
(1,440 - 459 me/l) x 23.0 mg/me = 22,600 mg/l
Dissolved solids
The dissolved solids determination can be used t o estimate the accuracy of

the resistivity determination. The specific gravity determination, and the
evaporation method can be used t o double check the calculated total dis-
solved solids. Theoretically, if all the dissolved solids are accurately deter-
mined, their sum will equal the weight of the residue left after evaporation
of the water. The dissolved solids include all the solid material in solution
which is ionized, or which is not ionized but does not include suspended
material, colloids, or gases.
The residue method involves evaporating a filtered sample t o dryness fol-
lowed by drying the residue in an oven at 180°C for 1 hour. The cooled
residue is weighed and the total dissolved solids are calculated; e.g., if 100 ml
of brine is evaporated and the residue weighs 3.0 g, then the dissolved solids
equal 30,000 mg/l. The evaporation method is subject to errors when
hygroscopic material such as calcium chloride is in the water, as is usually
the case in oilfield waters.
The calculation method simply involves adding the sum of all the analyzed
constituents as follows:
Constituent Concentration(mg/l)
Na+ 13,500
K+ 400
Li+ 10
Ca+ 2,000
Mg+ 1,200
c1- 23,500
Br- - 500
HC03 1,200
S04-2 1,200
Total dissolved solids 43,500
Spent acid
Hydrochloric acid is the oldest and most common solution used in oil-well
acidizing (Halliburton Company, 1970)..Many additives and other acids may
be used in conjunction with HC1, for example HF and HAc. Normally, 15%
HCl is used; however, other strengths are quite common. These acid solu-
tions are pumped into carbonate formations t o dissolve and remove a part of
the formation.
After reacting with carbonate rocks or being “spent” on the formation,
the solutions are returned t o the surface by various means. Often, they are
mixed with formation water, and an operator may want t o know when the
spent acid has been recovered, or if formation water or a mixture of solution
and water is being produced.
When 15% HC1 is completely spent on CaC03 or M g C 0 3 , the resulting
solution will contain 90,000 mg/l Ca or Ca equivalent. The normal formation
water contains only about 10,000 mg/l Ca or Ca equivalent. The procedure is
based on these differences.
Reagents and equipment. The necessary reagents and equipment include:

Calcium carbonate, 10-mesh.
NH4OH, reagent.
Whatman No.31 filter paper.
Plastic funnel.
150-ml beakers.
Graduated cylinder, 25 ml.
1-ml syringe or pipette, preferably plastic.
0.5% Eriochrome Black T indicator.
2N NaOH solution.
CDTA solution (disodium dihydrogen 1,2-cyclohexanediamine-tetraace-
tate): dissolve 100.0 g CDTA in 900 ml water and dilute to 1 liter. 1 ml
equals 9.0 mg Ca.
OTHER METHODS 119
Buffer solution: 67.5 g NH4C1, 570 ml NH40H made to 1 liter with

distilled water.
Procedure:
(1) Determine the pH of the returned water. If pH is below 4, the
presence of HC1 is indicated.
(2) Pour 10-15 ml of the sample into a beaker containing 10 g of
10-mesh CaC03. Bring t o a boil, remove from the hotplate, allow to settle
for about 5 minutes, and filter.
(3) Pipet 1.0 ml of the filtrate into a beaker containing 50 ml of HZO.
Heat to boiling, add 1 ml NH40H while stirring, remove from heat, let settle
for a few minutes, and filter through Whatman No.31 paper. Wash the
beaker and filter twice, using 25.0 ml H20 for each wash.
(4)Add 0.5 g Eriochrome Black T indicator t o the filtrate and 10 ml of
the buffer solution (pH should be 10). Titrate with standard CDTA solution
(1ml = 9.0 mg Ca) t o a permanent clear blue endpoint. Record the milliliters
of CDTA used. Refer t o a curve to determine the percent spent acid in the
sample.
(5) To determine a blank, take 1.0 ml of the formation water through the
procedure, starting at step 3 and determine 0% spent acid, or the blank
correction.
Curve construction
It is desirable t o construct a curve tpercent spent acid versus milliliter

CDTA) for the determination of spent acid. On rectangular graph paper, plot
Example: I m l formation water = 1.3 ml CDTA

I ml return water = 5.8 ml CDTA
p H r e f w n woter = 6.0
100 - I ml CDTA = 9.0 mg Ca
Fig. 3.12. Graph for use in calculating the amount of spent mineral acid in a water
sample.
the blank titration (formation water) as 0% spent acid. Draw a straight line
from this point through the intersection of the 100% spent acid and the
10.0-ml CDTA lines as illustrated in Fig. 3.12. This procedure corrects for
any Ca+’ or Mg+2 present in the dilution water.
In cases where it is impossible t o obtain formation water for the 0%spent
acid, a reasonable approximation can be made by titrating 100 ml of the
water used for washing and dilution. To this volume of CDTA, add 1.3 ml.
This value can then be used for the 0%spent acid point on the plot.
Free HCl
When free HC1 is indicated (pH below 4), and it is t o be determined, an

additional sample is required. Withdraw 1.0 ml of clear sample. Start with
step 3 and follow the procedure. The free HC1 is determined by the differ-
ence of the two titrations:
% free HC1= (A -B) x 1.5
where A = ml standard CDTA to titrate CaCO, treated sample, and B = ml
standard CDTA t o titrate sample.
Acetic acid solutions
Generally, acetic acid solutions are mixtures of acetic acid and HC1.
Example: 10% HCI t 5% acetic acid

I m l = 7.0 ml CDTA
I m l formation waler = 1.3 ml CDTA
I ml return water = 4 . 2 ml CDTA
I ml CDTA = 9.0 mg Ca
Return canlains 51% rpent acid
ml, CDTA
Fig. 3.13. Graph for use in calculating the amount of spent mineral and organic acid in a
water sample.
REFERENCES 121
Various proportions of each are common. The determination is complicated

by the fact that acetic acid will not completely spend itself on calcium and
magnesium carbonates. At a pH of 5-6, considerable free acetic acid is still
present in the solution and this necessitates a modification of the procedure.
In this case, it is necessary to have a representative sample, or t o prepare a
sample of the original acid mixture used on the acid job. Take 10-15 ml of
the treating acid and 10-15 ml of the returned water through the same
procedure as outlined for HC1.
Again a plot is constructed, percent spent acid versus milliliters CDTA.
Plot the milliliters CDTA used by the formation water as 0%spent acid and
the milliliters CDTA used by the injected acid sample as 100%spent acid as
illustrated in Fig.3.13. Connect these points by a straight line. From the
curve, determine the percent spent acid in the sample of returned water.
Other acid mixtures are sometimes used in oil wells. The handling of these
are usually too complicated for a rapid field determination.
References
American Petroleum Institute, 1968. API Recommended Practice for Analysis o f Oilfield
Waters. Subcommittee on Analysis of Oilfield Waters, API RP 45, 2nd ed., 49 pp.
Angino, E.E. and Billings, G.K., 1967. Atomic Absorption Spectrometry in Geology.
American Elsevier, New York, N.Y., 144 pp.
Ballinger, D.G., Booth, R.L., Midgett, M.R., Kroner, R.C., Kopp, J.F., Lichtenberg, J.J.,
Winter, J.A., Dressman, R.C., Eichelberger, J.W. and Longbottom, J.E., 1972. Hand-
book f o r Analytical Quality Control in Water and Wastewater Laboratories. National
Environmental Research Center, Cincinnati, Ohio, 107 pp.
Bogomolov, G.V., Kudelskii, A.V. and Kozlov, M.F., 1970. Ammonium as one of the
indications of oil-gas content. Dokl. Akad. Nauk S.S.S.R., 195:938-940 (in Russian).
Brooks, R.R., Presley, B.J. and Kaplan, I.R., 1967. APDC-MIBK extraction system for
the determination of trace elements in saline waters by atomic absorption spectro-
photometry. Talanta, 14:809-816.
Burriel-Marti, F. and Ramirez-Munoz, J., 1957. Flame Photometry. American Elsevier,
New York, N.Y., 531 pp.
Collins, A.G., 1962. Methods of analyzing oilfield waters: flame-spectrophotometric
determination of potassium, lithium, strontium, barium, and manganese. US. Bur.
Min. Rep. Invest., No. 6047,18 pp.
Collins, A.G., 1964. Eh and pH of oilfield waters. Prod. Monthly, 29:ll-12.
Collins, A.G., 1965. Methods of analyzing oilfield waters: cesium and rubidium. U.S.Bur.
Min. Rep. Invest., No. 6641,18 pp.
Collins, A.G., 1967. Emission spectrometric determination of barium, boron, iron,
manganese, and strontium in oilfield waters. Appl. Spectrosc., 21 :16-19.
Collins, A.G., 1969. Solubilities of some silicate minerals in saline waters. U.S. O f f .Saline
Water Res. Dev. Progr. Rep., No. 472, 27 pp.
Collins, A.G., Castagno, J.L. and Marcy, V.M., 1969. Potentiometric determination of
ammonium in oilfield brines. Environ. Sci. Technol., 3:274-275.
Collins, A.G., Waters, C.J. and Pearson, C.A., 1964. Methods of analyzing oilfield waters:
selenium and tellurium. U.S.Bur. Min. Rep. Invest., No.6474, 19 pp.
Collins, A.G., Pearson, C., Attaway, D.H. and Ebrey, T.G., 1962. Methods of analyzing
oilfield waters metallics: copper, nickel, lead, iron, manganese, zinc, and cadmium.
US.Bur. Min. Rep. Invest., No. 6087,24 pp.
Collins, A.G., Pearson, C., Attaway, D.H. and Watkins, J.W., 1961. Methods of analyzing
oilfield waters: iodide, bromide, alkalinity, acidity, borate boron, total boron, organic
boron, potassium, calcium, magnesium, iron, fluorides, and arsenic. US. Bur. Min.
Rep. Invest., No.5819, 39 pp.
Craig, H., 1961a. Isotopic variations in meteoric waters. Science, 133:1702-1703.
Craig, H., 1961b. Standards for reporting concentrations of deuterium and oxygen-18 in
natural waters. Science, 133:1833-1834.
Dean, J.A., 1960. Flame Photometry. McGraw-Hill, New York, N.Y., 354 pp.
Diehl, H. and Smith, G.F., 1958. The Copper Reagents: Cuproine, Neocuproine,
Bathocuproine. G. Frederick Smith Chemical, Columbus, Ohio, 48 pp.
Dunlap, H.F. and Hawthorne, R.R., 1951. The calculation of water resistivities from
chemical analyses. J. Pet. Technol., 7:17.
Epstein, S . and Mayeda, T., 1953. Variation of "0 content of waters from natural
sources. Geochim. Cosmochim. Acta. 4:213-224.
Fabricand, B.P., Imbimbo, E.S., Brey, M.E. and Watson, J.A., 1966. Atomic absorption
analysis of lithium, magnesium, potassium, rubidium, and strontium in ocean waters.
J. Geophys. R e s , 71:3917-3921.
Fisher, F.L., Ibert, E.R. and Beckman, H.F., 1958, Inorganic nitrate, nitrite, or nitrate-
nitrite. Anal. Chem., 30:1972-1974.
Friedman, I., 1953. Deuterium content of natural waters and other substances. Geochim.
Cosmochim Acta, 4:213-224.
Friedman, I. and Woodcock, A.H. 1957. Determination of deuterium/hydrogen ratios in
Hawaiian waters. Tellus, 9:553-556.
Furman, N.H., 1962. Standard Methods o f Chemical Analysis. D. Van Nostrand,
Princeton, N.J., 6th ed., 332 pp.
Garrels, R.M. and Christ, C.L. 1965. Solutions, Minerals, and Equilibria. Harper and Row,
Gautier, A., 1903. The arsenic content of some biologic materials. Compt. Rend., Acad.
Sci. Fr., 137:232.
George, W.O. and Hastings, W.W., 1951. Nitrate in the groundwaters of Texas. A m .
Geophys. Union Trans., 32:450-456.
Gorgy, S., Rakestraw, N.W. and Cox, D.L., 1948. Arsenic in the sea. J. Mar. Res.,
7 :2 2-41;
Halliburton Company, 1970. Chemical Research and Development. Halliburton Services,
Procedures 110.14 and 110.15, unpublished.
Herrmann, R. and Alkemade, C.T.J., 1963. Chemical Analysis by Flame Photometry.
Interscience, New York, N.Y., 644 pp.
Hodgman, C.D., Weast, R.C., Shankland, R.S. and Selby, S.M., 1962. Handbook of
Chemistry and Physics. Chemical Rubber, Cleveland, Ohio, 44th ed., 3604 pp.
Jones, P.J., 1944. Properties of water found in reservoirs, 111. Oil Gas J.,
43( 28):205-209.
Krauskopf, K.B., 1956. Dissolution and precipitation of silica a t low temperatures.
Geochirn Cosmochirn Acta, 1O:l-26.
Latimer, W.M., 1952. Oxidation Potentials. Prentice-Hall, New York, N.Y., 2nd ed., 392
PP.
Marsh, G.A., 1951. Portable dissolved oxygen meter for use with oilfield brines. Anal.
Chern, 23:1427.
Mellon, M.G., 1950. Analytical Absorption Spectroscopy. John Wiley and Sons, New
York, N.Y., 618 pp.
Mellon, M.G.,. 1956. Quantitative Analyses. Thomas F. Crowell, New York, N.Y., 694 pp.
Platte, J.A. and Marcy, V.M., 1959. Photometric determination of zinc with zincon:
application to water containing heavy metals. Anal. Chem., 31 :1226-1228.
Potter, E.C., 1956. Electrochemistry. MacMillan, New York, N.Y., 418 pp.
REFERENCES 123
Pourbaix, M.J., 1949. Thermodynamics of Dilute Aqueous Solutions. Edward Arnold,

London, 136 pp.
Rainwater, F.H. and Thatcher, L.L., 1960. Methods for collection and analysis of water
samples. U S . Geol. Surv. Water Supply Paper, No.1454, p.70.
Rakestraw, N.W. and Lutz, F.B. 1933. Determination of arsenic in sea water. Biol. Bull.,
65:397.
Ramirez-Munoz, J., 1968. Atomic Absorption Spectroscopy and Analysis by Atomic
Absorption Flame Photometry. American Elsevier, New York, N.Y., 315 pp.
Robinson, J.W., 1966. Atomic Absorption Spectroscopy. Marcel Dekker, New York,
N.Y., 204 pp.
Rosin, J., 1955. Reagent Chemicals and Standards. D. Van Nostrand, New York, N.Y.,
561 pp.
Sandell, E.B., 1959. Colorimetric Determination of Traces o f Metals. Interscience, New
York, N.Y., 1032 pp.
Schrink, D.R., 1965. Determination of silica in sea water using solvent extraction. Anal.
Chem., 37:764-765.
Scribner, B.F. and M. Margoshes, 1961. Excitation of solutions in a gas-stabilized arc
source. Natl. Bur. Standards Rep., No.7342, 8 pp.
Smales, A.A. and Pate, B.D., 1952. The determination of sub-microgram quantities of
arsenic by radioactivation, 11. The determination of arsenic in sea water. Analyst,
7 7 :188-195.
Stratton, G. and Whitehead, H.C., 1962. Colorimetric determination of arsenic in water
with silver diethyldithiocarbamate. J. A m . Water Works Assoc., 54:861-863.
Watkins, J.W., 1954. Analytical methods of testing waters to be injected into subsurface
oil-productive strata. U.S. Bur. Min. Rep. Invest., No.5031, 29 pp.
Welcher, F.J., 1957. The Analytical Uses o f Ethylenediaminetetraacetic Acid. D. Van
Nostrand, Princeton, N.J., 356 pp.
White, D.E., Brannock, W.W. and Murata, K.J., 1956. Silica in hot-spring waters. Geo-
chim. Cosmochim. Acta, 10:27-59.
Willard, H.H., Merritt, Jr., L.L. and Dean, J.A., 1965. Instrumental Methods of Analysis.
D. Van Nostrand Co., Princeton, N.J., 4th ed., 250 pp.
Wyllie, M.R.J., 1963. The Fundamentals of Well Log Interpretation. Academic Press, New
York, N.Y., 3rd ed., 238 pp.
Zobell, C.E., 1946. Studies on redox potential of marine sediments. Bull. A m . Assoc. Pet.
Geol., 30:477-513.
Chapter 4. INTERPRETATION OF CHEMICAL ANALYSES OF OIL-
FIELD WATERS
Water analyses may be used to identify the source of a water. In the

oilfield one of the prime uses of these analyses is to determine the source of
extraneous water in an oil well, so that casing can be set and cemented to
prevent such water from flooding the oil or gas horizons. In some wells a
leak may develop in the casing or cement, and water analyses are used to
identify the water-bearing horizon so that the leaking area can be repaired.
With the present emphasis on water pollution prevention, it is very impor-
tant to locate the source of a polluting brine, so that remedial action can be
taken.
Comparisons of water-analysis data are tedious and time-consuming; there-
fore, graphical methods are commonly used for positive, rapid identification.
A number of systems have been developed, all of which have some merit.
Calculating probable compounds
The hypothetical combinations of dissolved constituents found in waters

are commonly calculated by combining the positive and negative radicals in
the following order:
calcium bicarbonate
magnesium sulfate
sodium chloride
potassium nitrate
Calcium is combined with bicarbonate, and if more calcium is available
than that consumed by bicarbonate, it is combined with sulfate, chloride,
and nitrate until exhausted. Conversely, any excess bicarbonate is combined
with magnesium, sodium, and potassium until consumed. Other radicals can
and should be added for most petroleum reservoir waters. These include
lithium, strontium, barium, iron, borate, phosphate, bromide, and iodide.
They can be grouped in the appropriate column and then in the calculations
each positive and negative radical is totally combined, the next following
radical is combined until both the cations and anions are exhausted. If the
analysis is correct, the cations and anions will be present in approximately
equivalent amounts.
To calculate the hypothetical combinations, the reacting values of the
positive and negative radicals or ions are calculated as follows: reacting
126 INTERPRETATION OF CHEMICAL ANALYSES
TABLE 4.1
Reaction coefficients
Cation Anion
Calcium 0.0499 bicarbonate 0.0164

Magnesium 0.0823 sulfate 0.0208
Iron 0.0358 chloride 0.0282
Sodium 0.0435
TABLE 4.11
Reacting values (RV)
Cation (mg/l) RV Anion (mg/l) RV
Ca 4,000x 0.0499 = 199.6 HC03 500 x 0.0164 = 8.2

Mg 3,000x 0.0823 = 246.8 so4 200 x 0.0208 = 4.2
Fe 100 x 0.0358 = 3.6 C1 30,000 x 0.0282 = 846.3
Na 9,400 x 0.0435 = 408.9
858.9 858.7
TABLE 4.111
Reacting value distribution
Ca as calcium bicarbonate 8.2

Ca as calcium sulfate 4.2
Ca as calcium chloride 187.2
Mg as magnesium chloride 246.8
Fe as iron chloride 3.6
Na as sodium chloride 408.9
858.9
values (RV) or equivalents per million (epm) = mg/l of ion x valence of ion/
molecular weight of ion.
The term valence of ion/molecular weight of ion is called “reaction coeffi-
cient” and the positive and negative ions have values as shown in Table 4.1.
Table 4.11 indicates how the results of a water analysis are converted t o
reacting values.
The reacting values are a measure of the cations and anions dissolved in
the water. The 4,000 mg/l of calcium with a reacting value of 199.6 can
combine with all the bicarbonate, all the sulfate, and 187.2 epm of the
chloride. Magnesium will combine with 246.8,iron with 3.6, and sodium
with 408.9 epm of chloride. Thus the reacting values can be considered to be
distributed as shown in Table 4.111.
DETERMINING A SOUGHT COMPOUND 127
TABLE 4.IV
Combination factors
Reaction values given Compound sought Combination factor
Ca o r C 0 3 CaC03 50.1
Ca or SO4 CaS04 68.1
Ca or C1 CaClz 55.5
Mg or C 0 3 MgCO3 42.2
Mg or SO4 MgS04 60.1
Mg or C1 MgClz 47.6
Fe or C03 FeC03 57.8
Fe o r S 0 4 FeS04 76.0
Fe orC1 Fa12 63.4
Na or C03 Na~C03 53.1
Na or SO4 Naz SO4 71.0
Na or C1 NaCl 58.4
TABLE 4.V
Hypothetical combinations
Ca(HC03 ) to CaC03 8.2 x 50.1 = 411 CaC03*

CaS04 4 . 2 68.1
~ = 286CaS04
CaC12 187.2 x 55.5 = 10,390 CaC12
MgCh 246.8 x 47.6 = 11,748 MgClz
FeClz 3.6 x 63.4 = 228 FeCIz
NaCl 858.9 x 58.4 = 50,160 NaCl
*In mg/l.
Determining a sought compound
It is necessary to multiply the reacting value by a combination factor to

determine a hypothetical compound. This factor is necessary to convert the
reported radical into the desired compound. For example, the factor for
converting Ca to CaCO, is 2.50 and the reaction coefficient for Ca is 0.0499.
Therefore, the combination factor to convert the reacting value for Ca to
CaCO, is 2.50 + 0.0499 = 50.1. Table 4.IV illustrates some combination
factors.
The combination factors given in Table 4.IV can be used to calculate the
hypothetical combinations shown in Table 4.V, using the analysis shown in
Table 4.111.
Graphic plots
Graphic plots of the reacting values can be made to illustrate the relative
amount of each radical present. The graphical presentation is an aid t o rapid
identification of a water, and classification as t o its type, and there are
several methods that have been developed.
Tickell diagram
The Tickell (1921) diagram was developed using a 6-axis system or star
diagram. Percentage reaction values of the ions are plotted on the axes. The
percentage values are calculated by summing the epm’s of all the ions,
dividing the epm of a given ion by the sum of the total epm’s, and multi-
plying by 100.
Na Ca+Mg Na Ca+Mg
2-
ci \ So4
(a) CI RV=49.92% (b) h 9 2 ma / I i tar
Fig. 4.1. Tickell (a) and modified Tickell (b) diagram for Gulf Coast water, sample No.1.
Na Ca+Mg Ca + Mg
s
(a) CI ~ v = 4 9 . 2 9 % (b) 1i07 ma/ lltar
Fig. 4.2. Tickell (a) and modified Tickell (b) diagram for Anadarko Basin water, sample
No. 2.
GRAPHIC PLOTS 129
so4 so4
(a) C I RV=49.92 % (b) 5,708 m e / l i t e r
Fig. 4.3. Tickell (a) and modified Tickell (b) diagram for Williston Basin water, sample
N0.3.
No Ca+Ma Co+Mg
c i\ so4
(b) 1.769 me / liter
Fig. 4.4. Tickell (a) and modified Tickell (b) diagram for Gulf Coast and Anadarko Basin
waters, mixed 1:l.
Na Co+Mg Na Ca+Mg
CI so4 (b) C I‘7 $.

2870 me / i i t r r
Fig. 4.6. Tickell (a) and modified Tickell (b) diagram for Gulf Coast, Williston, and
Anadarko Basin waters, mixed 1 :1 :1.
Fig. 4.1.illustrates the Tickell diagram using reaction values in percentage

in the diagram on the left, and total reaction values in the diagram on the
right. The plots of total reaction values, rather than of percentage reaction
values, are often more useful in water identification because the percentage
values do not take into account the actual 'ion concentrations. Water dif-
fering only in concentrations of dissolved constituents cannot be distin-
guished.
To illustrate differences in patterns for different waters, Fig. 4.1-5 were
prepared using the Tickell method. Fig. 4.1 represents a water from the Gulf
Coast Basin, taken from the Wilcox formation of Eocene age. Fig. 4.2 is of a
sample from the Mer?.mec formation of Mississipian age in the Anadarko
Basin. Fig. 4.3 is of sample from a Devonian age formation in the Williston
Basin. Fig. 4.4 represents a 1:l mixture of waters of the Gulf Coast and
Anadarko Basins, and Fig. 4.5 is a 1:1:1 mixture of all three waters.
REISTLE SYSTEM
Fig. 4.6. Water-analysis interpretation, Reistle system - sample numbers correspond to

the samples of Fig. 4.1-3.
GRAPHIC PLOTS 131
Reistle diagram
Reistle (1927) devised a method of plotting water analyses using the ion
concentrations as shown in Fig. 4.6. The data are plotted on a vertical
diagram, with the cations plotted above the central zero line and the anions
below. This type of diagram often is useful in making regional correlations or
studying lateral variations in the water of a single formation, because several
analyses can be plotted on a large sheet of paper.
St iff diagra m
Stiff (1951) plotted the reaction values of the ions on a system of rectan-
gular coordinates as illustrated in Fig. 4.7. The cations are plotted to the left
and the anions to the right of a vertical zero line. The end points then are
connected by straight lines to form a closed diagram, sometimes called a
“butterfly” diagram. To emphasize a constituent that may be a key t o
interpretation, the scales may be varied by changing the denominator of the
Fig. 4.7. Water-analysis interpretation, Stiff method - sample numbers correspond to the
samples of Fig. 4.1-3.
ion fraction usually in multiples of 10. However, when looking at a group of

waters all must be plotted on the same scale.
Many investigators believe that this is the best method of comparing
oilfield water analyses. The method is simple, and nontechnical personnel
can be easily trained t o construct the diagrams.
Other methods
Several other water identification diagrams have been developed, primarily

for use with fresh waters, and they will not be discussed here. The Piper
(1953)diagram and the Stiff (1951)diagram were adapted to automatic data
processing by Morgan et al. (1966),and Morgan and McNellis (1969).The
Piper (1953)diagram uses a multiple trilinear plot t o depict the water analy-
sis, and this quaternary diagram shows the chemical composition of the
water in terms of cations and anions. Angino and Morgan (1966)applied the
automated Stiff and Piper diagrams to some oilfield brines and obtained
good results.
References
Angino, E.E. and Morgan, C.O., 1966. Application of pattern analysis t o the classification
of oilfield brines. Kans. State Geol. Sum.,Comput. Contrib., No.7, pp.53-56.
Morgan, C.O. and McNellis, J.M., 1969. Stiff diagrams of water-quality data programmed
for the digital computer. Kuns. State Geol. Sum., Spec. Distrib. Publ., No.43, 27 pp.
Morgan, C.O., Dingman, R.J. and McNellis, J.M., 1966. Digital computer methods for
water-quality data. Ground Water, 4:35-42.
Piper, A.M., 1953. A graphic procedure in the geochemical interpretation of water analy-
ses. US.Geol. Surv. Ground Water Note, No.12, 1 4 pp.
Reistle, C.E., 1927. Identification of oilfield waters by chemical analysis. U.S.Bur. Min.
Tech. Paper, No.404, 25 pp.
Stiff, H.A., 1951. The interpretation of chemical water analysis by means of patterns. J.
Pet. Technol., 3:15-17.
Tickell, F.G., 1921. A method for graphical interpretation of water analysis. Calif. State
Oil Gas Superv., 6:5-11.
Chapter 5. SIGNIFICANCE OF SOME INORGANIC CONSTITUENTS
AND PHYSICAL PROPERTIES OF OILFIELD WATERS
In general, the concentrations of the constituents in various natural solids

of reservoir rocks must be considered along with the amounts that are found
in associated oilfield waters. Some possible chemical reactions between host
rock and reservoir water may deplete or enrich the concentration of the
constituents in oilfield waters. Another important factor is the solubility of a
constituent.
The ionic potential, determined by dividing the ionic radius by the va-
lence, influences the solubility of elements. For example, elements with low
ionic potential are more likely t o remain in true ionic solution. Elements
commonly found in oilfield waters have the following ionic potentials:
sodium, 0.95; calcium, 0.50; magnesium, 0.33; chlorine, 1.81; bromine,
1.95; and iodine, 2.16. Apparently the cation (magnesium) and the anion
(chlorine) would be the most likely to remain in true ionic solution; how-
ever, several other variables occur during diagenesis which lead to depletion
or enrichment of constituents in waters.
Lithium
Lithium is the lightest alkali metal; it has a distinctly smaller radius,

0.60 8,than the other alkalies and is the smallest of all singly charged
cations. It is one of the less abundant elements, and its abundance in the
earth’s crust is about 6.5 x wt.% (Fleischer, 1962). Here again, it is an
exception because in general, the lighter elements tend to be more abundant
than the heavier elements. It is lithophilic in that it tends t o be associated
with the silicate phase in rocks (Ahrens, 1965); however, because of its small
size, it supposedly cannot replace the abundant alkali metals in mica.
It and the other alkali metals exist in a uniform positive one state of
oxidation and are inherently ionic. Their chemical behavior depends almost
entirely upon electron loss, and their chemistry is simpler than that of any of
the other metallic elements (Moeller, 1954).
Lithium is potentially toxic to plants (Hem, 1970), yet it is regularly
found in plant ashes, which indicates that it normally is present in soil waters
(Goldschmidt, 1958). Coal ashes of Neurode, Silesia, contained up to 198
ppm lithium, whereas soils in northeast Scotland contain 30-5,000 ppm.
The content of lithium in sediments ranges up to 6 ppm in quartzites and
sandstones, up to 15 ppm in calcareous rocks, and up t o 120 ppm in clays
and shales.
134 INORGANIC CONSTITUENTS AND PHYSICAL PROPERTIES
TABLE 5.1
Properties of the alkali metals
Property Lithium Sodium Potassium Rubidium Cesium
Atomic number 3 11 19 37 55
Nonhydrated
radius (A) 0.60 0.95 1.33 1.48 1.69
Hydrated
radius (A) 3.82 3.58 3.31 - -
Outer
electronic
configuration 1s' 2s' 2s22p6 3s' 3s' 3p6 4s' 4s24p6 5s' 5s' 5p6 6s'
Atomic weight 6.939 22.990 39.102 85.47 132.905
Ionization
potential (V) 5.390 5.138 4.339 4.176 3.893
TABLE 5.11
Five relative concentration changes of some dissolved ions during evaporation of sea water
and brine*
Constituents Concentrations (mg/l)
Sea water CaSO4 NaCl MgS04 KCI MgC12
Lithium 0.2 2 11 12 27 34
Sodium 11,000 98,000 140,000 70,000 13,000 12,000
Potassium 350 3,600 23,000 37,000 26,000 1,200
Rubidium 0.1 1 6 8 14 10
Magnesium 1,300 13,000 74,000 80,000 130,000 153,000
Calcium 400 1,700 100 10 0 0
Strontium 7 60 10 1 0 0
Boron 5 40 300 310 750 850
Chloride 19,000 178,000 275,000 277,000 360,000 425,000
Bromide 65 600 4,000 4,300 8,600 10,000
Iodide 0.05 2 5 7 8 8
*Approximate mg/l. Columns headed sea water, CaS04, etc., represent stages in sea water
evaporation. For example, sea water contains 0.2 mg/l of lithium, after calcium sulfate
has precipitated the residual brine contains about 2 mg/l of lithium, after sodium chloride
has precipitated the residual brine contains about 11 mg/l of lithium, the residual brine
contains about 12 mg/l of lithium after magnesium sulfate precipitates, 27 mg/l of lithium
after potassium chloride precipitates, and 34 mg/l of lithium after magnesium chloride
precipitates.
LITHIUM 135
The hydrated radius of lithium is 3.82 a,

as shown in Table 5.1 (Moeller,
1954). The ionic potential is 0.60, and the polarization is 1.67. The polariza-
tion is quite high and is a measure of its replacing power in an exchange
system. Apparently it can replace strontium, calcium, and magnesium since
their polarizations are 1.77, 2.02, and 3.08, respectively.
Some surface waters of the volcanic sodium chloride type are enriched in
lithium (White, 1957). Lithium from Searles Lake brine is recovered as
Li2NaP04 (Brasted, 1957). The content of lithium in oilfield waters is
usually less than 10 mg/l but in some Smackover formation waters from east
Texas, concentrations up t o 500 mg/l are present. When a brine containing
lithium goes through an evaporite sequence, lithium is one of the elements
whose concentration does not decrease, as illustrated in Table 5.11, in the
liquid phase as various minerals precipitate (Collins, 1970). Fig. 5.1 illus-
trates the enrichment of lithium as compared t o an evaporite sequence in
some subsurface brines from Tertiary, Cretaceous, and Jurassic age sedi-
ments. Fig. 5.2 illustrates a similar enrichment for some brines taken from
Pennsylvanian and Mississippian age sediments (Collins, 1969a). Possibly
lithium was liberated and potassium was depleted by exchange reactions
with clay minerals, degradation of lithium containing minerals, or simply a
leaching of minerals, primarily silicates, which contain lithium. Lithium sub-
stitutes in the structure of several common minerals and forms few minerals
of its own. If the minerals in which it has substituted should degrade or
break down with depth, the lithium might be resolubilized, thus increasing
its concentration in the aqueous phase. White et al. (1963) postulated that
because the lithium concentration in magmatic waters is related to volcanic
LITHIUM, mgll
Fig. 5.1. Comparison of the lithium concentrations in some Tertiary (T),Cretaceous (C),
and Jurassic (J) age formation waters from Louisiana with an evaporating sea water.
u 10 20 3
LITHIUM, mg/l
Fig. 5.2. Comparison of the lithium concentrations in some Mississippian (M) and
Pennsylvanian (P) age formation waters from Oklahoma with an evaporating sea water.
emanations, the increase in the lithium content of deeper waters might be

related to the same cause.
Sodium
The most abundant member of the alkali-metal group is sodium, ranking

number 6 with respect t o all the metallic elements. The radius of the sodium
ion is 0.95 A, and its geochemistry is controlled to some extent by calcium
because of the similarity of their ionic radii. Its abundance in the earth's
crust is about 2.8 wt.% (Fleischer, 1962). Table 5.1 shows that its outer
electronic configuration is 2s' 2p6 3s' , with a first ionization potential of
5.138 V, indicating that its single outer electron is less firmly held than in
the lithium atom with a first ionization potential of 5.390 V. The ionization
potential is a measure of the chemical reactivity - the lower the potential,
the greater the reactivity. Table 5.1 (Moeller, 1954) also illustrates some qf
its other properties.
According t o Ahrens (1965),sodium is lithophilic, and many distinctly
lithophile elements have valence electrons outside a closed shell of eight
electrons. The ionic radius decreases as the charge on the cation increases.
Sodium does readily participate in solid solution relationships because its
radius is small, making replacement of cations with 30% larger radii difficult.
The amounts of sodium in argillaceous sediments and marine shales are
about 1,000ppm and 1,300ppm, respectively (Goldschmidt, 1958).
SODIUM 137
Sodium in solution tends to stay in solution; it does not readily precipi-

tate with an anion, and it is less easily adsorbed by clay minerals than are
cesium, rubidium, potassium, lithium, barium, and magnesium. The major
source of sodium in sea water can be attributed t o the weathering of rocks.
Some sodium probably was derived through volcanic activity. The ocean and
evaporite sediments contain the bulk of the sodium. Igneous rocks contain
appreciably more sodium than sedimentary rocks with the exception of
evaporites.
Sea water contains about 11,000 mg/l of sodium, as illustrated in Table
5.11. The concentration of sodium increases in brine as it evaporates, t o
about 140,000 mg/l, when halite precipitates. Most oilfield waters contain
more sodium than any other cation, and most oilfield waters are believed to
be of marine origin. Fig.5.3 is a log-log plot of the chloride concentration
versus sodium of some subsurface brines taken from sediments of Tertiary,
Cretaceous, and Jurassic age. The straight line is a plot of chloride versus
sodium concentrations for some evaporite waters, and indicates the enrich-
ment of sodium ions until halite (NaC1) precipitates - at a chloride concen-
tration of about 140,000 mg/l (compared t o that of normal sea water,
19,000 mg/l). The plot of the concentrations of sodium versus chloride for
these subsurface brines falls very near the normal evaporite curve, indicating
that the concentration mechanism may be related to an evaporite process
(Collins, 1970). Fig. 5.4 is a similar plot for some subsurface brines taken from
sediments of Pennsylvanian and Mississippian age (Collins, 1969a). Several of
these samples are somewhat depleted in sodium which indicates that
SODIUM, g/l
Fig. 5.3. Sodium versus chloride concentrations for some formation waters taken from
Tertiary (T), Cretaceous (C), and Jurassic (J) zge sediments and compared to evaporating
sea water.
500F --/
200t-
0 - Nor ma1 evaoor it e
associated b i n e
/
I- /"
SODIUM, mg/l
Fig. 5.4. Sodium versus chloride concentrations for some formation waters taken from
Pennsylvanian (P) and Mississippian (M) age formation sediments and compared t o
evaporating sea water.
diagenetic processes, such as ion-exchange or ultra-filtration reactions in-

volving clays and/or carbonates, may operate to deplete the sodium concen-
tration in waters in older sediments.
Potassium
The second most abundant member of the alkali-metal group is potassium;

its abundance in the crust of the earth is about 2.55 wt.% (Fleischer, 1962).
Like the other alkali metals, it is lithophilic, and with its large ionic radius
(see Table 5.1) it participates in forming solid solutions and forms its own
minerals, such as feldspar and mica. The potassium feldspars are resistant to
leaching by water, which may account for the low potassium concentrations
in many natural waters. Clay minerals readily adsorb potassium, and in illite
it is incorporated into the crystal structure in such a manner that it cannot
be removed by ion-exchange reactions (Lyon and Buckman, 1960).
Potassium is less easily hydrated than sodium, and is more easily adsorbed
by colloids; therefore, it is retained in sediments and soils in greater abun-
dance than sodium. It is an essential element t o plants and animals. Accord-
ing to Gol&chmidt (1958),potassium in pulverized potassium feldspars is
absolutely unavailable t o plants.
The concentrations of potassium in carbonates, sandstones, and shales is
about 2,700, 10,700,and 26,600 ppm, respectively (Mason, 1966). Potas-
POTASSIUM 139
~~
.lvv
200 -
- 100 -
1 I I I IIll
POTASSIUM, g/ I
Fig. 5.5. Potassium versus chloride concentrations for some formation waters taken from
Tertiary (T), Cretaceous (C), and Jurassic (J) age sediments and compared to evaporating
sea water.
sium concentrates primarily in hydrolysates (clay minerals), such as illite and

glauconite, and in evaporites. Table 5.11 illustrates how the concentration of
potassium in the aqueous phase increases until sylvite (KC1) precipitates. The
concentration of potassium in some subsurface brines usually is depleted
with respect to an evaporite-associated sea water. Fig.5.5 illustrates the rela-
tion of potassium in some subsurface brines taken from sediments of Terti-
500 -
-
-
,'
-Nmal evaporite curve
-
-
5--
-
m
POTASSIUM, mg/l
Fig. 5.6. Comparison of the potassium concentrations in some Pennsylvanian (P) and
Mississippian (M) age formation waters from Oklahoma with an evaporating sea water.
ary, Cretaceous, and Jurassic ages to an evaporite-associated sea water

(Collins, 1970). Fig.5.6 illustrates the same relation for some subsurface
brines taken from Pennsylvanian and Mississippian age sediments (Collins,
1969a).
The depletion of potassium in subsurface brines might be caused by its
uptake by clays. For example, montmorillonite-type clay minerals system-
atically change to illite with increasing depth of burial, due to thermal
diagenesis; and, as a result of this transformation, they lose interlayer
(bound) water (Burst, 1969). This change appears t o begin at a temperature
above 90°C. (This freed interlayer water can be readily expelled, and its
movement probably is important in the first migration stage of hydrocar-
bons.) Laboratory experiments at elevated temperatures and pressures in-
dicate that montmorillonite loses its interlayer water and transforms into
illite in the presence of potassium-enriched water (Khitarov and Pugin,
1966). The structural variations of the expandable minerals in clays appar-
ently are also influenced by the potassium content of the associated waters.
Rubidium
Rubidium, like the other alkali metals is lithophilic, and its abundance in
the earth’s crust is about 3.0 x wt.%, which is greater than that of
lithium (Fleischer, 1962). It tends to be removed from solution more readily
than lithium, primarily because of its ability to replace potassium in mineral
structures. Table 5.11 indicates that it precipitates from an evaporite along
with sylvite to a greater extent than lithium, and it has a high chemical
reactivity. The radius of its ion, 1.48 a, is only about 10% larger than the
potassium ion, so it can be accommodated into the same crystal lattices.
Because of this, it forms no minerals of its own.
Rubidium and cesium occur sympathetically in nature; that is, both are
commonly found in amazonite, vorobyevite, and beryl (Goldschmidt, 1958).
Rubidium is a member of series NH4-K-Rb-Cs, and members of this series
are more similar in their chemical and physical properties than are the mem-
bers of any other group, with the exception of the halogens. Rubidium
concentrates in the late crystallates, particularly those of granitic derivation,
and it has a greater tendency t o be adsorbed by clays than has potassium. It
is removed from igneous rocks by water leaching and then adsorbed by
hydrolysate sediments and soils.
Shales contain about 250 ppm of rubidium; deep-sea red clays, about 400
ppm; and some glauconites, about 500 ppm (Goldschmidt, 1958).Sea water
contains about 0.12 mg/l of rubidium; subsurface brines contain up t o 4
mg/l. Higher concentrations of rubidium probably can be found in brines
associated with rocks containing potassium minerals, such as microcline
feldspars, or lepidolite mica.
CESIUM 141
Cesium
Cesium is the heaviest alkali metal and also the rarest, with an abundance
of about 7 x wt.% in the earth’s crust (Fleischer, 1962). It has an ionic
radius of 1.69 8,which is distinctly larger than potassium, and it cannot
replace potassium in minerals as easily as rubidium; probably because of this,
it forms its own minerals. It is leached from igneous and metamorphic rocks
by water during weathering, and is adsorbed by hydrolysate sediments and
soils more readily than rubidium or potassium. Its low ionization potential
indicates that it has the greatest chemical reactivity of the alkali metals.
Cesium and rubidium were discovered in 1860 by Robert Bunsen by use of
spectral analysis, a method which he and Kirchhoff invented.
Cesium concentrates primarily like rubidium, in marine argillaceous sedi-
ments. Some shales contain about 15 ppm; deep-sea red clays, 20 ppm; and
glauconite, 15 ppm of cesium (Goldschmidt, 1958). Sea water contains
5x mg/l of cesium, and some subsurface brines contain up to 1mg/l.
Beryllium
Beryllium is a member of the alkaline earth group in the periodic chart of

the elements, but few of its properties are similar t o the more abundant
members, such as magnesium, calcium, and strontium. Beryllium, like
lithium, is a light element with an atomic weight of 9.012 (Table 5.111; see
also Moeller, 1954), and like lithium, it is an exception t o the rule that light
elements are more abundant than heavy elements. The earth’s crust contains
about 6 x wt.% of beryllium (Fleischer, 1962).
In sedimentary rocks, beryllium is restricted primarily to hydrolysates and
especially to bauxites enriched in aluminum (Goldschmidt, 1958). Shales
contain about 6 ppm, and some coal ashes contain up to 8,000 ppm, al-
though generally only about 4 ppm. The concentration of beryllium in sea
TABLE 5.111
Properties of the alkaline earth metals
Property Beryllium Magnesium Calcium Strontium Barium
Atomic number 4 12 20 38 56
Ionic radius
(A 1 0.31 0.65 0.99 1.13 1.35
Outer electronic
configuration 1s’ 2s’ 2s’ 2p6 3s’ 3s2 3p6 4s2 4s’ 4p6 5s2 5s2 5p66s’
Atomic weight 9.012 24.31 40.08 87.62 137.34
Ionization
potential (V) 9.320 7.644 6.111 5.692 5.210
water is about 5 x lo-' mg/l, and some subsurface brines contain 0.02-4.2
mg/l. Since beryllium is highly toxic, waters containing it should be handled
with caution.
Magnesium
One of the more abundant members of the alkaline earth group of metals,
magnesium makes up about 2.1 wt.% (Fleischer, 1962) of the crust of the
earth.
Magnesium is dissolved during chemical weathering, mainly as the chloride
and sulfate. Ferromagnesian minerals in igneous rocks and magnesium car-
bonate in carbonate rocks are generally considered t o be the principal
sources of magnesium in natural waters. Carbon dioxide plays an important
role in the dissolution of magnesium from silicate and carbonate minerals.
Waters associated with either granite or siliceous sand may contain less than
5 mg/l of magnesium, whereas those associated with either dolomite or
limestone may contain over 2,000 mg/l of magnesium.
Elements commonly found in oilfield waters have the following ionic
potentials: sodium, 0.95; calcium, 0.50; magnesium, 0.33; chlorine, 1.81;
bromine, 1.95; and iodine, 2.16. Apparently the cation (magnesium) and the
anion (chlorine) would be the most likely to remain in true ionic solution;
however, several other variables occur during diagenesis which lead to deple-
tion of magnesium in waters.
Depletion of magnesium in some waters probably is a result of the replace-
ment reaction t o form dolomite, CaMg(C0, ) 2 . Whole mountain masses are
made of dolomite, which is formed by the regular substitution in the calcite
2oo t J
C
/
?$
Normal evaporite curve
'so0 500 rpoo 2,000 5ooO lop00 2

0m
, 5Q(
MAGNESIUM, mg I I
Fig. 5.7. Comparison of the magnesium concentrations in some Tertiary (T), Cretaceous
(C), and Jurassic (J) age formation waters from Louisiana with an evaporating sea water.
CALCIUM 143
c
500
M
M
r
20
10
1,000 I 0,000 lO0,OoO
MAGNESIUM, mg/l
Fig. 5.8. Comparison of the magnesium concentrations of some Pennsylvanian (P) and
crystal lattice of alternate ions of calcium and magnesium. The large differ-
ences in the ionic radii of Ca (0.99A) and Mg (0.65A) are the reason for this
diadochy.
Magnesium ions in aqueous solution have a large attraction for water
molecules and probably are surrounded by six water molecules in octahedral
arrangement. This may account for the paucity of magnesium in soils,
because the small cation becomes large by hydration. Sodium has a similar
reaction, but potassium, which does not, is readily adsorbed by soil colloids.
Shales, sandstones, and carbonates contain 15,000,7,000,and 47,000
ppm of magnesium, respectively (Mason, 1966). Subsurface brines contain
from less than 100 mg/l t o more than 30,000 mg/l; however, many subsur-
face brines are depleted in magnesium if compared to a sea water evaporite
sequence, (Table 5.11). Sea water contains about 1,300 mg/l. Fig. 5.7 is a
plot of chloride versus magnesium for some subsurface brines taken from
Tertiary, Cretaceous, and Jurassic age sediments. The position of the normal
evaporite curve indicates that all of these waters were depleted in magnesium
with respect to this curve (Collins, 1970). Fig. 5.8 is a plot showing similar
depletion of some subsurface brines taken from some sediments of Pennsyl-
vanian and Mississippian age.
Calcium
The abundance of calcium in the crust of the earth is about 3.55 wt.%
(Fleischer, 1962),making it the most abundant of the alkaline earth metals,
but only in the crust; in the earth as a whole, magnesium is much more
abundant. Calcium is dissolved as bicarbonate as a result of chemical
weathering of calcium-bearing minerals. Waters associated with limestone,
dolomite, gypsum, or gypsiferous shale usually contain an abundance of
calcium, but waters associated with granite or silicious sand may contain less
than 10 mg/l of calcium. Slight changes in the pH of waters containing
calcium bicarbonate will cause calcium carbonate to precipitate, and calcium
carbonate is one of the most common deposits found in plugged oilfield
lines, equipment, and reservoirs.
Precipitation of calcium carbonate in the sea is the prime mode of the
origin of limestone. The solubility of calcium carbonate in sea water in-
creases with salinity and increasing partial pressure of carbon dioxide, but it
decreases with increasing pH, calcium content, and temperature. The
solubility of calcium sulfate decreases with increasing temperature.
Shales, sandstones, and carbonate rocks contain about 22,100, 39,100,
and 302,300ppm of calcium, respectively (Mason, 1966).Sea water contains
400 mg/l and subsurface brines often contain 2,000-3,000 mg/l, with some
as high as 30,000 mg/l. Fig. 5.9 is a plot of chloride versus calcium concen-
trations for some subsurface waters taken from Tertiary, Cretaceous, and
Jurassic age sediments. The amount of calcium in these waters increases with
increasing salinity, and the waters from the older sediments appear to con-
tain more calcium. Fig. 5.10 is a similar plot for some subsurface brines
taken from sediments of Pennsylvanian and Mississippian age. These samples
all appear to be enriched in calcium relative t o the evaporite curve, and the
concentration of calcium appears to increase with increasing salinity.
200 -
- 100- -
\
0 -
1 I 1 I I I111
500 1 , m 2 p 5poo lop00 29ooo
CALCIUM, mg/l
Fig. 5.9. Comparison of the calcium concentrations of some Tertiary (T), Cretaceous (C),
STRONTIUM 145
&-\- Normal evaporite curve
P
P
Ii M
M
201 / 1
00
CALCIUM, mg/ I
Fig. 5.10. Comparison of the calcium concentrations of some Pennsylvanian (P) and
MEsissippian (M) age formation waters from Oklahoma with an evaporating sea water.
Strontium
Strontium, a minor element compared t o calcium and magnesium com-

prises about 0.03 wt.% of the earth's crust (Fleischer, 1962). Table 5.111
illustrates some of its properties, and it resembles calcium chemically. Stron-
tium has a tendency to work upward during fractional crystallizaticn be-
cause of its relatively large radius (Goldschmidt, 1958).It occurs abundantly
with potassium in volcanic rocks, alkali rocks, and pegmatites.
Dissolved strontium results from water leaching of rocks, and it has been
postulated that the strontium in petroleum-associated waters also may be a
byproduct of the organic decay processes which originally formed
petroleum. Strontium is only a microconstituent in most terrestrial animals,
but several species of marine animals contain considerable quantities of
strontium in their skeletons (Odum, 1951).
Table 5.11 indicates that strontium may reach a concentration of 60 mg/l
during sea-water evaporation, and then most of it precipitates with calcium
sulfate. The amount of sulfate in the water influences the amount of stron-
tium that remains in solution. Data by Sillhn and Martell (1964)indicate
that if the sulfate activity in a water is 100 mg/l, the strontium activity can
be about 28 mg/l. Davis and Collins (1971)studied the solubility of stron-
tium sulfate in strong electrolyte solutions and found that 958 mg/l of
strontium is soluble in a synthetic brine solution of ionic strength 3.05,
containing ions of sodium, calcium, magnesium, potassium, chloride,

bromide, and iodide. Calcium chloride concentration apparently has a very
pronounced effect upon the solubility of strontium sulfate.
Celestite and strontianite occur commonly in sediments. Carbonate sedi-
ments contain up t o 1,200 ppm of strontium; dolomites, usually less than
“““I I
C -Cretaceous
J -Jurassic
C J
c cc 2,000
C
cC
T
I IIII I I IIL
1
0 20 50 100 2 a
STRONTIUM, mgll
Fig. 5.11. Comparison of the strontium concentrations of some Tertiary (T), Cretaceous
(C), and Jurassic (J) age formation waters from Louisiana with an evaporating sea water.
500 -
-
200 -
-
\
P
I Ill
50 100 ZOO 500 1,000
STRONTIUM, mg/l
Fig. 5.12. Comparison of the strontium concentrations of some Pennsylvanian (P) and
BARIUM 147
170 ppm; and secondary gypsum, up t o 1,100 ppm (Goldschmidt, 1958).

Sea water contains about 8 mg/l of strontium, but subsurface brines contain
up to 3,500 mg/l. Fig. 5.11 is a plot of chloride versus strontium content for
some subsurface brines taken from some Tertiary, Cretaceous, and Jurassic
age sediments. Most of these samples were enriched in strontium compared
to the evaporite-associated water, and it is possible that a mechanism similar
to dolomitization could cause the enrichment. In comparison t o calcium, the
strontium appears to be increasingly accumulated; for example, only five
samples (from Tertiary sediments) fell within the normal evaporite curve.
Fig. 5.12 is a similar plot for some subsurface brines showing similar results
taken from sediments of Mississippian and Pennsylvanian age.
Barium
Barium, like strontium, is a minor element, comprising 0.04 wt.%, of the

earth’s crust; it is more concentrated in igneous rocks and less concentrated
in sedimentary rocks (Fleischer, 1962). It, like the other alkaline earth
metals, is predominantly lithophile. Table 5.111 illustrates some of the
properties of barium; its ionic radius, 1.35 A, permits it t o replace potas-
sium, but usually not calcium and even less commonly magnesium. Barium
forms more of its own minerals than does strontium. Barium is readily
absorbed by colloids, like potassium, and is therefore retained by soils or
precipitated with hydrolysates; it is also concentrated in deep-sea manganese
nodules (Hem, 1970).
Barium dissolves as bicarbonate, chroride, or sulfate during weathering
processes, and migrates in aqueous solutions as these compounds. The
solubility of barium sulfate increases when hydrochloric acid or chlorides of
the alkali or other alkaline earth metals are present in solution. The proper-
ties of barium are similar t o those of strontium. Both precipitate through
loss of carbon dioxide from a bicarbonate-bearing solution, or as sulfates by
the action of sulfuric acid, sulfides, or sulfates. Strontium, however, is less
likely t o be absorbed by clays than barium, because its ionic radius is smaller
and its ionic potential is higher.
Encrustation deposits taken from plugged pipes of waterflood systems for
secondary recovery of oil, where barium is present, usually contain barium,
calcium, strontium, iron, and traces of other metals. Barium may cause
problems in waterflood systems by reacting with the chromate-type oxygen-
corrosion inhibitors, forming water-insoluble barium chromate.
The amount of barium found in sandstones, shales, and carbonates is
about 180, 450, and 90 ppm, respectively (Goldschmidt, 1958).Sea water
contains about 0.03 mg/l, and subsurface brines may contain more than 100
mg/l; however, many brines contain less than 10 mg/l. Davis and Collins
(1971)found that 59 mg/l of barium sulfate is soluble in a synthetic brine
with an innic strength of 3.0487, containing sodium, calcium, magnesium,
TABLE 5.IV
Properties of aluminum. copper. iron, lead, manganese, and zinc
property Aluminum Copper Iron Lead Manganese Zinc
Atomic number 13 27 26 82 25 30
Ionic radius (A) 0.50 0.96(+1) 0.76(+2) 1.20(+2) 0.80(+2) 0.74
. ,
0.691+21 . .
0.64(+3) 0.84(+4)
. . 0.46c+7 1..
Outer electronic
configuration 2s22p63s23p1 3s23p63d'04s' 3s23p63d64s' 4d'05s'5p64f'5d106s'6p' 3s' 3p6 3d54s' 3s2'3p63d" 4s'
Atomic weight 26.98 63.54 55.54 207.19 54.938 65.37
Ionization potential (V) 5.984 1.723 1.165 7.415 1.168 9.391
MANGANESE 149
potassium, chloride, bromide, and iodide ions. Many analyses performed by

wet chemical methods indicate rather high concentrations of barium in some
subsurface brines. Some of these high results probably should be attributed
to strontium plus barium rather than barium only, because satisfactory
separation of the two in wet chemical methods is very difficult to accom-
plish.
Manganese
Manganese is a member of the VII B group of elements and is well known

for its multiplicity of oxidation states. Essentially it is cationic, and the
Mn+4 oxidation state usually is found in sediments. Its (+2) ionic radius is
0.80 8,while the ferric iron radius is 0.76 8 (see Table 5.IV); reasonable
amounts of interchange in crystal lattices between these two ions are possi-
ble. The abundance of manganese is about 0.1 wt.% of the earth’s crust
(Fleischer, 1962).
Manganese is present in many oilfield brines because it is readily dissolved
by waters containing carbon dioxide and sulfate. Except for titanium, man-
ganese is the most abundant trace element in igneous rocks. Nearly all
mineral groups of petrological importance contain manganese. During
weathering, manganese is dissolved mainly as the bicarbonate. Decomposition
of the bicarbonate leads to the formation of M d 4 compounds. In a reducing
type of environment Mn+ compounds migrate in aqueous solutions. Mn+
compounds are less mobile, and Mn+4 compounds precipitate from aqueous
solutions. In general, manganese remains in solution at a low redox potential
and precipitates at a high redox potential.
According to Goldberg (1963), manganese oxide nodules on the ocean
bottom occur in both shallow water and deep water environments. He attri-
butes these deposits to slow oxidation of dissolved manganese in areas where
the waters contact an oxide surface. In most subsurface brines, the manga-
nese is in the reduced form (Mn+*)because the redox potential is low and
the pH is less than 7.0. Any in subsurface brines probably would be
suspended with particulate matter or complexed by organic compounds,
rather than in ionic solution.
Shales and carbonates contain about 850 ppm and 1,100 ppm, respec-
tively, of manganese (Mason, 1966). Sea water contains about 0.002 mg/l,
and many subsurface brines contain 1.0 to 6.0 mg/l of manganese.
Iron
Iron is a member of the VIII group of elements and is predominantly

siderophile. However, because it has an affinity for sulfur, it is also thiophile;
and because it commonly enters into silicate minerals, it is lithophile as well.
It is an ubiquitous element, with an abundance of about 5 wt.% of the
earth’s crust (Fleischer, 1962).
150 INORGANIC CONSTITUENTS A N D PHYSICAL PROPERTIES
Iron, cobalt, and nickel possess atomic radii that differ only about 2% or
less, so that the crystal chemistry of the three are related. The divalent ions
of nickel, magnesium, cobalt, and iron have similar ionic radii; consequently,
their chemistries in the sequence of isomorphous crystallization of mixtures
are similar. The trivalent ions of iron and cobalt are similar in size, but the
high oxidation potential of cobalt prevents much replacement (Goldschmidt,
1958).
The solubility of iron compounds in ground waters is a function of the
type of iron compound involved, the amounts and types of other ions in
solution, the pH, and the Eh. According t o Larson and King (1954), 100
ppm of ferrous iron can stay in solution at pH 8 and pH 7; the theoretical
maximum is about 10,000 ppm. The effects of many other ions, plus tem-
perature and pressure differentials, such as those common to oilfield waters,
have not been thoroughly studied. When a ground water in which ferrous
iron is dissolved contacts the atmosphere, the following reaction can occur:
2Fe2++ 4HCO3- + H20 + 1 / 2 0 2 + 2Fe(OH), + 4C02

Sandstone contains iron oxide, iron carbonate, and iron hydroxide, and
shales and carbonate rocks contain oxides, carbonates, and sulfides of iron.
Oilfield waters with characteristic low redox potentials dissolve some iron
from the surrounding rock. The iron occurs in such waters at two levels of
oxidation, ferrous or ferric.
Knowledge of the amount and type of iron compounds in oilfield waters
is used to estimate the amount of corrosion that is occurring in the produc-
tion system, and t o determine the type of treatment required if the water is
t o be used for waterflooding. This knowledge also enables determination of
the Eh of the in situ water, because the Eh can be calculated from the Fe+2
and Fe+ values.
Shales, sandstones, and carbonates contain about 47,200, 9,800, and
3,800 ppm, respectively, of iron (Mason, 1966). Sea water contains about
0.01 mg/l, and subsurface brines contain from traces to over 1,000 mg/l of
iron.
Copper
Copper is a member of the VIII group of elements, and it is character-
istically thiophile; the largest concentrations of it are found in various sulfur
compounds. The earth’s crust contains about 0.01 wt.% of copper (Fleischer,
1962). Its compounds are dissolved easily during weathering, if the pH of the
solution is less than 4.5. Many of the water-soluble copper compounds are
salts of organic acids such as acetic, citric, and naphthenic. Much of the
copper that is dissolved is precipitated afterward as sulfide. Traces of copper
remain in the oceans, but its content is kept low because of the adsorption
on, or combination with, marine organisms. Miholic (1947) presented an age
ZINC 151
division for mineral waters based on the presence of heavy metals in waters
associated with joints and faults caused by tectonic movements of different
geological ages. He placed copper as the predominant heavy metal in the
Caledonian Group of the Orogenic Epoch (post-Silurian). Biochemical pro-
cesses are known to be responsible for enriching a deposit in metals such as
uranium, copper, and vanadium; therefore, this classification is restricted to
waters of igneous origin.
Most shales and carbonates contain about 45 and 4 ppm, respectively, of
copper, with sandstones containing less than 1 ppm (Mason, 1966). Sea
water contains about 0.003 mg/l, and most subsurface brines analyzed in this
laboratory contained from less than 0.5 mg/l up to about 3 mg/l. The solu-
bility of copper generally decreases with decreasing redox potential and
increases with increasing redox potential if reduced sulfur is present. Most
subsurface oilfield brines have relatively low redox potentials.
zinc
Zinc is a member of the I1 B group of elements and is predominantly

thiophile. Its abundance in the crust of the earth is about 0.013 wt.%
(Fleischer, 1962). Its geochemistry results from the similarity of its divalent
ionic radius and the radii of Mg+’, Ni+?, Co+’, Fe+’, and Mn+’
(Goldschmidt, 1958).
Zinc is dissolved readily as sulfate or chloride from acid rocks, such as
granite, during weathering. Conversely, zinc is not dissolved easily from
limestone with which it is deposited. Most alkaline waters do not extract
zinc; however, a solution of NH,, NH,NO,, and NaC10, can extract and
hold small quantities of zinc; the more acidic the water, the greater the
amount of zinc extracted. Zinc is precipitated as the sulfide, oxide, carbon-
ate, or silicate. Traces of zinc are found in sea water, but eventually zinc is
deposited in carbonated sediments or in bottom muds or sapropels as sulfide.
Shales, sandstones, and carbonates contain about 95, 16, and 20 ppm,
respectively, of zinc (Mason, 1966). Sea water contains about 0.01 mg/l, and
subsurface brines contain traces to more than 500 mg/l of zinc.
Mercury
Mercury is a member of the I1 B group of elements, which also includes

zinc and cadmium. It is relatively abundant for a heavy element, but still
must be considered scarce, with an abundance of about 4 x lo-’ wt.% of
the crust of the earth (Fleischer, 1962). Most commercial deposits of
mercury are of hydrothermal origin and are related to magmatic rocks; the
commercial ore is cinnabar, HgS, or the liquid metal itself (Goldschmidt,
1958). Mercury is predominantly thiophile, and its geochemistry is control-
led by the fact that it is volatile, with a boiling point of 357”C, and can be
reduced to the metal by ferrous iron. Therefore, in a magmatic environment
the temperature and the redox potential control its occurrence. It is trans-
ported in hot springs (White et al., 1963).
Shales, sandstones, and carbonates contain about 0.4, 0.03, and 0.04 ppm,
respectively, of mercury. Sea water contains 3 x lo-’ mg/l, and subsurface
oilfield brines contain 0-0.15 mg/l. The samples containing 0.15 mg/l of
mercury were found in relatively dilute brines taken from the Cymric and
the Rio Bravo oilfields in California. Free mercury is found in the oils
produced from these fields, and the ages of the producing formations range
from Eocene t o Pleistocene.
The mercury content of natural waters has been used t o locate cinnabar
deposits (Dall’Aglio, 1968). The amounts of mercury in waters appear t o
increase with increasing bicarbonate concentration. Karasik et al. (1965)
found that saline waters containing 200,000 mg/l of chloride contain very
small amounts of mercury, which suggests that anionic complexes such as
HgC14-* may not be important transporters of mercury. Brackish waters con-
taining up t o 3,000 mg/l dissolved solids, up to 400 mg/l of bicarbonate, and
the iodide ion sometimes contain up to 10 ppb of mercury, while stronger
brines contain <0.1 ppb of mercury, which suggests that mercury may be
transported as Hg14-* in brackish waters.
Lead
Lead is a member of the IV A group of elements; it is ubiquitous in the

earth, but its abundance in the crust is only about 0.002 wt.% (Fleischer,
1962). It is extracted from its minerals during weathering and migrates in the
form of soluble-stable compounds. It is particularly soluble in acetic and
other acids. Because the bicarbonate form is more soluble than the carbon-
ate, lead can be transported as the bicarbonate. Most of the lead is
precipitated from waters before they reach the sea. Hemley (1953) studied
lead sulfide solubility related to ore deposition from saline waters. He con-
cluded that lead-complex concentrations increase with increasing concen-
trations of bivalent sulfur and decrease at pH values above 7. The solubility
of lead is limited primarily by the solubility restrictions of its sulfide and
sulfate in reducing and oxidizing systems. How its solubility is influenced by
many other ions, such as those found in a brine, has not been sufficiently
studied.
Shales, sandstones, and carbonates contain about 20, 7, and 9 ppm of
lead, respectively. Sea water contains about 0.003 mg/l, and subsurface
brines contain trace amounts to more than 100 mg/l of lead.
Cadmium
Cadmium is a member of the I1 B group of elements and may be consid-
ered one of the rarer elements; its abundance is about 3 x lo-’ wt.% of the
earth’s crust (Fleischer, 1962). It is strongly thiophile, but its chemistry
BORON 153
differs from that of zinc in that it will precipitate from a strong acid solu-
tion, whereas zinc will not. There are few independent cadmium minerals,
and its distribution is mainly that of a “guest” atom or ion in minerals. It
frequently is present in lead-zinc deposits and occurs in solid solution in
hypogene sulfides. A main carrier of cadmium is sphalerite, and oxidation of
sphalerite or other sulfides containing cadmium will release the soluble
cadmium sulfate.
Shales and carbonates contain about 0.3 and 0.035 ppm of cadmium,
respectively, and sandstones contain less than 0.01 ppm (Mason, 1966). Sea
water contains about 0.0001 mg/l, and the subsurface oilfield brines may
contain from 0 to about 0.001 mg/l of cadmium. Subsurface brines of the
sulfate type in contact with lead-zinc deposits probably contain higher
concentrations of cadmium.
Boron
Boron is a member of the I11 A group of elements, and it is an oxyphile

and lithophile element. Its abundance in the crust of the earth is about 0.001
wt.% (Fleischer, 1962). It has small atomic and ionic radii.
Knowledge of the presence of boron compounds in oilfield waters is im-
portant for several reasons. Boron is useful in identifying the sources of
brines intrusive t o oil wells, or in fresh-water lakes or streams. In concen-
trations exceeding 100 mg/l, it affects electric log deflections. Boron is
present in oilfield brines as boric acid, inorganic borates, and organic borates.
When it is present as undissociated boric acid, it is an important buffer
mechanism, being second only to the carbonate system. It may be precipi-
tated as the relatively insoluble calcium and magnesium borates.
Kazmina (1951) calculated the borate-chloride coefficient of some
Russian oilfield waters. With a plot of the borate-chloride coefficient in
logarithmic coordinates as a function of chloride content, he distinguished
genetic groups of natural waters found in oil-bearing regions.
Mitgarts (1956) studied the significance of boron and other elements in
petroleum prospecting. In general, boron, together with bromine and iodine,
is always associated with waters accompanying petroleum. Like chlorine, it
can be considered an element of marine origin. The solubility of most boron
compounds, the hydrolytic cleavage of boron salts, and their ability t o be
occluded and coprecipitated with other compounds account for the exten-
sive migration of boron. Soluble-complex boron compounds in brines and
connate waters probably are there as a result of the decay of the same plants
and animals that were the source of petroleum.
respectively, of boron. Sea water contains about 4.8 mg/l, and subsurface
oilfield water contains from trace amounts to more than 100 mg/l. Fig. 5.13
is a plot of chloride versus boron concentrations of some oilfield brines taken
from some sediments of Tertiary, Cretaceous, and Jurassic age. The plot
---I-
200 Normal evaporite curve, / %
T T
a
50
0
J
I
" 30
"k/~ / c
J
C "
c
T
T
BORON,mg/I
Fig. 5.13.Comparison of the boron concentrations of some Tertiary (T), Cretaceous (C),
I P
BORON, m g / l
Fig. 5.14. Comparison of the boron concentrations of some waters from Pennsylvanian
(P) and Mississippian (M) age sediments with an evaporating sea water.
indicates that the majority of these brines are enriched in boron relative to a
normal evaporite-formed brine, and that the samples that were depleted in
boron may have contained dissolved halite. Fig. 5.14 is a similar plot for some
samples taken from some Pennsylvanian and Mississippian age sediments.
Boron is one of the elements whose concentration in the aqueous phase
increases as a brine is evaporated, as illustrated in Table 5.11.
ALUMINUM 155
Aluminum
Aluminum is the third most abundant element in the earth’s crust, but its
concentration in natural waters usually is less than 1 mg/l. The ionic radius
a
of trivalent aluminum is 0.57 (Goldschmidt, 1958),and it usually behaves
as a cation when 6-coordinated with oxygen compounds. However, when
4-coordinated, it usually acts like the central atom of an anion. The
4-coordination usually, but not exclusively, is associated with minerals
formed at high temperatures, but the 6-coordination is associated with
minerals formed at low temperatures, which includes most sediments in the
petroleum environment.
The clay minerals illite, kaolinite, and montmorillonite often contain
about 13.5, 21, and 11%aluminum, respectively. Quartzites, sandstones,
limestones, and shales contain about 0.7, 3.0, 0.6, and 10% aluminum,
respectively. During weathering silica will leach out and leave aluminum
hydroxide behind (Pirsson and Knopf, 1947), and sedimentation processes
leave only about 0.4 mg/l aluminum in sea water.
According t o Hem (1970), the cation A P 3 predominates in solutions with
a pH of 4.0 or less. Above pH 4.5, polymerization gives rise to an aluminum
species with a gibbsite (aluminum hydroxide) structural pattern. Above pH
7.0, the dissolved form is the anion A1 (OH),-.
The pH of the water is the main control of the amount of alumium that is
likely to be present in natural waters. A water with a pH less than 4.0 may
contain 1%or more of aluminum; for example, waters associated with acid
mine drainage. Oilfield waters contain trace amounts t o more than 100 mg/l
of aluminum.
A 1ha 1inity
Alkalinity is defined as the capacity of a solution to neutralize an acid,

usually t o a pH of 4.5. A solution with a neutral pH of 7.0 may have a
considerable amount of alkalinity; therefore, alkalinity is a capacity function,
in contrast to pH, which is an intensity function. The alkalinity-pH ranges
originally coincided with methyl orange and phenolphthalein color end
points. The potentiometric titration produces more accurate alkalinity
results, and it utilizes an end point where the most abrupt pH change occurs
while specific increments of a standard acid are added.
Alkalinity usually is caused by the presence of bicarbonate, carbonate, or
hydroxyl ions in a water; however, the weak acids such as silicic, phosphoric,
and boric can contribute titratable alkalinity species. Carbon dioxide, which
is dissolved in circulating waters as bicarbonate or carbonate as a result of
the carbon cycle, is the prime source of alkalinity in shallow ground waters.
However, in deep subsurface brines, additional carbon dioxide probably is
dissolved as a result of diagenesis of inorganic and organic compounds.
Most oilfield waters contain no hydroxyl ions, and most of them contain
no carbonate ions, but they do contain bicarbonate ions. Some oilfield

waters in the Rocky Mountain area are alkaline and contain both primary
and secondary alkalinity, where primary alkalinity is that associated with the
alkali metals and secondary alkalinity is that associated with the alkaline
earth metals. For example, the Green River formation waters that are in or
near trona beds may contain more than 20,000 mg/l of carbonate and 5,000
mg/l of bicarbonate. Most oilfield waters from other areas contain from
about 100 to 2,000 mg/l of bicarbonate.
Acidity
The basis of acidity is the solvated hydrogen ion H 3 0 + ,which is found in

nature. Volcanic emanations produce HF, HCl, and H2SO4, probably for-
med by reactions between water and constituents associated with the
magma. Waters associated with peat may contain organic acids, rain waters
may contain carbonic acid, and waters associated with reducing conditions
and anaerobic bacteria may contain H2 S.
Acidity, as contrasted to alkalinity, is the capacity of a solution to
neutralize a base, usually from below pH 4.5 t o pH 7.0. Most oilfield brines
normally do not contain acidity. New wells or reworked wells often are
acidified or "acidized" with a strong mineral acid or a combination of
mineral and organic acids. This treatment causes the produced water to
contain a certain amount of acidity until all of the acid is neutralized or
diluted. Because of the large quantities of acids used in some treatments, it
m a y take 6 months or more for the water produced from a treated well t o
return to normal. Organic acids and organic acid salts commonly are found
in oilfield waters, and the concentration ranges from trace amounts to more
than 3,000 mg/l.
Silica
Silicon is the second most abundant element in the earth's crust, which
contains about 27 wt.% of it (Fleischer, 1962). It always occurs in a com-
bined form. Most of the silicon compounds involve structures with oxygen,
and there are about a thousand silicate minerals in the earth's crust; however,
those which are predominant are relatively few in number.
The solubility of silica in water is a function of temperature, pressure, pH,
and other ions in solution. Most silica in natural water probably is in the
form of monomolecular silicic acid, H4 Si04 or Si(OH)4. Collins (196913)
studied the solubility of a serpentine in solutions of calcium chloride and
sodium chloride a t temperatures from 30" to 200°C and pressures from 176
t o 1,055 kg/cm2. The solubility calculated as silicon molarity in solution
increased with increasing concentrations of sodium chloride, increasing pres-
sure, and increasing temperature up to about 125°C. Between about 125"
and 2OO0C, the solubility decreased with increasing temperature. The solu-
AMMONIUM NITROGEN 157
bility of silicon in the presence of calcium chloride solutions decreased with

increasing Concentration of calcium chloride and with increasing temperature
above about 100°C. The solubility increased with increasing pressure and
with increasing temperature between 30' and about 100°C. Subsurface
petroleum-associated brines usually contain less than 30 mg/l of dissolved
silica; Rittenhouse et al. (1969) report that their silica content ranges from
about 1 to 500 ppm as silicon, and that some low salinity waters contained a
higher median content of silica than more saline waters in other areas.
Ammonium nitrogen
Ammonium contains nitrogen in the N-3 oxidation state, a reduced form.

Nitrogen can occur in all of its states of oxidation, ranging from -3 t o +5.
Oxidation of the reduced forms produces nitrogen gas, N 2 , and other
nitrogen species up to nitrate, NO3-. Ammonia, NH3, forms during the
anaerobic decay of organic nitrogenous material. The petroleum genetic
environment produces ammonia, which transforms to ammonium, NH4, in
many petroleum-associated waters because the redox potential is too low to
oxidize the ammonia to nitrate. The ammonium ion is too weak t o be
successfully titrated; however, Collins et al. (1969), developed a technique
using formaldehyde, whereby a produced strong acid can be titrated.
TABLE 5.V
Ammonium content of 10 subsurface brine samples
Sample State Formation
1 Utah Navaho 30-143 0

2 Utah Green River 852-1,719 71
3 Utah Lower Green River 914-1,737 91
4 Okla. Morrow 2,713-2,715 0
5 Okla. Rue 1,958-1,963 116
6 Utah Uinta Surface, Green River 717-1,111 2,069
7 Utah Green River 68 1-1,5 5 6 5
8 Okla. Hunton 2,696-2,77 1 434
9 Okla. Oswego 1,928-1,951 233
10 Okla. Chester 2,408-2.437 23
The NH4N content of several oilfield brines was determined and a wide
variation in concentration was found. Table 5.V illustrates the amounts of
NH4N found in 10 samples taken from subsurface rocks in Oklahoma and
Utah.
Phosphorus
Phosphorus occurs in the earth’s crust almost exclusively as the

ion, and a large percentage of it is contained in the apatite group of minerals,
which primarily are related to igneous rocks. The crust of the earth contains
about 0.12 wt.% of phosphorus. It is a member of the V A group of elements
with oxidation states ranging from -3 to +5. In contrast t o nitrogen,
phosphoric acid and phosphates are not oxidizing agents.
The phosphorus species present in most natural waters probably is the
phosphate anion, and it usually is reported as an equivalent amount of the
orthophosphate ion (PO4 )-3, the final dissociation product of phosphoric
acid, H3PO4. This dissociation occurs in four steps, giving four possible
phosphate forms: H3PO4, H2PO4-, HP04-2, and P 0 4 - 3 . In the alkalinity
titration, any HP04-2 is converted to H2P04- and appears as bicarbonate.
respectively, of phosphorus. Sea water contains about 0.07 mg/l. A detailed
study of the content of phosphorus in subsurface brines has not been made,
but of the few that have been analyzed, most have contained less than 1
mg/l.
Arsenic
Arsenic is a member of the V A group of elements and probably occurs in

nature mainly in the form of arsenides and sulfarsenides; it rarely occurs in
its elemental form. It is comparatively rare, and the earth’s crust contains
about 0.0005 wt.% of it (Fleischer, 1962). In an acidic environment, the
oxidized ion, A s O ~ - ~is, mobile, and mineral arsenates tend t o be solu-
bilized. The arsenates usually are formed in oxidation zones in contact with
atmosphere and free oxygen, and arsenic will precipitate with ferric iron
hydroxide. Glauconitic sediments have been found which contain up t o 70
ppm of arsenic (Goldschmidt, 1958).
Subsurface oilfield brines may contain arsenic as HAs02- or H2As04,
depending upon the Eh and pH. A low Eh may favor the HAs02- form.
respectively, of arsenic (Mason, 1966). Sea water contains about 0.003 mg/l
and subsurface oilfield brines contain from 0 to 10 mg/l. Compounds con-
taining arsenic sometimes are used in corrosion inhibitors; therefore, infor-
mation concerning well treatments should be obtained before assuming that
any arsenic found occurs naturally.
Oxygen
Oxygen is the most abundant element in the earth’s crust, which contains
about 49 wt.% of it (Fleischer, 1962). I t is capable of existing in many types
of combinations, and even though it is highly active, it occurs extensively in
SULFUR 159
the free form. Most combined oxygen is ionic; however, it forms a covalent
molecule with hydrogen, namely water. It also forms complex oxy-salts with
various metals. The oxygen content of rocks decreases with depth.
The solubility of oxygen in water is primarily a function of temperature
and pressure, and surface waters at ambient conditions may contain 7.63
mg/l a t 3OoC and 11.33 mg/l at 10°C (Hem, 1970). The amounts of
dissolved oxygen in subsurface petroleum-associated waters is usually low,
and in most in situ conditions it is undetectable because of the low redox
potential of the environment. It can cause corrosion problems in the well
pipes, but in most cases it is atmospheric oxygen that mixes with the pro-
duced brine during production operations that causes oxygen corrosion.
Sulfur
Sulfur is a member of the VI A group of elements and is widely dispersed

in sedimentary and igneous rocks as metallic sulfides. The crust of the earth
contains about 0.05 wt.% of sulfur (Fleischer, 1962). Free sulfur often is
related to volcanic activity and can be deposited directly as a sublimate.
Many commercial deposits, however, are associated with sedimentary
gypsum, and probably result from biogenic activity such as that. of anaerobic
bacteria. Large deposits of sulfur are found in caprocks of anhydrite over-
lying some salt domes.
Hydrogen sulfide, often found in oilfield waters, is formed by anaerobic
bacteria. One such species of bacteria is the Desulphouibrio,which obtains
its oxygen from sulfate ions, causing them to be reduced to hydrogen
sulfide.
Sulfur in surface water usually occurs in the form ( S 6 )
complexed with
oxygen as the sulfate anion S04-2. As previously mentioned, the conversion
of oxidized sulfur t o a reduced form commonly involves a biogenic process,
and such a reduction may not occur unless these bacteria are present. The Eh
of subsurface oilfield brines usually is somewhat reducing, and the sulfur
species in such environments can include hydrogen sulfide (H2 S), sulfite
and thionates (S406-’). Detailed studies of the sulfur species in
subsurface brines have not been made, and it is likely that other forms of
sulfur are present in some brines. The temperature, pressure, Eh, pH, and
other constituents in solution all influence the types of dissolved sulfur that
occur in oilfield brines.
Shales, sandstones, and carbonates contain about 2,400, 240, and 1,200
ppm, respectively, of sulfur (Mason, 1966). Sea water contains 900 mg/l of
sulfur as sulfate, and subsurface oilfield brines contain from none up to
several thousand milligrams per liter. The amount of sulfate in the brine is
influenced by bacterial activity and by how much calcium, strontium, and
barium is present. If these three cations are present in relatively high concen-
trations, the amount of sulfate present will be low. However, some brines
containing high concentrations of magnesium and low concentrations of the
other alkaline earth metals may contain high concentrations of sulfate.
Selenium
Selenium is a member of the VI A group of elements and occurs in -2,0,

+4,and +6 valence states, respectively. It is a scarce element with an abun-
dance of about 9 x 10" wt.% of the crust of the earth (Fleischer, 1962).
Large areas of North America are underlain by seleniferous rocks and soils.
These seleniferous rocks are of sedimentary origin and range in age from
Late Paleozoic to Holocene. Selenium is the only known element that can be
absorbed by plants in sufficient amounts to make them lethal when eaten by
animals (Trelease, 1945).Fig. 5.15 illustrates the distribution of seleniferous
vegetation.
Sandstones, shales, and carbonates contain about 0.6, 0.05,and 0.08 ppm,
respectively, of selenium. Sea water contains about 0.004 mg/l of selenium.
A few subsurface oilfield brines from areas where selenium is present in soils
were analyzed at this laboratory, but no selenium was detected in the brines
analyzed. Most brines are present in a petroleum environment under
reducing conditions, and in such an environment, selenium likely is reduced
to the element and precipitated. However, in areas where outcrop water
flows through petroleum-bearing formations, it is possible that selenium in
the form of t)e anion Se03-2 may be present.
Fig. 5.15. Distribution of seleniferous vegetation in the United States.

FLUORINE 161
Fluorine
Fluorine is a member of the VII A group of elements and is the most

electronegative of all the elements. Its ionic radius is 1.33 A, which is about
the size of OH- and O-*;therefore, it enters a variety of minerals. The
earth’s crust contains about 0.03 wt.% of fluorine (Fleischer, 1962). In
solutions, fluorine usually forms the fluoride F- ion; at a low pH, the HFo
form might occur. It also can form strong complexes with aluminum, beryl-
lium, and ferric iron.
Fluorine occurs in several minerals, but the only common industrial
source is fluorspar (CaF2). It occurs as HF or SiF in volcanic emanations,
and even as the free element in (stinkfluss) “stinking fluorspar” of Wolssen-
dorf, Bavaria. The solubility of calcium fluoride (fluorite) in water at 25OC is
about 8.7 ppm of fluoride (Aumeras, 1927);this solubility could be affected
by other dissolved constituents. Sodium fluoride is very soluble, and magne-
sium fluoride is more soluble than calcium fluoride; therefore, a petroleum-
associated water that is deficient in calcium and has been in contact with
rocks containing fluoride minerals will contain appreciable quantities of
fluoride.
Shales, sandstones, and carbonates contain about 740, 270 and 330 ppm
of fluorine, respectively (Mason, 1966). Sea water contains about 1.3 mg/l,
and natural waters with a dissolved solids concentration of less than 1,000
mg/l usually contain less than 1 mg/l of fluoride. However, concentrations
up t o 50 mg/l have been reported (Hem, 1970). Not many subsurface
petroleum-associated brines have been analyzed for fluoride, but a few are
known t o contain up t o 5 mg/l.
Chlorine
Chlorine is a member of the VII A group of elements and is the most

important member of the group with respect to water. The crust of the earth
contains about 0.19 wt.% of chlorine (Fleischer, 1962);some estimates place
the fluorine abundance above the chlorine abundance. Volcanic activity
produces the gas hydrogen chloride and sometimes chlorine, but much less
frequently. The caliche evaporite deposits in Chile contain the perchlorate
ion C104-; however, the mechanism by which it formed is not clear. Several
minerals contain the chloride ion.
The chloride ion does not form low-solubility salts. I t is not easily
adsorbed on clays or other mineral surfaces. It is not significant in oxidation
and reduction reactions, and it forms no important solute complexes.
Chloride is very mobile in the hydrosphere, yet it is relatively scarce in the
earth’s crust. It is the predominant anion in sea water and in most
petroleum-associated waters. It is found in all natural waters, and its average
concentration in rainwater is about 3 mg/l (Hem, 1970). Chloride salts are
very soluble; therefore, chloride is usually not removed from solution except
during freezing or evaporation processes and in hyperfiltration, as water
moves through some types of clay beds (White, 1965).
respectively, of chloride. Sea water contains about 19,000 mg/l of chloride,
the principal anion in the sea. The chloride content of the hydrosphere is
much greater than can be accounted for by weathering of rocks, and it has
been postulated that the primordial atmosphere may have been rich in
chlorine compounds. The volcanic emission of chlorine gases appears a more
plausible explanation, however.
Oilfield brines usually contain relatively high concentrations of chloride;
in some brines the concentration may be 200,000 mg/l or more. Chloride
usually is the predominant anion in oilfield brines. Table 5.1 illustrates how
its concentration can increase in an evaporite-associated brine. Evaporation
probably is the only geochemical process which appreciably affects the
chloride content of the seas.
Bromine
Bromine is a member of the VII A group of elements and it behaves

somewhat similarly t o chlorine. The crust of the earth contains about 0.0005
wt.% of bromine (Fleischer, 1962). It usually occurs as the ion bromide Br-,
and it does not form its own minerals when sea water evaporates (Valyashko,
1956). It forms an isomorphous admixture with chloride in the solid phases.
The order of crystallization (see Table 5.1) is halite (NaCl), sylvite (KCl),
carnallite (MgC12-KC1*6H20), and/or kainite (MgS04 *KC1=3H2 0), and at
the eutectic point, bischofite (MgCl? -6H2 0). Each of these chlorides
entrains bromide in the solid phase. This distribution accounts for the rela-
tive enrichment of bromide in the liquid phase because with each crystalliza-
tion more bromide is left in solution than is entrained in the solid phase.
Mun and Bazilevich (1962) reported that, in fresh-water lakes, bromide
accumulates in the muds, that its concentration is proportional t o the
organic-matter concentration in the sediments, and that it is not influenced
by the pelitic fraction. In muds of salt lakes, the higher the bromide concen-
trations in the brine, the higher it is in the muds. In general, the bromide
content in the pore solutions increased with depth, but the bromide content
in muds decreased with depth, owing to more complete decomposition of
organic bromine compounds.
Bromide is two t o three times more concentrated in carnallite than in
sylvite and five to ten times more than in halite (Myagkov and Burmistrov,
1964). Apparently, concentration and dilution are responsible for the com-
plex distribution of bromide in rocks of a carnallite zone. The determining
factor in the replacement of chloride by bromide is the mineral composition
rather than the bromide concentration in the brine.
BROMINE 163
Braitsch and Herrmann (1963) found that the absolute bromide content of
rocks can be used to determine primary and secondary paragenesis. Distribu-
tion of bromide between solution and crystals of halite, sylvite, carnallite,
and bischofite, and the effects of other ions plus temperatures between 25"
and 83OC, confirm this. This method was also applied to determine the
temperature of primary potash deposits. An investigation of the bromide/
sodium chloride relation in salt deposits revealed that bromide can be used
to determine the stratigraphy of evaporite-salt deposits (Baar, 1963).
Derivation of theoretical profiles of bromide thickness versus salt
thickness indicated that, with constant inflow, evaporation, and reflux, the
thickness profiles were all monotomic logarithmic functions. The irregular
and high bromide concentrations of some salt deposits were attributed to
inflow of bromide-rich bitterns from an adjacent potash basin (Holser,
1966).
Shales, sandstones, and carbonates contain about 4,1,and 6 ppm, respec-
tively, of bromide (Mason, 1966). Sea water contains about 65 mg/l of
bromide, and subsurface petroleum-associated brines contain from less than
50 to more than 6,000 mg/l of bromide. Fig. 5.16 illustrates the bromide
concentration plotted versus the chloride concentration for some subsurface
brines taken from Tertiary, Cretaceous, and Jurassic age sediments. This plot
indicates that the waters from these Tertiary age sediments are depleted in
bromide relative to a normal evaporite brine, whereas those from the
Cretaceous and Jurassic age sediments are enriched'in bromide.
BROMIDE, mg / I
Fig. 5.16. Comparison of the bromide concentrations in some formation waters from
Tertiary (T), Cretaceous (C), and Jurassic (J) age sediments from Louisiana with an
evaporating sea water.
P
M
PI
300 1,000 3000
I
10
BROMIDE, m g l l
Fig. 5.17. Comparison of the bromide concentrations in some formation waters from
Pennsylvanian (P) and Mississippian (M) age sediments from Oklahoma with an evapo-
rating sea water.
Fig. 5.17 is a similar plot for some brines taken from some Pennsylvanian
and Mississippian age sediments. The bromide concentrations in these brines
do not appear t o be significantly different.
Brines containing 1,500 to 8,000 mg/l of bromide, with calcium and
magnesium chloride as the major constituents, are formed by evaporation of
sea water and associated sedimentation rather than by dissolution of salts.
Increase in temperature causes a phase shift in the solid and brine phases,
resulting in an increase of bromide in solution.
Iodine
Iodine is a member of the VII A group of elements, and of the four

members discussed in this chapter, it is the least abundant, since it comprises
only about 3 x lo-' wt.% of the earth's crust (Fleischer, 1962). I t forms
three minerals of its own; namely, iodoargyrite (AgI), iodoembolite
[Ag(Cl,Br,I)], and miersite [(Ag,Cu)I]. Marine plants, such as kelp and plank-
ton algae, concentrate iodine.
The distribution of iodide in marine and oceanic silts and interstitial
waters indicates that near-shore ocean Sediments contain more iodide than
deep-sea sediments. Red clays and calcareous sediments contain less iodide
than organic-bearing argillaceous sediments. The iodide concentration in the
marine and oceanic sediments decreases with depth, but the iodide concen-
IODINE 165
tration in the interstitial waters increases with depth (Shishkina and Pavolva,
1965).
The iodide in bottom water layers and in the interstitial water of muds in
some Japanese lakes was found t o be selectively captured by flocculated
iron, manganese, and aluminum hydroxides which sank to anaerobic layers
(Sugawara et al., 1956). Reduction of the hydroxides releases iodide to the
bottom waters. However, the release of iodide is incomplete, and the
flocculates reach the bottom muds where the Eh is even more negative,
resulting in high accumulation of iodide in interstitial water of muds.
The primary source of organic matter in marine and oceanic basins is
photosynthesis by plankton algae. Algae are directly or indirectly the food
resource of all the remaining life in the basins, and the proliferation rate
differential and the types of feeding organisms influence the sediment deposi-
tion rate as well as the amount of iodide and bromide in the sediment
(Bordovskii, 1965).
Shales, sandstones, and carbonates contain about 2.2, 1.7, and 1.2 ppm,
respectively, of iodide (Mason, 1966). Sea water contains about 0.05 mg/l,
and most subsurface petroleum-associated brines contain less than 10 mg/l;
however, some have been found to contain up t o 1,400 mg/l.
Fig. 5.18 is a plot of the chloride concentrations versus the iodide concen-
trations for some brines taken from some Pennsylvanian and Mississippian
age sediments. Iodide is tremendously enriched in all of these brines com-
pared to the normal evaporite-associated brine. Some mechanisms such as
leaching or solubilization of iodine, iodate, or iodide compounds, ion fil-
tration, anion exchange, and desorption had t o occur, t o account for this
enrichment of iodide in the aqueous phase. A similar plot for some waters
taken from Tertiary, Cretaceous, and Jurassic age sediments gave similar
results except that these particular brines were not as heavily enriched in
iodide.
The iodide concentration of some subsurface waters is dependent on the
proximity of argillaceous deposits containing organic matter, rather than on
dissolved mineralization. Gas may play an important part in the accumula-
tion of iodide in subsurface waters. Some gas structures are bounded by
iodide-rich waters, and the iodide content is depleted at a distance from the
gas structure (Ovchinnikov, 1960).
Studies of some reservoirs, Holocene to Miocene in age, in lagoonal sedi-
mentary basins of thick sediments with wide areal extent, indicate that a
genetic relation exists between iodide in the formation waters and the
accompanying natural gas (Marsden and Kawai, 1965). Possibly the high
concentrations of iodide are the result of concentration by algae and other
marine organisms from ancient sea waters; their remains became part of the
sediments, and later the iodide was solubilized. However, because the iodide
usually is strongly incorporated in the sediment, such sediments must con-
tain large quantities of iodide, and other mechanisms must operate to solu-
b i k e the iodide in associated waters.
1 , 5 0 0 3
1,200
I#OOO
~ w mevaporite
i CUM ,
, '
/
/1
/-
\
/
/
cn 500 /
w'
2 200
sI loot vs
M P
P
O t MM
50E
20
M
M F P
M
MhP
IODIDE, mg/l
Fig. 5.18. Comparison of the iodide concentrations of some formation waters from
Pennsylvanian (P) and Mississippian (M) age sediments from Oklahoma with an evapo-
rating sea water.
Theoretically, only iodate is thermodynamically stable in sea water

(SillCn, 1961). The exact form of iodine in oilfield brines has not been
investigated. These forms probably will vary with the salinity, Eh, and other
factors. Sugawara and Terada (1957)established that both iodide and iodate
are present in comparable amounts in sea water. Biologists found that
iodine-concentrating algae ultilize only the iodate form (Shaw, 1962).
Significance of some physical properties
Redox potential
The redox potential often is abbreviated as Eh, and may also be referred
to as oxidation potential, oxidation-reduction potential, or pE. It is expres-
sed in volts, and at equilibrium it is related to the proportions of oxidized
and reduced species present. Standard equations of chemical thermo-
dynamics express the relationships.
Eo is the standard potential of a redox system when unit activities of
participating substances are present under standard conditions. Eo is related
to standard free energy change in a reaction by the equation:
where n is the number of unit negative charges (electrons) shown in the

PHYSICAL PROPERTIES 167
redox reaction and f is the Faraday constant in units that give a potential in
volts (94,484 absolute coulombs). Standard free energy values are given in
texts such as that of Latimer (1952).
When the system is not under standard conditions, the redox potential is
expressed by the Nernst equation:
Eh = E o +R
-
nf
T
log
(oxidized species)
(reduced species)
where R is the gas content (1.987 cal. degree mole), and T is the temperature
in degrees Kelvin. Geochemical literature and biochemical literature, such as
that of Pourbaix (1950), present increasing positive potential values to repre-
sent increasing oxidizing systems and decreasing potential values to represent
reducing systems. The sign of Eh used in this manner is opposite to standard
American practice in electrochemistry.
Zobell (1946) established basic procedures for measuring the Eh of
geologic-related materials. The Zobell solution containing 0.003M potassium
ferrocyanide and 0.003M potassium ferricyanide in a 0.1M potassium chlo-
ride solution has an Eh of 0.428 V at 25OC. Minor temperature variations can
be calculated using the equation:
Eh = 0.428+).0022 ( t - 25)
where t = temperature of the sample in degrees Celsius.

Garrels and Christ (1965) describe procedures for determining Eh
equilibria of mineral substances. Particularly useful are the procedures de-
scribed for constructing diagrams showing fields of stability for various
mineral substances as functions of pH and Eh. Fig. 5.19 is an Eh/pH
diagram. Such stability field diagrams might be constructed for the sub-
stances comprising petroleum and should be of considerable help in under-
standing the mechanisms of origin, accumulation, and chemical stability of
petroleum. Unfortunately, this approach does not yield simple results
because most oxidation reactions involving hydrocarbons and other
petroleum constituents are not reversible in the usual sense. Furthermore,
thermodynamic data are available for only a small fraction of the large
number of reactions and products that are possible.
Attempts t o obtain useful results from Eh measurements in natural media
involve numerous difficulties. In a natural medium, such as petroleum-
associated water, there are many variables, none of which is controlled,
which individually or collectively may have little or great influence on Eh
measurements made on the water. Many chemical substances, such as ferric
or ferrous ions, various organic oxidation-reduction systems, sulfides, and
sulfates, may be present in the water in large or small amounts. Even con-
trolled systems in the laboratory often produce unaccounted-for variances.
In the field, the lack of knowledge of actual participating species may seri-
ously impair proper interpretation of Eh readings. Eh measurements made
1,200 - H2O
000 -
600 1 I
< 400-
I /Modern sea water
z I
W
200 - Acid I Alkaline
Fa++
I
on poorly poised media (media with poor Eh stability), such as some oilfield
waters, involve additional uncertainties. Response of electrodes in such solu-
tions is sluggish, and electrodes are easily contaminated with trace amounts
of substances which will produce invalid readings.
In the natural environment, reactions occur that involve protons and elec-
trons, such as:
FeS04 + 2H20 * S04-2 + FeO-OH + 3H+ + e-
Such reactions depend upon both the pH and the Eh of the system, and
the equilibrium line of such reactions is Eh = E o - 59 a/n pH mV, where a is
the number of protons.
Knowledge of the redox potential is useful in studies of how compounds
such as uranium (Naumor, 1959), iron, sulfur (Hem, 1960), and other miner-
als (Cloke, 1966; Pirson, 1968) are transported in aqueous systems. The
solubility of some elements and compounds is dependent upon the redox
potential and the pH of their environment. The Eh/pH diagram shown in
Fig. 5.19 can be used to predict that ferrous ions are the more common form
of dissolved iron and that ferric ions will precipitate in an oxidizing environ-
ment if the pH is above 1. Similar diagrams can be drawn for other con-
stituents.
Some water associated with petroleum is “connate” water, and Fig. 5.19
indicates that such water has a negative Eh; this has been proven in various
field studies (Buckley et al., 1958). The Eh of some petroleum-associated
waters in the Anadarko Basin ranged from -270 mV to -300 mV (Collins,
1969a).
Knowledge of the Eh is useful in determining how t o treat a water before
it is injected into a subsurface formation (Ostroff, 1965). For example, the
Eh of the water will be oxidizing if the water is open to the atmosphere, but
if it is kept in a closed system in an oil-production operation, the Eh should
not change appreciably as it is brought to the surface and then reinjected. In
such a situation, the Eh value is useful in determining how much iron will
stay in solution and not deposit in the well bore.
Organisms that consume oxygen cause a lowering of the redox potential.
In buried sediments, it is the aerobic bacteria that attract organic con-
stituents which remove the free oxygen from the interstitial water. Sedi-
ments laid down in a shoreline environment will differ in degree of oxidation
as compared t o those laid down in a deep-sea environment (Pirson, 1968).
For example, the Eh of the shoreline sediments may range from -50 t o 0
mV, but the Eh of deep-sea sediments may range from -150 to -100 mV.
The aerobic bacteria die when the free oxygen is totally consumed; the
anaerobic bacteria attack the sulfate ion, which is the second most important
anion in the sea water. During this attack, the sulfate reduces t o sulfite and
then t o sulfide; the Eh drops to -600 mV; H2S is liberated, and CaC03
precipitates as the pH rises above 8.5 (Dapples, 1959).
The term pH means the logarithm (base 1 0 ) of the reciprocal of the

hydrogen-ion concentration, and the pH of pure water at 25OC is 7.0, which
means that there is lo-’ ‘mole per liter of H+ in solution. When other
constituents are solubilized by water, the pH probably will change because
the chemical equilibrium shifts as new ions combine with H+ or OH-. The
presence of slightly dissociated acids or bases will tend to buffer the solu-
tion, and the addition of H+ or OH- will shift the pH only a small amount
until the acids or bases are changed to salts.
The pH of oilfield waters usually is controlled by the carbon dioxide-
bicarbonate system. Because the solubility of carbon dioxide is directly
proportional to temperature and pressure, the pH measurement should be
made in the field if a close-to-natural-conditions value is desired. The pH of
the water is not used for water identification or correlation purposes, but it
will indicate possible scale-forming or corrosion tendencies of a water. The
pH also may indicate the presence of drilling-mud filtrate or well-treatment

chemicals.
A detailed study indicates that virtually no environment exists on or near
the earth’s surface where the pH/Eh conditions are incompatible with
organic life (Baas Becking et al., 1960). Because COz is the main byproduct
of organic oxidation and the building material of plant and much bacterial
life, it must be expected t o play a dominant role. It dissolves in HzO,
producing the bicarbonate ion and a free hydrogen ion. The concentration of
the hydrogen ion is 1 x moles per liter (pH 7) at 25OC in pure water,
but when saturated with COz, it rises t o 1 x lo-’ moles per liter (pH 5).
The equilibrium conditions of carbon dioxide, carbonic acid, and the bicar-
bonate ion are:
Hz 0 + COz * HzCO, * HC03- + H+ * 2H+ + C03-’

and the pH of each equilibrium in ocean water is pH 5, pH 6.3, and pH 10.3.
This reaction moves to the right with increasing temperature in a closed
system. In the presence of organic constituents, the equilibria are modified,
and the pH range can extend from 2 to 12.
Fig. 5.20. Changes in pH as a result of the addition of carbonate ions to distilled water
and water solutions containing sodium and chloride ions.
The pH of concentrated brines usually is less than 7.0, and the pH will rise
during laboratory storage, indicating that the pH of the water in the reservoir
probably is appreciably lower than many published values. Addition of the
carbonate ion to sodium chloride solutions will raise the pH, as illustrated in
Fig. 5.20. If calcium were present, calcium carbonate would precipitate. The
reason why the pH of most oilfield waters rises during storage in the labora-
tory is because of the formation of carbonate ions as a result of bicarbonate
decomposition.
Ionic radii
Table 5.VI contains data concerning the radii of the nonhydrated ions, the
hydrated ions, the ionic potential, and the polarization. The size of the ions
is of interest concerning the mobilities or the relative transport coefficient of
a given ion through a clayshale membrane system or the replacement
coefficient in a clay ion-exchange system. The ionic potential is of interest
because elements with low ionic potentials are the most likely to remain in
true solution. The polarization, which is equivalent to the valency divided by
the ion radius, is of interest because the larger the polarization, the lower the
replacing power in an exchange system (Collins, 1970).
The ionic potentials of the constituents involved in the diagenesis are
important (Hem, 1960). Those that stay in true ionic solution to rather high
pH levels include Na+, K+, Mg+’, Fe+’, Mn+’, Ca+’, Sr+*,and Ba+’ ;they
are the soluble cations, and their ionic potentials range from 0.3 to 1.3,
where the ionic potential is the ratio between the ionic charge and the ionic
radius. Constituents that are precipitated by hydrolysis are those with ionic
potentials of 3-12 and include such ions as A P 3 , Fe+3, S P 4 , and M r P 4 .
Constituents which form soluble complex ions and usually go into true ionic
solution include B+3, C 4 ,N + 5 , P+’, S 6 ,and Mn+’ ;their ionic potentials
are over 12. In general, the hydroxides of the soluble cations possess ionic
bonds; therefore, they are soluble. The hydrolysates, or those ions precipi-
tated by hydrolysis from hydroxyl bonds, and the soluble complex ions both
have hydrogen bonds.
TABLE 5.VI
Radii, valence, ionic potentials, and polarization
Constituents Nonhydrated Valence Hydrated Ionic Polarization

radius (A) radius (A) potential
Lithium 0.60 +1 3.82 0.60 1.67

Sodium 0.95 +1 3.58 0.95 1.05
Potassium 1.33 +1 3.31 1.33 .75
Calcium 0.99 +2 4.12 0.50 2.02
Magnesium 0.65 +2 4.28 0.33 3.08
Strontium 1.13 +2 4.12 0.57 1.77
Barium 1.35 +2 4.04 0.68 1.48
Boron 0.23 +3 - 0.08 -
Chloride 1.81 -1 3.32 1.81 0.55
Bromide 1.95 -1 3.30 1.95 0.51
Iodide 2.16 -1 3.31 2.16 0.46
Sulfate 2.90 -2 3.79 1.45 0.69
Density
Equations that were developed for sea water can be applied to oilfield
waters to obtain approximate values for engineering studies. The density
values (ao) at O°C and atmospheric pressure are related to the chlorinity (CZ)
as follows:
-0.069 + 1.4708 CZ - 0.00157 C12 + 0.0000398 C13 = UO
where CZ = chlorinity (see Table 3.111).

The density is very dependent upon temperature:
where D = a complex function of ao, and temperature and D values can be

obtained from Knudsen’s Hydrographic Tables (Knudsen, 1901).
Vapor pressure
The relative lowering of the vapor pressure of oilfield water can be calcu-
lated with the following equation:
Ap/po = 0.538 x S
where p o = the vapor pressure of distilled water at the same temperature,

and S = the salinity (see Table 3.111) (Kellog and Company, 1956, 1966,
1968).
Boiling point
A first approximation of the boiling point elevation can be calculated

from:
At = 0.0158S
where S = the salinity.
Freezing point
An empirical equation which can be used t o estimate the freezing points

is :
t = 4.0086 - 0.064633 ((TO) - 0.0001055 ( 0 0 ) ~
See “Density” for an explanation of terms.

Viscosity
The viscosity will increase with decreasing temperature and with increas-
ing salinity. The viscosities of sodium chloride solutions of the same ionic
strength can be used to estimate oilfield water vicosities.
Osmotic pressure
A relationship between osmotic pressure ( P o ) and the depression of the

freezing point at 0°C is (in atmospheres):
The osmotic pressure at other temperatures can be estimated (Kellog and

Company, 1956,1966,1968):
Po x (1+ 0.00367t)
Specific heat
The values for the specific heat, c p , of oilfield waters can be approxi-
mated from the values of an equivalent sodium chloride solution.
Thermal conductivity
The thermal conductivity coefficient, A, can be calculated from thermal

capacities because the ratio of thermal conductivities of two materials is the
same as that of the thermal capacities of equal volumes. The values for X at
various temperatures are available in a “Saline Water Conversion Technical
Data Book” (Kellog and Company, 1956,1966,1968).
Surface tension
The surface tension of an oilfield water increases with decreasing tempera-

ture and with increasing salinity. An empirical formula which can be used t o
calculate it is:
75.64 - 0.144t + 0.0399 Cl = surface tension (dynes/cm*)
where t = temperature in Celcius, and CZ = the chlorinity (see Table 3.111;

Kellog and Company, 1956,1966,1968).
References
Ahrens, L.H., 1965. Distribution o f the Elements in Our Planet. McGraw-Hill, New York,
N.Y., 110 pp.
Aumeras, M., 1927. Studies of ionic equilibria, 11. Equilibrium of calcium fluoride in dilute
hydrochloric. J. Chem. Phys., 24:548-571.
Baar, C.A., 1963. How to use the bromine test to determine the stratigraphical position in
rock salt series. Neues Jahrb. Mineral. Geol. Palaontol., Monatsh., 7( 1):145-153.
Baas Becking, L.G.M., Kaplan, I.R. and Moore, D., 1960. Limits of the natural environ-
ment in terms of pH and oxidation-reduction potentials. J. Geol., 68:243-284.
Bordovskii, O.K., 1965. Source of organic matter in marine basins. Mar. Geol.,
3( 1/2):5-31.
Braitsch, 0. and Herrmann, A.G., 1963. Zur Geochemie des Broms in Salinaren Sedi-
menten, Teil I. Experimentelle Bestimmung der Br-Verteilung in verschiedenen
naturlichen Salzsystemen. Geochim. Cosmochim. Acta, 27 :361-391.
Brasted, R.C., 1957. The halogens. In: M.C. Sneed, J.L. Lewis and R.C. Brasted (Editors),
Comprehensive Inorganic Chemistry, 3. Van Nostrand, New York, N.Y., 250 pp.
Buckley, S.E., Hocott, C.R. and Taggart, Jr., M.S., 1958. Distribution of dissolved
hydrocarbons in subsurface waters. In: L.G. Weeks (Editor), Habitat of Oil. American
Association of Petroleum Geologists, Tulsa, Okla., pp.850-882.
Burst, J.F., 1969. Diagenesis of Gulf Coast clayey sediments and its possible relation to
petroleum migration. Bull. A m . Assoc. Pet. Geol., 53:73-93.
Cloke, P.L., 1966. The geochemical application of E h - p H diagrams. J. Geolog. Ed.,
14:140-1 48.
Collins, A.G., 1969a. Chemistry of some Anadarko Basin brines containing high concen-
trations of iodide. Chem. Geol., 4:169-187.
Collins, A.G., 1969b. Solubilities of some silicate minerals in saline waters. U.S. Off.
Saline Water Res. Dev. Progr. Rep., No. 472, 27 pp.
Collins, A.G., 1970. Geochemistry of some petroleum-associated waters from Louisiana.
US.Bur. Min. Rep. Invest., No.7326, 31 pp.
Collins, A.G., Castagno, J.L. and Marcy, V.M., 1969. Potentiometric determination of
ammonium nitrogen in oilfield brines. Environ. Sci. Technol., 3:274-275.
Dall’Aglio, M., 1968. The abundance of mercury in 300 natural water samples from
Tuscany and Latium (Central Italy). In: L.H. Ahrens (Editor), Origin and Distribution
o f the Elements. Pergamon Press, Oxford, p.1065-1081.
Dapples, E.C., 1959. The behavior of silica in diagenesis. In: H.A. Ireland (Editor), Silica
in Sediments - SOC.Econ. Paleontol. Mineral., Spec. Publ., No.7, pp.36-55.
Davis, J.W. and Collins, A.G., 1971. Solubility of barium and strontium sulfates in strong
electrolyte solutions. Environ. Sci. Technol., 5:1039-1043.
Fleischer, M., 1962. Recent estimates of the abundances of the elements in the earth’s
crust. U.S. Geol. Surv. Circ., No.285, 9 pp.
Garrels, R.M. and Christ, C.L., 1965. Solutions, Minerals, and Equilibria. Harper and
Row, New York, N.Y., 450 pp.
Goldberg, E.D., 1963. The oceans as a chemical system. In: M.N. Hill (Editor), Composi-
tion of Sea Water: Compamtive and Descriptive Oceanography. Interscience, New
York, N.Y., 2:3-25.
Goldschmidt, V.M., 1958. Geochemistry. Oxford University Press, London, 730 pp.
Hem, J.D., 1960. Some chemical relationships among sulfur species and dissolved iron :
chemistry of iron in natural water. U.S.Geol. Surv., Water Supply Paper, No, 145942,
pp.57-73.
Hem, J.D., 1970. Study and interpretation of the chemical characteristics of natural
waters. U.S. Geol. Suru., Water Supply Paper, No. 1473, 363 pp.
Hemley, J.J., 1953. A study of lead sulfide solubility and its relation to ore deposition.
Econ. Geol., 48:113-137.
REFERENCES 175
Holser, W.T., 1966. Bromide geochemistry of salt rocks. In: Jon L. Rau (Editor), 2nd
Symposium on Salt. The Northern Ohio Geological Society, Cleveland, Ohio,
1:248-275.
Karasik, M.A., Goncharov, Yu. and Vasilevskaya, A.Ye., 1965. Mercury in waters and
brines of the Permian salt deposits of Donbas. Geochem. Int., 2:82-87.
Kazmina, T.I., 1951. The boron-chloride coefficient in oilfield waters. Dokl. Akad. Nauk
S.S.S.R . , 77 :301.
Kellog, M.W. and Company, 1956, 1966, 1968. Saline Water Conversion Technical Data
Book and Supplements 1 and 2. (Catalog No. 1, 1.7712: Sa3, Supp.1, Supp.2, Supt. of
Documents), U.S. Government Printing Office, Washington, D.C.
Khitarov, N.I. and Pugin, V.A., 1966. Behavior of montmorillonite under elevated
temperatures and pressures. Geochem. Int., 3:621-626.
Knudsen, M., 1901. Hydrographic Tables. G.E.C. Gadd, Copenhagen, 6 3 pp.
Larson, T.E. and King, R.M., 1954. Corrosion by water at low flow velocity. J. A m .
Water Works Assoc., 46 :1-9.
Latimer, W.M., 1952. Oxidation Potentials. Prentice-Hall, New York, N.Y., 2nd ed., 352
PP.
Lyon, T.L. and Buckman, H.D., 1960. The Nature and Properties o f Soils. Macmillan, C.,
Marsden, S.S. and Kawai, K., 1965. “SuiyBsei-Ten’nengasu” - a special type of Japanese
natural gas deposit. Bull. Am. Assoc. Petrol. Geol., 49:286-295.
Mason, B., 1966. Principles of Geochemistry. John Wiley and Sons, New York, N.Y., 329
PP.
Miholic, S., 1947. Ore deposits and geologic age. Econ. Geol., 42:713.
Mitgarts, B.B., 1956. The composition of subterraneous waters in its value for petroleum
prospecting, according to the data from the Fergana region. Vses. Nauch. Issled., Geol.
Inst., Vopr. Neftepoiskovi Gidrogeol., 18:58-93 (in Russian).
Moeller, T., 1954. Inorganic Chemistry. John Wiley and Sons, New York, N.Y., 966 pp.
Mun, A.I. and Bazilevich, Z.A., 1962. Distribution of bromine in lacustrine bottom muds.
Geochemistry, 2:199-205.
Myagkov, V.F. and Burmistrov, D.F., 1964. Distribution of bromine in the carnallite beds
of the Verkhnekamsk deposit. Geochem. Int., 1:701-703.
Naumor, G.B., 1959. Transportation of uranium in hydrothermal solution as a carbonate.
Geochemistry, 1:5-20.
Odum, H.T., 1951. Notes on the strontium content of sea water, celestite radiolaria, and
strontianite snail shells. Science, 114:211.
Ostroff, A.G., 1965. Introduction t o Oilfield Technology. Prentice-Hall, Englewood
Cliffs, N.J., 412 pp.
Ovchinnikov, N.V., 1960. Patterns in the alteration of the chemical composition of
subsurface waters of the Azqv-Kuban Trough and the distribution of iodine and
bromine therein. Izv. Vyssh. Uchebn., Zaved., Geol. Razvedka, 3:134-138.
Pirson, S.J., 1968. Redox log interprets reservoir potential. Oil Gas J., 66:69-75.
Pirsson, L.V. and Knopf, A., 1947. Rocks and Rock Minerals. John Wiley and Sons, New
York, N.Y. 3rd ed., 179 pp.
Pourbaix, M.J.N., 1950. Thermodynamics o f Dilute Aqueous Solutions. Longmans, Green
and Co., New York, N.Y., 151 pp.
Rittenhouse, G., Fulton, Jr., R.B., Grabowski, R.J. and Bernard, J.L., 1969. Minor
elements in oilfield waters. Chem. Geol., 4:189-209.
Shaw, T.I., 1962. Physiology and biochemistry of algae. In: R.A. Lewin (Editor),
Halogens. Academic Press, New York, N.Y., pp. 247-253.
Shishkina, O.V. and Pavolva, G.A., 1965. Iodine distribution in marine and oceanic muds
and in their pore fluids. Geochem. Int., 2:559-565.
Sillbn, L.G., 1961. Physical chemistry of sea waters. A m . Assoc. Adv. Sci. Publ., No.67,
pp. 549-581.
Sill&, L.G. and Martell, A.E., 1964. Stability constants of metal-ion complexes. Chem.
Soc. Lond., Spec. Publ., No.17, 754 pp.
Sugawara, K. and Terada, K., 1957. Iodine distribution in the western Pacific Ocean. J.
Earth Sci., 5:81-102.
Sugawara, K., Koyama, T. and Terada, K., 1956. Co-precipitation of iodide ions by some
metallic oxides with special reference to iodide accumulation in bottom water layers
and interstitial waters of muds in some Japanese lakes. UNESCO/NSRIC/65,presented
at Int. Congr. Theor. Appl. Limnol., Helsinki, Finland, 1956, 11 pp.
Trelease, S.F., 1945. Selenium in soils, plants, and animals. Soil Sci., 160:125-131.
Valyashko, M.G., 1956. Geochemistry of bromine in processes of halogenesis and use of
bromine content as a genetic and prospecting criterion. Geochemistry, 6:33-49.
White, D.E., 1957. Magmatic, connate and metamorphic waters. Bull., Geol. Soc. A m . ,
68 :1669.
White, D.E., 1965. Saline waters of sedimentary rocks. In: A. Young and J.E. Galley
(Editors), Fluids in Subsurface Environments - A m . Assoc. Pet. Geol., Mem. 4 ,
pp.342-366.
White, D.E., Hem, J.D. and Waring, G.A., 1963. Data of geochemistry. U.S. Geol. Surv.
Prof. Paper, No.440-F, pp. FI-F67.
Zobell, C.E., 1946. Studies on the redox potential of marine sediments. Bull. A m . Assoc.
Pet. Geol., 30:477-513.
Chapter 6. ORGANIC CONSTITUENTS IN SALINE WATERS
Water is a peculiar solvent and has been considered t o possess at ambient

temperature a quasi-crystalline, open structure which will allow solute mole-
cules to fill the space between the lattice points (Eley, 1939). As a nonpolar
molecule dissolves in water at ambient temperature, the structure of the
water in its immediate vicinity becomes more crystalline, or a microscopic
“iceberg” surrounds the solute (Frank and Evans, 1945). Water also has been
considered t o be an equilibrium mixture of an icelike and a close-packed
structure, and with a molecule of gas as a solute, it reacts with the icelike
structure filling one of the cavities to form a gas-hydrate and shifting the
equilibrium from the close-packed structure to the icelike structure (Namoit,
1961).
Another theory is that water is composed of clusters of highly hydrogen-
bonded molecules which are surrounded by a closely packed structure of
monomeric water. These flickering clusters form and dissolve perpetually as
a result of local energy fluctuations. Therefore, a water molecule can have a
solute molecule as a neighbor along with its four H-bonded water neighbors.
Interactions between the solute and water molecules will depress the energy
level of the tetrabonded water molecule. However, large numbers of water
molecules surround an unbonded molecule, and if it acquires a solute
neighbor after the latter replaces a water molecule, the energy level is raised.
Changes in the energy levels cause a shift of water molecules between various
levels in accordance with the Boltzmann distribution law, giving an increase
in the “icelikeness” and an increase in the clusters of water molecules near
the surface of the solute molecule (Nemethy and Scheraga, 1962).
When a hydrocarbon molecule transfers from the pure liquid to the solu-
tion hydrocarbon-water, interactions are established while hydrocarbon-
hydrocarbon interactions are broken. The amount, kind, and state
(suspended, dissolved, or colloidal) of organic matter in petroleum-associated
waters is important in determining the origin and migration of petroleum,
and in problems concerning pollution of fresh waters by petroleum-
associated waters. Probably the most plausible theory concerning the origin
of petroleum is that it originated from organic constituents which are
recognized as remnants or degradation products of living organisms of past
ages; these organic source materials entered fine-grained aquatic sediments
where biochemical and chemical conversions and fractionations occurred
(Erdman, 1965). As increased sedimentation took place, the resulting over-
burden pressure and compaction caused the interstitial water, which con-
178 ORGANIC CONSTITUENTS IN SALINE WATERS
tained minute quantities of hydrocarbon, to be squeezed out of the con-

solidating sediments, and subsequently the oil was accumulated in sands and
left behind in structural traps (Kidwell and Hunt, 1958). Obviously, the
waters associated with petroleum play a very important part in the origin,
migration, accumulation, and subsequent production of petroleum - the
accumulation and production of petroleum being totally dependent upon
hydraulic flow in response to geostatic and hydrostatic pressures.
Consider briefly that 99% of the oil found and produced typically occurs
within the pore spaces of sedimentary rocks (Hedberg, 1964).About 59% of
the production comes from sandstone reservoirs, 40% from carbonates, and
1% from other types of rock. Petroleum in igneous and metamorphic rocks
occurs primarily in fracture pore spaces and probably has migrated to these
rocks from its place of origin.
The solubilities of petroleum hydrocarbons in water increase with
temperature and decrease as the salinity of the water increases. A tempera-
ture drop from 150” t o 25°C reduces the solubility of petroleum in water by
a factor of 4.5-20.5. Such a mechanism can account for the accumulation
of petroleum because as upward moving subsurface waters containing
dissolved hydrocarbons decrease in temperature and pressure and meet more
saline waters, they will release dissolved hydrocarbons (Price, 1973).
Information concerning dissolved organic matter in sea water was
published as early as 1892 (Duursma, 1965). Palmitic acid, stearic acid,
acrolein, and organic nitrogen were tentatively identified. The dissolved
organic matter was found t o be about 2 mg/l in the open sea, increasing to
about 15 mg/l in water taken near the coast of Greece, all of which was
attributed t o saponification of the fats of dead organic organisms. Phyto-
plankton organisms comprise most living marine organic matter, 10% of
which eventually becomes animal matter. The bulk of the organic particulate
matter in the sea results from dead animal matter, but the dissolved organic
constituents appear to be derived from dead phytoplankton and detritus
rather than excretions from living cells.
Decomposition of organic matter results primarily from the activities of
heterotrophic bacteria. Organic matter decomposes more rapidly in a near-
shore environment, where there is an abundance of such matter and bacteria,
than in a deep-sea environment, where both the matter and bacteria are
diluted. The dissolved organic matter can be classified into groups as follows:
(1)nitrogen-free (for example, carbohydrates); (2)nitrogen-containing (for
example, proteins); (3)lipids (for example, esters of fatty acids); and (4)
complexes comprised of mixtures of the preceding three groups (for
example, humic acids).
Nitrogen-free organic compounds
Many petroleum-associated waters contain methane; however, in Japan,

there is a type of natural-gas deposit called “suiy6sei-tenynengasuy’,
a dry gas,
NITROGEN-FREE ORGANIC COMPOUNDS 179
which is found dissolved in subsurface brines (Marsden and Kawai, 1965).

The major reservoirs in which it is found are marine or lagoonal sedimentary
basins with thick sediments and of wide areal extent. Some of the associated
brines contain more than 80 mg/l of iodide, which is the only commercial
source of iodine in Japan. Some of these brines also contain dissolved
ethane, propane, isobutane, butane, isopentane, and pentane.
An interesting note is that large Soviet deposits of natural gas in a solid
state totaling about 15 trillion m3 were reported to the U.S.S.R. Committee
for Inventions and Discoveries. According to U.S.S.R. investigators, mole-
cules of ground water attract molecules of natural gas and convert them to a
hydrate, which resembles silvery-grey ice, where the pressure is 250 atm and
the temperature 25°C or less. 1m3 of the hydrate contains up t o 200 m3 of
natural gas. These solid hydrate deposits are found in permafrost zones at
depths to 2,500 m. Because of the high electrical resistance, they are
discoverable by geophysical methods. The hydrate can be converted to gas
by sinking a well and reducing the pressure and/or pumping a catalyst such
as methyl alcohol into it (Anonymous, 1970).
The solubility of the hydrocarbons benzene, toluene, o-xylene, rn-xylene,
p-xylene, naphthalene, biphenyl, diphenylmethane, and phenanthrene was
found t o increase with increasing silver-ion concentration, indicating that a
slightly soluble 1-1complex formed (Andrews and Keefer, 1949). Evidence
was obtained that two water-soluble complexes formed with silver and each
aromatic hydrocarbon tested. Potassium nitrate causes a reduction in the
solubility of aromatics in aqueous solutions (salting-out effect), but silver
nitrate increases the solubility of toluene about 73% compared to its solu-
bility in pure water. Apparently this effect with silver ions results from the
formation of n-complexes between the benzene ring and the cation.
Benzene hydrocarbons exhibit a minimum solubility in water near 18°C
which corresponds to a zero heat of solution. The actual volume occupied by
a hydrocarbon with one benzene ring in water solution apparently influences
its degree of solubility, and the larger the molecule the less soluble it is in
water (Bohon and Claussen, 1951). However, naphthalene and biphenyl,
which are larger in size and are multiring compounds, were 7 to 10 times
more soluble, indicating that some property of the benzene ring may in-
fluence the solubility. I t was postulated that a positive heat of solution
resulted in the heat of cavity formation, while a negative heat of solution
resulted from the formation of icelike structures around the dissolved
hydrocarbons and/or a n-electron complex of the aromatic nucleus where
the n-electrons functioned as a base and the water as an acid. The heat of
cavitation would predominate above 18°C and would cancel the negative
heat reaction at 18°C ,and below 18°C the negative heat would be larger.
A study of the effect that the salts sodium fluoride, sodium chloride,
lithium chloride, ammonium chloride, sodium iodide, cesium chloride,
tetramethylammonium bromide, etc., have upon the activity coefficient of
benzene in aqueous solutions indicates that the salting-out effect varies con-
siderably among the electrolytes (McDevit and Long, 1952). A limiting law
for determining the influence of electrolytes on the activity of nonpolar
solutes was developed, which related the magnitude of the salt effect t o the
volume changes associated with salt and water mixing.
Molecular hydrogen was found in oilfield waters in the Lower Volga region
(Zinger, 1962). Up to 43%of the dissolved gas in these waters was hydrogen;
other gases dissolved in the waters were methane, ethane, butane, pentane,
carbon dioxide, nitrogen, helium, and argon. The pH of these waters was as
low as 3.4, and the iron content was as high as 1,100 ppm.
The solubility of methane increases with pressure and decreases with in-
creased salt concentration at ambient temperature in NaC1-H2 0 and
CaC12-H20 systems (Duffy et al., 1961). From the experimental data, it
was estimated that 1 cubic foot of sedimentary rock with 20% porosity
buried 300 m deep and saturated with a brine containing 50,000 ppm of
NaCl could accommodate 0.3 mole of methane in solution.
A gas-liquid partition chromatographic technique was used t o determine
the solubilities of C1-C9 paraffin and branched-chain paraffins, four
cycloparaffins, and five aromatic hydrocarbons in water (McAuliffe, 1963).
Later this study was extended to seventeen paraffins, seventeen olefins, nine
acetylenes, seven cycloparaffins, seven cycloolefins, and six aromatic hydro-
carbons (McAuliffe, 1966). The data indicated that the solubilities of the
hydrocarbons in water increased as unsaturate bonds were added t o the
molecule, with ring closure, with addition of unsaturate bonds in the ring,
and with addition of bonds which decreased the hydrocarbon molar volume.
Branching increased the solubility in water for paraffin, olefin, and acetylene
hydrocarbons but not for the cycloparaffin, cycloolefin, and aromatic hydro-
carbons. Plots of the log of the solubility in water were a linear function of
hydrocarbon molar volume for each homologous series of hydrocarbons.
A capillary-cell method was used to measure the diffusion of methane,
ethane, propane, and n-butane in water (Witherspoon and Saraf, 1964). At
25OC, the results indicated that the diffusion coefficients times los cm2/sec
were 1.88, 1.52, 1.21, and 0.96, respectively, for methane, ethane, propane,
and n-butane. The coefficients increased with higher temperatures.
Near the critical solution temperature and about 300 atm, the solubility
versus pressure curves for some hydrocarbon-water systems show a sharp
maximum. However, pressure has a negative effect on solubility beyond this
maximum, and a second two-phase region appears. The five binary
hydrocarbon-water systems studied were benzene, n-heptane, n-pentane,
2-methyl-pentane, and toluene (Connolly, 1967).
The accommodation of CI2-C& n-alkanes in distilled water was deter-
mined as a function of hydrocarbon supply, settling time, filtration pore-
size, and mode of introduction (Peake and Hodgson, 1967). Apparently it is
possible to accommodate hydrocarbons in water at levels higher than solubil-
NITROGEN-FREE ORGANIC COMPOUNDS 181
ity levels, and such accommodation systems are stable for several days.
Preferential accommodation of alkanes in the C16-C20range was found at
the expense of other 'alkanes with lower and higher carbon numbers.
A gas chromatographic method for the determination of petrol in water
was developed whereby the petrol was extracted from the water into
nitrobenzene and the extract was analyzed using a column polyethylene
glycol 1,500 on silanized Chromosorb W (Jeltes-and Veldink, 1967). The
methods were sensitive t o 0.1 mg/l, and for concentrations > 0.5 mg/l the
precision was about 5% for the major components.
Low-molecular-weight hydrocarbons in the C1-C4range were detected in
sea water. Generally the concentration tended to decrease with depth
(Swinnerton and Linnenbom, 1967). Methane was the most abundant
hydrocarbon found, but smaller amounts of ethane, ethylene, propane,
propylene, n-butane, isobutane, and some butenes also were detected and
measured.
Hundreds of drill-stem samples of brine from water-bearing subsurface
formations in the Gulf coastal area of the United States were analyzed to
determine their amounts and kinds of hydrocarbons (Buckley et al., 1958).
The chief constituent of the dissolved gases usually was methane, with mea-
surable amounts of ethane, propane, and butane present. The concentration
of the dissolved hydrocarbons generally increased with depth in a given
formation and also increased basinward with regional and local variations. In
close proximity t o some oilfields, the waters were enriched in dissolved
hydrocarbons, and up to 14 standard cubic feet of dissolved gas per barrel of
water was observed in some locations.
The ratio of toluene t o benzene in 27 crude oils from various sources
ranged from 2.0 t o 11.3. Toluene is less soluble than benzene in distilled
water, where the ratio is about 0.3 (McAuliffe, 1966). A method of
prospecting for petroleum, utilizing information concerning the amount of
benzene dissolved in subsurface waters, was patented (Coggeshall and
Hanson, 1956). Gas chromatographic methods proved t o be good for deter-
mining the amount of benzene and other hydrocarbons in the petroleum-
associated waters (Zarrella et al., 1967). Collected information indicates that
the concentration of benzene in petroleum-associated water varies with
different types of hydrocarbon accumulations, that the benzene concen-
tration decreases with increasing distance from the hydrocarbon accumula-
tion, and that benzene is specific for detecting the occurrence of petroleum
hydrocarbon accumulation in a given geologic horizon. A brine sample taken
from a horizon separated by 27 m of shale from an oil pool contained 0.02
ppm of benzene, indicating that low-permeability shale prevents movement
of hydrocarbons.
Chromatographic techniques were developed for the determination of
sugars and phenols in sea waters and in sediments (Degens and Reuter,
1964). Biogeochemical differences were observed between the sugars in the
sea and in the sediments.
Wilson e t al. (1970) found that ethylene, propylene, and carbon

monoxide are produced in illuminated sea water to which dissolved phyto-
plankton was added. Higher saturated gaseous hydrocarbons and methane
were not produced.
Bonoli and Witherspoon (1968) measured the diffusion coefficients of
methane, ethane, propane, n-butane, n-pentane, benzene, toluene, ethyl-
benzene, cyclopentane, methylcyclopentane, and cyclohexane in pure water
at temperatures ranging from 2" to 60°C using the capillary-cell method. The
effect of sodium chloride was studied, and the largest decrease in diffusion
coefficients was found for the paraffin hydrocarbons. They attributed the
decrease to the effects of ions in water acting as structure breakers as well as
obstacles to diffusion because of obstructions and hydrations.
Hydrocarbons containing nitrogen
Chromatographic techniques were developed for determining humic acids,

amino acids, and indoles in saline waters and in sediments (Degens and
Reuter, 1964). Arginine was found in the particulate matter in sea water and
decreased in concentration with depth. Relatively abundant concentrations
of ornithine, serine, and glycine were found in sea water.
The total concentrations of amino acids found in some petroleum-
associated waters ranged from 20 to 230 pg/l (Degens et al., 1964). In
general, the amino acid content increased with salinity. Adjustment of the
salinity of the brines t o that of modern sea water indicated a similarity
between the amino acid spectra in the two. High concentrations of serine
and the presence of threonine and phenylalanine and glutamic and aspartic
acids were found in the petroleum-associated waters. It was postulated that
the amino acids occurred in the petroleum waters in a combined state as
nonproteinaceous acid complexes and that the solubility of these complexes
probably is a function of salinity. This postulate was based on information
which indicated that serine is thermally unstable. More recent information
indicates that serine, lysine, threonine, glycocol, histine, isoleucine, and
leucine are fairly stable up t o 180°C (Califet and Louis, 1965).
Liquid-exchange chromatography was used to determine the amounts of
amino acids in some saline waters (Siege1 and Degens, 1966). The results
indicated the bulk of the amino acids dissolved in the sea are tied up in
complexes and are not in a free form.
A study of the organic solutes in sea water led t o the conclusion that
coprecipitation methods are the most versatile for their isolation (Chapman
and Rae, 1967). Some of the organics that can be isolated by this method
include glucose, glutamic acid, aspartic acid, citric acid, succinic acid, glycol-
late, glycine, and lysine. The percent of recovery of these solutes by this
method varied from 16 t o 90%. The method involved the coprecipitation of
these organic solutes with iron or copper.
Most of the nitrogen in humic acid is located in the large and intermediate
FATTY ACIDS 183
size molecules, with the smallest molecules being practically nitrogen-free.

This is attributed to the fact that the link involving nitrogen is more
susceptible t o oxidation than the rest of the molecule. Amino acids can be
released from humic acid by acid hydrolysis, alkaline hydrolysis, and reduc-
tion with sodium amalgam (Piper and Posner, 1968).
A widely used procedure for concentrating and recovering trace organics is
the carbon-adsorption method developed by Braus et al. (1951).A modifi-
cation of this method by Robinson et al. (1967) allowed recovery of
organics using three activated carbon filters in series, with the final two
receiving acidified water.
Krause (1962)investigated the decomposition of organic matter in natural
waters and found that immediately after the death of an organism that
amino acids and keto acids appeared in the water. After 24 hours of aerobic
decomposition, a qualitative and quantitative maximum was reached by both
groups, and the amino acids present were alanine, aspartic, glutamic, glycine,
leucine, lysine, methionine, phenylalanine, serine, tyrosine, and raline; and
the keto acids present were pyruvic, oc-ketoglutaric, oxaloacetic, and
glyoxylic. After 10 days, the only acids that remained of the amino group
were glutamic, glycine, lysine, and serine; and of the keto group, pyruvic and
oc-ketoglutaric.
Litchfield and Prescott (1970)analyzed sea water, and pond water, and
spent algal media and found aspartic acid in all of the samples. Other amino
acids frequently found were serine, glycine, alanine, and arginine. Techni-
ques employed in the analysis were dansylation, extraction, and thin-layer
chromatography .
Fatty acids
Ralston and Hoerr (1942) determined the solubilities of the normal

saturated fatty acids from caproic to stearic acid, whose number of carbon
atoms ranges from 6 to 18 in water, ethanol, acetone, 2-butanone, benzene,
and glacial acetic acid from '0 to about 60'C. In general, the solubilities
increased with increasing temperature.
Free fatty acids and hydroxyl ions form when soaps hydrolyze. The rate
and percentage of hydrolysis is pH dependent, generally the potassium soaps
are more hydrolyzed than the corresponding sodium soaps, and free fatty
acid never separates as such from pure soap solutions unless reacted with an
excess of acid such as carbon dioxide (McBain et al., 1948).
Quantitative recovery of organic constituents from saline waters without
alteration is difficult. Temperature and pressure changes, bacterial actions,
adsorption, and the high inorganic/organic constituents ratio in most
petroleum-associated waters are some reasons why quantitative recovery is
difficult. Some of these factors apply also to sea waters, and Jeffery and
Hood (1958) evaluated five methods which proved effective for isolation
of portions of the soluble organic compounds in sea water. They were: (1)
dialysis or electrodialysis; (2) ion exchange;, (3) solvent extraction; (4)

coprecipitation; and (5) carbon adsorption. Their results showed that the
total organic material was most efficiently removed by electrodialysis or by
coprecipitation with ferric hydroxide.
A study of the differential uptake of organic compounds by montmoril-
lonite and kaolinite revealed that montmorillonite adsorbed the compounds
more efficiently than kaolinite. The compounds studied were aspartic acid,
alanine, glucose, and sucrose (Williams, 1960). Approximately 13%of the
aspartic acid was removed from solution by montmorillonite, while kaolinite
removed only about 2%.
The following saturated, monosaturated, and diunsaturated long-chain
fatty acids were found in sea water: saturated Cl0, C12,CI4,C16,CIS,CzO,and
CZ2; diunsaturated CIS; and monounsaturated CI6 and c18 (Emery and
Koerner, 1961). Also isolated were CIS,C1,, and C19 acids which might or
might not have been originally present.
A gas chromatographic method was developed for the determination of
trace amounts of the following fatty acids in water: n-valeric, isovaleric,
n-butyric, isobutyric, propionic, and acetic (Emery and Koerner, 1961). The
gas chromatograph was equipped with a flame ionization detector and a
column of Tween 80 on Chromosorb W.
The fatty acids lauric, myristic, palmitic, stearic, hyristoleic, palmitoleic,
oleic, linoleic, and linolenic were identified in sea water using solvent extrac-
tion, esterification, and gas-liquid chromatography (Slowey et al., 1962).
Samples of deep-sea water contained less unsaturated acids and shorter-chain
acids than surface samples.
Saturated straight-chain fatty acids were found in petroleum-associated
waters from two reservoirs. The carbon numbers were CI4through C30.The
same acids were identified in a shale-core sample from a petroleum reservoir.
The even-numbered acids predominated over the odd-numbered acids in the
amounts found in every case. The identification methods consisted of extrac-
tion by refluxing, esterification, gas chromatography, and mass spectrometry
(Cooper, 1962).
A gas chromatographic method capable of separating unesterified fatty
acids was developed (Metcalfe, 1963). Acids up to CzOwere identified using
a thermal conductivity detector and a column composed of phosphoric acid-
treated ethylene glycol succinate polyester on Chromosorb W.
Bordovskii (1965) studied the sources of organic matter in marine basins,
the sedimentation of organic matter in water, and the transformation of
organic matter in sediments and its early diagenesis. He also pointed out that
the organic matter in water is present in true solution, as colloidal organic
detritus, and as live organisms in suspension. Bacteria play an important part
in altering the composition of the organic material in the aqueous phase as
well as in the sediments.
Wangersky (1965) found that organic carbon was present in freshly
distilled water and that it survived triple distillation and distillation from
NAPHTHENIC A N D HUMIC ACIDS 185
alkaline permanganate. He correlated this with algal growth in city reservoirs

and suggested that the organic carbon content of distilled water must be
considered by anyone growing organisms in distilled water. He also found
that bubbling sea water caused organic compounds to form aggregates on the
surface and further theorized that such reactions may be related to the origin
of life.
Kabot and Ettre (1963) developed gas chromatographic methods capable
of determining free fatty acids. They analyzed different mixtures of the
normal fatty acids using both packed and Golay columns in conjunction
with a flame ionization detector. They concluded that the quantitative
analysis of free fatty acids is possible.
Naphthenic and humic acids
Davis (1968) examined the organic fractions of artesian well waters from a
Texas oil-bearing Eocene age aquifer, using infrared and chromatographic
methods. He found that the water coproduced with oil contained 1,000
times more naphthenic acids than water located updip from the oil. He also
found a phthalic acid ester dissolved in the petroleum-associated water but
concluded that it may be common t o ground waters in general.
Shaborova et al. (1961) state that “the presence in subsurface waters of
organic acids in the form of salts of various metals or in a free state indicates
a current process of leaching of organic matter from the enclosing rock. The
presence of organic acids in subsurface waters is one of the evidences for the
existence in the earth’s crust of chemical processes of decomposition of
preserved organic matter. In turn, the organic acids are broken down into
simpler compounds by decarboxilization. It is known that decarboxilization
of organic acids is accompanied by the formation of hydrocarbons. In
nature, this process is a real geochemical factor. Consequently, the organic
acids and their salts that are dissolved in subsurface waters can be regarded as
one of the sources for the generation of hydrocarbons.”
Using a steam distillation method, organic acids were found in concen-
trations from 663 to 2,242 mg/l in subsurface waters taken from a Kazhim
stratigraphic well. The average molecular weight of the acids was from 46 to
58, and the waters taken from Devonian age sediments contained higher
concentrations of the acids than waters taken from Carboniferous age sedi-
ments.
Lochte et al. (1949) analyzed waters produced with high-pressure gas
wells and identified the following acids: acetic, propionic, isobutyric,
n-butyric, isovaleric, n-valeric, n-hexanoic, and other C6isomers. Crude oils
were treated with ammonia solution followed by electroprecipitation of the
aqueous phase to remove naphthenic acids (Agaev, 1961). Further isolation
of the naphthenic acids was accomplished by heating the aqueous phase to
decompose the ammonium salts and remove ammonia and water.
Oden (1919) recognized fulvic acid, humus acid, and hymatomelanic acid
in soils. Later, Page and Dutoit (1930) modified the name humus acid to
humic acid. Sestini (1898) demonstrated that the humic acids are of com-
plex composition and contain ethereal and anhydride components in
addition to alkyl, hydroxyl, and ketonic groups.
Burges (1960) suggested that humic acid is a single chemical substance or
a group of similar substances, and that primarily it is nonnitrogenous.
Steelink et al. (1960) fused soil humic acids and found the following degra-
dation products: catechol, profocatechuic acid, and resorcinol. Steelink and
Tollin (1962) determined the presence of two free-radical species in humic
acid using an electron paramagnetic-resonance spectrometer. They believed
that one could be a semiquinone radical and the other a quinhydrone radical.
Fulvic acids, humic acids, and hymatomelanic acids have been found in
natural waters (Wilson, 1959; Black and Christman, 1963; Packman, 1964).
The brown color, characteristic of many natural waters, is attributed to
complex organic compounds which probably are derived from water-soluble
peptizable components of soil humus.
A method that can be used t o determine the organic acids in petroleum-
associated waters was published by the Natural Gasoline Association of
America (1953). The water is treated with lime water to convert the organic
acids to their calcium. salts, which are titrated with a standard mineral acid.
Determination of oil in water
The following method was developed by Nalco Chemical Company (1971)

and is applicable t o waters and brines where the oily matter is hydrocarbons
or hydrocarbon derivatives and all liquid or unctuous substances that have a
boiling point above 90°C and are extractable from waters or brines at pH 5.0
or lower using benzene, chloroform, or carbon tetrachloride.
The sample is extracted with a fluorocarbon solvent which is evaporated
off in a specially designed vessel and the residual oil measured volumetrically
in a microsyringe.
Pear-shoped l o p , capacity opprox.17 m l ,

o f f s e t odditionol opening so th0t"popped"
somple w i l l be r e t o i n e d
S y r i n g e , 5 0 0 kl, 10-pl divisions
Fig. 6.1.Microsyringe-evaporatingflask.
DETERMINATION OF OIL IN WATER 187
Apparatus. The necessary apparatus consists of:

(1)Microsyringe - evaporating flask (see Fig. 6.1): this assembly consists
of a single-neck flask of approximately 20 ml volume which tapers opposite
and slightly offset to the neck into a microsyringe equipped with a gas-tight
Teflon-tipped plunger and calibrated to measure 0-500 pl.
(2) 1,000-ml pear-shaped separatory funnel.
(3) Hotplate or hot-water bath: capable of being controlled in the range of
45O-55'~ at * ~ O C .
(4) 500-ml Berzelius, tall form beaker.

(1)50%hydrochloric acid solution, reagent-grade.
(2) pH paper indicating strip or pH meter.
(3) 1,1,2-trichlorotrifluoroethane(Freon TF) reagent-grade, purified,
48OC boiling point.
Sampling. Collect a composite or spot sample representative of stream to be

measured. Volume to be taken will be dependent on content of oily material
and should be in the range of 1-5 liters. Sample should be caught in glass
container.
Procedure. Extraction: adjust pH of entire sample t o pH 5 or below using

hydrochloric acid added in small increments. Thoroughly mix the sample
and allow it t o stand 15 minutes. Measure the volume of entire sample and
transfer to separatory funnel. Add portion of 1,1,2-trichlorotrifluoroethane
extraction fluid (see Note 1) to sample container, thoroughly rinsing any
adhering oil material. Add this and balance of fluorocarbon to separatory
funnel. Shake thoroughly for 5 minutes and let stand t o separate layers.
Draw off fluorocarbon layer into suitable beaker, filtering any entrained
solids, if necessary, and warm gently to boiling point (see Note 2). Continue
boiling until volume remaining can be contained in measuring flask.
Transfer to measuring flask with fluorocarbon rinse of beaker, and
immerse the flask and contents into 500-ml beaker partially filled with water
and warmed to 65°C on hotplate or in hot-water bath. Be sure open neck of
flask is clear of upper edge of beaker (can be maintained by extension of
syringe piston). Continue until volume is reduced to that of syringe volume.
Draw fluid into syringe and increase heat slowly t o remove last traces of
solvent, indicated by lack of bubbles forming in the syringe column.
Measure amount of oil material in graduated syringe using graduations
midway in syringe.
Note 1: for single extractions fluorocarbon volume should be one-tenth of
the original sample volume. In double extractions for better accuracy and
reproducibility use two volumes of fluorocarbon, each onetwentieth of the
original sample volume.
Note 2: although fluorocarbon is essentially considered nontoxic, the
evaporation of the solvent should be done in a well-ventilated area or under

an exhaust hood with adequate draft to handle high-density vapors.
Calculation :
1-11measured = mg/l oily matter
c
mi sampIe
Organic acids in oilfield brines
This method measures the bulk of the organic acid salts in oilfield waters.
Reagents and apparatus. Acetone; NaOH, 0.02N; HC1, 0.05N; acetic acid, 10
mg/l; a hotplate; and a standard pH meter.
Procedure. Pipet 25 ml of filtered brine into a 250-ml beaker; add 25 ml of

acetone and adjust the pH t o precisely 6. Titrate the sample to a pH of 3.5,
and record the amount of 0.05N HC1, used in the adjustment from pH 6 to
3.5. Boil the solution for 5 minutes. Cool and titrate back to pH 6 with
0.02N NaOH. Make a blank determination for NaOH and HC1. To calculate,
subtract blank from NaOH and HC1 titrations.
Calculation:
(ml HC1 x HC1 N ) - (ml NaOH x NaOH N )
x 60,030 = mg/l organic acids
ml sample
as acetic acids.
References
Agaev, A.A., 1961. Separation of ammonium salts of naphthenic acids from crude oil in
an electric field. Izv. Vyssh. Uchebn. Zaved., N e f t Gaz, 4:95-98 (in Russian).
Andrews, L.J. and Keefer, R.M., 1949. Cation complexes of compounds containing
carbon-carbon double bonds, 11. The solubility of cuprous chloride in aqueous maleic
acid solutions. J. A m, Chem. Soc., 71 :2379-2380.
Anonymous, 1970. Solid natural gas discovered. Ind. Res., 12:70.
Black, A.P. and Christman, R.F., 1963. Chemical characteristics of fulvic acids. J. A m .
Water Works Assoc., 55:897-912.
Bohon, R.L. and Claussen, W.F., 1951. The solubility of aromatic hydrocarbons in water.
J. A m . Chem. Soc., 73:1571-1578.
Bonoli, L. and Witherspoon, P.A., 1968. Diffusion of paraffin, cycloparaffin, and aromat-
ic hydrocarbons in water and some effects of salt concentration. In: Advances in Or-
ganic Geochemistry - Proc. 4th Int. Meet. Org. Geochem., Amsterdam. Pergamon
Press, New York, N.Y., 9:16-18.
Bordovskii, O.K., 1965. Accumulation and transformation of organic substance in marine
sediments. Mar. Geol., 3:3-114.
Braus, H.,Middleton, F.M. and Walton, G., 1951. Organic chemical compounds in raw
and filtered surface waters. Anal. Chem., 23:1160-1164.
REFERENCES 189
Buckley, S.E., Hocott, C.R. and Taggart, Jr., M.S., 1958. Distribution of dissolved
hydrocarbons-in subsurface waters. In: L.G. Weeks (Editor), Habitat o f Oil. American
Association of Petroleum Geologists, Tulsa, Okla., pp. 850-882.
Burges, A., 1960. The nature and distribution of humic acid. Sci R o c . R . Dublin SOC.
Ser. A , 4:53-51.
Califet, Y. and Louis, M., 1965. Contribution to the knowledge concerning the stability
of amino acids contained in sedimentary rocks. Compt. Rend., Acad. Sci. Fr.,
261 :3645-3646.
Chapman, G. and Rae, A.C., 1967. Isolation of organic solutes from sea water by
coprecipitation. Nature, 214:627-628.
Coggeshall, N.D. and Hanson, W.E., 1956. Method of geochemical prospecting. US.
Patent, No. 2,767,320.
Connolly, J.F., 1967. Solubility of hydrocarbons in water near the critical solution tem-
peratures. J. Chem. Eng. Data, 11:13-16.
Cooper, J.E., 1962. Fatty acids in recent and ancient sediments and petroleum reservoir
waters. Nature, 193:744-746.
Davis, J.B., 1968. Distribution of naphthenic acids in an oil-bearing aquifer. Presented at
Annual Geol. SOC. A m . and Assoc. SOC. Meet., Mexico, D.F., November, 1968.
Program, p. 7 1.
Degens, E.T. and Reuter, J.H., 1964. Analytical techniques in the field of organic
geochemistry. In: U. Colombo and G.F. Hobson (Editors), Advances in Organic
Geochemistry. Pergamon Press, New York, N.Y., pp.377-402.
Degens, E.T., Hunt, J.M., Reuter, J.H. and Reed, W.E., 1964. Data on the distribution of
amino acids and oxygen isotopes in petroleum brine waters of various geologic ages.
Sedimentology, 3 :199-225.
Duffy, J.R., Smith, N.O. and Nagy, B., 1961. Solubility of natural gases in aqueous salt
solutions, I. Liquidus surfaces in the System CH4-Hz O-NaC1-CaCl2 at room temper-
atures and at pressures below 1,000 psia. Geochim. Cosmochim. Acta, 24:23-31.
Duursma, E.K., 1965. The dissolved organic constituents of sea water. In: J.P. Riley and
G. Skirrow (Editors), Chemical Oceanography. Academic Press, New York, N.Y.,
1:433-475.
Eley, D.D., 1939. Solubility of gases, 1. Inert gases in water. Trans. Faraday SOC.,
35 :1281-1293.
Emery, E.M. and Koerner, W.E., 1961. Gas chromatographic determination of trace
amounts of the lower fatty acids in water. Anal. Chem., 33:146-147.
Erdman, J.G., 1965. Petroleum - its origin in the earth. In: A. Young and J.E. Galley
(Editors), Fluids in Subsurface Environments - A m . Assoc. Pet. Geol., Mem. 4 ,
pp.20-52.
Frank, H.S. and Evans, M.E., 1945. Free volume and entropy in condensed systems. J.
Chem. Phys., 13:507-532.
Hedberg, H.G., 1964. Geologic Hspects of origin of petroleum. Bull. A m . Assoc. Pet.
Geol., 48 (11):1755-1803.
Jeffery, L.M. and Hood, D.W., 1958. Organic matter in sea water; an evaluation of various
methods for isolation. J. Mar. Res., 17:247-271.
Jeltes, R. and Veldink, R., 1967. The gas chromatographic determination of petrol in
water. J. Chromatogr., 27:242-245.
Kabot, F.J. and Ettre, L.S., 1963. Gas chromatography of free fatty acids, 111. Quantita-
tive aspects. J. Gas Chromatogr., 1:7-10.
Kidwell, A.L. and Hunt, J.M., 1958. Migration of oil in sediments. In: L.G. Weeks
(Editor), Habitat of Oil. American Association of Petroleum Geologists, Tulsa, Okla.,
pp.7 90-8 17.
Krause, H.R., 1962. Investigation of the decomposition of organic matter in natural
waters. F A 0 Fish. Biol. Rep., No.34, 1 4 pp.
Litchfield, C.D. and Prescott, J.M., 1970. Analysis by dansylation of amino acids
dissolved in marine and freshwaters. Limnol. Oceanogr., 15:250-256.
Lochte, H.L., Burnam, C.W. and Meyer, H.W.H., 1949. Organic acids produced by high
pressure gas wells The Pet. Eng., 21 :C34-C40.
Marsden, S.S. and Kawai, K., 1965. "Suiy6sei-Ten'nengasu" - a special type of Japanese
natural gas deposit. Bull. A m . Assoc. Pet. Geol., 49:286-295.
McAuliffe, C., 1963.Solubility in water of C l - C g hydrocarbons. Nature, 200:1092-1093.
McAuliffe, C., 1966. Solubility in water of paraffin, cycloparaffin, olefin, acetylene,
cycloolefin, and aromatic hydrocarbons. J. Phys. Chem., 70:1267-1275.
McBain, J.W., Laurent, P. and John, L.M., 1948. The hydrolysis of soap solutions, 111.
Values of pH and the absence of fatty acid as free liquid or solid. J. A m . Oil Chem.
SOC.,25:77-84.
McDevit, W.F. and Long, F.A., 1952. The activity coefficient of benzene in aqueous salt
solutions. J. A m . C h e m SOC.,74:1773-1777.
Metcalfe, L.D., 1963. The gas chromatography of fatty acids and related long chain
compounds on phosphoric acid treated columns. J. Gas Chromatogr., 1 :7-11.
Nalco Chemical Company, 1971. Analytical Method f o r the Determination o f Oil in
Water. Unpublished method (permission obtained through H.C. Noe).
Namoit, A.Y., 1961.Solubility of nonpolar gases in water. Zh. Strukt. Khim., 2:408-417.
Natural Gasoline Association of America, 1953. Condensate Well Corrosion. NGAA,
Tulsa, Okla, 203 pp.
Nemethy, G. and Scheraga, A.A., 1962. Structure of water and hydrophobic bonding in
proteins, 1. A model for the thermodynamic properties of liquid water. J. Chem.
Phys., 36:3382-3400.
Oden, S., 1919. Humic acids, studies in their chemistry, physics, and soil science.
Kolloidchem. Beih., 11:75-260.
Packman, R.F., 1964. Studies of organic color in natural water. h o c . SOC. Water Treat.
Exam., 13:316-329.
Page, H.J. and Dutoit, M.M.S., 1930. Studies on the carbon and nitrogen cycles in the
soil, 111. The formation of natural humic matter. J. Agric. Sci., 20:478-488.
Peake, E. and Hodgson, G.W., 1967. Alkanes in aqueous systems, 11. The accommodation
of C12-C36n-alkanes in distilled water. J. A m . Oil Chem. SOC.,44:696-702.
Piper, T.J. and Posner, A.M., 1968. On the amino acids found in humic acid. Soil Sci.,
106 :188-192.
Price, L.C., 1973. Solubility of petroleum in water as function of temperature and
salinity and its significance in primary migration. Bull. A m . Assoc. Pet. Geol., 57 :Sol.
Ralston, A.W. and Hoerr, C.W., 1942.The solubilities of the normal saturated fatty acids.
J. Org. Chem., 7:546-555.
Robinson, L.R., O'Conner, J.T. and Engelbrecht, R.S., 1967.Organic materials in Illinois
groundwaters. J. A m . Water Works Assoc., 59:227-236.
Sestini, F., 1898. The nitrogen content associated with humic acids in soil. Landw. Vers.
Sta., 51:153-160.
Shaborova, N.T., Tunyak, A.P. and Nektarova, M.B., 1961. Study of organic acids in
underground waters. Geol. Nefti Gaza, 5:50-51 (in Russian).
Siegel, A. and Degens, E.T., 1966. Concentration of dissolved amino acids from saline
waters by ligand-exchange chromatography. Science, 151 :1098-1101.
Slowey, J.F., Jeffery, L.M. and Hood, D.W., 1962.The fatty-acid content of ocean water.
Geochim Cosmochim. Acta, 26:607-616.
Steelink, C. and Tollin, G., 1962. Humic acid contains stable free radicals. Chem. Eng.
News, 40:53-54.
Steelink, C., Berry, J.W., Ho, A. and Nordby, H.E., 1960.Alkaline degradation products
of soil humic acid. Sci. Proc. R. Dublin SOC.,Ser. A , 1(4):59-67.
Swinnerton, J.W. and Linnenbom, V.J., 1967. Gaseous hydrocarbons in sea water: deter-
mination. Science, 156 :1119-1120.
REFERENCES 191
Wangersky, P.J., 1965.The organic chemistry of sea water. A m . Sci., 53~358-374.

Williams, P.M., 1960. Organic Acids Found in Pacific Ocean Waters. Ph.D. Dissertation,
Univ. of Calif., Los Angeles, Calif., 74 pp.
Wilson, A.L., 1959. Determination of fulvic acids in water. J. Appl. Chem., 9:501-510.
Wilson, D.F., Swinnerton, J.W. and Lamontagne, R.A., 1970. Production of carbon
monoxide and gaseous hydrocarbons in sea water: relation to dissolved organic carbon.
Science, 168:1577-1 579.
Witherspoon, P.A. and Saraf, D.N., 1964. Diffusion of methane, ethane, propane, and
n-butane in water. Presented at A m . Chem. SOC.,Div. Pet. Chem. Meet., Chicago, Ill.,
1964, Preprints, pp.303-311.
Zarrella, W.M., Mousseau, R.J.,Coggeshall, N.D., Norris, M.S. and Schrayer, G.J., 1967.
Acta, 31 :1155-1166.
Zinger, AS., 1962. Molecular hydrogen in gas dissolved in waters of oil-gas fields, Lower
Volga region. Geochemistry, 10 :lo15-1023.
Chapter 7. ORIGIN OF OILFIELD WATERS
The five spheres of the earth are the lithosphere (rocks), the pedosphere
(soils, till, and other surficial materials), the hydrosphere (natural waters),
the atmosphere (gases), and the biosphere (living organisms). Oilfield waters
are a part of the hydrosphere, and petroleum is a product of the biosphere.
The total volume of water in the hydrosphere is about 1,338 x 10'' liters,
and about 8.4 x 10" liters is ground water (Skinner, 1969). Most of the
water, 1,300 x 10" liters, is in the oceans. Less than 50% of the total
ground water is in strata below 1 km. The total amount of water in sedimen-
tary rocks and associated with liquid and gaseous hydrocarbons is less than
1.3 x 1 O I 8 liters.
All of the petroleum recovered t o date has been taken from oil wells
drilled into the upper 8 km of the earth's crust. The average thickness of the
0 km
I 7 km
400km
1,000 k m
2.900km
5,000km
6,371 k m
Fig. 7.1. Regions of the interior of the earth.

194 ORIGIN OF OILFIELD WATERS
earth’s crust is about 1 7 km, ranging from 5 km under the oceans t o about
35 km under the continents (Clark and Ringwood, 1964). Fig. 7.1 illustrates
the various regions of the interior of the earth, with the distance from the
surface of the crust to the center of the inner core being 6,731 km. In this
discussion we are concerned only with the crust to a distance of 0.08% of
the depth to the center of the earth.
Hydrocarbons are believed t o have originated from organic material in
sedimentary material which was produced by weathering and erosion of the
earth’s surface. This eroded material is carried away by water, ice, or wind
and redeposited, ultimately forming sedimentary rocks. The major sedimen-
tary, minerals are clays, quartz, calcite, gypsum, anhydrite, dolomite, and
haiite. Most of the large bodies of sedimentary rocks were formed in marine
environments; smaller sedimentary deposits formed in lakebeds and river
floodplains.
Definitions of some water terms
Meteoric water. White (1957) defined it as water that was recently involved
in atmosphere circulation and further that “the age of meteoric groundwater
is slight when compared with the age of the enclosing rocks and is not more
than a small part of a geologic period.”
Sea water. The composition of sea water may vary somewhat, but in general
will have a composition relative t o the following (in mg/l): chloride -
19,375, bromide - 67, sulfate - 2,712, potassium - 387, sodium - 10,760,
magnesium - 1,294, calcium - 413, and strontium - 8.
Table 7.1 (Anonymous, 1964) gives a more comprehensive picture of the
constituents found in sea water. The analyses given in Table 7.1. are in parts
per million.
Interstitial water. Interstitial water is the water contained in the small pores
of spaces between the minute grains or units of rock. Interstitial waters are:
(1) syngenetic (formed at the same time as the enclosing rocks); or (2)
epigenetic (originated by subsequent infiltration into rocks).
Connate water. The term connate implies born, produced, or originated

together, connascent. Therefore, connate water probably should be con-
sidered t o be an interstitial water of syngenetic origin. White (1957) called
connate water of this definition a fossil water, i.e., water that has been out
of contact with the atmosphere for at least a large part of a geologic period.
As White (1957) pointed out the implication that connate waters are only
those “born with” the enclosing rocks is an undesirable restriction.
Diagenetic water. Diagenetic waters are those waters that have changed
chemically and physically, both before, during, and after sediment consolida-
SEDIMENTARY ROCKS 195
TABLE 7.1
Average composition of sea water
Element Amount Element Amount Element Amount

(PPm) (PPm 1 (PPm)
Chlorine 18,980 Zinc 0.01 Tungsten 1 lo4

Sodium 10,560 Molybdenium 0.01 Germanium 1 lo4
Magnesium 1,270 Selenium 0.004 Xenon 1 lo4
Sulfur 880 Copper 0.003 Chromium 5x 10-
Calcium 400 Arsenic 0.003 Beryllium 5 10-~
Potassium 380 Tin 0.003 Scandium 4x
Bromine 65 Lead 0.003 Mercury 3x 10-
Carbon 28 Uranium 0.003 Niobium 1
Oxygen 8 Vanadium 0.002 Thallium 1 10-~
Strontium 8 Manganese 0.002 Helium 5x 10"
Titanium 0.001 Gold 4x lod
Boron 4.8 Thorium 0.0007 Praseodymium 2x
Silicon 3.0 Cobalt 0.0005 Gadolinium 2
Fluorine 1.3 Nickel 0.0005 Dysprosium 2
Nitrogen 0.8 Gallium 0.0005 Erbium 2
Argon 0.6 Cesium 0.0005 Ytterbium 2
Lithium 0.2 Antimony 0.0005 Samarium 2
Cerium 0.0004 Holmium 8x
Rubidium 0.12 Yttrium 0.0003 E uro pi um 4x
Phosphorus 0.07 Neon 0.0003 Thulium 4x
Iodine 0.05 Krypton 0.0003 Lutetium 4x
Barium 0.03 Lanthanum 0.0003 Radium 3 x lo-"
Indium 0.02 Silver 0.0003 Protactinium 2 x lo-'*
Aluminum 0.01 Bismuth 0.0002 Radon 9
Iron 0.01 Cadmium 0.0001
tion. Some of the reactions that occur in or t o diagenetic waters include

bacterial, ion exchange, replacement (dolomitization), infiltration by
permeation, and membrane filtration.
Formation water. Formation water as here defined is water that naturally

occurs in the rocks and is present in them immediately before drilling (Case,
1955).
Juvenile water. Water that is in primary magma or derived from primary

magma is juvenile water (White, 1957).
Sedimentary rocks
At least a portion of the water found in petroleum reservoirs consisting of

sedimentary rocks w a s deposited with the sediments before they were trans-
formed into rock. As the sediment compacted to form rock, the composition
TABLE 7.11
Average composition of igneous and some types of sedimentary rocks (ppm)
Element Igneous rocks Sedimentary rocks Evaporites

(halite)
resistates hydrolyzates precipitates
Si 277,200 367,500 272,800 24,200 90
A1 81,300 25,300 81,900 4,300 20
Fe 50,000 9,900 47,300 4,000 11
Ca 36,300 39,500 22,300 304,500 930
Na 28,300 3,300 9,700 370 325,000
K 25,900 11,000 27,000 2,700 800
Mg 20,900 7,100 14,800 47,700 460
Ti 4,400 960 4,300 - 0.8
P 1,180 350 740 175 -
Mn 1,000 trace 620 385 1
F 600-900 - 510 250 20
S 520 2,800 2,600 1,100 770
C 320 13,800 15,300 113,500 70
c1 314 trace - 200 586,000
Rb 310 273 300 0 -
Sr 300 < 26 170 425-765 -
Ba 250 170 460 120 2
Cr 200 68-200 410-680 2 -
Zn 132 < 20 200-1,000 < 50 1
Ni 80 2-8 24 0 0
cu 70 - 192 20.2 3
Li 65 17 46 < 26 -
N 46 - - - -
Sn 40 - 40 - -
co 23 0 8 0 -
Pb 16 20 20 5-1 0 2
Th 11.5 (? )
-
6.1 10.1
-
1.1 <
-
0.2
cs 7 12
Be 6 (3 - <4 0 -
As 5 - - 5 - -
U 4 1.2 1.2 1.3 0.01
B 3 9-31 310 3 <2
Br 1.62 - < 0.2 - 60
I 0.3 - - 0.07-0.55 < 2
Cd 0.15 0 0.3 - -
Se 0.09 - 0.6 < 0.1 < 0.5
Hg 0.08-0.5 0.1 0.3 0.03 -
Ra 1.3 x 0.7 x 1.08 x lod 0.42 x lod -
of the interstitial water changed because of reactions with the rock. A

simplistic view of sedimentary rocks and their relation to oilfield waters
should include consideration of weathering; erosion; transportation
mechanisms; sorting of weathered products; depositional environments of
clastics, carbonates, evaporites, organic matter, and silica; sediment com-
pactions; sediment diagenesis; and petroleum and natural gas.
The volume of the earth is about 1,100 billion km3 and the volume of the
oceans is about 1.3 .billion km3; however, the oceans with an area of 360
million km2 cover 70% of the surface of the earth. The average composition
of some of the igneous and sedimentary rocks of the earth’s crust is shown in
Table 7.11, which was taken from Clarke and Washington (1924) and
Rankama and Sahama (1950). The resistate rocks referred to in Table 7.11
are composed of residues not chemically decomposed in the weathering of
the parent rocks. Hydrolyzate rocks are the insoluble products formed by
chemical reactions during weathering of parent rocks. Precipitate rocks are
those formed by chemical precipitation of minerals from aqueous solution.
Evaporite rocks are marine evaporites which were produced when the water
in which they were dissolved was evaporated.
Sedimentary rocks comprise about 5% of the lithosphere, while the
igneous rocks form 95%. The three main types of sedimentary rocks are
shale, sandstone, and iimestone, and their relative abundance determined
from geochemical data ranges from 70 t o 83% shale, from 8 to 16%sand-
stone; and from 5 t o 14% limestone (Pettijohn, 1957). Levorsen (1966)
noted that oil and gas are found in reservoir rocks consisting primarily of
sandstones, limestones, and dolomites.
Weathering
Weathering is a most important factor in producing the source material for

the creation of sedimentary rocks. Processes that cause weathering are
chemical, physical, and biological (Ross, 1943).
The weathering of rock by physical methods includes temperature changes
brought about by climate changes. Examples are the breaking of rock by
thermal expansion (heat), the breaking of rock by the expansion of freezing
water in the pores or cracks, or the mechanical breaking of rock as a glacier
moves over it. Breaking the rock causes the surface area to become larger
without significantly changing the chemical composition.
Biological weathering includes the cracking of rock as a result of plant
roots and the action of acids derived from plants, animals, and bacteria. The
biotic factor includes bacteria, algae, protista, protozoa, higher animals and
plants, during both life and subsequent necrotic decomposition which
furnish Ht ions, colloids, complexing agents, and dispersants.
Chemical weathering involves the action of water upon the parent rock
and upon the products of physical and biological weathering. In chemical
weathering the composition of the source rock is changed by solution,
hydrolysis, oxidation, and reduction reactions. The Ht ion when concen-
trated in aqueous solution is a very important energy factor because it will
cause rapid chemical reactions with parent rocks. The redox potential in-
fluences the rate of removal of elements, such as iron and manganese from
the parent rock, and if it is a reducing potential, these elements are more
likely to remain in solution after solubilization.
Erosion
Erosion is the opposite of deposition (the processes are reversible), but

erosion must occur before deposition can proceed. The products of weath-
ering are eroded and transported to a new location by the action of water
and wind. The water serves to transport the majority of these products, and
it can transport them by dissolution, suspension, or pushing of larger
particles.
Transpor ta t ion mechanisms
Both wind and water can transport the products of weathering, however,
this discussion will consider only water. The transport mechanisms con-
sidered are chemistry, physics, and hydraulics.
Perhaps the primary solvents of weathered products are carbonated water,
organic acids, and sulfate solutions. Elements that dissolve readily in car-
bonated waters are lithium, sodium, potassium, magnesium, calcium, stron-
tium, iron, manganese, phosphorus, and others. The organic acids will
dissolve iron and manganese, while sulfate solutions will dissolve copper,
iron, and manganese compounds.
The chemistry of the water is a prime factor in the dissolution of the rock;
if the pH is acidic, the transition group metals are more likely to dissolve,
while if it is basic, elements such as silica are more likely to dissolve. Salts of
the alkali and alkaline earth metals will dissolve if the pH is either acidic or
basic; however, if the pH is above 10, some of the alkaline earths such as
magnesium will precipitate. The pH of the water is influenced by the dissolu-
tion of carbon dioxide. For example, as carbon dioxide dissolves in water,
the pH will change. The pH of pure water in equilibrium with carbon dioxide
can be calculated and is 5.65 (Hem, 1970). The pH is calculated using the
mass-law equations in which the activity of water is unity in dilute solution,
and h , = constant equal to the product of the activities of H+ and OH-.
Introduction of another phase such as calcite into the water carbon
dioxide system will change the pH. Garrels and Christ (1965) calculated that
such a system in equilibrium with the atmosphere will have a pH of 8.4.
Additional ions such as those found in ocean water will produce other pH
values. For example, if the system is ocean water in equilibrium with carbon
dioxide, the pH at each equilibrium step is approximately:
H20+C02 *- H2C03
HC03-+H+
7PH 5)
H2 co3
A
(PH 6-31
HC03+H+ =+ 2H++C03-2 (pH 10.3)
Some chemicals when dissolved in water act as buffers, where a buffer is

defined as something that .produces an effect which inhibits a large pH
change when an acid or a base is added t o the water. Therefore, as a water
becomes more concentrated with certain dissolved solids, the pH can be

more stable because of buffer effects.
A solution containing an activity of orthosilicic acid greater than its solu-
bility product ( K ) is oversaturated and SiOz will precipitate. If the activity
s.p
of orthosilicic acid is less than its K s p , SiQz can dissolve until (H4Si0,)
= Ksp. The same applies for other constituents.
Complex compounds and ion pairs are important chemical transport
mechanisms. An example of a complex compound is a metal chelate such as
copper chelated with ethylenediamine tetraacetate, with the interacting
ligands immediately adjacent to the metal cation. An ion pair is formed if
the coordinated water is retained in forming the complex (Stumm and
Morgan, 1970).
The redox potential (Eh) also influences the transport of metals in solu-
tion. Most surface waters in contact with the atmosphere will have an
oxidizing potential, Fig. 7.2. However, many subsurface waters have low
redox potentials, and metals such as iron and manganese are transported as
reduced species.
600 -
I
400-
I /M,odern sea water
r I
W
200 - Acid I Alkaline
Fe++ I
-400 -
-600 -
I I I I I H2, I
-eooo 2 4 6 8 10 12 14
PH
Fig. 7.2. Diagram of some EhlpH relationships.

Sorting of weathered products
Products of weathering are transported by water by rolling along the

streambed, by suspension of the smaller particles, and by solution of soluble
components. As the larger rocks and pebbles roll along the streambed, they
are abraded by bumping against each other and by abrasive action with the
rocks in the streambed. The size of the clastics, which are detritus trans-
ported mechanically t o the point of sedimentation and portions solubilized
by water before sedimentation, decreases in the downcurrent direction
(Pettijohn, 1957), and this change in grain size is primarily a sorting effect.
The effect is noted in both fluvial and marine deposits.
Mobile belt
f
Geosynclinal trough Borderland
r
For eland
(stable) Shelf area
Fig. 7.3. Idealized depositional basin. (After Moore, 1969.)
Knowledge of the sorting of the clastics is used in reconstructing the

ancient environment (Visher, 1965).This knowledge can be applied to ex-
ploration for petroleum and other valuable minerals. A simplistic deposi-
tional basin is shown in Fig. 7.3; the deposited clastics will be found on the
borderland side of the basin and not on the foreland side.
Depositional environments of clastics
Depositional environments of the clastics include eolian, fluvial, regressive

marine, transgressive marine, deltaic, bathyal-abyssal, and lacustrine. Eolian
deposits are sands that are drifted and arranged by currents of air or wind.
Fluvial deposits are those related t o streams, rivers, and ponds. Water in
these environments usually contains less than 10,000 mg/l of dissolved
solids.
Regressive marine deposits are land-derived sediments that are transported
seaward and settle in the ocean. The salinity of the water transporting these
sediments will vary, it is fresher at its source and becomes more brackish as it
nears the sea. The dissolved solids in contemporary sea water are about
35,000 mg/l, while some estuary waters contain about 20,000 mg/l of dis-
solved solids.
Transgressive marine deposits usually are small in volume compared t o
fluvial and regressive-marine deposits. Such deposits are formed mainly by
erosion and redeposition of sediment deposits. The salinity of the sea may
change somewhat during transgression but probably not much.
Deltaic deposits result from a combination of environmental factors and
are related t o both fluvial and regressive marine processes. During flood
times the rivers transport tremendous volumes. of material, both clastic and
organic, into delta areas. Deltaic deposition is a very important factor in the
formation of petroleum because of the tremendous amount of organic ma-
terial deposited.
Bathyal-abyssal deposits are formed in deep-water areas in the sea, and
turbidity currents are responsible for most of the clastic deposition (Emery,
1960). Lacustrine deposits are those that are formed in lakes. If the lake is a
fresh-water lake, the dissolved solids may be less than 1,000 mg/l; in a
salt-water lake, the dissolved solids may be greater than 35,000 mg/l.
Consider a simplistic sedimentation area where the borderland area is the
prime source of sediments. The coarse- t o fine-grained clastics which are
weathering products of the high-mountains borderland are deposited near
their source. The clastics are detritus transported mechanically to the point
of sedimentation and are not solubilized by the water before deposition.
Primarily they are the sandstones and shales (clays). They will not be found
on the for.eland side of a depositional basin. Clay deposition can be detrital
or authigenic; illite often is detrital. There are at least two dozen clay
minerals, many of which occur in very minute grains and most of which
cannot be resolved by high-power petrographic microscopes. The electron
microscope, X-ray diffraction, and differential thermal analysis are used to
determine the type of clay.
The clays are very important in relation t o petroleum and gas because
they are the major component in the shales from which petroleum and gas
are generated. The clays also possess base exchange properties which will
react with constituents in water and petroleum. The detrital clays settle in
low-energy waters and they settle more rapidly from a saline water than from
a fresh water.
Depositional environments of the carbonates
Limestones and dolomites are the dominant carbonate reservoir rocks,

while the sandstones are the dominant clastic reservoir rocks (Ham, 1962).
The carbonates were precipitated at the place where the rocks first formed,
while clastics were primarily transported grains.
Plumley et al. (1962) classified the carbonates according t o an energy
index of the water from which they precipitated. They divided them into
five types. Type I is deposited in quiet water; it consists of calcite, 15-5096
clay, and < 5% detrital quartz. Type I1 is deposited in intermittently
agitated water and consists of calcite, < 25% clay, and < 50%detrital quartz.
Type I11 is deposited in slightly agitated water and it consists of calcite with
up to 50%detrital quartz. Type IV is deposited in moderately agitated water
and consists of calcite with up to 50% detrital quartz. Types I11 and IV are
similar in percentage of minerals; however, they are further differentiated
according t o grain size, sorting, roundness, and fossils, as are the other types.
Type V is deposited in strongly agitated (high-energy) water and it consists of
calcite, < 5% clay, and < 25% detrital quartz.
The calcium in the carbonates is released during rock weathering and goes
into solution as bicarbonate. The solubility of calcium carbonate in water is
also dependent upon the amount of carbon dioxide in solution. If the
amount of dissolve carbon dioxide decreases, calcium carbonate is precipi-
tated; therefore, the amount of calcium carbonate precipitation increases in
warm water because the amount of carbon dioxide in solution is less than in
cold water. Considerable amounts of carbonate precipitation occur in warm
environments (Illing, 1954). Aquatic plants also absorb carbon dioxide and
cause carbonate precipitation. The deposited carbonates can be pure or
mixed with sand, clay, iron, manganese, and organic matter.
Modern carbonate deposition occurs as deep-water oozes, and reefs, and
on shallow shelves. The deepwater oozes form along the flanks of ocean
basins at depths of less than 6,000 m. They do not form at depths greater
than 6,000 m because the calcium carbonate solubility increases with the
increased pressure. On the flanks of the basins, terrigenous material mixes
with the calcium carbonate. Often the mixture is 65--89% percent calcium
carbonate with silt making up the remainder (Gevirtz and Friedman, 1966).
Reef carbonates develop in open oceans on shallow platforms forming
atolls, as isolated patches on the ocean shelves, or along the margins of shelf
areas as fringing reefs. Fringing reefs generally occur in tropical regions on
the western side of the ocean basin. Reefs form as a result of living organisms
which form the framework of the sediment (Ginsburg and Lowenstam,
1958).
Present-day shelf carbonates are developing in Florida Bay, on the Bahama
Banks, on the Australian shelf, and on shelves off British Honduras, Yucatan,
and India. The precipitation often occurs as shallow carbonate mud banks.
The sediment in the shallow shelf areas often consists of about 10%skeletal
material mixed with oolites, mud aggregates, grapestone, aragonite needles,
calcareous algae, etc. The average rate of carbonate accumulation on the
Bahama Banks is 50 mg cm-2 yr-I (Broecker and Takahashi, 1966). The
salinity of the water ranges from 36,000 mg/l of dissolved solids t o 46,500
mg/l. The more saline waters occur in lagoonal areas. The pH ranges from
about 8.0 to 8.2 and is lowest at the end of the day because of C02
extraction from the water by marine plants.
In the simplistic depositional basin shown in Fig. 7.3 (Moore, 1969),
limestones and reef limestones will be deposited on the foreland side, formed
from water soluble constituents that precipitated from the saline solutions.
Depositional environments of evaporites
Calcium carbonate precipitates from sea water after it begins to evaporate

(Usiglio, 1849). Removal of calcium ions from solution increases the Mg/Ca
ratio in the residual brine, and the precipitated calcium carbonate reacts with
the magnesium enriched brine t o form dolomite (Deffeyes et al., 1964). The
common order of evaporite deposition in a basin cut off from the open sea is
limestone > dolomite > gypsum > halite > potash. Evaporite deposition can
be stopped in a basin by a change in climatic regimen or tectonism. If it were
changed and new water were allowed to enter, the already deposited salts
probably would be effectively protected by the superadjacent high-density
brine, and the lighter waters either meteoric or sea water would float on top,
developing euxinic conditions similar to a situation found in the Black Sea.
Toxic conditions in euxinic areas tend t o preserve deposited organic
matter. The preserved organic matter later can be transformed to hydrocar-
bons and petroleum. Hypersaline lagoons often are very prolific in the
production of organic matter (Phleger and Ewing, 1962). Algal pads and an
abundance of organic organisms are characteristic of a high-pH, high-salinity
environment (Carpelan, 1957) such as exists before euxinic conditions
develop.
The deposition of evaporites occurs when water evaporates under re-
stricted environmental conditions. The restrictions usually are caused by
tectonism such as an uplift, regression of the strand line leaving a relict sea,
or biological building of a reef.
Sloss (1953) outlined five environments from which evaporites deposit;
they are normal marine, euxinic, penesaline, saline, and supersaline. Primary
carbonates and dolomitized carbonates precipitate from the normal marine
environment where the water contains about 35,000 mg/l of dissolved solids.
Limestones rich in organic matter and black shales often are related t o
euxinic environments.
In the penesaline environment carbonates and primary dolomite form.
The salinity of the water is toxic to normal marine life but not sufficiently
saline so that halite precipitates. The dissolved solids in the water range from
about 250,000 mg/l t o 350,QOOmg/l.
From a saline environment carbonates, sulfates, and halites will precipi-
tate. From a supersaline environment potassium compounds will precipitate,
and the amount of dissolved solids in the solution will be about 500,000
mg/l.
A simplistic model of an evaporite basin is a closed basin initially filled
with sea water and evaporated to dryness. In the isochemical system a layer
of calcium carbonate is deposited over the entire basin floor as evaporation
proceeds. Next, gypsum is deposited and when the water is reduced to about
10%of the original volume, halite precipitates. Halite will deposit only in the
deeper parts of the basin and if the solution goes t o dryness the more soluble
salts will deposit in any remaining depressions and on top of the halite.
Deffeyes et al. (1964) proposed that dolomitization of limestone results

from the evaporation of sea water and precipitation of gypsum causing the
ratio of magnesium to calcium in the water t o increase. The concentrated
water flows downward, because it is more dense, into the underlying sedi-
ments where it reacts with limestone to form dolomite.
Modern evaporite deposits are thin and cover relatively small areas of the
earth; however, the ancient environments indicate that these depositions
were widespread in the United States (Krumbein, 1951) and in the world
(Lotze, 1938). The majority of the major evaporite bodies are of marine
origin, and range in age from Cambrian to Tertiary. They form in arid marine
climates where water lost by evaporation equals or exceeds that supplied by
rainfall, rivers, or the open sea. They also form in deep-water environment
(Brongersma-Sanders, 1971).
The data in Table 7.111 illustrate how the concentrations of some of the
dissolved constituents change as sea water is evaporated to dryness (Collins,
1969a). As sea water is evaporated to dryness and the chloride concentration
approaches 178,000 mg/l, calcium sulfate precipitates. Halite precipitates
when the chloride concentration approaches 27 5,000 mg/l, magnesium
sulfate precipitates during the next stage as the chloride concentration
approaches 277,000 mg/l. The concentrations of the ions in solution at these
various stages approximate those shown in Table 7.111. The data in Table
7.111 indicate that, as sea water evaporates, the concentrations of lithium,
magnesium, boron, chloride, bromide and iodide in the residual liquor in-
crease, and that the concentrations of sodium, potassium, rubidium, calcium,
and strontium decrease. In most depositional areas, the brines never reached
the concentration necessary for the deposition of potassium and magnesium
TABLE 7.111
Concentration changes during evaporation of sea water and brine
Element Sea water CaS04.l NaCl 4 MgS04.l KCI 4 MgClz.1
Lithium 0.2 2 11 12 27 34
Sodium 11,000 98,000 140,000 70,000 13,000 12,000
Potassium 350 3,600 23,000 37,000 26,000 1,200
Rubidium 0.1 1 6 8 14 10
Magnesium 1,300 13,000 74,000 80,000 130,000 153,000
Calcium 400 1,700 100 10 0 0
Strontium 7 60 10 1 0 0
Boron 5 40 300 310 750 850
Chloride 19,000 178,000 275,000 277,000 360,000 425,000
Bromide 65 600 4,000 4,300 8,600 10,000
Iodide 0.05 2 5 7 8 8
Total 295,000 517,000 469,000 538,000 602,000

salts; or if they did, these salts were later removed by leaching so that their
occurrence is relatively rare.
Holser (1963) analyzed some brine inclusions in halite from Permian age
evaporites. He found that the Br/Cl and Mg/Cl ratios in many of the brine
inclusions are similar to those found in the late stages of halite deposition.
He concluded that some of the inclusions were connate bitterns with few
diagenetic changes, and that the Br/Mg ratio of sea water has remained
relatively constant since Permian time. Some diagenetic changes were evident
in a few of the inclusions in which a large ratio of Ca/C1 and a low ratio of
SO4/Cl compared t o sea water were found.
Sediments commonly associated with evaporites are red beds, quartzose
sandstones, subgraywacke sandstones, carbonate rocks, and marine shales
(Krumbein, 1951). Normal marine evaporite successions are found in inter-
cratonic basins such as the Michigan and Williston Basins. Euxinic black
shales sometimes are associated with evaporites. Low redox potentials have
been found in modern evaporite (Morris and Dickey, 1957; Quaide, 1958).
Examples of modern depositional evaporites are the Karaboghaz Gulf on the
eastern side of the Caspian Sea, the Great Bitter Lake of Suez, the Rann of
Cutch in northwest India (Grabau, 1920), and the Persian Gulf sabkhas
(Evans et al., 1963; Butler, 1969).
Deposition of organic matter
The organic matter can be biogenic (produced by living systems) or

abiogenic (not produced by living systems). The source of biogenic matter
can be both terrestrial and marine; for example, considerable plant and
animal debris is collected from the land by streams and rivers and carried to
the sea, while in the sea large quantities of plant and animal matter live and
die.
The organic matter that is deposited with sediment usually decomposes if
the conditions are right; however, if the environment is reducing some of it
may be preserved. The preserved organic matter is transformed into other
organic compounds (Kvenvolden, 1964). During sediment diagenesis the
organic matter is transformqd to insoluble organic matter (kerogen) and
soluble petroleum hydrocarbon (Hunt and Jamieson, 1958). Chemical,
bacterial, and catalytic reactions are involved in these conversions. Tempera-
ture and pressure affect the reaction rates. Some of the chemical reactions
are as follows:
(1)Oxidation : C2Hm + 0 2 -+nC02 + '/znH,O

(2) Reduction: R'OH + H2 + R'H +H20
(3) Elimination: R'COOH + R'H + co2
(4) Polymerization: (small units) + big molecule
(5) Cracking: c-c-c-c-c-c~oc-c-c- + c-c
The organic matter produced by photosynthesis in the oceans is estimated

t o be sufficient t o produce 11 million metric tons of hydrocarbon precursors
annually (Riley, 1944). A very small amount of this organic material pre-
served in sedimentary rocks each year through geologic time would supply
all of the known oil and gas fields plus many undiscovered giant fields.
Shales consisting of organic material, siltstones, claystones, and limy mud
mixtures are found in the trough of a basin (Fig.7.3.). A simplistic idealized
view of the trough area is that organic matter deposited in the trough is
preserved in the stagnant low-Eh environment. If rapid subsidence occurs,
the organic material later is transformed into hydrocarbons which move up
and out of the trough into stratigraphic traps on the foreland side or struc-
tural traps on the borderland side of the basin.
The time interval between deposition of the organic material and con-
version to petroleum is millions of years, during which time the trough or
ocean basin is filled with sediment and buried, and the sediment is com-
pacted to rock. Some of the water in which the sediments deposited will
remain in the rocks as interstitial water.
Deposition of silica
Most of the silica in sediments is of the detrital variety but some is

authigenic. Silica is dissolved by waters with high pH potentials and precipi-
tated from water with a low pH. The precipitated silica often acts as a
cement.
Sediment compact ion
Sediments compact or consolidate in response t o an imposed load, and in

the natural environment, the load is the weight of overlying sediments
(Weller, 1959). Compaction of a sediment results in a reduction of the
interstitial volume concurrent with expulsion of interstitial water and defor-
mation of the sediment skeleton (the solid granular framework exclusive of
bound interstitial water). The grains and the interstitial water are almost
incompressible, and the rate of expulsion of interstitial water is about
identical with the rate of compaction (Taylor, 1956).
Terzaghi and Peck (1968) studied the expulsion of pore water from un-
consolidated clays, and determined that the rate at which clay compaction
occurs is dependent upon the clay permeability, its volume compressibility,
and the square of the thickness of the bed which is compacted. Shales
decrease in porosity when compacted; for example, their porosity can
decrease about 80%as they compact during burial.
White (1957) noted that very large quantities of water are removed from
sediments during their compaction. For example, water-saturated shale
decreasing in porosity from 20 t o 10% loses 10" liters of water per km3 of
sediment. Such sediment could yield per km3 a water supply of 20 liters per
minute for 10,000 years. Compaction water is further considered in a sub-

sequent chapter concerned with the accumulation of petroleum.
The resistance t o flow of bound interstitial water is greater than the
resistance to deformation of the sediment framework during the first stages
of compaction. As pressure is increased and sediment compacts, the sedi-
ment framework increases resistance to deformation, which also governs the
deformation rate of the bound-water films and the outflow of the interstitial
water. Rosenqvist (1962)found that bound water has a higher viscosity than
unbound insterstitial water. Permeability decreases with compaction, and
this together with the increased water viscosity leads to additional resistance
to expulsion of bound interstitial water during subsequent compaction.
Porosity decreases during compaction, and at infinite depth it would
become infinitely small. Porosity and permeability are two important factors
in determining the amount of petroleum and/or water that can be recovered
from a given reservoir or aquifer (Pollard and Reichertz, 1952;Caraway and
Gates, 1959). The porosity of an aquifer indicates how much fluid the
aquifer can hold, and the permeability indicates how fluid can move through
the aquifer. If the porosity is high but the permeability is low, the reservoir
may contain large amounts of oil, gas, or water, but they cannot be
recovered unless special techniques are used to increase the permeability.
Sediment diagenesis
Mineralogical and chemical changes occur in the sediments as they com-

pact. The mineralogic composition of recent marine sediments and ancient
marine sedimentary rocks are different, as is the composition of the intersti-
tial water in the recent sediments compared with the waters in ancient
stratigraphic units. Chemical reactions occur between the sediments and
their interstitial water (Chave, 1960). It has been shown that chemical
changes can be measured in recent sediments, e.g., below the sediment-
water interface, changes in pH and Eh result from degradation of sulfate ions
by bacteria (Emery and Rittenberg, 1952).
Numerous chemical inhomogeneities occur within a single core sample of
recent sediments (Degens and Chilingar, 1967). The magnesium concen-
tration in the interstitial w&er decreases slightly with depth, while the con-
centrations of calcium and potassium increase (Siever et al., 1965). The
interstitial waters from continental shelf sediments have higher chloride con-
centrations and higher ratios of Ca/C1, K/Cl, and Rb/C1 than the overlying sea
water; the ratios of Li/Cl and Mg/Cl are about the same as in the overlying
sea water except that the Li/C1 ratios are higher in the innershelf samples
than in the outershelf samples. The Sr/Cl ratio is higher in the overlying sea
water, while the pH and Eh values are lower in the sediments than in the
overlying sea water (Friedman et al., 1968).
A detailed study indicates that virtually no environment exists on or near
the earth’s surface where the pH/Eh conditions are incompatible with
organic life (Baas Becking et al., 1960). Because COz is the main byproduct
of organic oxidation and the building material of plant and much bacterial
life, it plays a dominant role. Carbon dioxide dissolves in water, producing
the bicarbonate ion and a free hydrogen ion. The concentration of the
hydrogen ion is lo-’ equiv./l (pH 7) at 2OoC in pure water, but when
saturated with COz it rises t o lo-’ (pH 5). This reaction moves to the right
with increasing temperature in a closed system. In the presence of organic
constituents, the equilibrium is modified, and the pH range can extend from
2 to 12.
The ionic potentials of the constituents involved in diagenesis are impor-
tant (Cloke, 1966). Those that stay in true ionic solution up t o rather high
pH levels are Na+, K+, Mg+’, Fe+’, Mn+2,Ca+’, Sr+’, Ba+’, etc.; they are
the soluble cations, and their ionic potentials range from 0 t o 3, where the
ionic potential is the ratio between the ionic charge and the ionic radius.
Constituents that are precipitated by hydrolysis are those with ionic poten-
tials from 3 t o 12 and include such ions as A P 3 , Fe+3, S P 4 , and
Constituents that form soluble complex ions and usually go into true ionic
’,
solution include B+3 , V 4 , N+ P+’ , S6, and Mn+’ ; their ionic potentials
are over 12. In general, the hydroxides of the soluble cations possess ionic
bonds; therefore, they are soluble, the hydrolysates or those precipitated by
hydrolysis form hydroxyl bonds, and the soluble complex ions have hydro-
gen bonds.
Organisms that consume oxygen cause a lowering of the redox potential,
and in buried sediments it is the aerobic bacteria that attract the organic
constituents and remove the free oxygen from the interstitial water. Sedi-
ments laid down in a shoreline environment often differ in degree of oxida-
tion from those laid down in a deep-sea environment (Pirson, 1968). For
example, the Eh of the shoreline sediments may range from -50 t o 0 mV
while the Eh of deep-sea sediments may range from -150 to -100 mV. The
aerobic bacteria die when the free oxygen is totally consumed, and the
anaerobic bacteria attack the sulfate ion which is the second most important
anion in the sea water. During this attack, the sulfate is reduced to sulfite
and then t o sulfide. Also the Eh drops t o -600 mV (Fig.7.2). Sulfide is
liberated and CaC03 precipitates as the pH rises above 8.5 (Dapples, 1959).
Sulfur has two stable isotopes, * S and S, with a mass differential of 6%.
The isotopes are fractionated during the change of S04-2 to Sv2, and SZ
is enriched in the more energetic 32S isotope. The average ratio of 3zS/34S
in normal sea water sulfate is about 21.76 (Ault, 1959). The sulfate isotopes
are useful in interpreting ancient diagenetic stages.
Reactions occur during sediment diagenesis and affect the composition of
the interstitial water. Calcite is precipitated if the pH is high, or it is dis-
solved if the pH is low. Dolomitization occurs as follows:
2CaC03 + MgClz * CaMg(C03)2 + CaClz

Other exchange reactions occur, whereby montmorillonite transforms to

illite (Burst,‘1969). Ions are adsorbed by negatively charged clays from the
sea water (Krauskopf; 1956). Dissolved salts hydrolyze; e.g., olivine hydrol-
yzes to serpentine as follows:
2Mg2 SiO, + 3H20 =+3Mg * 2Si02 2H20 + Mg(OH)2
Hydration and dehydration reactions occur, such as the gypsum-anhydrite

relation :
CaS0, + 2 H 2 0 + CaSO, * 2H20
As depth of burial increases, many mineralogical changes take place in the

sediments. In the Gulf Coast Tertiary, the clay mineral montmorillonite
gradually disappears at depths between 2,500 and 3,000 m (Burst, 1969). It is
replaced by mixed layer and illitic clay minerals. This change involves
chemical alteration and also the release of water of crystallization. This
change appears t o be temperature dependent with the reactions starting at
about 100°C.
Other mineral changes occur in the sandstones such as the deposition of
secondary silica overgrowths on the quartz grains causing resultant loss in
porosity. Available data differ on the loss of porosity with burial depth
(Atwater and Miller, 1965; Philipp et al., 1963). Much of the silica may
come from outside the porous sand, such as from shales whose pore water is
traversing the sand as a result of compaction. Authigenic clay minerals such
as kaolinite also form in the pores.
Interstitial water in deeply buried sediments sometimes is at a pressure
close t o the weight of the overburden. This pressure may be sufficient t o
burst the rock and allow the water t o move out through the fissures which it
forms. These fissures are principally vertical and extend upward into zones
of lower temperature. The water forming and subsequently filling the fis-
sures usually is a mixture of salty interstitial sedimentary pore water and
water from the clay minerals released by their recrystallization. If it is hot,
and saturated with silica and other minerals, it could force its way upward
and as it cools deposit quartz, feldspar, calcite, and other minerals. Possibly
many hydrothermal ore veins were formed by interstitial sedimentary waters
rather than by “juvenile” waters.
These hydrothermal veins contain metallic minerals composed of com-
pounds containing copper, zinc, lead, gold, and silver. The process is a
geothermal convection cell and it is able to concentrate and segregate useful
minerals. The process has many points of resemblance to the concentration
and segregation of petroleum - the principal difference being that the geo-
thermal convection cells operate at higher temperatures.
Petroleum and natural gas
The total amount of organic matter dispersed in the sedimentary rocks of

the earth has been estimated to be about 2,700 trillion metric tons; of this
amount 50 trillion metric tons are dispersed petroleum hydrocarbons, of
which 0.5 trillion metric tons exist in petroleum reservoirs (Hunt, 1968).
The types of hydrocarbons in a petroleum often indicate its origin; for
example, if it contains predominately odd-numbered n-alkanes in the low
molecular-weight range, it probably was formed from marine organisms.
Petroleums from the Uinta Basin contain a predominance of odd-numbered
hydrocarbons in the vax fraction, indicating a nonmarine organic source.
Waxes derived from land plants contain a predominance of hydrocarbons
with carbon numbers of C27,CZ9,and C31,while hydrocarbons derived from
marine plankton may contain more hydrocarbons with carbon numbers of
c17, and c19.
The water in the sediments containing the organic matter contains many
dissolved organic constituents such as salts of the humic, fatty, and
naphthenic acids, sugars, heterocyclics, and aromatic oxygen compounds.
Degens et al. (1964) observed that as the salt concentration in petroleum-
associated waters increases, the concentration of dissolved amino acids in-
creases.
Petroleum is generated in organic-rich shales, but the mechanisms whereby
it migrates from the shales and concentrates in porous reservoir rocks are not
understood. Petroleum precursors leave the shale with the water as the water
is expelled by compaction.
As burial proceeds, pressures and temperatures increase. With increasing
temperature, chemical changes in the solids are accelerated and the organic
matter first generates petroleum, which ultimately is converted to methane
and finally graphite. The clay minerals continue their recrystallization, and
finally metamorphism to slates, phyllites, and schists occurs: These processes
involve a continuing loss of porosity with the release of additional pore
water.
The solubility of petroleum hydrocarbons in water increases with in-
creasing temperature and pressure. However, at ambient temperature and
pressure the solubility in pure water is rather low (McAuliffe, 1969). Water-
wet shale has no permeability t o immiscible fluids such as gas or oil, so the
petroleum or petroleum precursors probably do not move as droplets. Peake
and Hodgson (1966) report “accommodations” of specific hydrocarbons in
water up to about 30 ppm. Cartmill and Dickey (1970) found that a
colloidal suspension was able to pass through water-wet sands, but the tiny
droplets coalesced at points where the grain size decreased. Neruchev and
Kovacheva (1965) offered some evidence that the amount of extractable
hydrocarbon decreases in shales for the first few meters away from the
reservoir rocks, as if removed by some flushing action.
Bruderer (1956) suggested that oil deposits originated from sea water
SEDIMENTARY ROCKS 21 1
containing dissolved hydrocarbons, which subsequently deposited them in

sedimentary reservoir ,rocks. Al’tovskii et al. (1961) believe that petroleum
forms in subsurface formation waters and moves to traps as an emulsion or
in aqueous solution. Movement of hydrocarbons from the source bed to the
reservoir via aqueous solution or as molecular films was postulated by Brod
(1960). Baker (1960) suggested that, as the sediments compact, the express-
ed water may contain solubilizers capable of releasing sediment hydrocar-
bons into aqueous colloidal solution, and subsequent changes in the aqueous
equilibria cause the colloids to become oil droplets in reservoirs.
The soluble salts of adenosine triphosphate are capable of solubilizing
numerous inorganic and organic compounds in neutral or slightly alkaline
medium and of keeping these compounds in solution (Mandl et al., 1952).
Hydrocarbon solubilization processes are essential to the migration of
petroleum and acquired evidence indicates that nature provides solubilizing
agents. Numerous solvents are evaluated by Mandl (1953), many of which
are useful agents for solubilizing relatively insoluble inorganic and organic
compounds.
Organic as well as inorganic compounds will enter the aqueous phase to
the limit of their solubilities. Interactions of the solubilized compounds
affect the solubility product of other solubilized compounds. For example,
the presence of dissolved sodium chloride may inhibit or increase the solu-
bility of a normal alkane. For any given system, equilibria of all components
will be attained only at a given temperature and pressure. If the temperature
changes, if the pressure changes, or if more inorganic or organic constituents
are added or subtracted, the equilibria will change causing solubilization or
deposition.
Hydrocarbons can exist in the aqueous phase as emulsions or colloids, as
suspended particles, or in true solution. Suspended particles will settle from
an aqueous phase because of gravity, whereas colloids will remain in suspen-
sion. Suspended particles possess colligative effects, but colloidal sols do not
(Van Nostrand Press, 1958). A dissolved particle usually is smaller than 5
mp, while colloidal particles usually range in size from 0.25 mp to 6 mp.
Phenols and alkalinaphthenates capable of acting as emulsifying agents
and causing oil-in-water emulsion t o migrate have been detected in crude oils
and in the associated brines (Neumann and Jobelius, 1967). It has been
demonstrated that finite amounts of paraffinic, aromatic, napthenic hydro-
carbons and other organic derivatives are dispersed in recent marine sedi-
ments (Smith, 1954). Surfactant organic acids stabilize hydrocarbon
particles of relatively large size and contribute t o solubilization and subse-
quent migration of hydrocarbons from sediments (Baker, 1960). A shift in
the equilibria caused by a pH or Eh change, adsorption of the surfactant,
etc., results in deposition of the pseudo-soluble hydrocarbon. Therefore,
solubilization and mobilization of hydrocarbons from sediments into the
aqueous phase occurs when the equilibria are shifted t o the correct con-
ditions. Subsequent migration of the hydrocarbons in the aqueous phase
occurs if the equilibria in the aqueous phase remain constant while the water
moves through the sedimentary rocks. Deposition or accumulation of
hydrocarbons occurs if the equilibria in the aqueous phase shifted, causing
desolubilization or precipitation of the hydrocarbons.
Temperature gradients in sedimentary basins usually are about 1°C per 46
m of depth, and the rate of heat flow to the surface is approximately
1.2 x loV6 cal cmV2 sec-I (Birch, 1954). Temperature is believed to be a
primary cause in the conversion of organic matter in rocks t o petroleum
(Philippi, 1965), it is also believed that lipids are the major precursors of
petroleum and that most petroleum is generated by chemical reactions oc-
curring at temperatures above 115°C.
Nonmarine sources are recognized for many crude oils, in contrast to the
once general belief that such sources are unfavorable for the generation of
petroleum. Perhaps the best known examples in the United States are the
nonmarine sequences in the Eocene of the Uinta Basin in Utah. Other
examples of oil and gas with continental source sediments are basins such as
the Dzungaria, Tsaidam, Tarim, Turfan, Ordos, Pre-Nan Shan, and Sungliao
of China (Meyerhoff, 1970). There is considerable nonmarine Tertiary age
strata in the Cook Inlet-Kena Basin in coastal Alaska.
TABLE 7.IV
Tertiary system - highest concentration of a constituent found, average concentration.

and number of samples analyzed
Constituent Concentration (mg/l) Number of samples
highest average
Lithium 27 4 169
Sodium 103,000 39,000 379
Potassium 1,200 2 20 176
Rubidium 0.6 0.24 11
Cesium 0.4 0.20 9
Calcium 38,800 2,530 37 6
Magnesium 5,800 530 368
Strontium 420 130 142
Barium 240 60 140
Boron 450 36 170
Copper 1 0.63 3
Chloride 201,300 64,600 380
Bromide 1,300 85 323
Iodide 35 28 322
Bicarbonate 3,600 560 3 64
Carbonate 300 75 8
Sulfate 8,400 3 20 139
Organic acid
as acetic 1,900 140 53
Ammonium 2,700 230 64
COMPOSIDION OF OILFIELD WATERS 213
TABLE 7.V
Cretaceous system - highest concentration of a constituent found, average concentration,

highest average
Lithium 13 4 26
Sodium 88,600 31,000 987
Potassium 580 130 38
Cesium 0.10 0.10 1
Calcium 37,400 7,000 987
Magnesium 8,000 900 987
Barium 67 0 40 34
Boron 70 20 38
Chloride 187,000 62,OOb 987
Bromide 1,760 550 173
Iodide 190 25 172
Bicarbonate 1,660 260 864
Sulfate 7,100 280 776
Ammonium 35 23 2
Composition of oilfield waters
To illustrate the variety of oilfield waters found in subsurface petroleum-

bearing formations, samples were taken, and analyzed, from formations of
the following ages: Tertiary, Cretaceous, Jurassic, Permian, Pennsylvanian,
Mississippian, Devonian, Silurian, Ordovician, and Cambrian. Insufficient
samples from Triassic age strata were available. These data are given in Tables
7.IV-XIII, and all of the samples were taken from sedimentary basins in the
United States. Each table gives the highest value found in this study in
milligrams per liter for a given constituent, the average values, and the num-
ber of samples used t o estimate the average value. The constituents deter-
mined for most of the brines include lithium, sodium, potassium, rubidium,
cesium, calcium, magnesium, strontium, barium, boron, copper, chloride,
bromide, iodide, bicarbonate, carbonate, sulfate, organic acid as acetic
(actually organic acid salts but calculated as acetic acid), and ammonium.
Comparison of the lithium content in the samples from Tertiary age strata
(Table 7.IV) with the lithium content of sea water (Table 7.1), indicates that
the average lithium content of 169 oilfield waters is enriched by a factor of
20. At least one sample of oilfield water contained 27 mg/l of lithium or an
enrichment factor of 135 compared t o sea water. Table 7.11 indicates that
igneous rocks contain up t o 65 ppm of lithium and sedimentary rocks con-
tain up to 46 ppm of lithium.
TABLE 7.VI
Jurassic system - highest concentration of a constituent found, average concentration,

highest average
-
Lithium 400 10 80
Sodium 120,000 57,300 85
Potassium 900 140 9
Cesium 0.10 0.10 1
Calcium 56,300 25,800 85
Magnesium 5,200 2,500 84
Strontium 2,080 320 9
Barium 50 10 7
Boron 50 13 9
Chloride 210,000 141,000 85
Bromide 6,000 1,200 80
Iodide 40 16 8
Sulfate 1,480 21 0 78
Organic acid
as acetic 12 12 1
TABLE 7.VII
Permian system - highest concentration of a constituent found, average concentration,

-
highest average
Lithium 6 3 3
Sodium 109,000 47,000 54
Potassium 40 5 170 3
Rubidium 2 0.80 3
Cesium 0.20 0.13 3
Calcium 22,800 8,600 54
Magnesium 5,800 2,000 53
Strontium 10 7 3
Boron 20 8 3
Copper 0.88 0.88 1
Chloride 177,000 92,700 64
Bromide 68 46 3
Iodide 3 3 1
Carbonate 36 36 1
Sulfate 3,400 7 30 41
Organic acid
as acetic 2 20 170 2
Ammonium 24 24 3
COMPOSITION OF OILFIELD WATERS 21 5
TABLE 7.VIII
Pennsylvanian system - highest concentration of a constituent found, average concentra-

tion, and number of samples analyzed
highest average
Lithium 35 7 45
Sodium 101,000 43,000 951
Potassium 710 170 57
Rubidium 2.30 0.55 25
Cesium 8.50 0.15 19
Calcium 205,000 9,100 9 50
Magnesium 15,000 1,900 947
Barium 640 30 41
Boron 70 15 54
Manganese 105 60 2
Chloride 270,000 87,600 950
Bromide 3,900 490 57
Iodide 1,410 210 52
Carbonate 70 40 2
Sulfate 5,400 430 7 56
Organic acid
as acetic 2,300 430 44
Ammonium 3,300 300 51
Table 7.V indicates that oilfield water samples taken from Cretaceous age
strata were enriched in lithium with respect to sea-water. The highest lithium
concentration found in 26 samples was 13 mg/l.
Table 7.VI indicates that oilfield waters taken from Jurassic age strata
contain up t o 400 mg/l of lithium, which, compared with sea water (Table
7.1), represents a concentration factor of 2,000. Compared t o the hydro-
lyzates in sedimentary rocks (Table 7.11), the concentration factor is about
9.
The lithium concentration in oilfield waters taken from Permian age strata
averaged 3 mg/l (Table 7.VII). Only three samples were available for use in
determining this average.
Table 7.VIII indicates that the lithium concentration averaged 7 mg/l in
45 samples taken from Pennsylvanian age strata. This represents a concen-
tration factor of 35 compared with sea water (Table 7.1).
Table 7.IX indicates that the lithium concentration in oilfield waters
taken from Mississippian age strata is enriched by a factor of 45 compared
with sea water. Table 7.X indicates a similar enrichment factor of 250 for
oilfield waters taken from Devonian age strata. For waters from Silurian age
strata (Table 7.XI) the enrichment factor found was 185; for the Ordovician
TABLE 7.IX
Mississippian system - highest concentration of a constituent found, average concentra-

tion, and number of samples analyzed
highest average
Lithium 55 9 81
Sodium 115,800 41,500 210
Potassium 5,000 430 80
Rubidium 5 1 47
Cesium 2 0.40 37
Calcium 37,800 8,900 209
Magnesium 11,200 1,600 202
Barium 20 5 44
Boron 240 40 86
Copper 3 3 2
Manganese 36 12 5
Chloride 206,000 85,000 210
Bromide 1,800 410 88
Iodide 620 110 89
Carbonate 450 450 1
Sulfate 3,500 540 191
Organic acid
as acetic 3,070 370 84
Ammonium 700 210 83
age (Table 7.XII), it w a s 100; and for the Cambrian age (Table 7.X111), it was
85.
The data in Tables 7.VI-XI11 indicate that waters taken from sediments
that formed during the various geologic ages do not all have the same chemi-
cal composition and that the waters have evolved considerably in comparison
to modern sea water composition (Table 7.1). The manner whereby this
evolution occurred is not completely understood; however, recent studies
have shed some light on the problem. Note the amount of organic acid as
acetic found in waters taken from the sedimentary rocks (Tables 7.VI-XIII).
The organic acids are present in the oilfield waters as organic acid salts.
These organic compounds possibly are a precursor of petroleum and serve as
a transportation mechanism for migration. The exact composition of each
organic acid salt has not been determined. Knowledge of the composition of
these organic acid salts would aid in geochemical studies of petroleum.
Rittenhouse et al. (1969) studied the minor elements in 823 oilfield-water
samples taken from subsurface formations in the United States and Canada.
The data that they found are shown in Table 7.XIV as 25% quartiles, median
concentrations, and 75% quartiles. The dissolved solids are given in grams per
COMPOSITION OF OILFIELD WATERS 217
TABLE 7.X
Devonian system - highest concentration of a constituent found, average concentration,

highest average
~~
Lithi um 170 50 29
Sodium 101,000 48,000 85
Potassium 11,600 3,100 30
Rubidium 11 4 12
Cesium 1.4 0.5 11
Calcium 129,000 18,000 85
Magnesium 26,000 2,900 82
Strontium 2,300 1,000 8
Barium 120 40 7
Boron 90 30 30
Copper 2 2 1
Manganese 200 175 2
Chloride 259,000 115,000 85
Bromide 3,500 1,060 32
Iodide 120 30 32
Carbonate 60 30 2
Sulfate 1,700 450 74
Organic acid
as acetic 67 0 130 27
Ammonium 560 110 32
liter, the data followed by p are given in parts per billion (ppb), and the
other data are given in parts per million. They analyzed samples from several
basins as illustrated in Table 7.XIV, and the elements analyzed included
lithium, magnesium, manganese, nickel, cobalt, chromium, copper, potas-
sium, tin, strontium, titanium, vanadium, and zirconium.
Rittenhouse et al. (1969) concluded that elements in oilfield waters com-
monly are present in the following concentrations:
5% Na, C1
5% or ppm Ca, SO4
> 100 ppm K, Sr
1-100 ppm Al, B, Ba, Fe, Li
ppb (most oilfield waters) Cr, Cu, Mn, Ni, Sn, Ti, Zr
ppb (some oilfield waters) Be, Co, Ga, Ge, Pb, V, W, Zn
They found no relationship between the constituents in the brine and the
minerals in the aquifer rocks except for potassium. They postulated that
exchange reactions occurred between the clays in the rocks and potassium in
the water to control the dissolved potasssium.
TABLE 7.XI
Silurian system - highest concentration of a constituent found, average concentration,

highest average
-
Lithium 90 37 8
Sodium 89,000 49,100 14
Potassium 8,400 1,900 11
Rubidium 8 4 2
Cesium 0.4 0.4 2
Calcium 41,000 21,000 14
Magnesium 12,000 4,300 12
Strontium 880 7 30 2
Barium 15 15 1
Boron 90 30 10
Chloride 195,000 122,000 14
Bromide 1,700 520 11
Iodide 30 17 10
Bicarbonate 27 0 115 11
Sulfate 3,500 830 13
Organic acid
as acetic 220 90 9
Ammonium 20 0 80 10
TABLE 7.XII
Ordovician system - highest concentration of a constituent found, average concentration,

highest average
Lithium 70 20 15
Sodium 89,100 31,000 609
Rubidium 6 2 11
Cesium 0.5 0.2 9
Calcium 39,000 6,100 609
Magnesium 10,900 1,300 607
Barium 10 6 10
Boron 80 20 18
Manganese 56 56 1
Chloride 205,600 62,000 609
Bromide 7 20 300 19
Iodide 70 25 16
Carbonate 60 25 26
Sulfate 7,600 1,070 583
Organic acid
as acetic 3,300 5 20 14
Ammonium 630 140 16
RESEARCH STUDIES 21 9
TABLE 7.XIII
Cambrian system - highest concentration of a constituent round, average concentration,

_______--__ ~- - ~ __
highest average
._____ ~ _ _ _
Lithium 40 17 8
Sodium 43,000 23,400 23
Rubidium 3.3 3.3 1
Cesium 0.6 0.6 1
Calcium 14,500 4,000 23
Magnesium 8,800 1,300 22
Strontium 360 125 7
Boron 13 7 5
Chloride 95,000 46,100 23
Bromide 1,170 520 5
Iodide 40 18 3
Sulfate 2,600 1,170 22
Organic acid
as acetic 50 30 3
Ammonium 120 60 3
Compared with sea water the 823 brines were enriched in manganese,
lithium, chromium, and strontium, and depleted in tin, nickel, magnesium,
and potassium. Generally the silicon content varied inversely with the dis-
solved solids content. This agrees with a study of the solubilities of silicate
minerals where Collins (1969b) found that in general the silicon solubilities
decreased with increasing concentrations of dissolved salts at ambient con-
ditions.
Research studies related to the origin of oilfield brines
Tables 7.IV-XIV indicate that the compositions of oilfield brines are not
consistent, and that they are not formed by the simple evaporation or dilu-
tion of sea water. Oilfield brines are found in deep formations that some-
times contain fresher water nearer surface outcrop areas, in formations con-
taining evaporites or in close proximity to soluble minerals, and in forma-
tions close to surface saline waters.
The amounts and ratios of the constituents dissolved in oilfield waters are
dependent upon the origin of the water and what has occurred t o the water
since entering the subsurface environment. For example, some subsurface
waters found in deep sediments were trapped during sedimentation, while
other subsurface waters have infiltrated from the surface through outcrops.
N
ES
0
TABLE 7.XIV
Minor elements in 823 oilfield brine samples in United States and Canada*'
Number Lithium Magnesium Manganese Nickel Tin

of
samples q25 md q75 q25 md q75 q25 md q75 q25 md q75 q25 md q75
Illinois Basin 22 10 15 25 3,000 6,000 8,000 8Op 175p 750p ND ND ND < 1P 5P

Louisiana and Texas
Gulf Coast 79 ND ND 4 15 250 550 8OOp 3.5 >5,OOOp < lp < 3p ND < 1P 1P
East Texas 88 ND ND ND 150 250 800 1 , 8 0 0 ~ 3.3 >5,OOOp < 1P 3p ND 3P 1lP
North Texas 24 ND ND 15 3,000 5,000 6,000 25 45 90 15p 150p ND 12p 25p
West Texas and New
Mexico 148 3 15 25 500 1,000 1,650 2OOp 1.8 >5,OOOp < lp < l p ND < 1P 3P
Permian only 74 2 10 25 500 1,000 2,000 18Op 1.7 >5,OOOp < 1P 3p ND < 1P 3P
Pennsylvanian only 34 3 10 20 500 1,000 1,500 500p 2.8 >5,OOOp < lp < l p <Ip < 1P 3P
Silurian and Devonian
only 15 4 10 25 200 400 560 30p 300p >5,OOOp < lp < l p ND IP 5P
Ordovician and Cambrian
only 21 10 15 25 500 800 1,000 150p 400p >5,OOOp < lp < l p <1 IP 4P
Anadarko Basin*' 118 ND 10 35 900 1,550 3,000 600p 5.6 >5,OOOp 6p 15p ND 2P 4P
Williston Basin, post-
Paleozoic 25 ND ND 10 10 250 2,000 9Op 300p 450p < 3p < 3p ND < lp < lp
Williston Basin,
Paleozoic 55 18 35 50 300 600 2,000 2OOp 660p 1,200~ ND < 3p ND < lp < lp
Powder River Basin 22 ND ND 2 10 40 225 300p 450p 2,000~ < 3p < 3p ND < lp < lp
Other Wyoming 28 ND ND 45 20 100 200 60p 300p 1,000~ ND 3p ND < lp < lp
Colorado 18 ND ND ND < 10 30 300 9Op 300p 750p < 3p < 3p ND <lop <lop
California 116 ND ND ND 35 90 175 300p 950p 2,800~ lop 35p ND 2.5~ 4.5~
Sea Water 0.1 1,272 lp-lop 5.4p 3P
Estimated detection -
limit - 2 10 1P
1P _ 1P_ _ ~ - ___
*' Medium (md - Rittenhouse et al., 1969)and quartile (9) concentrations in each area; ND = below detection limits; p = concentration in ppb, otherwise ppm.
** No data, less sensitive methods of analysis used.
*3 Includes Oklahoma Platform and Ardmore Basin.
TABLE 7.XIV (continued)*'
Number Dissolved solids (gll) Cobalt Chromium Copper Potassium

of - ~
-
samples md q75 md q75 q75 md q75 q25 md q75
- ~ ~ ~
Illinois Basin 22 70 98 119 ND ND ND 3P <lop lop 75p 180 300 400

Louisiana and Texas Gulf
Coast 79 30 69 131 ND ND <5P < 2P <25p <25p <25p 160 300 400
East Texas 88 27 66 116 ND*~ ND*~ ND*~ 2P < 1P < 1 < 1 ND <50 300
North Texas 24 173 222 241 ND ND ND 35P 25P 150p 450p ND 300 1,000
West Texas and New
Mexico 148 61 111 173 ND ND ND 4P ND 1P 10P 120 350 500
Permian only 74 70 143 215 ND ND ND 4P ND 2P 10P 160 400 750
Pennsylvanian only 34 80 115 168 ND ND ND 4P ND < 1P 5P 200 300 400
only 15 42 55 72 ND ND ND 4P < 1P 4~ 1 5 ~ 170 300 450
Ordovician and
Cambrian only 21 53 67 128 ND ND <5P 3P N 4 4~ 1 5 ~ 200 400 650
Anadarko Basin*3 118 51 137 203 ND ND ND 2 5 ~ < 1P lop 25p 20 250 500
Williston Basin, post-
Paleozoic 25 9 59 88 ND <5P <5P < 2P <25p <25p 2Op 200 300 400
Williston Basin,
Paleozoic 55 115 173 296 ND ND ND 1 5 ~ ND 3P 5P .400 800 <5,000
Powder River Basin 22 3 5 11 <5P <5P <5P < 6~ <25p <25p 70p 200 300 400
Other Wyoming 28 4 5 11 ND ND ND 20 ND ND 30 ND 300 700
Colorado 18 3 5 15 ND <5P <5P ND < 6; <25p <25p <25p 200 300 400
California 116 5 18 30 ND ND 2P 5p 15p 2p 5p 2Op ND 45 70
Sea water - 35 0.27~ 0.04~-0.07~ lp-15p 380
Estimated detection
limit - - 1P IP 1P 50
_ _ - .-. ~~
*' Medium (md - Rittenhouse et al., 1969)and quartile (4)concentrations in each area; ND = below detection 1imits;p = concentration in ppb,otherwise ppm.
*' No data, less sensitive methods of analysis used.
to
to
to
Number Strontium Titanium Vanadium Zirconium

of
samples q25 md q75 q25 md q75 q25 md q25 md q75
__. ~ -
_____.
Illinois Basin 22 140 300 400 < l o p <1 op <lop ND ND <lop <lop
Louisiana and Texas
Gulf Coast 79 45 85 200 ND <lop <lop ND <lop <lop <lop
East Texas 88 75 350 750 ND ND < 1P ND ND ND ND
North Texas 24 150 450 700 < l o p 7P 2OP ND ND <lop lop
West Texas and New
Mexico 148 1 5 200 400 ND <lop <lop ND ND ND ND
Permian only 74 65 90 300 ND <lop <lop ND ND ND <lop
Pennsylvanian only 34 180 300 450 ND <lop <lop ND ND ND ND
only 15 75 90 300 < l o p <lop <lop ND ND ND ND
Ordovician and
Cambrian only 21 100 250 400 ND <lop <lop ND ND ND ND
Anadarko Basin 118 90 300 650 ND <lop <lop ND ND <lop 2Op
Williston Basin,
postPaleozoic 25 20 1 0 0 200 ND ND <lop ND ND ND ND
Williston Basin,
Paleozoic 55 50 95 450 < l o p <lop 25P ND ND ND <lop
Powder River Basin 22 ND 25 50 < l o p <lop <lop ND <lop <lop <lop
Other Wyoming 28 10 20 45 ND <lop <I op ND ND ND ND
Colorado 18 7 20 60 < l o p <1 op <lop ND ND <lop <lop
California 116 ND 1 0 22 < l o p <lop 7P ND ND ND ND
Sea water - 13 present ND
Estimated detection
limit - 16 10P 1OP
~ - -.
*' Medium (md; from Rittenhouse e t al., 1969) and quartile ( 4 ) concentrations in each area; ND = below detection limits; p = concentration
in ppb, otherwise ppm.
** No data, less sensitive methods of analysis used.
RESEARCH STUDIES 223
Some waters are mixtures of the infiltration water and trapped ancient sea
water.. Also, the rocks containing the waters often contain soluble con-
stituents which dissolve in the waters or contain chemicals which will
exchange with chemicals dissolved in the waters causing alterations of the
dissolved constituents.
The amounts of dissolved constituents found in oilfield waters range from
less than 10,000 mg/l t o more than 350,000 mg/l. This salinity distribution
is dependent upon several factors including hydraulic gradients, depth of
occurrence, distance from outcrops, mobility of the dissolved chemical
elements, soluble material in the associated rocks, ion exchange reactions,
and clay membrane filtration.
Concentration of sea water can occur by surface evaporation, and there
are at least three independent processes that can cause major changes in
buried, isolated sea water:
(1) Dilution with meteoric or fresher waters which have entered outcrops.
(2) Reactions with minerals in the sediments and sedimentary rocks (the
reactions are often temperature and pressure dependent).
(3) Membrane filtration through clays and shales as a result of pressure
and osmosis.
Playa deposits
Jones et al. (1969) studied the composition of brines in shallow, fine-

grained playa deposits in the Great Basin. The concentrations of dissolved
solids in the water in these sediments often were as much as five times
greater than in the water in the associated lakes. They attributed the concen-
tration processes t o capillary evaporation and entrapment of fossil brines
when the salinity of the lake water was greater (lake nearly dry).
Continental Slope drill holes
Manheim and Bischoff (1969) analyzed pore waters from drill holes on
the Continental Slope of the northern Gulf of Mexico. A relationship be-
tween the salinities of the waters and the proximity of diapiric structures
was found. This indicated that salts are leached from salt-bearing sediments
to increase the salinities of the pore waters. In some samples the high
bromide and potassium concentrations suggested that late-stage evaporitic
minerals such as carnallite and polyhalite were leached from salt bodies.
They postulated that molecular diffusion is a major mechanism which in-
fluences the distribution of salt in the pore waters. Similar conclusions have
been made for saline waters in other areas.
Relation to petroleum accumulations
Van Everdingen (1968) suggested that major circulation systems of for-

mation water exist in the Western Canada Sedimentary Basin, that the flow
systems affected accumulations of hydrocarbons in the basin, and further
that pressure and salinity variations might be explained by membrane
properties of the shales. He saw a need for studies of the hydrodynamics of
the basin.
Hydrodynamics and geochemistry of the Paradox Basin were studied by
Hanshaw and Hill (1969). The ground-water movement in the basin is
generally southwestward from the high outcrop areas in western Colorado,
flowing toward the Colorado River discharge areas. Hydrodynamic con-
ditions exist in lower Paleozoic strata which are favorable to accumulations
of petroleum in stratigraphic traps. Paleozoic aquifers in northwestern New
Mexico have very high potentiometric surfaces and these aquifers may be the
outflow receptors of an osmotic membrane system operating within the San
Juan Basin. This regional study was excellent and of value in exploration for
petroleum and gas.
Parker (1969) studied brines and waters in five aquifers of Cretaceous age
in the East Texas Basin. He found that the composition of the waters in the
older, more deeply buried aquifers were modified more than waters in
younger, less deeply buried aquifers. Most of the modifications were made
by exchange reactions, dilution by meteoric waters, and loss of sulfate be-
cause of bacterial reduction. Hydrodynamic movement of the waters in the
Woodbine formation contributed t o the giant oil accumulation in the East
Texas Basin. Much of the stratigraphic trapped oil probably was trapped in
part because of this type of flow.
Sabkha sediments and transport of valuable ores
Bush (1970) discussed the origin of chloride-rich brines from Sabkha sedi-
ments and how they are related t o inclusion brines and lead-zinc deposits of
the type found in the Mississippi Valley. He noted that Helgeson (1964) and
Barnes and Czamanske (1967) have shown that chloride-rich scrlutions can
transport lead and zinc as chloride complexes. According t o Bush (1970),
Sabkha brines free of sulfur are expelled by sediment compaction, migrate,
and become enriched in base metals until they contact a zone of higher
temperature and pressure. In this zone, sulfides are present as a result either
of inorganic reduction of sulfate, anaerobic reduction, or hydrocarbon
reduction of anhydrite. The sulfides cause the base metals to precipitate.
Formation brines are the medium in which several metals, in addition to
hydrocarbons, migrate prior to deposition in ore deposits. A current theory
is that the metals travel primarily as chloride complexes in solutions that are
depleted in reduced sulfur species (Dunham, 1970). The metals subsequently
are precipitated when a source of reduced sulfur is met. An example of a
source of reduced sulfur is an area where anaerobic bacteria are reducing
sulfate. This occurs in waters near petroleum-bearing formations, and such
waters in carbonate reservoirs often contain considerable amounts of sulfide.
Brine classification
A study of the evolution of subsurface brines in Israel by Bentor (1969)

led him to classify brines into four groups. The first group consists of brines
similar t o sea water except for an increased concentration of calcium and a
decreased concentration of magnesium, which he attributed to dolomitiza-
tion. The second group was similar t o sea water but contained two to three
times higher concentrations of dissolved salts, deficient in sulfate and
magnesium, and enriched in bromide and iodide. The sulfates were lost by
organic reduction, the magnesium was lost by exchange reactions with clays,
and bromide and iodide were added by organic sources. The third group was
a high-salinity calcium chloride-type brine formed by surface evaporation
and later modified in the subsurface by differential ultrafiltration, The
fourth group was a highly saline, calcium chloride type with Ca/Na ratios
greater than one. This group was divided into two subgroups where the first
subgroup is a highly saline and highly differentiated Early Paleozoic brine,
while in the second subgroup they are old Paleozoic brines which were
submitted t o an additional cycle of surface concentration by evaporation.
Ion association
Truesdell and Jones (1969) studied ion association in brines and found
that, except for the chloride ion, the major simple ions form ion pairs, while
the minor and trace metals in brines form coordination complexes. Selective
ion electrodes can be used to determine directly the ionic activities of
sodium, potassium, chloride, fluoride, and sulfide in brines. Experimental
data were used t o calculate chemical models for ion association and coordi-
nation complexes in brines. These models are useful in explaining the
chemical behavior of brines.
Relation t o lithology
Kramer (1969) used factor analysis to study the relationships of the brines
to the type of rock from which they were taken. His results indicated that
the major ions in most brines are sodium, calcium, and chloride; brines are
enriched in calcium and bicarbonate and are deficient in magnesium and
sulfate relative t o sea water. The factor groupings did not reflect the lithol-
ogy of the rocks from which the brines were taken, indicating that such a
relationship does not exist or is difficult to detect. The brine analyses used in
the study were primarily macro analyses and did not include pH, minor, or
trace constituents. A study of this type would benefit significantly if the
following conditions were met: (1) use only the best available sampling
methods; (2) use field analysis techniques; (3)use positive lithology identifi-
cation; and (4) use only the best available laboratory methods of analysis.
Kramer (1969) did not have these controls because he used only the
published data of various laboratories.
Carpenter and Miller (1969) used statistical and thermodynamic methods
in an effort t o determine the origin of the dissolved chemical constituents in
saline subsurface waters in north-central and northwestern Missouri. Statisti-
cal analysis of scatter diagrams indicated that the concentrations of lithium,
sodium, potassium, and bromide and the ion activity ratios of K+/H+,
Ca+’/Mg+*, and Sr+’ /Ba+’ in the waters are influenced by reactions with
constituents in the aquifer rocks. They concluded that the ion ratios are of
little value in determining the origin of the waters because the concen-
trations of the dissolved constituents in the waters had reacted with minerals
in the aquifer rocks. This study was excellent because it did show that the
concentrations of constituents in the water are controlled to some extent by
reactions with the aquifer rocks. Additional work of this kind is needed in
the study of deep brines.
A study of brines from the Sylvania formation in the Michigan Basin
indicated that evaporation and dolomitization were two dominant controls
for their dissolved concentrations of calcium, magnesium, sodium, stron-
tium, and bromide (Egleson and Querio, 1969). Mechanisms responsible for
concentrations of elements such as potassium, lithium, rubidium, ammonia,
boron, and iodide were believed t o be reactions with sedimentary rocks,
leaching of organic constituents, and bioconcentration.
Relation to depth and salinity
A study of the chemical composition of some selected Kansas brines

indicated that in general the concentrations of calcium, sodium, and chloride
increase with increasing salinity, while the sulfate concentrations decrease
(Dingman and Angino, 1969). However, the Ca/C1 ratio, concentrations of
calcium and salinity, did not generally increase with geologic age or with
depth of the aquifer.
Dickey (1969) surveyed the analyses of oilfield waters from many areas of
the United States and concluded that in general the Ca/Mg ratio increases
with increasing salinity while the ratio Na/(Ca + Mg) decreases irregularly
with increasing. salinity and depth. This observation is compatible with the
findings of several investigators. The dominant anion in subsurface waters
usually changes with depth; in near-surface waters, it is sulfate; at depths
exceeding 520 m it is bicarbonate; and in deep brines, it is chloride
(Chebotarev, 1964). The Ca/Na ratio usually increases with depth and age of
the associated rocks, while the Mg/Na ratio decreases.
Iodide
Collins (1969a) studied the chemistry of some oilfield brines from the
Anadarko Basin which contain high concentrations of iodide. The concen-
trations of bromide in many of these brines are lower than the iodide which
is unusual. Localized sedimantary rock deposits enriched in organic iodine
are the source of the high iodide concentrations in these brines (Collins et
al., 1971).
Hot brines
Hot brines containing minor and trace amounts of several metallic

elements in addition to macro concentrations of some alkalies, alkaline
earths, and chloride are found in drill holes in southern California, in the
Caspian Sea, and in deeps in the Red Sea. These brines were formed from
evaporites dissolved by meteoric water, and the metallic elements were
leached from country rocks by the hot brines (Tooms, 1970). Laboratory
reactions of 2M and 4M sodium chloride with andesite and shale at
300"-500°C have produced solutions containing metallic elements in con-
centrations similar to the hot brines (Ellis, 1968).
Comparison of oilfield brines with evaporated sea water
Bromide does not form its own minerals when sea water evaporates. It
forms an isomorphous admixture with chloride in the precipitates
(Valyashko, 1956; Braitsch and Herrmann, 1963). As sea water evaporates,
the carbonates precipitate first, followed by the sulfates. Little or no
bromide precipitates, or if it does, it is occluded with these.
Halite (NaCl) begins t o precipitate when the chloride concentration is
about 275,000 mg/l (Table 7.111) compared with that of normal sea water,
19,000 mg/l. Some bromide is entrained with chloride in the precipitate.
300
2oo t Normal evaporite curve/

.-
c
C
Fig. 7.4. Use of the bromide ion to differentiate some Tertiary (T), Cretaceous (C), and
Jurassic (J) age brines.
TABLE 7 .XV
Bioconcentrated bromide and iodide in seaweeds and corals
Bromide (ppm) Iodide (ppm)
Seaweed
Laminaria digitata (dry matter) 1,380 5 10-8,000
Laminaria saccharina (dry matter) 340 2,000
Desmaresta (ash) 6,800 5,200
Corals*
Gorgonia uerrucosa 16,200 69,200
Gorgonellidae 19,800 22,100
Isididae 7,400 20,300
* After Vinogradov (1953).
However, with each crystallization, more bromide is left in solution than is

entrained in the precipitate.
Sylvite (KC1) begins to precipitate when the chloride concentration is
about 360,000 mg/l (Table 7.111), followed by carnallite (MgC12*KCI *6H20 )
and bischoffite (MgC12 *6H, 0).During evaporation the concentration rate of
bromide in solution increases. The change in the slope of the curve in Fig.
7.4 illustrates this approximately.
Other concentration mechanisms operate t o account for the high bromide
concentrations (6,000 mg/l and up) found in some brines. One of the
mechanisms is related to bioconcentrators such as seaweeds and corals. The
seaweeds and corals concentrate the bromide, they die, and are buried with
the sediments. Later the bromide is leached by the surrounding waters. Table
7.XV illustrates some of the concentrations of bromide and iodide that
Vinogradov (1953) found in various seaweeds and corals.
Laboratory experiments have demonstrated that bromide is accommo-
dated in the halite crystal lattice and replaces chloride in solid solution
(Borchert and Muir, L964). The weight percentage of bromide in solid solu-
tion in the halite lattice is related to its weight percentage in the parent brine
as :
wt.% Br (in halite)
C = wt.% Br (in solution)
where C = the pa@ition coefficient.
In most natural environments C = 0.14 (Braitsch and Herrmann, 1964).
In a marine salt sequence the wt.% Br/NaCl rises from about 0.007 wt.% at
the bottom to 0.02 wt.% at the beginning of potassium precipitation. How-
ever, the bromide concentration with a given natural halite sequence may
vary considerably, even though theoretically it should increase continuously
from the bottom t o the top of the depositional strata. These variations can
be attributed t o inflow of fresh sea water during the deposition or subsequent
leaching after deposition. Rittenhouse (1967) developed a method to classify
oilfield waters based upon the bromide concentrations.
Fig. 7.4 is a log-log plot of chloride versus bromide concentrations for
some Louisiana oilfield waters. The T, C, and J on the figure refer t o Terti-
ary, Cretaceous, and Jurassic, indicating the ages of the rocks from which the
waters were taken. The normal evaporite curve was plotted by using data
from Table 7.111. The data in the figure indicate that most of the Tertiary
waters are deficient in bromide when compared to an evaporite water,
whereas the Cretaceous and Jurassic waters are enriched in bromide (Collins,
1967).
The Tertiary waters contain dissolved halite, which accounts for their low
bromide concentration, while the waters that are enriched in bromide con-
tain bitterns or have leached bromide from sediments that were enriched in
bioconcentrated bromide.
The bromide content of oilfield brines can be used t o distinguish between
brines that originated because of evaporation of sea water and those formed
by the dissolution of evaporite minerals. This can be done by using Fig. 7.4.
If the bromide concentration falls to the right of the normal evaporite curve,
the brine contains evaporated sea water, while if it falls t o the left of the
curve, it contains dissolved evaporite minerals.
Fig.7.5 illustrates how closely the concentration of sodium of some
Louisiana oilfield waters taken from Tertiary, Cretaceous, and Jurassic age
rocks follow the sodium concentration of a brine associated with normal
evaporation (Collins, 1970).
300 0 ,
200
- 100
W Normal evaporite curve
50
0
J
-
20
lo
5,000 10,000 20,000 ' &bob I llob!OOO
' 1
500 300
SODIUM, m g / l
Fig. 7.5. Relationships of the chloride concentrations to sodium concentrations in a
normal evaporite brine to oilfield brines taken from formations of Tertiary (T), Cre-
taceous (C), and Jurassic (J) age in the United States,
23 0 ORIGIN OF OILFIELD WATERS
Ion exchange
Ion exchange reactions on clay minerals are reversible and they follow the
law of mass action. The number of exchange sites governs the reaction, and
other important factors include temperature, pressure, solution concen-
trations, and bonding strength of exchangeable ions. Ion exchange between
clay minerals and a brine will stop when equilibrium is attained.
As the waters move in their subsurface environment, their dissolved ions
have a tendency to exchange with those in the rocks. There are two extreme
types of adsorption in addition to intermediate types of adsorption. The
extreme types are: (1) a physical adsorption or Van der Waals adsorption
with weak bonding between the adsorbent and the constituent adsorbed; and
(2) a chemical adsorption with strong valence bonds.
Cations can be fixed at the surface and in the interior of minerals. These
fixed cations can exchange with cations in the water. Under the right
physical conditions of the adsorbent, similar exchange can occur with the
anions. Some of the constituents in formations that are capable of exchange
and adsorption are argillaceous minerals, zeolites, ferric hydroxide, and cer-
tain organic compounds.
Particle size influences the exchange rates and capacities if the solids are
clays such as illite and kaolinite. The rate increases with decreasing particle
size. However, if a larger mineral has a lattice, the exchange can easily occur
on the plates. The concentration of exchangeable ions in the adsorbent and
in the water is important. More exchange usually occurs when the solution is
highly concentrated.
According t o Grim (1952), the replacing power of some ions in clays is:
(1) In NH, ,kaolinite:
Cs > Rb > K > Ba > Sr > Ca > Mg> H > Na > Li
(2) In NH, ,montmorillonite:
Cs > Rb > K > H > Sr > Ba > Mg> C a > Na > Li
These two clays often are present in sedimentary rocks and the replacing
order indicates that lithium and sodium are more likely to be left in solution,
while cesium and rubidium are more likely t o be removed from solution.
Fig. 7.6 is a plot of the chloride content versus the lithium content of
some oilfield waters taken from the Smackover formation. The .lithium
enrichment results at least in part from exchange reactions on clays. Lithium
has a small radius, a low atomic number, a larger hydrated radius than
sodium, and a larger polarization than sodium. Because of these, its replacing
power in the lattices of clay minerals is low (Kelley, 1948). Other ions such
as barium, strontium, calcium, magnesium, cesium, rubidium, potassium, and
sodium will preferentially replace lithium in clay minerals, thus releasing
lithium t o solutions. Furthermore, the solubility products of most lithium
1,000 C
000 -
- A Louisiana
600 Mississippi
Alabama
400 -
A
o Arkansas
Texas
0 0
0
V
0
40 -
0
20- d
10 I I 1 I I , I I I I I I I I O I , I I 1 I I
300
200
- 100
\
m
w
2 50
0
4
3 30
20 C- J
T
T
,I 1 I II,] I 1 1 1 I I I L
19 3 100 200 500 1,000 2,000 5,000 10 100
POTASSIUM, m g / I
Fig. 7.7. Relationships of the concentrations of chloride and potassium in a normal
evaporite-formed brine to oilfield brines taken from formations of Tertiary (T), Creta-
ceous (C), and Jurassic (J)age in the United States.
compounds are higher than those of other alkalies and alkaline earths. There-
fore lithium tends to stay in solution.
Fig. 7.7 compares the potassium concentration of some Louisiana oilfield
waters with those of waters subjected to evaporation. All of these waters are
depleted in potassium with respect to a brine subjected t o evaporation,
indicating that potassium was lost to the associated sediments during
diagenesis. It has been shown that a tendency exists for potassium t o be

adsorbed and fixed by clay minerals, mica, and potassium feldspar in normal
low-temperature processes (White, 1965; Khitarov and Pugin, 1966; Grim,
1952).
The data in Tables 7..III-XIV indicate that the concentration of calcium
in oilfield waters generally is enriched relative t o sea water. Cation exchange
reactions with clays accounts for some of this enrichment:
2Na+ (solution) + Ca (clay) + Ca+* (solution) + 2Na (clay)
Collins (1972) found that the ratio Na/(CL + Mg) tends t o decrease as the
dissolved solids concentration increases in some oilfield waters from the East
Texas Basin. This depletion of sodium with respect to calcium plus magne-
sium was attributed to diagenesis of the waters and it correlated with an
index of base exchange (Schoeller, 1955), indicating that the alkali metals in
the waters exchanged with alkaline earth metals on the argillaceous minerals
t o decrease the dissolved alkali metals and increase the dissolved alkaline
earth metals.
Fig. 7.8 is a plot of the calcium concentration in some oilfield waters
taken from the Smackover formation. All of these waters are enriched in
calcium relative to the evaporated sea water.
Krejci-Graf (1963) found that solutions predominantly concentrated in
chloride can force an exchange of calcium and bromide from clay minerals
for sodium and chloride from the solution. If this type of reaction occurred
Fig. 7.8. Relationships of the concentrations of chloride t o calcium in an evaporite-

formed brine to oilfield brines taken from the Smackover formation in five states of the
United states.
1,000
000 -
600 - 6 Louisiana
- Mississippi
400 - A Alabama
- 0 Arkansas
0 Texas
- 200 -
0
\
w
0 100,
5 80
I 60
V
40
20
101I<I I I I I 1 1 1 1 1 1 ' I ' I ' l ' ' ' 1 ' I '

100 200 400 1,000 2,000 4,000 10,000 40,000
BROMIDE, mg/l
Fig. 7.9. Relationships o f the concentrations of chloride t o bromide in an evaporite-

formed brine t o oilfield brines taken from the Smackover formation in five states of the
United States.
t o the Smackover brines, it explains their enrichment of calcium and

bromide.
Kozin (1960) wrote about a "reverse" exchange of anions when the
cations exchange on clays:
C1- (solution) + Br (clay) + Br- (solution) + C1 (clay)
Such a reaction also helps to account for the bromide enrichment found in
most oilfield waters taken from the Smackover formation (Fig. 7.9).
A similar reaction for iodide:
C1- (solution) + I (clay) -,I- (solution) + C1 (clay)
would help explain the tremendous enrichment of iodide in oilfield brines

(Collins, 1969a) with respect t o sea water as demonstrated in Tables
7.1V-XIII.
Fig. 7.10 shows that boron usually is enriched relative to the normal evap-
orite curve in Smackover oilfield brines. Boron, like lithium, has a small
radius, a low atomic number, and large polarization. Therefore, its replacing
power in the lattices of clay minerals is low. Also, boron does not have a
tendency to enter silicate lattices of the common rock-forming minerals.
Because of these factors, it usually remains in solution until late-stage crys-
tallization.
1,000 -
000 -
600 - A Louisiana
- Mississippi
400- A Alabama
0 Arkansas
0 Texas
-
-w 200
\
:
LL
100:
00-
2
0
60-
40 -
20 -
BORON, mg/l
Fig. 7.10. Relationships of the concentrations of chloride to boron in an evaporite-

United States.
Mineral formation
A study of some brines taken from Devonian age reservoir rocks indicated
that dolomitization probably is the most important mechanism in deter-
mining the calcium, strontium, and magnesium content of these brines
(Egleson and Querio, 1969). It also was concluded that the relative amounts
of ammonium, iodide, and lithium in these brines were too high to be
derived directly from sea water, and the ammonium and iodide probably
were enriched in the brines as a result of bioconcentration and subsequent
leaching of organic debris.
Dolostone deposits owe their origin t o hypersaline brines (Friedman and
Sanders, 1967). Some dolomite, including diagenetic and epigenetic forms,
originates from subsurface brines. In the geologic columns in several oil-
TABLE 7.XVI
Approximate sea-water composition before and after gypsum precipitation (mg/l)
Ion Before precipitation After precipitation
Calcium 390 0
Magnesium 1,300 1,300
Bromide 65 65
Sulfate 2,580 2,580
TABLE 7.XVII
Approximate sea-water composition after dolomitization or bacterial reduction (mg/l)
Ion After dolomitization After bacterial reduction
Calcium 0 0
Magnesium 883 1,300
Bromide 65 65
Sulfate 0 0
productive basins, mixtures of dolomite and anhydrite occur, which in-

dicates that sulfate may have been removed from the associated waters by
dolomitization as well as by bacterial reduction.
Table 7.XVI illustrates the approximate amounts of calcium, magnesium,
bromide, and sulfate that could exist in a water’before and after precipita-
tion of gypsum.
Assuming that the residual sulfate (1,644 mg/l) was removed by the
dolomitization reaction:
MgC12 + 2CaC0, += CaC12 + CaMg(C0,- )*-

CaC12- + MgS04- += CaS04 + MgCi2
MgSO, + 2CaC03 + CaS04 + CaMg(C0, ) 2
then the Mg/Br ratio would be about 883/65 = 13.6, as illustrated by the
data in Table 7.XVII. However, if the residual sulfate was removed by bacte-
rial reduction :
C,H, + Na2 SO4 + Na2C03 + H 2 S + C 0 2 + H 2 0
the Mg/Br ratio would be about 1300/65 = 20.

Magnesium will react with CaC03 (calcite) t o form dolomite, thus in-
creasing the concentration of calcium in the brine. However, the total cal-
cium plus magnesium in the brine should remain constant. This can be
calculated as (24.31/40.08) x mg/l calcium + mg/l magnesium = total equiv-
alent magnesium or Mg’. The ratio Mg’/Mg will vary, depending upon the
availability of calcite, and the ratio should be indicative of the degree of
dolomitization.
For example, brines that are in equilibrium with sandstones should have a
relatively low Mg’/Mg ratio, those in equilibrium with dolomite should have
higher ratios, and those in equilibrium with limestone should have the
highest ratios. The average ratio for some Smackover brines is 7 (Table
7.XVIII), which indicates that the brines were in equilibrium with limestone
and dolomite. Brines from some Tertiary age rocks which were primarily
TABLE 7.XVIII
Concentration ratios and excess factor ratios for some constituents in Smackover brines
Constituent Average composition (mg/l) Concentration Excess Number of

ratio*' factor*' Smackover
sea water Smackover samples
brines
Lithium 0.2 17 4 870 18.1 71

Sodium 10,600 66,975 6 0.1 283
Potassium 380 2,841 8 0.2 82
Calcium 400 34,534 86 1.8 284
Magnesium 1,300 3,465 3 0.1 280
Strontium 8 1,924 241 5 85
Barium 0.03 23 767 16 73
Boron 4.8 134 28 0.6 71
Copper 0.003 1.1 359 7.5 64
Iron 0.01 41 4,049 84.2 90
Manganese 0.002 30 14,957 31 1 69
Chloride 19,000 171,686 9 0.2 284
Bromide 65 3,126 48 1 74
Iodide 0.05 25 501 10.4 73
Sulfate 2,690 446 0.2 0.003 27 1
Mg'* 1,543 24,362 16 0.3 284
*' Amount in brine/amount in sea water.

** Concentration ratio of a given constituent/concentrationof bromide.
*3 Mg' = (24.31/40.08) x mg/l Ca + mg/l Mg.
sandstone (Table 7.XIX) had an average ratio of 2.8, while brines from some
Cretaceous age rocks had an average ratio of 6.0 (Table 7.XX).
Bromide does not form its own minerals when sea water evaporates. Some
of it is lost from solution because it forms an isomorphous admixture with
chloride with the halite precipitate. However, more bromide is left in solu-
tion than is entrained in the precipitate. Therefore, relative t o chloride, the
bromide concentration in the brine increases exponentially. Bemuse of this,
the bromide concentration in the brine is a good indicator of the degree of
sea water concentration, assuming that appreciable quantities of biogenic
bromide have not been introduced.
Table 7.XVIII presents data that were obtained by comparing the average
composition of some Smackover brines with that of sea water. The concen-
tration ratio was calculated by taking the mean average for a given con-
stituent in the Smackover brines and dividing it by the amount of the con-
stituent found in normal sea water. The excess factor was determined by
dividing the concentration ratio of a constituent by the concentration ratio
of bromide. The calculation for Mg' or total equivalent magnesium was
previously explained, and the number of Smackover samples indicates how
TABLE 7.XIX
Concentration ratios of some constituents in some brines taken from Tertiary age rocks
Constituent Average composition (mg/l) Concentration Excess

ratio*' factor * '
sea water Tertiary brines
Lithium 0.2 3 15 12.5

Sodium 10,600 37,539 3.5 2.9
Potassium 380 226 0.6 0.5
Calcium 400 2,077 5.2 4.3
Magnesium 1,300 686 0.5 0.4
Strontium 8 148 18.6 15.5
Barium 0.03 73 2,439 2,033
Boron 4.8 20 4.1 3.4
Chloride 19,000 63,992 3.4 2.8
Bromide 65 79 1.2 1
Iodide 0.05 21 426 355
Sulfate 2,690 104 0.03 0.03
Mg' 1,543 1,947 1.3 1.1
*' Amount in brine/amount in sea water.

*' Concentration ratio of a given constituent/concentrationof brymide.
*3 Magnesium e,quivalent of calcium plus magnesium in brine: Mg = (24.31/40.08) x mg/l
Ca + mg/l Mg .
TABLE 7.XX
Concentration ratios of some constituents in some brines taken from Cretaceous age rocks
Constituent Average concentration (mg/l) Concentration Excess

ratio*' factor*'
sea water Cretaceous brines
Lithium 0.2 4 20 4.5

Sodium 10,600 28,462 2.7 0.6
Potassium 380 193 0.5 0.1
Calcium 400 4,999 12.5 2.8
Magnesium 1,300 606 0.5 0.1
Strontium 8 346 43.3 9.8
Barium 0.03 48.3 1,608 365
Boron 4.8 27.5 5.7 1.3
Chloride 19,000 54,910 2.9 0.7
Bromide 65 287 4.4 1
Iodide 0.05 37 737 168
Sulfate 2,690 206 0.1 0.02
Mg'*3 1,543 3,643 2.4 0.5
*' Amount in brinelamount in sea water.

*' Concentration ratio of a given constituentlconcentrationof brymide.
*' Magnesium equivalent of calcium plus magnesium in brine: Mg = (24.31/40.08) x mg/l
Ca + mg/l Mg.
many samples were used in the calculation. For example, 71 Smackover

brines were analyzed for lithium, while 283 were analyzed for sodium.
The concentration ratios (Table 7.XVIII) indicate that all of the deter-
mined constituents in the Smackover brines were enriched with respect to
sea water except sulfate. However, the excess factor ratios indicate that
sodium, potassium, magnesium, chloride, sulfate, and total equivalent mag-
nesium were depleted in the Smackover brines, while lithium, calcium, stron-
tium, barium, copper, iron, manganese, and iodide were enriched. Further,
these ratios indicate that the Smackover brines have been altered consider-
ably if it is assumed that they originally were sea water.
The concentration ratio 48 for bromide (Table 7.XVIII) is one of the
highest that this author has seen. For example, bromide concentration ratios
of 1.2 (Table 7.XIX), 4.4(Table 7.XX), and 8.8 and 7.2 were found for brines
from Tertiary, Cretaceous, Pennsylvanian, and Mississippian age rocks
(Collins, 1967, 1969a, 1970). The concentration ratios and excess factors in
Tables 7.XIX and XX indicate several constituents are enriched and several
are depleted in these brines also.
Almost one-third of the magnesium in sea water and subsequent bitterns
can be removed during the dolomitization reaction. The formation of
chlorite from montmorillonite requires about 9.2 moles of MgO per mole of
chlorite (Eckhardt, 1958):
1.7 A1203 0.9 MgO 8 Si02 - 2 H 2 0 + 9.2 MgO + 6 H 2 0 +=
10.1 MgO 1.7 A1203 6.4 Si02 * 8 H 2 0 + 1.6 Si02
1,000 -
800 -
600 - a Louisiana
Mississippi
400 - A Alabama
0 Arkansas
-cn
\ 200 - o a */ 0 Texas
W*
g 100:
5 80- ‘Normal evaporite
I
0
60 - curve
40 -
0
20 -
10 I I I I I I I I I 1 1 6 1 I I I I I I 1 1 8 1 I I I
00
Fig. 7.11. Relationships of the concentrations of chloride to magnesium in an evaporite-

United States.
1,000
000
Louisiana
600 Mississippi
400 Alabama
Arkansas
Texas
- 2 00
\
0
0 0 0 4, 0
: 00 100 0 0
2
0
60 0
0
40
0
20
00
Fig. 7.12. Relationships of chloride to strontium in an evaporite-formed brine to oilfield

brines taken from the Smackover formation in five states of the United States.
Such a reaction could remove large amounts of magnesium from waters.

Hiltabrand (1970) has shown that contemporary argillaceous sediments can
remove 100 mg/l of magnesium from sea waters.
Fig. 7.11 is a plot of the chloride concentrations versus the magnesium
concentrations in some oilfield waters taken from the Smackover formation.
The figure indicates that the Smackover waters are depleted in the concen-
tration of magnesium with respect t o an evaporite-formed brine. Tables
7.111-XI11 indicate that in general oilfield waters taken from rocks of other
formations also are depleted in magnesium. The data also show that general-
ly as the dissolved magnesium decreases the dissolved calcium increases. This
is related t o the formation of minerals such as chlorite or dolomite and t o
exchange reactions with argillaceous minerals. It is not a result of solubility
because most magnesium compounds are more soluble than calcium com-
pounds.
Fig. 7.12 is a plot of the concentration of chloride versus the concen-
tration of strontium found in some oilfield brines taken from the Smackover
formation. This figure indicates that the strontium concentration is enriched
in the Smackover brines relative to sea water. Reactions that account for
some of this enrichment are:
2SrC03 + MgCl, +. SrMg(C03), + SrCl,

SrMg(C0, ), + MgCl, +. 2MgC0, + SrC1,
The data in Fig. 7.7 and Tables 7.111-XIV indicate that the concentration
of potassium in oilfield waters generally is depleted relative to sea water.
Montmorillonite-type minerals systematically change t o illite with depth
in Gulf Coast shales (Burst, 1969). As a result of this transformation, the

montmorillonite-type minerals lose interlayer water. Laboratory experiments
at elevated temperatures and pressures indicate that montmorillonite loses its
interlayer water and transforms into illite in the presence of potassium-
enriched water (Khitarov and Pugin, 1966). The structural variations of the
expandable minerals in clays also are apparently influenced by the potassium
content of the associated waters. This indicates that oilfield waters tend to
become depleted in potassium content where this reaction occurs.
Reactions between brines and minerals to form silicates that account for
the depletion of dissolved alkali metals are:
*
3A1, Si2O5(OH), + 2K+ 2KA13Si, O l 0(OH), + 3H20 + 2H+
KA13Si030,0(OH)2+ 6Si02 + 2K+ + 3KA1Si308 + 2H+
A12S i 2 0 5(OH), + 4sio2 + 2Na+ =+2NaA1Si308 + H2 0 + 2H+
These reactions account not only for the depletion of potassium or

sodium, but also for a decrease in pH because of the release of hydrogen
ions. The decrease in pH enables the water t o dissolve metallic metals, to
convert bicarbonate to carbon dioxide, or to convert bisulfide t o sulfide. The
Smackover brines often contain relatively high concentrations of sulfide.
Several investigators have attempted t o determine what mechanism is
responsible for the increased concentration of calcium and depletion of mag-
nesium relative to sea water in many subsurface brines. Chave (1960) and
Von Engelhardt (1960) compared ocean water with subsurface brines con-
taining high concentrations of calcium, and demonstrated that dolomitiza-
tion cannot account for all of the calcium in the brine solutions. Von
Engelhardt (1960) noted that even the formation of chlorite utilizing magne-
sium with exchange of sodium and calcium does not account for all of the
soluble calcium; however, exchange reactions with other clays were not con-
sidered. Kramer (1963) assumed that calcium was more abundant in ancient
oceans, but White (1965) found this relation t o be untenable and suggested
that shale-membrane filtration accounts for increased concentrations of
calcium in some brines. Additional data are needed before more definite
conclusions can be made. The amounts and ratios of calcium and magnesium
vary from one formation water to another as well as within one formation at
different geographic areas. Mineral formation, exchange reactions, leaching,
and shale-membrane filtration all can alter the composition of the brine.
However, in a specific area, one type of reaction may predominate.
Mem brane-concen tra ted brines
Essentially the postulate that clays and shales act as membranes t o filter
dissolved solids from waters results from the fact that synthetic membranes
are used to desalinate waters by reverse osmosis. Conceivably, compacted
clays and shales may perform as imperfect semipermeable membranes. Solu-
tions of salts of different concentrations separated by a semipermeable

membrane will cause water from the lower salt-concentration side t o move
through the membrahe t o the higher concentration side, producing a greater
pressure on the high-concentration side. The pressure differential is the
osmotic pressure of the system and can account for abnormal pressures
found in some reservoirs.
Reverse osmosis occurs when hydraulic pressure in excess of the osmotic
pressure is applied t o the high-concentration side, which forces water
through the membrane to the low-concentration side. The system is not
100% effective and some dissolved solids move through the membrane
(Kimura and Souriragan, 1967).
Such a system requires rather high pressure differentials in nature t o
produce the highly concentrated brines found in some formations. The
osmotic pressure could produce pressure differentials in formations, but the
pressure comes t o equilibrium as the two solutions equilibrate. The reverse
osmosis system works only as long as the excess hydraulic pressure is
applied. In the absence of the excess hydraulic pressure, the system comes t o
equilibrium.
Larson (1967) reported some desalination results for water with reverse
osmosis using cellulose-acetate membranes. With a brackish water containing
about 4,300 mg/l of dissolved solids, input pressure of 42 kg/cm2 and
temperature of 15.g0C, the ion rejection rates were as high as 99.9%. The
rejection order based on the percent rejected was:
Ca+’ + Mg+’ > HC03-2 + SO4-’ > C1-> Na+ > NO3-
Assuming that this mechanism operates in a shale filtration system, the order
of ion concentration on the high brine concentration side would be the
same. The ion concentrations on the fresher water side would be the reverse
or :
NO3- > Na+ > Cl- > SO4-’ + HC03-* > Ca+’ + Mg+’
Other investigators have obtained similar results. For example, Loeb and
Manjikian (1965) found a rejection order of SO4-’ > Mg+’ > Ca+’ > Na+
> HC03- > C1- > NO3-. Michaels et al. (1965) found a rejection order of
Ca+2 > Li+ > Na+ > K+ for the pressure independent portion of salt
transport in cellulose acetate reverse osmosis desalination membranes. This
correlates with the size of the hydrated ion radii because calcium is the
largest and potassium the smallest. Further, this indicates that the pore size
of the membrane is a controlling factor.
The data ‘of Larson (1967) showed that sulfate and carbonate scale
formed on the high-pressure side of the membrane and if not removed would
cause flow to decrease or stop. The pH on the output or fresh-water side of
the membrane decreased.
Russell (1933)considered several processes which could produce subsur-

face brines more concentrated than sea water. He concluded that evapora-
tion of the water by natural gas generally is not important, water evapora-
tion in coarse-grained rocks generally is not important, gravitational settling
of dissolved solids is not greatly important, rocks containing considerable
amounts of feldspars and other unstable minerals take up large quantities of
hydration water, clays adsorb bases and later expel them into solution
causing concentration, and osmosis may occur through semipermeable mem-
branes.
DeSitter (1947)noted that oilfield waters are altered as a result of two
prominent diagenetic phases. During the first phase magnesium, calcium,
sulfate, and carbonate precipitate from the original sea water. During the
second phase the concentration of magnesium and calcium ions increases
along with the concentration of other dissolved solids. He reasoned that the
second phase occurred because of filtration through semipermeable shales.
The filtration results because of sediment compaction until a semiper-
meable membrane develops which allows water molecules t o pass through
but retards salt ions. Thus, the more concentrated brines are found where
sediment compaction and water flow distance were the greatest. This usually
occurs in the deepest portion of a basin.
McKelvey et al. (1957) forced aqueous saline solutions through ion-
exchange resins and found that the effluent solutions contained less dis-
solved salts than the influent solutions. Effluents from cation-exchange
resins were found to contain Na/K ratios similar t o those in the influent;
however, the Mg/Ca ratios were at first higher than in the influent but with
additional squeezing the ratio decreased t o much lower values. They
postulated that similar reactions occur during the compaction of sediments
to change the concentrations of constituents dissolved in waters.
Pressures of 7 kg/cm2 to 105 kg/cm2 were applied t o force sodium
chloride solutions through cation-exchange membranes. The results indicated
that the membranes desalted the saline solutions, producing a filtrate con-
taining less salt than the influent. This salt filtering effect was attributed t o
the electrical properties of the membrane.
Milne et al. (1964)determined the filtering efficiencies of sodium chloride
solutions by bentonite membranes. The filtration efficiencies were 94% at
140 kg/cm2 and 88% at 703 kg/cm2 with 0.5N sodium chloride. Increased
salinity caused less efficient filtration because filtration efficiencies of 94%
for 0.W sodium chloride and 66% for 4N sodium chloride at a pressure of
352 kg/cm2 were obtained. A similar mechanism could operate in the sub-
surface to create concentrated brines.
Young and Low (1965)performed an experiment using natural rock and
demonstrated that osmotic flow of water through shale and siltstone occurs.
The osmotic pressures produced were less than theoretical and they were
attributed to microcracks in the natural rock which caused them t o be less
effective than a perfect membrane.
Bredehoeft et al. (1963) developed a mathematical model t o predict the

distribution of ions within a formation. They assumed that a hydrostatic
head differential opetates between the margin and center of a geologic basin,
producing a water movement upward through confining low permeability
beds. If these low permeability beds contain clay membranes to restrict the
passage of ions, the waters on the upflow, or more permeable, side become
more concentrated in dissolved solids. They theorized that this process
produced the concentrated brines found in the Illinois Basin, and that their
model added weight t o the membrane theory of brine concentration. A
major drawback to the model is the tremendous pressures that are necessary
to produce a movement of water upward through confining low permeability
beds.
Graf et al. (1965) found that isotopic fractionation occurred when waters
passed through shale micropores in the Illinois, Michigan, Alberta, and Gulf
Coast Basins. Their study did not yield sufficient evidence t o estimate the
total fraction of water movement in the basins subsequent t o sediment com-
paction. The 6 " 0 concentrations in brines did not indicate a direct correla-
tion with ancient oceans.
A study of the 6D and 6l80 in formation waters indicated that the water
was predominantly meteoric, little exchange or fractionation had occurred
to alter the deuterium, but extensive exchange between the water and rock
had altered the oxygen (Clayton et al., 1966). They postulated that forma-
tion waters in the Gulf Coast Basin lost their original connate water because
of sediment compaction and flushing, and that the present water is meteoric
water which came in through outcrops.
This study was good; however, basic studies concerning the fractionation
and exchange of isotopes between water, hydrocarbons, and rocks need to
be made. Results of such studies should enable more positive interpretations.
A simplistic model was derived t o determine the amounts of fresh water
and sea water necessary t o create the brine compositions now present in the
Illinois and Michigan Basins (Graf et al., 1966). The model assumes: (1)
perfect efficiency of shale ultrafilters; (2) complete bacterial reduction of
sulfate with replacement in solution of equivalent bicarbonate; (3) complete
removal of bicarbonate and equivalent sodium by shale ultrafiltration; and
(4) magnesium reaction with calcium carbonate to form dolomite. The
dolomitization reaction furnished more soluble calcium than is possible for
the Illinois Basin, so another calculation was made assuming complete loss of
magnesium t o clay minerals with no return of calcium.
The calculations indicated that less fresh water passed through the rocks
of the Illinois Basin than those of the Michigan Basin. These data conflicted
somewhat with Clayton et al. (1966) in that they argued that the water
molecules now in the Illinois Basin originated as fresh water, while the data
of Graf et al. (1966) indicated that too few volumes of fresh water passed
through the Illinois Basin t o alter the brine significantly.
A study of the hydrodynamics of the Illinois Basin indicated that in
recent times, before pumpage, the differences in vertical head in the deep
aquifers were insufficient t o cause upward flow through shale, resulting in
ultrafiltration (Bond, 1972). In fact the head differentials were barely suf-
ficient to enable upward flow through an open conduit.
Berry (1969) outlined the relative factors that influence membrane filtra-
tion in geologic environments. The membrane properties of shales are caused
by the electrfcal properties of their clays and organic materials. Clays
predominantly are cation exchangers with singly charged SiO- and
AlOSi-% sites and minor anion exchanges with replaceable OH- ions.
Divalent cations are adsorbed in preference to monovalent cations and
sodium is hyperfiltratcd with respect t o lithium and strontium with respect
to calcium, because of preferential adsorption of ions with ionic potentials
most similar t o the ionic potential of the exchange site. The selectivity of
hyperfiltration for the halogens is C1 > Br > I > F because of their substitu-
tion for O H in the clays. Thus, in waters concentrated by this process the
Ca/Na, Na/Li, Sr/Ca, Cl/Br, Br/I, and I/F ratios should increase. These ratio
increases have been found in some brine systems, but by no means in all
systems.
Billings et al. (1969) found five types of formation waters in the Western
Canada Sedimentary Basin and postulated the origin of two of the types.
One type of water was formed by selective membrane filtration which pro-
duced waters containing high concentrations of dissolved solids. A second
type was a mixture of membrane-concentrated formation water and bitterns
formed after the precipitation of halite but before the precipitation of
sylvite. They theorized that the alkalies were filtered selectively by clay-
shale membranes, producing a concentrated brine, and that the relative con-
centration pattern is Rb > K > Na > Li. This pattern is the reverse of what
occurs by ion exchange but is similar to the surface mobilities of cations
along clay surfaces.
A detailed study of the Western Canada Sedimentary Basin, including a
determination of the rock volume and pore volume (Hitchon, 1968), the
effect of topography upon the fluid flow (Hitchon, 1969a), and the effect of
geology upon the fluid flow (Hitchon, 1969b), strongly suggested that
thermal, electro-osmotic, and chemico-osmotic forces are operating within
the basin to affect the fluid energy gradients. Pressure differentials of about
98 kg/cm*along with salinity differences of 200,000 mg/l between forma-
tions in close proximity were found which suggest that chemico-osmotic
forces are occurring.
Hitchon and Friedman (1969) used chemical analyses and stable-isotope
analyses for hydrogen and oxygen for surface waters, shallow ground waters,
and deep ground waters in a study of the origin of formation waters in the
Western Canada Sedimentary Basin. They postulated that surface waters have
mixed with diagenetically altered sea water to form the formation waters.
Using mass balance data for the deuterium and dissolved solid contents of
the formation waters, they calculated not only how much fresh water is
CONCLUSIONS 245
present in the modified sea water but also observed how it redistributed the
dissolved solids t o prQduce salinity variations.
They concluded that formation waters result from mixing of surface
waters with modified marine or nonmarine water in the subsurface rocks,
that exchange of oxygen isotopes between the water and rock caused differ-
ent water types in different basins, and that formation waters that have
passed through shale ultrafilters are more depleted in deuterium.
A study of the Surat Basin showed that most of its hydrocarbon accumu-
lations are associated with quasi-stagnant waters. The salinities of these
quasi-stagnant waters were higher than were the salinities of the waters in the
more dynamic recharge areas. The investigators postulated that these high
salinity waters were formed by membrane filtration because of cross-
formational flow and also that the hydrocarbon accumulations in these
quasi-stagnant areas resulted from release of hydrocarbons mobilized by a
moving water. The hydrocarbons were released because of the higher
salinities of the waters in the quasi-stagnant areas (Hitchon and Hays, 1971).
A study of waters in sedimentary rocks of Neogene age in the northern
Gulf of Mexico Basin was made by Jones (1969). The hydrologic conditions
currently found in these sediments are similar to conditions that previously
occurred in older sedimentary basins. Osmotic flow has a dominant influence
upon the hydrology of normally and abnormally pressured aquifer systems
in the northern Gulf Basin.
Jones (1969) found that many forces such as gravity, sediment diagenesis,
different water salinities, ionic and molecular diffusion, different electrical
potentials of sediments, thermal potentials, pressure, and osmotic membrane
filtration affect the hydrology in this basin.
Fowler (1970) found that salinity variations within the Frio sands in the
Chocolate Bayou field, Brazoria County, Texas, are the result of selective
concentration of ions by shales acting as membranes. In this field, pressures
seem to reflect the flow paths of the waters, and the greatest changes in
pressures are found across shaly sections. Analyses of water samples from
this field over a 28-year period indicate decreasing salinity with production
time caused by dilution of the original brines by waters squeezed from the
shales adjacent to the aquifers.
Chilingarian and Rieke (1969) reviewed the processes which can alter the
chemical composition of formation waters. They concluded that most of the
original water was sea water, and that the concentration process in many
cases results from compaction and membrane filtration rather than evapora-
tion. Their experimental results indicated that solutions squeezed out of
rocks during compaction progressively decrease in dissolved solids concen-
trations with increasing depth.
Conclusions
The origin of oilfield waters is related to many natural processes. Initially,

meteoric water reacted with weathered rock, soil, and organic matter. The
excess waters that did not penetrate the rock or soil caused the rock and soil
to erode and channels formed through which the water could move more
easily. Forces of gravity caused the water to move from areas of high poten-
tial to areas of low potential, and as the waters moved, the concentrations of
dissolved solids in them increased. Some of these waters found their way to
lakes and the sea. As they entered the lakes or seas their movement slowed,
causing some of the suspended particles in them t o deposit. Mixing of the
waters with the more saline waters in the sea caused dissolved carbonate and
organic compounds to precipitate.
Evaporation of the sea and lake waters caused other compounds such as
sulfates t o precipitate. The pH of the waters changed slightly because of
reactions with the atmosphere, the sediments, and other waters. Each pH
change caused precipitation of compounds or dissolution of new com-
pounds.
Some of the waters became highly concentrated in dissolved solids in the
more shallow marine environments. Evaporites formed in these lagoons,
pans, and exposed supratidal sabkhas. Evaporites also formed in deep-water
basins when the salinity of the water at the bottom of the basin became
sufficiently high.
The sediments were buried as additional sediments were deposited on
them, and water surrounding the sediment particles also was buried. As the
depth of burial increased, the sediments compacted and some of the water
was squeezed out. Both the squeezed-out water and the remaining interstitial
water reacted with minerals in the sediments t o change the composition of
the dissolved solids in the water and the composition of the sediments.
Mechanisms that cause the oilfield waters t o differ in composition from
water originally deposited with the sediments include ion exchange, infil-
trating waters, sediment leaching, mineral formation, sulfate reduction, and
ultrafiltration through clay-shale membranes.
References
Al'tovskii, M.E., Kuznetsova, Z.I. and Shvets, V.M.,1961. Origin of Oil and Oil Deposits
(English Transl. by Consultants Bureau). Plenum Press, New York, N.Y., 107 pp.
Anonymous, 1964. Chemistry of the oceans. Chem. Eng. News, 42:12A.
Atwater, G.I. and Miller, E.E., 1965. The effect of decrease in porosity with depth on
future development of oil and gas reserves in South Louisiana. Presented at Annual
Meet., A m . Assoc. Pet. Geol., New Orleans, La., 1965 -Bull. A m . Assoc. Pet. Geol.,
49:334.
Ault, W.U., 1959. Isotopic fractionation of sulfur in geochemical processes. In: P.H.
Abelson (Editor), Researches in Geochemistry. John Wiley and Sons, New York, N.Y.,
pp.241-259.
Baas Becking, L.G.M., Kaplan, I.R. and Moore, D., 1960. Limits of the natural environ-
ment in terms of pH and oxidation-reduction potentials. J. Geol., 68 :243-284.
REFERENCES 24 7
Baker, E.G., 1960. A hypothesis concerning the accumulation of sediment hydrocarbons

to form crude oil. Geochim. Cosmochim. Acta, 19:309-317.
Barnes, H.L. and Czamanske, G.K., 1967. Solubilities and transport of ore minerals. In:
H.L. Barnes (Editor), Geochemistry o f Hydrothermal Ore Deposits. Holt, Rinehart,
and Winston, New York, N.Y., pp.334-381.
Bentor, Y.K., 1969. On the evolution of subsurface brines in Israel. Chem. Geol.,
4 :83-1 10.
Berry, F.A.F., 1969. Relative factors influencing membrane filtration effects in geologic
environments. Chem. Geol., 4:295-301.
Billings, G.K., Hitchon, B. and Shaw, D.R., 1969. Geochemistry and origin of formation
waters in the Western Canada Sedimentary Basin, 2. Alkali metals. Chem. Geol.,
4:211-223.
Birch, F., 1954. The present state of geothermal investigations. Geophysics, 19:645+359.
Bond, D.C., 1972. Hydrodynamics in deep aquifers of the Illinois Basin. Ill. State Geol.
Surv., Circ., No. 470, 7 2 pp.
Borchert, H. and Muir, R.O., 1964. Salt Deposits - The Origin, Metamorphism, and
Deformation o f Evaporite. D. Van Nostrand, London, 338 pp.
Braitsch, 0. and Herrmann, A.G., 1963. Zur Geochemie des Broms in Salinaren Sedi-
menten, I. Experimentelle Bestimmung der Br-Verteilung in Naturlichen Salzsystemen.
Geochim. Cosmochim. Acta, 27:361-391.
Braitsch, 0 . and Herrmann, A.G., 1964. Zur Geochemie des Broms in Salinaren Sedi-
menten, 11. Die Bildungstemperaturen Primarer Sylvin- und Carnallit-Gesteine.
Geochim. Cosmochim. Acta, 28 :1081-1 109.
Bredehoeft, J.D., Blyth, C.R., White, W.A. and Maxey, G.B., 1963. Possible mechanism
for concentration of brines in subsurface formations. Bull. A m . Assoc. Pet. Geol.,
47 :257-269.
Brod, I.O., 1960. On principal rules in the occurrence of oil and gas accumulations in the
world. Int. Geol. Rev., 2:922-1005.
Broecker, W.S. and Takahashi, T., 1966. Calcium carbonate precipitation on the Bahama
Banks. J. Geophys. Res., 71 :1575-1602.
Brongersma-Sanders, M., 1971. Origin of major cyclicity of evaporites and bituminous
rocks: an actualistic model. Mar. Geol., 11 :123-144.
Bruderer, W., 1956. Les ocbans souterrains fossiles et le phtrole. Assoc. Fr. Technol. Pet.
Bull., 120:535-556.
Burst, J.F., 1969. Diagenesis of Gulf Coast clayey sediments and its possible relation t o
petroleum migration. Bull. A m . Assoc. Pet. Geol., 53:73-93.
Bush, P.R., 1970. Chloride-rich brines from sabkha sediments and their role in ore forma-
tion. Inst. Min. Metall. Trans., 79:B137-B144.
Butler, G.P., 1969. Modern evaporite deposition and geochemistry of coexisting brines,
the Sabkha Trucial Coast, Arabian Gulf. J. Sediment. Petrol., 39 :70-89.
Caraway, W.H. and Gates, G.L., 1959. Methods for determining water contents of oil-
bearing formations. U.S. Bur. Min. Rep. Invest., No.5451, 81 pp.
Carpelan, L.H., 1957. Hydrobiology of the Alviso salt ponds. Ecology, 38:375-390.
Carpenter, A.B. and Miller, J.C., 1969. Geochemistry of saline subsurface water, Saline
County (Missouri). Chem. Geol., 4:135-167.
Cartmill, J.C. and Dickey, P.A., 1970. Flow of a disperse emulsion of crude oil in water
through porous media. Bull. A m . Assoc. Pet. Geol., 54:2438-2447.
Case, L.C., 1955. Origin and current usage of the term "connate water". Bull. A m . Assoc.
Pet. Geol., 39:1879-1882.
Chave, K.E., 1960. Evidence on history of sea water from chemistry of deeper subsurface
waters of ancient basins. Bull. A m . Assoc. Pet. Geol., 44:357-370.
Chebotarev, I.I., 1964. Metamorphism of natural waters in the crust of weathering.
Geochim. Cosmochim. Acta, 8:22-48, 137-170, 198-212.
Chilingarian, G.V. and Rieke, 111, H.H. 1969. Some chemical alterations of subsurface
waters during diagenesis. C h e m Geol., 4:235-252.
Clark, S.P. and Ringwood, A.E., 1964. Density disturbance and constitution of the
mantle. Rev. Geophys., 2:35-88.
Clarke, F.W. and Washington, H.S., 1924. The composition of the earth’s crust. US.
Geol. Surv. Prof.Paper, No.127, pp.1-112.
Clayton, R.N., Friedman, I., Graf, D.L., Mayeda, T.K., Meents, W.F. and Shimp, N.F.,
1966. The origin of saline formation waters, 1. Isotopic composition. J. Geophys.
Res., 71 :3869-3881.
Cloke, P.L., 1966. The geochemical application of E h - p H diagrams. J. Geol. Educ.,
1 4:140-148.
Collins, A.G., 1967. Geochemistry of some Tertiary and Cretaceous age oil-bearing forma-
tion waters. Environ. Sci. Technol., l :725-730.
Collins, A.G. 1969a. Chemistry of some Anadarko Basin brines containing high concen-
trations of iodide. C h e m Geol., 4:169-187.
Collins, A.G., 1969b. Solubilities of some silicate minerals in saline waters. U S . O f f .
Saline Water R e s Dev. Progr. R e p . , No.472, 27 pp.
Collins, A.G., 1970. Geochemistry of some petroleum-associated waters from Louisiana.
U.S. Bur. M i a Rep. Invest., No.7326, 31 pp.
Collins, A.G., 1972. Geochemical Classification o f Formation Waters f o r Use in Hydrocar-
bon Exploration and Production. M.S. Thesis, University of Tulsa, Tulsa, Okla., 63 pp.
Collins, A.G., Bennett, J.H. and Manuel, O.K., 1971. Iodine and algae in sedimentary
rocks associated with iodine-rich brines. Geol. SOC.A m . Bull., 82:2607-2610.
Dapples, E.C., 1959. The behavior of silica in diagenesis. In: Silica in Sediments ( A
Symposium) - SOC.E c o n Paleontol. Mineral., Spec. Publ., No.7, pp.36-54.
Deffeyes, K.S. Lucia, F.J. and Weyl, P.K., 1964. Dolomitization: observations on the
island of Bonaire, Netherlands Antilles. Science, 143 :678+379.
Degens, E.T. and Chilingar, G.V., 1967. Diagenesis of subsurface waters. In: G. Larsen
and G.V. Chilingar (Editors), Diagenesis in Sediments. Elsevier, Amsterdam,
pp.77-502.
Degens, E.T., Hunt, J.M., Reuter, J.H. and Reed, W.E., 1964. Data on the distribution of
amino acids and oxygen isotopes in petroleum brine waters of various geologic ages.
Sedimentology, 3 :199-225.
DeSitter, L.Y., 1947. Diagenesis of oilfield brines. Bull. A m . Assoc. Pet. Geol.,
31 :2030-2040.
Dickey, P.A., 1969. Increasing concentration of subsurface brine with depth. Chem.
Geol., 4:361-370.
Dingman, R.J. and Angino, E.E., 1969. Chemical composition of selected Kansas brines
as a aid to interpreting change in water chemistry with depth. Chem. Geol.,
4: 325-339.
Dunham, K.C., 1970. Mineralization by deep formation waters - a review. Znst. Metall.
Trans., 79:B127-B>36.
Eckhardt, F.J., 1958. Uber Chlorite in Sedimenten. Geol. Jahrb., 75:437-474.
Egleson, G.D. and Querio, C.W., 1969. Variation in the composition of brine from the
Sylvania formation near Midland, Michigan. Environ. Sci. Technol., 3:367-371.
Ellis, A.J.’, 1968. Natural hydrothermal systems and experimental hot waterhock interac-
tion: reactions with NaCl solutions and trace metal extraction. Geochim. Cosmochim.
Acta, 32:1356-1363.
Emery, K.O., 1960. The Sea O f f Southern California: A Modern Habitat o f Petroleum.
John Wiley and Sons, New York, N.Y., 366 pp.
Emery, K.O. and Rittenberg, S.C., 1952. Early diagenesis of California Basin sediments in
relation t o origin of oil. Bull. A m . Assoc. Pet. Geol., 36:735-806.
REFERENCES 249
Evans, G., Kinsman, D.D.J. and Shearman, D.J., 1963 A reconnaissance survey of the
environment of the recent carbonate sedimentation along Trucial Coast, Persian Gulf.
In: L.M.J.U. van Straaten (Editor), Developments in Sedimentology, 1 . Deltaic and
Shallow Marine Deposits. Elsevier, Amsterdam, pp. 129-1 35.
Fowler, Jr., W.A., 1970.Pressures, hydrocarbon accumulation, and salinities - Chocolate
Bayou field, Brazoria County, Texas. J. Pet. Technol., 22:411-423.
Friedman, G.M. and Sanders, J.E. 1967. Origin and occurrence of dolostones. In: G.V.
Chilingar, H.J. Bissel and R.W. Fairbridge (Editors), Carbonate Rocks - Origin, Occur-
rence and Classification. American Elsevier, New York, N.Y., pp.267-348.
Friedman, G.M. Fabricand, B.P., Imbimbo, E.S., Brey, M.E. and Sanders, J.E., 1968.
Chemical changes in interstitial waters from continental shelf sediments. J. Sediment.
Petrol., 38:1313-1319.
Garrels, R.M. and Christ, C.L., 1965. Solutions, Minerals, and Equilibria. Harper and
Row, New York, N.Y., 450 pp.
Gevirtz, J.L. and Friedman, G.M. 1966. Deep-sea carbonate sediments of the Red Sea
and their implications on marine lithification. J. Sediment. Petrol., 36:143-151.
Ginsburg, R.N. and Lowenstam, H.A., 1958. Influence of marine bottom communities on
the depositional environment of sediments. J. Geol., 66:310-318.
Grabau, A.W., 1920. Geology of Nonmetallic Mineral Deposits Other Than Silicate, 1 .
Principles of Salt Deposits. McGraw-Hill, New York, N.Y., 435 pp.
Graf, D.L., Friedman, I. and Meents, W.F., 1965. The origin of saline formation waters,
11. Isotopic fractionation by shale micropore systems. Ill. State Geol. Surv. Circ.,
No.393, 32 pp.
Graf, D.L., Meents, W.F., Friedman, I. and Shimp, N.F., 1966. The origin of saline
formation waters, 111. Calcium chloride waters. Ill. State Geol. Surv. Circ. No. 397,60
PP.
Grim, R.E., 1952. Clay Mineralogy. McGraw-Hill, New York, N.Y., 396 pp.
Ham, W.E. 1962. Classification of carbonate rocks. A m . Assoc. Pet. GeoL, Mem. 1 , 279
PP.
Hanshaw, B.B. and Hill, G.A., 1969. Geochemistry and hydrodynamics of the Paradox
Basin region, Utah, Colorado, and New Mexico. Chem. Geol., 4:263-294.
Helgeson, H.C., 1964. Complexing and Hydrothermal Ore Deposition. MacMillan, New
York, N.Y., 128 pp.
Hem, J.D., 1970. Study and interpretation of the chemical characteristics of natural
water. U.S. Geol. Surv. Water Supply Paper, No. 1473,363 pp.
Hiltabrand, R.R. 1970. Experimental Diagenesis o f Argillaceous Sediment. Ph.D. Disser-
tation, Louisiana State University, Baton Rouge, La., 152 pp., unpublished.
Hitchon, B., 1968. Rock volume and pore volume data fo; plains region of Western
Canada Sedimentary Basin between latitudes 49' and 60 N. Bull. A m . Assoc. Pet.
Geol., 52:2318-2323.
Hitchon, B. 1969a. Fluid flow in the Western Canada Sedimentary Basin, 1. Effect of
topography. Water Resour. Res., 5:186-195.
Hitchon, B., 196913. Fluid flow in the Western Canada Sedimentary Basin, 2. Effect of
geology. Water Resour. Res., 5:460-461.
Hitchon, B. and Friedman, I., 1969. Geochemistry and origin of formation waters in the
Western Canada Sedimentary Basin, I. Stable isotopes of hydrogen and oxygen.
Hitchon, B . and Hays, J., 1971. Hydrodynamics and hydrocarbon occurrences, Surat
Basin, Queens land, Australia. Water Resour. Res., 7:658-676.
Holser, W.T. 1963. Chemistry of brine inclusions in Permian salt from Hutchison,
Kansas. In: J.L. Rau (Editor), Symposium o n Salt. Northern Ohio Geological Society,
Cleveland, Ohio, pp.86-103.
Hunt, J.M., 1968. How gas and oil form and migrate. World Oil, 163:140-150.
Hunt, J.M. and Jamieson, G.W., 1958. Oil and organic matter in source rocks of petro-
leum. In: L.G. Weeks (Editor), Habitat of Oil. American Association of Petroleum
Geologists, Tulsa, Okla., pp. 735-746.
Illing, L.V., 1954. Bahaman calcareous sands. Bull. A m . Assoc. Pet. Geol., 38:l-95.
Jones, B.F., Vandenburgh, A.S., Truesdell, A.H., and Rettig, S.L., 1969. Interstitial brines
in playa sediments. Chem. Geol., 4:253-262.
Jones, P.H., 1969. Hydrology of Neogene deposits in the northern Gulf of Mexico Basin.
La. Water Resour. Res. Inst. Bull., GT-2, 105 pp.
Kelley, W.P., 1948. Cation Exchange in Soils. Reinhold, New York, N.Y., 144 pp.
Khitarov, N.I. and Pugin, V.A., 1966. Behavior of montmorillonite under elevated
temperatures and pressures. Geochem. Znt., 3:621-626.
Kimura, S . and Souriragan, S., 1967. Analysis of data in reverse osmosis with porous
cellulose acetate membranes used. A m . Inst. Chem. Eng., 13:497-503.
Kozin, A.N., 1960. Geochemistry of bromine and iodine of formation waters in the
Kuybyshev area o n the Volga. Pet. Geol., 4:llO-113.
Kramer, J.R., 1963. History of the composition of sea water - liquid inclusions com-
pared with a chemical equilibrium model. Geol. SOC.A m . , Spec. Paper, No. 73, 190 pp.
Kramer, J.R., 1969. Subsurface brines and mineral equilibria. Chem. Geol., 4:37-50.
Krauskopf, K.B., 1956. Factors controlling the concentrations of thirteen rare metals in
sea water. Geochim. Cosmochim. Acta, 9:l-32.
Krumbein, W.C., 1951. Occurrence and lithologic associations of evaporites in the United
States. J. Sediment Petrol., 21:63--81.
Krejci-Graf, K., 1963. Uber Rumanische Olfeldwasser. Geol. Mitt. Hydrogeol. Hydrochem.,
2:3 51-392.
Kvenvolden, K.A., 1964. Hydrocarbons in modern sediments and the origin of petroleum.
Min. Mag., Colo. School Min., 54:24-25.
Larson, T.J., 1967. Purification of subsurface waters by reverse osmosis. J. Am. Water
Works Assoc., 59:1527-1548.
Levorsen, A.I., 1966. Geology of Petroleum. W.H. Freeman, San Francisco, Calif., 2nd
ed., 724 pp.
Loeb, S. and Manjikian, S., 1965. Six-month field test of a reverse osmosis desalination
membrane. Ind, Eng. Chem. Process Design Dev., 4: 207-21 2.
Lotze, F., 1938. Wichtigen Lager Stattender “nicht-erze”, III. Steinsalz und Kalisalze
Geologie. Borntraeger, Berlin, 936 pp.
Mandl, I., 1953. Solubilization of insoluble matter in nature, 11. The part played by salts
of organic and inorganic acids occurring in nature. Biochim. Biophys. Acta,
10:540-569.
Mandl, I., Grauer, A. and Neuberg, C., 1952. Solubilization of insoluble matter in nature,
I. The part played by salts of adenosinetriphosphate. Biochim. Biophys. Acta,
8 :654-663.
Manheim, F.T. and Bischoff, J.L., 1969. Geochemistry of pore waters on the Continental
Slope of the northern Gulf of Mexico. Chem. Geol., 4:63-82.
McAuliffe, C., 1969. Determination of dissolved hydrocarbons in subsurface brines.
Chem. Geol., 4:225-234.
McKelvey, J.G., Spiegler, K.S. and Wyllie, M.R.J., 1957. Salt filtering by ion-exchange
grains and membranes. J. Phys. Chem., 61(2):174-178.
Meyerhoff, A.A., 1970. Development in Mainland China, 1949-1968. Bull. A m . Assoc.
Pet. Geol., 54:1567-1580.
Michaels, A.S., Bixler, H.J. and Hodges, Jr., R.M., 1965. Kinetics of water and salt
transport in cellulose acetate reverse osmosis desalination membranes. J. Colloid Sci.,
20: 1034-1056.
Milne, I.H., McKelvey, J.G. and Trump, R.P., 1964. Semi-permeability of bentonite mem-
branes to brines. Bull. A m . Assoc. Pet. Geol,, 48:103-105.
REFERENCES 251
Moore, C.A., 1969. The occurrence of oil in sedimentary basins. World Oil, 168:69-72.
Morris, R.C. and Dickey, P.A., 1957. Modern evaporite deposition in Peru. Bull. A m .
Assoc. Pet. Geol., 41:2467-2474.
Neruchev, S.G. and Kovacheva, I.S., 1965. The effect of geological conditions on the
amount of oil given up by source rocks. Dokl. Akad. Nauk U.S.S.R., 162:913-914.
Neumann, H.J. and Jobelius, H., 1967. Detection of emulsifying agents in crude oils and
oilfield waters as a contribution to the problem of oil migration. Erdol Kohle
Petrochem., 20:622-625.
Parker, J. W., 1969. Water history of Cretaceous aquifers, East Texas Basin. Chem. Geol.,
4:lll-133.
Peake, E. and Hodgson, G.W., 1966. Alkanes in aqueous systems, I. Exploratory investiga-
tions on the accommodation of Cz 0-c~3 n-alkanes in distilled water and occurrence
in natural water systems. J. A m . Oil Chem. SOC.,43:215-222.
Pettijohn, F.J., 1957. Sedimentary Rocks. Harper and Brothers, New York, N.Y., 2nd
ed., 718 pp.
Philipp, W., Drong, H.J., Fuchtbauer, H., Haddenhorst, H.G. and Jankowsky, W.J., 1963.
The history of migration in the Gifhorn Trough ( N W Germany), Sixth World Pet.
Congr., Frankfurt/Main, June, 1963, Sect. I, Paper, No. 19, pp. 457-481.
Philippi, G.T., 1965. On the depth, time and mechanism of petroleum generation.
Geochim. Cosmochim. Acta, 29: 1021-1049.
Phleger, F.B. and Ewing, G.C., 1962. Sedimentology and oceanography of coastal lagoons
in Baja, California, Mexico. Geol. SOC.A m . Bull., 73:145-181.
Pirson, S.J., 1968. Redox log interprets reservoir potential. Oil Gas J., 66:6*75.
Plumley, W.J., Risley, G.A., Graves, Jr., R.W. and Kaley, M.E., 1962. Energy index for
limestone interpretation and classification. In: W.E. Hem (Editor), Classification o f
Carbonate Rocks - A m . Assoc. Pet. Geol., Mem.1, pp.85-107.
Pollard, T.A. and Reichertz, P.O., 1952. Core-analysis practices - basic methods and new
developments. Bull. Am. Assoc. Pet. Geol., 36:230-252.
Quaide, W . , 1958. Claymineralsfrom salt concentration ponds. A m . J. Sci., 256:431-437.
Rankama, K. and Sahama, T.G., 1950. Geochemistry. Chicago University Press, Chicago,
Ill., 991 pp.
Riley, G.A., 1944. The carbon metabolism and photosynthetic efficiency of the earth as a
whole. J. A m . Sci., 32:134.
Rittenhouse, G., 1967. Bromine in oilfield waters and its use in determining possibilities
of origin of these waters. Bull. A m . Assoc. Pet. Geol., 51:2430-2440.
Rittenhouse, G., Fulton, R.B., Grabowski, R.J. and Bernard, J.L., 1969. Minor elements
in oilfield waters. Chem. Geol., 4:189-209.
Rosenqvist, I.T., 1962. The influence of physico-chemical factors upon the mechanical
properties of clays. Clays Clay Minerals, 9 :12-27.
Ross, C.S., 1943. Clays and soils in relation to geologic processes. J. Wash. Acad. Sci.,
33:225-235.
Russell, W.L., 1933. Subsurface concentration of chloride brines. Bull. A m . Assoc. Pet.
Geol., 17:1213-1228.
Schoeller, H., 1955. Geochemie des eaux souterraines. Rev. Inst. F'r. Pet., 10:181-213,
219-246, 507-552.
Siever, R., Beck, K.C. and Berner, R.A. 1965. Composition of interstitial waters of
modern sediments. J. Geol., 73:39-73.
Skinner, B.J., 1969. Earth Resources. Prentice-Hall, Englewood Cliffs, N.J., 149 pp.
Sloss, L.L., 1953. The significance of evaporites. J. Sediment. Petrol., 23:143-161.
Smith, P.V., 1954. Studies on origin of petroleum: occurrence of hydrocarbons in recent
sediments. Bull. A m . Assoc. Pet. Geol., 38:377-381.
Stumm, W. and Morgan, J.J., 1970. Aquatic Chemistry. Wiley-Interscience, Div. of John
Wiley and Sons, New York, N.Y., 583 pp.
Taylor, D.W., 1956. Fundamentals o f Soil Mechanics. John Wiley and Sons, New York,
N.Y., 700 pp.
Tenaghi, K. and Peck, R.B., 1968. Soil Mechanics in Engineering Pmctice. John Wiley
and Sons, New York, N.Y., 84 pp.
Topms, J.S., 1970. Review of knowledge of metalliferous brines and related deposits.
Inst. Min. Metall. Trans., 79:B116-B126.
Truesdell, A.H. and Jones, B.F., 1969. Ion association in natural brines. Chem. Geol.,
4:51-62.
Usiglio, J., 1849. Analyse de I’eau de la Mediterrank sur les c6tes de France. Ann. Chim.
Phys., 3:92-1 07, 27 :172-191.
Valyashko, M.G., 1956. The geochemistry of bromine in halogenesis processes and the
use of bromine content as a genetic and prospecting criterion. Geochemistry,
1 :33-49.
Van Everdingen, R.O., 1968. Studies on formation waters in western Canada; geo-
chemistry and hydrodynamics. Can. J. Earth Sci., 5:523-543.
Van Nostrand Press, 1958.Science Encyclopedia. Princeton, N.J., 2nd ed., pp. 371-1528.
Vinogradov, A.P., 1953. The elementary chemical composition of marine organisms.
Sears. Found. Mar. R e s , Mem., 11:85,91,107,216.
Visher, G.S., 1965. Use of vertical profile in environmental reconstruction, Bull. A m .
Assoc. Pet. Geol., 49:41-61.
Von Engelhardt, W., 1960. On the chemistry of the pore solution of sediments. Uppsala
Univ. Geol. Inst. Bull., 40: 189-204 (in German).
Weller, J.M., 1959. Compaction of sediments. Bull. A m . Assoc. Pet. Geol., 43:273-310.
White, D.E., 1957. Magmatic, connate, and metamorphic waters. Geol. SOC.A m . Bull.,
68:1659-1 682.
White, D.E., 1965. Saline waters of sedimentary rocks. In: A. Young and J.E. Galley
(Editors), Fluids in Subsurface Environments - A m . Assoc. Pet. Geol., Mem.4,
pp.342-366.
Young, A. and Low, P.F., 1965. Osmosis in argillaceous rocks. Bull. A m . Assoc. Pet.
Geol., 49:1004-1008.
Chapter 8. CLASSIFICATION OF OILFIELD WATERS
Classification of waters provides a basis for grouping closely related

waters. Because the grouping is chemical, it is dependent upon the dissolved
constituents found in the waters. Most of the classification systems devel-
oped t o date have considered only the dissolved major inorganic constituents
and have ignored the organic and the minor and trace inorganic constituents.
Waters as related to the earth are meteoric, surface, and subsurface. Sur-
face waters can be fresh or saline if the amounts of dissolved constituents in
the waters are used to classify them. For example, water from melting snow
on a mountain top usually will contain small amounts of dissolved mineral
matter and can be classified as fresh water, while water in an ocean will
contain about 35,000 mg/l dissolved minerals and is classified as saline.
Waters found in rivers connecting the mountain stream t o the ocean may
contain varying amounts of dissolved constituents and depending upon the
amounts can be classified as fresh or saline. In a similar manner, subsurface
waters are classified as fresh or saline. Merely classifying a water as either
fresh or saline does not provide a very useful classification. The dissolved
constituents that are used in many classification systems depend upon the
amounts or ratios of sodium, magnesium, calcium, carbonate, bicarbonate,
sulfate, and chloride found in the water. The reason for this is that these are
the ions that usually are determined or calculated in a water. (Sodium often
is calculated from the difference found in the stoichiometric balance of the
determined anions and cations.)
The amounts and ratios of these constituents in subsurface waters are
dependent upon the origin of the water and what has occurred t o the water
since entering the subsurface environment. For example, some subsurface
waters found in deep sediments were trapped during sedimentation, while
other subsurface waters have been diluted by infiltration of surface waters
through outcrops. Some waters have been replaced by infiltration water.
Also, rocks containing the waters often contain soluble constituents, which
dissolve in the waters or contain chemicals which will exchange with chemi-
cals dissolved in the waters causing alterations of the dissolved constituents.
The amounts of dissolved constituents found in subsurface waters can
range from a few milligrams per liter t o more than 350,000 mg/L This
salinity distribution is dependent upon several factors, including hydraulic
gradients, depth of occurrence, distance from outcrops, mobility of the
dissolved chemical elements, soluble material in the associated rocks, and the
exchange reactions.
254 CLASSIFICATION OF OILFIELD WATERS
Portions of three classification systems (Palmer, 1911; Sulin, 1946;

Schoeller, 1955) and Bojarski’s (1970) modification of Sulin’s system were
applied to about 4,000 formation waters (U.S. Bureau of Mines, 1965). The
waters were analyzed by standard methods (American Petroleum Institute,
1968). The results indicated that the classifications are useful in exploration
and production problems.
Palmer’s classification
Palmer (1911) observed that the basic characteristics of natural waters are
dependent upon their salinity (salts of strong acids) and alkalinity (salts of
weak acids). Salts that cause salinity are those that are not hydrolyzed, while
alkalinity is caused by free alkaline bases produced by the easily hydrolyz-
able salts of weak bases.
All positive ions (cations) including hydrogen can cause salinity, but of
the negative ions (anions), only the strong acids, (e.g., chloride, sulfate, and
nitrate) can cause salinity. Because salinity is dependent upon the combined
activity of the cations and anions and is limited by the reacting values of the
strong acids, its value is determined by multiplying the total value of the
strong acids by two.
Alkalinity is caused by free alkaline bases as a result of the hydrolytic
action of water on dissolved bicarbonates and other weak acid salts. The
alkalinity value is calculated by doubling the reacting values of the bases
which exceed the reacting values of the strong acids.
The ions that commonly are found in waters comprise three groups: (a)
alkalies (sodium, potassium, lithium), whose salts are easily soluble in water
and do not cause hardness; (b) alkaline earths (magnesium, calcium, stron-
tium, barium), whose salts cause hardness and many of which are sparingly
soluble; and (c) hydrogen, whose salts are acids and cause acidity.
Geologists know what “strong alkalies”, “alkaline earths”, “strong acid
radicles”, “weak acid radicles”, “ions”, and “reacting values” mean general-
ly. To compare several analyses it usually is easier if they are made on a
chemical basis. The. proportions of the various ions do not react in propor-
tion t o the various weights given in milligrams per liter but rather in propor-
tion to their “capacity for reaction”, or “reaction value”. The reacting value
of each ion is determined by multiplying the amount of each radicle by
weight (mg/l) by its “reaction coefficient”, which is the valence of a radicle
divided by its atomic weight.
The groups of the ions are determined by summing the reacting values of
their members, and according to the predominance of reacting values of the
groups, five special properties were designated by Palmer. To determine the
special properties, the reacting values of a group of cations or anions are
doubled so that the full value of a given special property is considered. The
terms “primary” and “secondary” were used t o qualify the general proper-
PALMER’S CLASSIFICATION 25 5
ties of the water; e.g., the principal soluble decomposition products of the
oldest rock formations are the alkalies (primary), while more recent rock
formations are the principal source of the alkaline earths (secondary), This
theory of Palmer’s that the terms primary and secondary are associated with
the age of the rock should not necessarily be considered undisputably true,
because primary salinity certainly can be acquired from other soluble
material than that derived directly from decomposition products of the
oldest rock formations.
The five special properties of water are:
(1) Primary salinity (alkali salinity); that is, salinity not t o exceed twice
the sum of the reacting values of the radicles of the alkalies.
(2) Secondary salinity (permanent hardness); that is, the excess (if any) of
salinity over primary salinity, not t o exceed twice the sum of the reacting
values of the radicles of the alkaline earths group.
(3) Tertiary salinity (acidity); that is, the excess (if any) of salinity over
primary and secondary salinity.
(4) Primary alkalinity (permanent alkalinity); that is, the excess (if any) of
twice the sum of the reacting values of the alkalies over salinity.
(5) Secondary alkalinity (temporary alkalinity); that is, the excess (if any)
of twice the sum of the reacting values of the radicles of the alkaline earths
group over secondary salinity.
Reacting values in percent are used in this system. The percentage values
are determined by summing the milliequivalents of all the ions, dividing the
milliequivalents of a given ion by the sum of the total milliequivalents, and
multiplying by 100. Waters are classified by numerical values of the relation-
ships of anions to the cations, where a , b , and d represent the percentage
values of the alkali cations, alkaline earth cations, and strong acid anions,
respectively. Any one of the following five conditions may exist: d may be
equal t o or less than a , greater than a and less than a + b y equal to a + b y or
greater than a + b . Using these conditions, waters are classified into five
classes:
Classl: d < a
2d = primary salinity
2(a - d ) = primary alkalinity
2b = secondary alkalinity
Class 2: d = a
2u or 2d = primary salinity
2b = secondary alkalinity
Class 3 : d > a ; d < (a + b )

2a = primary salinity
2(d - a ) = secondary salinity
2(a + b - d ) = secondary alkalinity
Class 4: d = (a + b )
2a = primary salinity
2b = secondary salinity
Class 5: d > ( a + b )
2u = primary salinity
2b = secondary salinity
2(d - a - b ) = tertiary salinity (acidity)
These five classes of water are found in nature. Examples of the first three
classes are various surface waters, sea water and brines represent class 4,
while mine drainage waters and waters of volcanic origin fall in class 5
(Palmer, 1911).
Rogers (1917, 1919) studied oilfield waters of the San Joaquin Valley,
California, and used the classification system of Palmer (1911). He found
that generally the surface waters of the San Joaquin Valley possess second-
ary salinity rather than primary alkalinity, contain more sulfate than
chloride, and contain low amounts of bicarbonate. With increasing depth,
the subsurface waters decrease in secondary salinity until primary alkalinity
becomes evident. Waters above an oil zone often contained hydrogen sulfide,
which was attributed t o reduction of sulfates by hydrocarbons, thus de-
creasing the amounts of sulfate and increasing the bicarbonate in the water,
which Rogers called an altered water. Further he found that, in these altered
waters in close proximity to hydrocarbon accumulations, chloride becomes
relatively and absolutely important because of the residual chloride from the
original (ancient) sea water chlorides as compared to waters above the oil
zone which often are freshened because of a more hydrodynamic situation.
Altered waters, according t o his definition, can have either primary alkalinity
or secondary salinity depending upon their amounts of carbonate and chlo-
ride, but normal waters have only secondary salinity.
Elliott (1953) used the Palmer system t o determine the chemical charac-
teristics of some Paleozoic age formation waters in Texas. He found that all
of the waters in the group that he studied (about 70) contained predomi-
nant, primary salinity. Many of these waters contained appreciable concen-
trations of sulfate; one contained 5,800 mg/l sulfate, and many contained
more than 2,000 mg/l. The calcium concentration ranged up to 13,000 mg/l
while the bicarbonate concentrations ranged up t o 800 mg/l.
Ostroff (1967) used the Palmer classification to classify waters from
several basins and t o compare this classification system with two other
systems. He found that the Palmer system groups some of the constituents
together that are not closely related chemically. Furthermore the system
does not consider ionic concentrations or saturation conditions related t o
sulfate or bicarbonate.
SULIN’S CLASSIFICATION 257
Sulin’s classification
Sulin (1946), a Russian geochemist, proposed a classification system based

upon various combinations of dissolved salts in the waters. The waters are
described according to chemical type, subdivided into group, subgroup, and
class. He found four basic environments of natural water distribution:
(1) Continental (terrestrial) conditions which promote the formation of
sulfate waters. Such conditions supply soluble sulfate constituents t o the
water and the genetic type of such a water is “sulfate-sodium”.
(2) Continental conditions which promote the formation of sodium bicar-
bonate waters. The genetic type is “bicarbonate-sodium”.
( 3 ) Marine conditions and the formation of a “chloride-magnesium ” type
of water.
(4) Deep subsurface conditions within the earth’s crust and the formation
of a “chloride-calcium ” type of water.
The first two types are characteristic of meteoric and/or artesian waters,
the third of marine environments and evaporite sequences, and the fourth of
deep stagnant conditions.
Types, groups, and subgroups
Water composition is expressed in milligram-equivalents of the separable

ions, and the composition is calculated per 100 g of water. The percent of
the sum of the equivalents is used t o exclude the degree of water mineraliza-
tion, and t o compare waters containing different amounts of dissolved solids.
The ratio Na/Cl expressed in the percent equivalent form determines the
genetic water type. If the value is greater than one, sodium predominates
over chloride and the excess sodium can be combined with sulfate or bicar-
bonate. Therefore waters with a Na/Cl ratio greater than one belong to the
bicarbonate-sodium or the sulfate-sodium types. Sulin calculated sodium as
the sum of all the alkalies (Li, K, Na etc.) and chloride as the sum of all the
halides (Cl, Br, I).
The ratio (Na - C1)/S04, if greater than one, indicates that the water is
the bicarbonate-sodium type, while if it is less than one it is the sulfate-
sodium type. Similarly the ratio (Cl- Na)/Na if less than one indicates the
chloride-magnesium type, but if greater than one it indicates the chloride-
calcium type.
Water classes
Subdivision of the groups of waters were made by Sulin (1946) using the
Palmer (1911) characteristics, because these characteristics express the dis-
solved constituents in the waters in a generalized format. For example, the
sum of the alkali chlorides and sulfates corresponds t o primary salinity, and
the sum of the alkaline earth chlorides and sulfates corresponds to secondary
salinity and the sodium bicarbonate-calcium stage. N o sodium bicarbonate is

present in sulfate-sodium, chloride-magnesium, or chloride-calcium types of
water; therefore, these types are classified as follows:
(1)Class A2 : secondary alkalinity predominates (alkaline earth carbonates
and bicarbonates).
(2) Class S2 : secondary salinity predominates (alkaline earth sulfates and
chlorides).
(3) Class S, : primary salinity predominates (alkali sulfates and chlorides).
(4) Class S,: tertiary salinity predominates (iron and aluminum sulfates
and chlorides and free strong acids).
Bicarbonate-sodium type waters contain sodium bicarbonate and are
classified as follows:
( 5 ) Class A2 : secondary alkalinity predominates (alkaline earth carbonates
and bicarbonates).
(6) Class A , : primary alkalinity predominates (alkali carbonates and bicar-
bonates).
(7) Class S, : primary salinity predominates (alkali chlorides and sulfates).
(8) Class A, : tertiary alkalinity predominates (iron and aluminum carbon-
ates and bicarbonates).
The water classification is expressed by use of a formula representing
decreasing values of the Palmer characteristics. For example, S, S2 A 2 indi-
cates that primary salinity is predominant and is followed by secondary
salinity and secondary alkalinity. Therefore, the classes are subdivided into
subclasses, and class S1 can include the subclass S, S2 A 2 , S1 A2 S2, S1 S2,
and S, . Table 8.1 outlines Sulin’s method of water characterization. Table
8.11 briefly outlines the relative values of the coefficients which determine
the four genetic types of waters.
The Palmer characteristics do not account for the interrelations between
chloride and sulfate and between calcium and magnesium. Therefore, Sulin
calculated the ratio SO4/C1 and Ca/Mg to establish additional subgroups. The
complete water characterization included the following: (a) water formula
given in Palmer characteristics; (b) coefficients in percent equivalents for
S04/C1 and Ca/Mg; (c) sum of the milligram equivalents per 100 g of water
(Z r ) t o illustrate the degree of water mineralization; and (d) the genetic
coefficients (Na - C1)/ SO4 and (Ca - Na)/Mg t o determine the water type,
and Na/C1 t o determine related genetic types of water.
Hydrochemical indicators of hydrocarbons
Sulin (1946) noted that certain properties of subsurface waters were

favorable indicators of hydrocarbon accumulations. The bicarbonate-sodium
and chloride-calcium types of waters are widely found in oilfields. However,
the chloride-calcium type is the more favorable indicator if it has the most
characteristic composition plus certain minor or micro constituents. In gen-
eral, he determined that hydrocarbon accumulations are most commonly
SULIN’S CLASSIFICATION 25 9
TABLE 8.1
Sulin’s method of water characterization

_____
Na/Cl >1
Sulfate-sodium type:
(Na+ - C1-)
<1 Bicarbonate-sodium type:
(Na+ - C1-) >
so,,-2 so4-2
Bicarbonate group Bicarbonate group
class A2 class A1
calcium subgroup sodium subgroup
magnesium subgroup class A2
Sulfate group calcium subgroup
class S1 magnesium subgroup
magnesium subgroup class S 1
sodium subgroup sodium subgroup
class Sp Sulfate group
calcium subgroup class S1
magnesium subgroup sodium subgroup
Chloride group Chloride group
class S 1 class S1
magnesium subgroup
sodium subgroup
Na/Cl <1
(C1-- Na+) - Na+) >
Chloride-magnesium type:
Mg+’
< 1 Chloride-calcium type: (Cl-Mg+2
Bicarbonate group Bicarbonate group

class Aq class A 2
calcium subgroup calcium subgroup
magnesium subgroup magnesium subgroup
Sulfate group Sulfate group
class s1 class S1
class S2 class s 2
Chloride group Chloride group
class S1 class S1
sodium subgroup sodium subgroup
class S2 class s 2
26 0 CLASSIFICATION OF OILFIELD WATERS
TABLE 8.11
Coefficients characterizing the genetic types of waters

~~
Type of water Na+/Cl- (Na’ - Cl-)/S04-2 (Cl-- Na+)/Mg+’
Chloride-calcium <1 <O >1

Chloride-magnesium <1 <O <1
Bicarbonate-sodium >1 >1 <O
Sulfate-sodium >1 <1 <O
related t o water types in this order: chloride-calcium > bicarbonate-sodium

> chloride-magnesium > sulfate-sodium. Most oilfield waters of the
chloride-calcium type belong to the S1 S2 A2 class with a few in the S2 S,
A2 class, while most oilfield bicarbonate-sodium waters belong to the S, A,
A2 and A , S1 A 2 classes.
Other significant indicators were grouped by Sulin; however, none of
them can assure the existence of a hydrocarbon deposit, and certainly they
cannot provide definite evidence of the size of the accumulation. The groups
are as follows:
Group I: direct hydrocarbon indicators; for example, naphthenic acid salts
and iodide. The naphthenic acids are more soluble in bicarbonate-sodium
type waters and are related t o the composition of the hydrocarbon accumu-
lation. Iodide is related to oil because it must have an organic origin. Sulin
also noted the dissolved gases in the waters and considered the heavier
hydrocarbons such as ethane and butane and the absence of oxygen as direct
indicators.
Group 11: highly mineralized chloride-calcium or bicarbonate-sodium
types of water containing reduced forms of sulfur are important indirect
indicators of oil. The sulfate content should be low to indicate interaction
with bituminous constituents and/or sulfate-reducing bacteria.
Group 111: in this group are constituents which have no genetic relation-
ship to hydrocarbons but appear characteristic of waters that are related t o
hydrocarbon accumulations. The constituents are bromide, boron, barium,
strontium, radium, and possibly fluoride.
Modification of Sulin’s system by Bojarski
Bojarski (1970) studied 400 water analyses and differentiated hydro-

chemical zones within basins in Poland that appear suitable for preservation
of hydrocarbon deposits. He distinguished the waters as follows:
(1) Waters of the bicarbonate-sodium type. Such waters occur in the
upper zone of a sedimentation basin, with “intense water exchange” (that is,
a hydrodynamic situation where the waters are moving at a relatively fast
geological rate), which promotes unfavorable conditions for the preservation
of petroleum and natural gas deposits. The waters are defined by the ratio
MODIFICATION OF SULIN’S SYSTEM 26 1
(Na-Cl)/SO, > l . As Sulin (1946) noted, if the ratio Na/Cl in epm is

greater than 1, the water contains more sodium than chloride and the excess
sodium can react with sulfate or bicarbonate ions. Therefore, such waters
belong to the bicarbonate-sodium or sulfate-sodium types. If the ratio
(Na - C1)/S04 is greater than 1,it indicates an excess of sodium with respect
t o both chloride and sulfate.
(2) Waters of the sulfate-sodium type with (Na - Cl)/S04 < 1. This ratio,
if less than 1, indicates that all of the sodium will react with chloride or
sulfate.
(3) Waters of the chloride-magnesium type with (Cl- Na)/Mg < 1. A ratio
of this type indicates that all of the chloride will react with sodium and
magnesiun. Such a water is characteristic of the transition zone between a
hydrodynamic area which is becoming more hydrostatic in the deeper part
of the basin, and the amount of dissolved bromide increases directly with the
(Cl- Na)/Mg ratio.
(4) Waters of the chloride-calcium type with (Cl- Na)/Mg > 1. This ratio
indicates an excess of chloride with respect t o sodium and magnesium, and
the excess will react with calcium. This type of water occurs in deeper zones
which are isolated from the influence of infiltration waters and are hydro-
static or almost hydrostatic.
Bojarski observed a large variation in the chemical composition in the
chloride-calcium type of water and subdivided this type as follows:
(a) The first class, chloride-calcium I with Na/Cl > 0.85 characterizes an
active hydrodynamic zone with considerable water movement. It is con-
sidered a zone of little prospect for the preservation of hydrocarbon
deposits.
(b) The second class, chloride-calcium I1 with Na/C1 = 0.85-0.75, charac-
terizes the transition zone between an active hydrodynamic zone and a more
stable hydrostatic zone of the sedimentation basin, which is generally con-
sidered a poor zone for hydrocarbon preservation.
(c) The third class, chloride-calcium I11 with Na/Cl = 0.75-0.65 (0.60),
characterizes favorable conditions for the preservation of hydrocarbon
deposits.It is designated as a fairly favorable environment for the preserva-
tion of hydrocarbons.
(d) The fourth class, chloride-calcium IV with Na/C1 = 0 . 6 5 4 . 5 0 , is
characterized by complete isolation of the hydrocarbon accumulations as
well as by the presence of residual waters. I t is considered a good zone for
the preservation of hydrocarbons.
(e) The fifth class, chloride-calcium V with Na/C1 < 0.50, is characterized
by the presence of ancient residual sea water which has been highly altered
since original deposition, both in the concentration of dissolved solids and in
the ratios of the dissolved constituents. Bojarski considers a zone of this type
to be one of the most likely areas where hydrocarbons are accumulated.
Additional characteristics of water associated with hydrocarbon accumula-
tions are as follows: (1) iodide > 1mg/l; (2)bromide > 300 mg/l (increasing
iodide and bromide concenbrations may point to a bitumen accumulation);

(3) ratio Cl/Br < 350; and (4) SO4 x lOO/Cl< 1.
In addition t o indicating the degree of alteration, bromide and iodide as
biophile constituents play a decisive role in the classification Bojarski
adopted. This followed because of the increased concentration of biophile
elements in the waters accompanying a petroleum deposit. The concen-
tration of iodide in the ground waters depends mainly on the organic sub-
stances, whereas the’concentration of bromide up t o a certain limit takes
place in an inorganic medium, but an increase in bromide must be evaluated
as a positive indication. In many waters accompanying petroleum deposits,
large amounts of bromide and smaller amounts of iodide were detected, or
vice versa. This probably is related t o the type of bituminous substances
which absorb the individual biophile elements in different amounts.
Chebotarev’s classification
Chebotarev (1955),an Australian geochemist, classifies waters on the basis

of dissolved bicarbonate, sulfate, and chloride, and he does not consider the
acid waters or those that contain free sulfuric or hydrochloric acid. His
fundamental assumption is that the anions are independent variables while
the cations are dependent.
The geochemical types of waters are related t o the products of weath-
ering. Table 8.111 illustrates the cycles and products that are produced by
weathering. During the first cycle the igneous rocks are weathered allowing
chloride, sulfate, calcium, sodium, silica, and magnesium to go into solution.
The second cycle is the weathering of sedimentary rocks with the solution of
more of the same products. The third cycle is the weathering of recent drift
and yields of the above constituents plus aluminum and iron.
Table 8.IV illustrates Chebotarev’s (1955) geochemical classification of
subsurface waters. The phase of weathering corresponds t o four phases of
the solution and redistribution of the chemical constituents in the earth’s
crust and correlates with their relative mobilities. He plotted the relative
mobilities of nine chemical constituents using the mobility of chloride as
100%. From this four phases were obtained, namely: (1)chloride and sulfate
100% t o about 58% mobility; (2) calcium, sodium, magnesium and potas-
sium 3% t o about 1.2% mobility; (3) silica about 0.20% mobility; and (4)
iron oxide and aluminum oxide, less than 0.05% mobility. The four phases
of weathering correspond to the products of weathering shown in Table 8.IV
and also to the cycles and products of weathering in Table 8.111. For
example the fourth phase in Table 8.IV corresponds with the first cycle in
Table 8.111.
The genetic types of water shown in the upper portion (A) of Table 8.IV
do not correspond directly with the weathering phases since the genetic
types overlap the phases. These genetic types are related t o the accumulation
products shown in Table 8.111. In the lower portion (B) of Table 8.IV are the
TABLE 8.111
Cycles and types of products of weathering (after Chebotarev, 1955)

b
rn
Cycles of
weathering:
First cycle (orthoeluvium)
from igneous and highly metamorphosed rocks
Second cycle (paraeluvium)
from sedimentary rocks
Third cycle (Neoluvium)
from Recent Drift
zrrl
is
residual products accumulative products residual products accumulative products residual products accumulative products b
=iJ
Types of (1) coarse detrital (1) chloride-sulphate (1) detrital (1) chloride-sulphate (1) solonez and (1) chloride-sulphate 0
products (chiefly alluvial) gypsum bearing z
of weathering (2) calcareous (2) calcareous (2) siallitic (2) calcareous (2) leached supra- (2) CaC03
(under vegetation (chiefly colluvial (supra-calcareous chloride-sulphate calcareous
cover) and proluvial) (siallitic)
(3)siallitic (3)siallitic (3) allitic (?) (3) unsaturated siallitic*2 (3) unsaturated siallitic (3) alumino- and ferri-
calcareous chloride- siallitic*’ siliceous system
sulphate
(4) a l l i t i ~ * ~
(lateritic crust of
weathering)
*’ A large quantity of silica and much of its calcium and sodium compounds are removed; the aluminosilicates pass gradually into residual aluminosilicic acids,
i.e., acids of the kaolin type.
*’ The action of carbonated atmospheric water is insufficient to replace the absorbed ions by hydrogen.
*3 The accumulation of sesquioxides at the expense of the leaching out of the alkalis, alkaline earth, and silica.
TABLE 8.IV
Geochemical classification of subsurface waters (after Chebotarev, 1955)
( A ) Relationship of the products of weathering to the genetic types of water
Presumable phase of weathering Products of weathering Genetic types of water
Fourth phase residual (orthoeluvium and detrital paraeluvium) bicarbonate (alkaline)

bicarbonate-chloride (alkaline-saline)
Third and partly second phases siallitic drift
chloride-bicarbonate (saline-alkaline)
Second and partly first phases calcareous accumulation
chloride-sulfate (saline)
First phase chloride-sulphate accumulation chloride (saline)
(B) Geochemical groups of Waters
Major group Class Genetic types of Reacting value in percent

of water water
H C 0 3 - + COB- Cl- SO4-’ Cl- + SO4-’ H C 0 , - + CI- HC03- + SOq-‘
Bicarbonate I bicarbonate >40 <lo - Na+ + K+ prevail in all types of

waters
C3+ and Mg” less than 2.5 in the
I1 bicarbonate-chloride 40-30 10-20 water of high saline facies and less
than 19.0 in low saline facies
111 chloride-bicarbonate 30-15 20-35
Sulphate IV sulphate-chloride 15-5 <25 >25 Na+ + K+ prevail in all types of

waters
sulphate - - >40 Ca+’ less than 4.5 in all waters
Mg” less than 4.0 in all waters
Chloride 111 chloride-bicarbonate 30-15 >20 - - Na+ + K+ prevail in all types of

waters
IV chloride-sulphate 15- 5 >20 - <2 5 Ca” less than 12.5 in the water
of high saline facies
V ch 1or ide < 5 >40 - <10 Mg+’ less than 6.0 in all types of
waters
CHEBOTAREV’S CLASSIFICATION 26 5
geochemical groups of waters. The three major groups of waters are divided
into genetic types, which are determined from the absolute concentrations
of the dissolved constituents expressed in reacting values in percent. The
bicarbonate group contains three genetic types of water, namely: (1)bicar-
bonate; (2) bicarbonate-chloride; and (3)chloride-bicarbonate. The amounts
of bicarbonate plus carbonate and chloride plus sulfate determines the
genetic type. The sulfate group is subdivided into two genetic types: (1)
sulfate-chloride; and (2) sulfate. The chloride group is divided into three
genetic types: (1)chloride-bicarbonate; (2) chloride-sulfate; and (3) chloride.
Chebotarev relates the genetic types to the products of weathering because
he believes that although the concentration of dissolved solids in subsurface
waters may vary substantially, the types of soluble salts remain largely
unchangeable .
The water classes are related to the products of weathering, rainfall, and
drainage conditions. Class I corresponds t o soluble products from the weath-
ering of orthoeluvium or igneous and highly metamorphosed rocks and their
silicate compounds. Class I1 waters are related t o products of weathering
from the same silicates and calcareous accumulations. Class I11 waters
primarily are related to weathered products from calcareous accumulations.
Class IV waters are related to weathering of alluvial, detrital, and gypsum
deposits. Class V waters are related to marine deposits plus weathering of the
products that derived the Class IV waters.
Table 8.IV shows the approximate reacting values in percent for waters
found in some oilfields (Chebotarev, 1955). For example, in such waters in
the bicarbonate group, the major cations are sodium plus potassium with the
reacting value percentage for calcium and magnesium less than 2.5 if the
saline facies are high. (The divisions between his saline facies are shown in
Table 8.V.)
Table 8.V shows the relationships of hydrodynamic zones t o the geo-
chemistry of the water and the geological environment. The zones are: (1)
recharge with active water exchange; (2) pressure with delayed water
exchange; and (3)accumulation with stagnant conditions.
The equilibrium of the chemical systems (those typical of the major geo-
chemical group of waters) is a criterion called the coefficient of water
exchange ( K e ) and is computed as:
Na(K)HC03 + (Ca,M&)(HC03)2
Ke =
Na(K)Cl + (Ca,Mg)C12 + Naz SO4 + (Ca,Mg)S04
The absolute and relative coefficients of water exchange for the three major
groups of waters are as follows:
Ke (absolute) Ke (relative)
Bicarbonate waters 1.55 96.9
Sulfate waters 0.11 6.9
Chloride waters 0.016 1.o
TABLE 8.V
Relationships of hydrodynamic zones to the geochemistry of water and the geological evironment (after Chebotarev, 1955)
Hydrodynamic zone Geochemistry of water Geological environment
recharge- water- class hydro-chemical approximate common structures relation to water depth examples
discharge exchange facies salinity terms for (ft)
cycle (ppm) water
Zone of active I and I1 low saline facies 180-2,400 fresh different intensive flush usually less everywhere
recharge exchange (some than 500
times
111)
transitional 2,400--11,400 brackish deep portions of delayed flush sometimes Great Artesian Basin,
(typical) facies structures with 5,000-7,000 Rocky Mountain oil-
peculiar geo- field and others
chemical en-
vironment
high saline facies 11,400-37.800 saline hampered flush
Zone of delayed 111 and low saline facies 400-2,500 fresh different inadequate flush usually less everywhere
pressure exchange IV than 1,000
trazs. Lional 2.500-7.400 brackish deeper portions circulation and sometimes precaucasian Basin,
(typical) facks of structures, drainage limited 3.000-4,000 South Dakota Basin,
folded zones some oilfield areas
high saline facies 7,400--19,300 saline
Zone of stagnant V low saline facies 1.500-20,000 fresh and different salt accumulation different chiefly in arid regions;
accumulation condition saline prevails upon deeper portions of many
leaching artesian basins: some
oilfields
transitional 20.000-90,000 saline and deeper portions sometimes

(typical) facies brines of structures, 8,000-1 3,000
highly folded
zones
high saline facies 90,000-300,000 brines water exchange many oilfield areas
manifests on geo- (Louisiana, Alberta, etc.)
loeical scale time
SCHOELLER’S SYSTEM 26 7
The absolute and relative coefficients were derived from assumed chemical
compositions of typical waters of the major geochemical groups. The abso-
lute value can be not lower for the group, but it can be relatively higher.
Several thousand analyses of waters associated with oil pools in the world
were used to formulate the typical waters. As the water-exchange conditions
deteriorate, the type of water changes, and the changes are related to altera-
tion or diagenesis of the waters (metamorphism). The data in Table 8.V give
hydrochemical facies, common names of various waters as related to
dissolved solids concentrations, geological structures, flush or water circula-
tion, depths, and examples where some of the water types are found.
The highly concentrated chloride waters are primarily associated with oil
occurrences; however, this is not always true. A prime determinant of the
chemical composition of oilfield waters is the hydrodynamic situation and
the type of trap. For example, an intensively flushed zone will contain a
different type of water from a zone with limited circulation. The type of
basin strongly influences the type of water that.is likely t o be found. For
example, an open basin probably will contain artesian waters of the bicar-
bonate group, a partly closed basin may contain artesian or subartesian
waters of the bicarbonate or sulfate groups, while a closed basin is more
likely t o contain bicarbonate waters on the flanks of the basin with sulfate
and chloride waters in the deeper areas.
Chebotarev (1955) applied his classification t o 917 subsurface waters in
oilfields in the world. The classification indicated that 73.7% of the waters
were of the chloride genetic type, 23.0% of the bicarbonate type, and 3.3%
of the sulfate type. Most of the sulfate and bicarbonate types were found in
the Rocky Mountain areas of the United States and probably were mixtures
containing infiltrating meteoric water.
Schoeller’s system
A French geochemist, Schoeller (1955),classified waters on the basis of

their dissolved constituents and in the following order of importance: (1)
chloride; (2) sulfate; (3) bicarbonate plus carbonate; (4) index of base
exchange (IBE); and (5) relationships of anions t o cations. This system
separates waters into six primary types based upon their amounts of dis-
solved chloride and four subgroups based on their concentrations of sulfate.
The amounts of bicarbonate and carbonate ions give additional differentia-
tion, and an index of base exchange indicates exchange of ions in the waters
with ions in associated clays. Table 8.VI outlines Schoeller’s classification.
As shown in Table 8.VI the chloride concentration separates waters into
six types. Waters from several oil-producing regions were classified and the
sequence C1> SO4 > HC03 was found to occur in very high chloride waters
and in sea waters, especially when they are saturated with CaS04. If the
waters are not saturated in CaS04, the sequence Cl- > HC0,- > S04-2 is
predominant, and in low-chloride waters the predominant sequence is
H C O ~ -> C T > S04-2.
26 8 CLASSIFICATION O F OILFIELD WATERS
TABLE 8.VI
Schoeller’s scheme for classifying petroleum reservoir waters*
Chloride concentration as Ci-
Very high if > 700

Marine if 420 - 700
High if 140 - 420
Average if 40 - 140
Low if 10 - 40
N o r m a l i f < 10
Sulfate concentration as s04-’
Very high if >

58
High if 24 - 58
Average if 6 24 -
Normalif< 6
Near saturation when J(S04-’) (Ca+’) > 70
Bicarbonate plus carbonate concentmtion as H C 0 3 - + c03-’
High if 7>
Normal if 2 7-
Low if 2<
However, he recommends using q ( H C 0 3 - + CO3-’ )‘(Ca+’ ) rather
than HC03- + C03-’
Index o f base exchange (IBE)
If Cl- > Na+ then IBE = (Cl- - Na+)/Cl-

If Na+ > Cl- then IBE = ( C T - Na+)/(S04-’ + HCO3- + C03-2)
Importance o f anions and cations
c1- > SO^' > co3-’

c1- > .co~-‘> so4-’
co3-’ > c1-> so4-2
co3-’ > so4-’ > C T
Na+ > Mg” > Ca+’
Na+ > Ca+ ’ > Mg+’
* All constituents are calculated in epm.
SCHOELLER’S SYSTEM 26 9
Also, in very high chloride waters only the sequence Na+ > Ca+’ > Mg+2
is found. As the C1- decreases, the sequence Na+ > Mg+’ > Ca+’ becomes
more frequent. In very high chloride waters SO4-’ > Ca+’, but in less
concentrated waters the opposite may occur. The sequence HC03-< Ca+’
always is found in very high chloride waters. In less concentrated chloride
waters, either HCO,- < Ca+’ or HC03- > Ca+’ may be found, while in
low chloride waters HC03- > Ca+’ is predominant.
Schoeller used an arbitrary value of 70 for JSO4- x Ca+’ to indicate
that a water is saturated with CaS04. (This is not necessarily true because
some waters, depending upon their other dissolved constituents, can contain
smaller or larger amounts.) He divided waters into four additional types
depending upon their amounts of sulfate.
Saturation with CaS04 was found to occur only in very high chloride
waters. The calcium concentration always is very high - ranging from 150 to
1,100 epm - in high chloride waters which have SO4-’ > 58 epm and
usually is less than 150 epm in high chloride waters where SO4-’ < 58 epm.
All petroleum waters even if saturated in CaS04 have a low S04/C1 ratio
which is attributed t o reduction of sulfates and high concentrations of
chloride. The ratio never exceeds one except in low or normal chloride
waters.
The third subgroup contains three additional types depending upon the
amounts of bicarbonate and carbonate in the waters. The preferred formula
for this calculation is Y(HCO,- + C03-’ )’ (Ca+’ ) which is proportional to
the gaseous pressure of C 0 2 in equilibrium with CaCO, in the water. As the
Cl- increases, the tendency is for Ca+’ t o increase and HC03- to decrease;
however, because the Ca+’ increases, the product of y(HCO,- + C03-’ )’
(Ca+’ ) does not vary greatly.
As the waters move in their subsurface environment their dissolved ions
’have a tendency to exchange with those in the rocks. Two extreme types of
adsorption can be noted in addition to intermediate types of adsorption. The
extreme types are a physical adsorption or the Van der Waals adsorption with
weak bonding between the adsorbent and the constituent adsorbed and a
chemical adsorption with strong valence bonds. Both of these adsorptions
can act simultaneously.
Cations can be fixed at the surface and in the interior of the associated
minerals. These fixed cations can exchange with the cations in the water.
When the exchange occurs, there is an exchange of bases. With the right
physical conditions of the adsorbent, similar exchange can occur with the
anions. Some of the formation constituents that are capable of exchange and
adsorption are argillaceous minerals, zeolites, ferric hydroxide, and certain
organic compounds.
Particle size influences rates and capacities, if the solids are clays such as
illite and kaolinite. The.rate increases with decreasing particle size. However,
if a larger mineral has a lattice, the exchange can easily occur on the plates.
The concentration of exchangeable ions in the adsorbent and in the water is

important. More exchange will occur when the solution is highly concen-
trated.
Schoeller (1955) used the formula:
1
(a-x)=K -
( a E x ) IP
to indicate the relationship that exists between the initial concentration, a ,
of the cations in milliequivalents in the unreacted water, and x , which equals
the final concentration of the cations in milliequivalents in the water after
equilibrium or reaction with the rocks. The amounts of cations exchanged
by passing from the liquid t o the rock or clay is a - x and the index of base
exchange (IBE) = (a - x ) / a . By substitution:
The IBE is used to indicate the ratio between the exchanged ions and the
same ions as they originally existed. For example, assume that in the original
water there were as many equivalents of C1- as (Na+ + K '
)
, and that when
the Na+ and K+ of the water exchanged with the alkaline earths in the rocks
alkaline exchange occurred, then:
IBE =
C1- - (Na+ + K+)
c1-
and this value is positive if the equivalents are Cl- > (Na' + K '
)
. Theoret-
ically all the halides should be included as C1- and all the alkalies as Na+ or
(Na+ + K '
)
.
However, when the alkaline earth ions in the water exchange for alkali
metal ions on the rocks then:
Cl- - (Na+ + K + )
IBE =
SO4- + HC03- + NO3-
and this value is negative if the equivalents are Cl- < (Na+ + K+).The lack
of equilibrium between the halides and the alkalies is not always a charac-
teristic of base exchange because sea water has a positive value without the
occurrence of base exchange. Negative values usually are observed for water
coming from altered crystalline rocks. Waters with an IBE equal t o or greater
than 0.129 can be true connate petroleum reservoir waters. Waters with a
negative IBE are waters of meteoric origin that have infiltrated into marine
sediments.
Comparison of petroleum-reservoir waters with other types of subsurface
waters revealed that the other waters have most of the same characteristics
a much higher SO4- concentration and a lower
or Kr. Waters that are in contact with organic matter
SCHOELLER’S SYSTEM 271
(other than petroleum), such as bitumens, lignites, and coals, resemble petro-
leum reservoir water, but the frequency of a Kr above normal is greater in
petroleum associated waters. Waters related - to magmatic reactions com-
monly possess high concentrations of HC03 .
Schoeller’s (1955) study of petroleum reservoir waters indicated that a
positive IBE is more frequent as the C1- increases. A negative IBE is more
frequent as the C1- decreases, and a negative value is predominant in low
and normal chloride waters associated with petroleum. In fact, this charac-
teristic appears specific for petroleum reservoir waters since in other subsur-
face waters a positive index occurs as frequently as a negative index.
Ancient sea water (connate water) deposited with the sediments usually
has an IBE > 0.129 and a Cl/Na > 1.17. Intruding meteoric water in sedi-
mentary marine rock has an IBE < 0.129 and Cl/Na < 1.17. Petroleum-
reservoir waters with an IBE greater than sea water 0.129 also have the
characteristics Cl/Na > 1.17, Cl/Ca < 26.8, Cl/Mg > 5.13, Mg/Ca < 5.24; a
very high value for $(HCO,-)’ (Ca+*) indicating sulfate reduction; low con-
centrations of HC03-; and frequent high concentrations of NH4 +.
Petroleum-reservoir waters containing infiltrating meteoric water mixed
with ancient sea water have an IBE less than sea water, 0.129, and the
characteristics Cl/Na < 1.17, the ratio Mg/Ca increases and approaches but
never equals 5.24, and the ratios Na/Ca and Na/Mg decrease as the dissolved
solids increase.
Gases in petroleum-reservoir waters
Schoeller (1955) noted that there should be equilibrium between the free
petroleum gases and those dissolved in the water. Considering the solubility
of the gases, those in the water should reflect the composition of the petro-
leum. Components characteristic of petroleum accumulations are ethane,
propane, butane, pentane, ethylene, and propylene. Associated components,
which may also be present in volcanic waters and in waters in contact with
other organic matter such as coal, peat, and lignite as well as in petroleum-
associated waters, are favorable components. These are methane, carbon
dioxide, organic nitrogen, hydrogen sulfide, helium, radon, and the absence
of oxygen. Other components or universal components found in all types of
waters are nitrogen and argon.
The top waters can contain gases such as Hz S, C 0 2 , and CH4 but because
they do not contact the petroleum deposit they are not similar in com-
position t o the bottom waters. The edge waters are in contact with the
petroleum and are characterized by higher amounts of HC03, sulfate reduc-
tion, and the presence of H2S , NH4, and small amounts of dissolved hydro-
carbons.
Oilfield brine analyses
About 4,000 oilfield brine analyses were classified. All of these analyses
are now in the U.S. Bureau of Mines (1965) open-file report on oilfield
brines. The data are on automatic data processing magnetic tape as well as in
a file of computer printout sheets listed by State, county, sedimentary basin,
formation, etc.
These brine data were collected by the Bureau of Mines because the value
of oilfield brine analyses in the study of various petroleum-related problems
was recognized early in the history of petroleum and natural gas. Before
1928 the Bureau of Mines had indicated in several reports (Ambrose, 1921;
Swigart and Schwarzenbek, 1921; Collom, 1922; Mills, 1925; Reistle, 1927)
ways in which the analyses could be used. In earlier years, confusion existed
because of greatly varying methods of analysis that were used. A paper
presenting the methods used by the Bureau of Mines was published by
Reistle and Lane (1928). This system of determining the characteristic con-
stituents of oilfield waters and of calculating and reporting results was
widely adopted by the petroleum industry. Later the Bureau of Mines
cooperated with several interested agencies, and a more detailed report with
more modern methods of analyzing oilfield waters was published (American
Petroleum Institute, 1968).
The Bureau of Mines has an oilfield water analysis laboratory at the
Bartlesville Energy Research Center, Bartlesville, Oklahoma. These 4,000
samples were analyzed at this laboratory.
Analysis methods
The specific gravity of each sample is determined so that a correct aliquot

size can be taken for a specific ion analysis. Chloride is determined by
titration with silver nitrate, carbonate and bicarbonate are determined by
titration with a standard acid, and a pH meter is used to determine the end
points. This alkalinity determination should be completed at the time of
sampling for accur?te results. However, most of the data for the 4,000
samples were obtained by analysis in the laboratory and were completed
within 6 t o more than 48 hours after sampling. Therefore, the alkalinity data
cannot be considered absolute but only relative.
The calcium was determined by titration of calcium oxalate with perman-
ganate until about 1957, about which time it was determined by com-
plexometric titration such as with disodium ethylenediametetraacetic acid
(EDTA) until about 1969; since then, it has been determined by atomic
absorption. Magnesium determination has a similar history. I t was precipi-
tated as the pyrophosphate until about 1957, and titrated with EDTA until
1969, from then t o now it has been determined by atomic absorption.
Sulfate was determined by precipitation as barium sulfate, and this meth-
od still is used. Sodium was determined by calculation from the difference
OILFIELD BRINE ANALYSES 273
between the reacting values of the assumed total anions and cations until
about 1960, after which time it was determined by flame photometry or
atomic absorption.
Several other analyses for dissolved constituents in oilfield brines are now
made by the Bureau of Mines. For example, potassium, lithium, rubidium,
and cesium are determined by atomic absorption or flame photometry;
strontium and barium by atomic absorption; manganese, iron, and boron by
atomic absorption or titrimetric methods; and bromide and iodide by
titrimetric methods.
The precision of the methods is as follows: alkalinity, 2-3% of the
amount present; sodium, 2--5% of the amount present; calcium and magne-
sium, 4--5% of the amount present; sulfate, 1-2% of the amount present;
chloride, 1% of the amount present. If sodium is calculated, the precision
value reflects the sum of the precision data for the data from which it is
calculated plus the undetermined dissolved constituents. The significant
figures for the analytical data are all the certain digits and only the first
doubtful digit. This number usually is limited to three significant figures
except for specific gravity, where four or five are common.
It often is recognized that the sampling method is as important as the
analytical method. This certainly is true of oilfield brines.
Field sampling methods
Most of the 4,000 samples were obtained only from wells where reason-
able assurance was evident that the formation brine was not contaminated
by drilling fluids or by intrusion of water from other formations. Wells were
selected on the basis of age, type of completion, and production of fluids.
Samples were not taken from some gasfields because of the likelihood of
dilution by water condensed from vapor carried up the hole with the gas.
Some samples were taken from gas-condensate wells that produced large
volumes of brine.
In many cases the electrical resistivity measurement was made on the
sample at the time the sample was taken, and resulting values were compared
with measurements from other samples from the same formation within that
field or nearby fields. Obvious discrepancies were eliminated by sampling
additional wells.
Nearly all samples were withdrawn at production wellheads, and the water
was separated from the oil in portable separators. A few samples were taken
of brines from formations that did not produce enough water t o permit
taking samples at individual wellheads. Such samples were taken from gun-
barrels or oil-water separators.
Samples were taken in clean, 1-gallon glass jugs that were first rinsed
several times with the water sampled and then filled, capped, and labeled. In
a few instances samples were obtained by the producer from comparatively
isolated small pools or fields and shipped t o the laboratory in 1-gallon poly-
ethylene jugs.
Application of the classification systems
Several investigators have applied the classification systems t o determine

their usefulness in studies related t o exploration and production of petro-
leum. Ostroff (1967) concluded that Palmer’s system is less useful than the
systems of Sulin and Schoeller. Further he concluded that the Sulin system
is more applicable to petroleum formation waters because many such waters
contain more than 2,000 epm of dissolved solids, and the Schoeller system
tends to lump these highly concentrated waters together. The index of base
exchange (IBE) in Schoeller’s system, however, appears to have merit for
certain interpretations as does theJ Ca+* x SO4-*.
Dickey (1966) concluded that the Palmer system does not correlate very
well with the geology of oil reservoirs, that Schoeller’s nomograph is useful,
but that the Sulin system appears t o conform better with geology. He also
noted that relating water types to geology and flow patterns should be useful
in exploration and that distinguishing between a stagnant water and artesian
related waters should be highly significant in a new oilfield development
area.
The Sulin system appears t o be more generally applicable than the other
systems in studies of waters from petroleum reservoirs. Because of this and
because the Schoeller system appears t o have merit in studying certain types
of oilfield waters, it was decided t o apply portions of the Sulin system, the
Schoeller system, and Bojarski’s modification of the Sulin system t o a study
of brines from various sedimentary basins of the United States that are
known to be related to petroleum and natural gas.
Calculations
The analyses of most oilfield waters are reported in units of milligrams per
liter (mg/l). The conversion of mg/l to equivalents per million (epm) is done
using the following formula:
mg/l of ion
= epm ion
atomic weight of ion
specific gravity of brine x
valence of ion
Table 8.VII provides formulas for calculating the epm for many of the
common constituents found in oilfield brines. If the constituent is reported
in parts per million (ppm), it is not necessary t o divide by the specific gravity
of the brine.
The sum of the epm’s (Z epm) shown in Table 8.VII are converted to
Zr/100 g of water for the Sulin calculations by moving the decimal to the
left in the Z epm and calling this Zr. The term s is used t o indicate the
percent of equivalent of a given constituent. The percentage equivalent (s) of
each ion is determined by dividing the equivalents per 100 g of water by the
total equivalent in 100 g of water. For example, if the r for sodium equals
APPLICATION OF THE CLASSIFICATION SYSTEMS 27 5
TABLE 8.VII
Formulas for converting milligrams per liter to equivalents per million for constituents
commonly found in oilfield waters
Lithium mg/l Li+ x 0.1442/sp. gr.* = epm Li+

Potassium mg/l K+ x O.O256/sp. gr. = epm K+
Sodium mg/l Na+ x O.O435/sp. gr. = epm Na+
Magnesium mg/l Mg+’ x O.O823/sp. gr. = epm Mg+’
Calcium mg/l Ca+’ x O.O499/sp. gr. = epm Ca+’
Strontium mg/l Sr+’ x O.O228/sp. gr. = epm Sr+’
Barium mg/l Ba+’ x O.O146/sp. gr. = epm Ba+’
Carbonate mg/l co3-’ x O.O333/sp. gr. = epm c03-’
Bicarbonate mg/l HC03- x O.O164/sp. gr. = epm HC03-
Sulfate mg/l SO^+ x 0.0208/sp.gr. = eprnS04-’
Chloride mg/l C1- x O.O282/sp. gr. = epm C1-
Bromide mg/l Br- x O.O125/sp. gr. = epm Br-
Iodide mg/l I- x O.O079/sp. gr. = epm I-
Zepm =
* Specific gravity.
200.6 and the Zr for the total equivalents equals 518.8, then
200.6/518.8 x 100 = 38.7, or 38.7 percentage equivalents for sodium.
The Sulin classification considers only the macro constituents; if ions such
as potassium, lithium, strontium, barium, bromide, and iodide are deter-
mined, they should be added t o their associated macro constituents t o be
properly considered in the analysis report. For example, when sodium,
potassium, and lithium are determined by atomic absorption the total mg/l
of each is reported. Therefore, the epm Na + epm K + epm Li should be
added to obtain the correct r value for sodium:
epm Na + epm K + epm Li
rNa=
10 10
epmCa + e p m S r + epmBa
rCa=
10 10
epm C1 epm Br epm I

rC1 =
+
10 10
The r values then are divided by the Zr and multipIied by 100 to obtain
the s value or percentage equivalents. The s values are used to determine the
type, class, and other Sulin values of the water as illustrated in Table 8.1,
where a = sNa, b = sCa + sMg, and d = sC1+ s S 0 , . The epm values are used
to determine the Schoeller characteristics such as the degree of chloridiza-
tion, the degree of sulfation, IBE, etc., illustrated in Table 8.VI.
27 6 CLASSIFICATION O F OILFIELD WATERS
TABLE 8.VIII
Classification of some oilfield waters from 1 0 formations in eight sedimentary basins
No. State Formation Basin Depth Concentration (e pm)
1 Kans. Arbuckle Cent .Kans. 1,050 634.6 60.6 133.0 7.7 49.0 773.8
3 Kans. Arbuckle Cent.Kans. 1,023 282.0 24.8 40.2 9.5 22.1 313.9
5 Kans. Arbuckle Cent.Kans. 992 430.3 38.8 65.4 8.5 17.5 510.9
6 Kans. Arbuckle Cant.Kans. 1,152 458.7 34.4 81.3 3.4 45.9 523.2
8 Kans. Arb u ck le Cent.Kans. 949 356.4 57.4 100.4 12.8 39.9 456.7
11 Kans. Lansing Cent. Kans. 928 1,759.6 266.6 373.9 1.8 0.0 2,398.9
12 Kans. Lansing Cent.Kans. 1,075 570.4 117.4 174.6 2.4 34.9 826.6
13 Kans. Lansing Cent.Kans. 966 1,899.9 266.4 442.1 0.3 0.8 2,605.0
14 Kans. Lansing Cent .Kans. 999 1,728.2 232.0 512.4 1.7 2.5 2,469.4
15 Kans. Lansing Cent. Kans. 902 1,903.8 274.5 478.8 1.1 0.0 2,655.6
16 Kans. Lansing Cent. Kans. 1,009 2,514.8 309.3 532.7 0.4 1.5 3,353.6
17 Kans. Lansing Cent .Kans. 1,063 1,854.7 225.2 449.8 0.5 1.1 2,529.6
18 Kans. Lansing Cent.Kans. 1,148 2,087.2 223.6 384.8 1.4 1.o 2,706.1
19 Kans. Lansing Cent.Kans. 1,172 2,414.5 251.0 519.0 0.7 2.4 3,17 9.9
20 Kans. Lansing Cent .Kans. 853 1,898.4 210.8 416.7 0.8 0.7 2,515.5
21 Okla. Wilcox Cherokee 1,228 2,366.5 193.3 810.1 1.1 7.7 3,360.6
23 Okla. Wilcox Cherokee 904 2,161.5 163.2 530.3 0.4 7.3 2,847.4
27 Okla. Wilcox Cherokee 582 1,587.2 151.2 287.5 1.8 1.4 2,021.6
29 Okla. Wilcox Cherokee 1,97 2 2,701.4 153.1 606.0 0.7 8.4 3,449.1
31 Ark. Nacatoch E.Texas 360 233.4 9.0 22.4 3.7 0.0 261.1
34 Texas Nacatoch E.Texas 905 788.2 9.4 28.7 3.1 0.0 824.6
35 Texas Nacatoch E.Texas 70 1 481.8 7.8 17.3 3.3 0.0 503.5
40 Texas Nacatoch E.Texas 181 490.6 3.6 7.8 2.8 0.6 498,4
>
* Chloride (epm): (VH)C = 700, (MC) = 420-700, (H)C = 140-420, (A)C = 40-140,
(L)C = 10-40, (N)C = <
10. Sulfate (epm): (VH) = > 58, ( H )
=: 24-58, (A ) = 6-24, (N )
=<6.
APPLICATION OF THE CLASSIFICATION SYSTEMS 277
Sulin Schoeller*
Zepm type class CT so4-’

1,659.0 Cl-Ca 80.7 20.0 0.18
1,523.6 Cl-Ca 52.6 17.4 0.21
692.6 Cl-Ca 29.8 15.4 0.10
1,585.4 C l - C a 77.8 8.7 0.23
1,071.6 C1-Ca 33.8 16.8 0.16
1,147.2 Cl-Ca 61.1 9.9 0.12
1,492.0 CI-Ca 65.6 14.6 0.19
1,024.0 Cl-Ca 62.3 25.2 0.22
1,105.0 CI-Ca 46.2 17.9 0.1 3
1,506.4 C1-Ca 71.4 52.5 0.16
4,800.9 CI-Ca 0.0 10.7 0.27
1,726.6 Cl-Ca 78.0 10.2 0.31
5,214.7 Cl-Ca 19.4 3.5 0.27
4,946.5 Cl-Ca 36.0 11.6 0.30
5,314.0 CI-Ca 0.0 8.5 0.28
6,712.5 Cl-Ca 28.7 4.4 0.25
5,061.1 Cl-Ca 22.5 5.1 0.27
5,404.4 Cl-Ca 20.4 9.1 0.23
6,367.7 Cl-Ca 35.9 6.3 0.24
5,043.1 Cl-Ca 17.8 6.6 0.25
6,739.4 Cl-Ca 79.1 9.9 0.30
5,854.9 Cl-Ca 72.3 3.6 0.25
5,710.4 Cl-Ca 62.3 4.7 0.24
6,021.0 Cl-Ca 84.3 8.5 0.21
5,188.0 C l - C a 59.9 7.1 0.23
5,330.0 Cl-Ca 82.7 6.4 0.25
4,051.1 Cl-Ca 20.5 9.9 0.21
6,291.3 C l - C a 55.7 8.7 0.22
6,918.8 Cl-Ca 71.4 7.1 0.22
6,892.3 Cl-Ca 73.6 9.4 0.23
529.8 CI-Ca 0.6 6.8 0.11
1,123.5 C l - C a 0.0 4.9 0.17
726.2 Cl-Ca 6.0 8.8 0.16
1,654.2 C k C a 0.0 6.5 0.04
1.013.8 Cl-Ca 0.0 5.7 0.04
594.5 Cl-Ca 3.6 4.6 0.07
1,037.4 Cl-Mg 1.6 18.1 0.01
635.2 C l - C a 3.6 4.4 0.06
1.027.8 Cl-Ca 4.5 13.2 0.1 1
1,004.0 C l - C a 2.2 4.0 0.02
TABLE 8.VIII (continued)
No. State Formation Basin Depth Concentration (epm)

(m)
Na+ Mg+’ Ca+’ HCOJ- S04-* Cl-
41 Ark. Paluxy E.Texas 1,115 1,246.8 90.6 254.1 0.4 0.0 1,594.4
42 Ark. Paluxy E.Texas 737 586.0 32.2 86.7 2.8 3.3 699.0
43 Ark. Paluxy E.Texas 1,297 1,310.3 107.1 278.5 2.7 .7 1,692.0
44 Ark. Paluxy E.Texas 884 934.9 64.8 197.6 1.5 1.1 1,194.9
45 Ark. Paluxy E.Texas 1,417 1,507.2 138.6 504.4 0.5 3.3 2,208.7
46 Texas Paluxy E.Texas 2,340 1,642.1 22.5 378.5 0.0 9.3 2,033.8
49 Texas Paluxy E.Texas 1,512 205.3 6.4 10.0 14.9 6.0 202.4
51 Ark. Rodessa E.Texas 1,897 1,971.8 157.2 669.7 0.0 8.0 2,789.9
54 Ark. Rodessa E.Texas 711 538.6 35.9 75.0 3.0 2.9 643.4
55 Texas Rodessa E.Texas 2,844 1,772.1 127.3 878.8 1.5 2.8 2,773.2
56 Texas Rodessa E.Texas 2,519 1,861.9 217.4 1,084.7 1.1 2.7 3,165.1
61 Texas Woodbine E.Texas 1,047 1,507.0 44.3 161.4 4.2 3.5 1,705.0
63 Texas Woodbine E.Texas 898 341.8 4.4 9.3 11.3 0.6 343.5
64 Texas Woodbine E.Texas 841 1,060.5 32.4 58.9 0.0 0.2 1,161.4
66 Texas Woodbine E.Texas 1,144 809.1 12.7 61.1 7.3 3.7 873.3
67 Texas Woodbine E.Texas 925 478.8 12.4 23.6 7.0 0.1 507.7
70 Texas Woodbrine E.Texas 1,332 1,268.0 30.2 81.9 8.3 4.0 1,367.4
71 Ala. Eutaw Interior Sal. 1,061 1,124.4 45.0 164.7 2.7 0.0 1,330.1
72 Ala. Eutaw Interior Sal. 972 1,102.9 25.6 160.5 2.0 0.0 1,288.9
73 Ala. Eutaw Interior Sal. 1,060 1,186.7 59.6 164.3 2.9 0.0 1,411.8
74 Miss. Eutaw Interior Sal. 1,444 1,733.6 82.7 287.3 0.9 0.0 2,108.5
77 Miss. Eutaw InteriorSal. 2,443 1,721.3 256.7 458.0 0.0 0.0 2,433.4
80 Miss. Eutaw InteriorSal. 2,469 2,153.7 87.7 518.9 0.0 1.0 2,765.6
* Chloride (epm): (VH)C =->700, (MC) = 420-700, (H)C = 140-420, (A)C = 40-140,
(L)C = 10-40, (N)C = < 10. Sulfate (epm): (VH) = > 58,(H) = 24-58, (A) = 6-24, (N)
- < 6.
_-
Sulin Schoeller* d c a + ' + so4-' V(HCO~-+ coz-2 ) ? ( ~ a +IBE
~)
Zepm type class CT SO^-'

3,186.6 CI-Ca 3.1 3.8 0.22
1,410.3 CI-Ca 17.1 8.8 0.16
3,391.5 Cl-Ca 14.1 12.8 0.23
2,395.0 CI-Ca 15.0 7.6 0.22
4,363.0 CI-Ca 41.0 5.4 0.29
4,086.4 Cl-Ca 59.5 0.0 0.19
3,543.9 Cl-Ca 27.9 16.6 0.14
4,066.7 Cl-Ca 60.9 13.6 0.26
445.2 SO4-N 7.7 13.1 0.0
4,121.6 CI-Ca 61.0 11.1 0.26
5,596.8 Cl-Ca 73.3 0.5 0.29
5,384.3 Cl-Ca 81.3 8.0 0.27
2,916.3 CI-Ca 78.7 0.4 0.26
1,299.0 CI-Ca 14.8 8.8 0.16
5,556.0 Cl-Ca 50.1 12.9 0.36
6,333.1 Cl-Ca 54.4 11.1 0.41
6,235.2 Cl-Ca 66.2 7.3 0.34
5,622.2 CI-Ca 79.0 13.5 0.30
5,818.1 CI-Ca 55.9 7.7 0.31
5,163.7 CI-Ca 57.4 8.4 0.27
3,425.6 Cl-Ca 23.9 14.2 0.12
2,705.0 CI-Ca 10.0 15.6 0.06
711.0 CI-Mg 2.4 10.6 0.oo
2,313.7 Cl-Ca 4.0 0.0 0.09
2,970.7 CI-Ca 27.6 5.4 0.14
1,767.3 C1-Ca 15.1 14.8 0.07
1,029.9 Cl-Ca 1.5 10.5 0.06
3,379.5 Cl-Ca 0.0 14.8 0.14
3,261.0 Cl-Ca 3.7 7.4 0.11
2,760.0 Cl-Ca 18.1 17.8 0.07
2,667.1 Cl-Ca 0.1 10.6 0.15
2,580.1 Cl-Ca 0.1 8.8 0.14
2,825.4 CI-Ca 0.1 11.1 0.16
4,213.1 Cl-Ca 0.1 6.2 0.18
4,789.9 Cl-Ca 101.6 20.3 0.15
3,777.7 Cl-Ca 0.0 5.7 0.1 7
4,869.6 Cl-Ca 0.0 0.0 0.29
4,704.8 Cl-Ca 0.0 17.8 0.18
3,789.1 Cl-Ca 0.0 11.0 0.17
5,527.1 Cl-Ca 22.7 0.0 0.22
TABLE 8.VIII (continued)
No. State Formation Basin Depth Concentration

(m)
Na+ Mg+2 Ca+2 c1-
-___
81 Miss. Wilcox Miocene 1,975 2,031.8 36.8 100.3 4.5 0.0 2,163.3
84 Miss, Wilcox Miocene 1,871 2,157.3 54.1 89.9 6.8 0.0 2,114.0
91 La. Smackover N .Louisiana 3,109 1,589.7 86.7 1,256.8 0.0 8.8 2,940.8
92 Ark. Smackover N.Louisiana 2,103 2,444.9 275.9 1,392.8 2.0 4.6 4,105.9
96 Ark. Smackover N. Louisiana 2,526 2,729.0 218.6 1,598.2 2.4 1.7 4,540.4
97 Ark. Smackover N.Louisiana 2,615 4,277.5 10.0 34.9 2.4 0.0 4,312.9
98 La. Smackover N.Louisiana 2,949 2,225.4 149.7 2,282.1 0.0 1.4 4,654.9
99 La. Smackover N.Louisiana 3,271 1,971.2 173.5 1,668.9 0.0 1.8 3,810.9
100 La. Smackover N.Louisiana 701 681.8 35.2 63.6 5.1 2.1 778.6
101 La. Wilcox W. Gulf 1,399 1,718.5 44.9 90.1 1.4 0.2 1,851.2
102 La. Wilcox W. Gulf 1,814 2,035.8 46.5 124.7 5.6 0.8 2,199.9
103 La. Wilcox W. Gulf 747 1,206.3 43.6 47.2 4.6 0.6 1,291.2
204 La. Wilcox W. Gulf 1,561 1,874.2 40.9 91.0 0.7 0.8 2,003.8
105 La. Wilcox W. Gulf 1,722 2,138.9 41.9 96.2 1.4 1.4 2,273.4
106 La. Wilcox W. Gulf 666 1,057.5 26.6 51.0 0.0 0.0 1,058.4
107 La. Wilcox W. Gulf 1,124 1,379.4 41.8 60.0 2.5 0.6 1,477.6
108 La. Wilcox W. Gulf 2,441 2,119.4 43.7 100.2 6.1 0.0 2,263.4
109 La. Wilcox W. Gulf 2,158 2,132.3 15.3 115.2 4.0 8.0 2,256.4
110 La. Wilcox W. Gulf 471 840.1 26.8 24.1 7.0 0.0 888.4
111 N.D. Silurian Williston 3,633 3,431.2 100.6 993.7 1.5 8.4 4,305.3
>
* Chloride (epm): (VH)C = 700, (MC) = 420-700, (H)C = 140-420, (A)C = 4040,
>
(L)C = 10-40, (N)C = < 10. Sulfate (epm): (VH) = 58, (H) = 24-58, (A) = 6-24, (N)
=<6.
APPLICATION OF T H E CLASSIFICATION SYSTEMS 281
Cepm type class ~ 1 - SO^-^

~
4,336.9 Cl-Ca 0.0 12.7 0.06

3,981.1 Cl-Ca Q.0 11.2 0.07
2,920.8 Cl-Ca 0.0 11.8 0.06
4,422.4 HCO, 0.0 16.1 0.0
-Na
4,267.3 Cl-Ca 26.2 11.4 0.7
4,223.9 Cl-Ca 0.0 9.0 0.06
4,403.4 Cl-Ca 0.1 18.9 0.07
4,747.3 Cl-Ca 5.4 11.1 0.07
2,747.7 Cl-Ca 0.0 17.0 0.06
2,878.2 Cl-Ca 0.0 11.5 0.08
5,883.0 Cl-Ca 103.6 0.0 0.46
8,226.2 Cl-Ca 80.0 17.7 0.40
8,840.1 Cl-Ca 78.0 17.9 0.43
9,306.8 Cl-Ca 53.5 21.1 0.40
8,335.6 Cl-Ca 78.0 0.6 0.42
4,454.9 Cl-Ca 53.4 21.3 0.40
8,637.9 Cl-Ca 0.0 5.9 0.01
9,313.8 Cl-Ca 57.5 0.7 0.52
7,626.5 Cl-Ca 55.5 0.6 0.48
1,566.7 Cl-Ca 11.4 11.6 0.12
3,706.6 Cl-Ca 5.1 5.6 0.07
4,413.6 Cl-Ca 10.2 15.8 0.07
2,593.7 C1-Ca 5.6 10.1 0.07
4,011.7 Cl-Ca 8.8 3.6 0.06
4,55 3.5 Cl-Ca 11.7 5.9 0.06
2,193.6 Cl-Mg 0.0 0.0 0.0
2,962.2 Cl-Ca 6.1 7.3 0.07
4,533.4 Cl-Ca 0.0 15.2 0.06
4,531.3 Cl-Ca 29.9 12.1 0.05
1,786.7 Cl-Ca 0.0 9.9 0.05
8,840.7 Cl-Ca 90.6 13.0 0.20
Interpretation o f the classification
The majority of the 4,000 oilfield waters that were classified fell in the S ,
S2 A2 class and were of the chloride-calcium type. Table 8.VIII illustrates
the classification data for some of the samples from the Arbuckle, and
Lansing Kansas City formations in the Central Kansas Basin; the Wilcox
formation in the Cherokee Basin; the Nacatoch, Paluxy, Rodessa, and
Woodbine formations in the East Texas Basin; the Eutaw formation in the
Interior Salt Basin; the Smackover formation in the North Louisiana Basin;
and the Wilcox formation in the western Gulf Basin.
Most of the samples of the bicarbonate-sodium and sulfate-sodium types
were found at relatively shallow depths. This could indicate that the waters
contained infiltrating water.
The Cl/Na ratio was determined for each sample as suggested by Bojarski.
This ratio was calculated from the epm values, and the ratios ranged from
0.33 to values greater than 1. Values greater than 0.85 are characteristic of
hydrodynamic waters, according t o Bojarski (1970),while the more altered
waters found in static environments have ratios of less than 0.50.
Schoeller's classification uses the chloride concentration t o separate the
waters into six types. The majority of the 4,000 oilfield waters analyzed for
this study were very high chloride waters (where the chloride epm is equal to
or greater than 700). The sulfate concentration of these waters according to
his classification was not as consistent. Many of them were normal with
sulfate epm less than 6. However, in several waters the sulfate epm was
higher than 24. Few waters contained sulfate in excess of 58 epm.
The JCa+' x SO4-' exceeded 70 in some waters, indicating that such
waters were nearly saturated with calcium sulfate. Schoeller did not consider
that the solubility of calcium sulfate increases in the presence of certain
other dissolved ions; therefore the value of 70 may not always indicate
saturation. However, in primary and secondary recovery operations this
value should be considered.
The 4 ( H C 0 3 - + C03-' )' (Ca+' ) was used by Schoeller to determine if a
water was saturated with calcium carbonate, and such a water should have a
value greater than 7. This is not entirely accurate, but the formula does
indicate if the water contains an excess of calcium, which decreases the
carbonate concentration. Many of the 4,000 waters evaluated had values
greater than 7. A distilled water thus saturated would deposit precipitated
calcium carbonate, but the activities of other ions dissolved in a brine cause
the solubility product t o be different in the brine.
The predominant cation sequence for these 4,000 oilfield brines was Na+
> Ca+? > Mg+'. The SO4/Cl ratio ranged from 0.00 to 0.34.The ratio 0.34
was found for a bicarbonate-sodium water sample, which may have con-
tained infiltrating water. None of the chloride-calcium type waters had a
S04/Cl ratio greater than 0.17, with many of them 0.00.
The IBE indicates that exchange of metal ions dissolved in the water have
exchanged with metal ions on the clays. (Schoeller made the arbitrary as-
sumption that the concentrations of sodium and chloride originally present
in the water were equal.) If the IBE is a positive number the exchange was
alkali metals in the water for alkaline earth metals on the clays, and if the
IBE is negative the exchange was alkaline earth metals in the water for alkali
metals on the clays. Very few negative numbers were evident when the IBE
was determined on the 4,000 oilfield water analyses. This indicates that most
of the formation waters associated with hydrocarbons had exchanged dis-
solved alkali metals for alkaline earth metals on the clays. The few samples
that yielded the negative IBE numbers were sulfate-sodium and bicarbonate-
sodium type waters which, according to Sulin, is indicative of terrestrial
derived waters.
Ratios
The ratios Na/(Ca + Mg) and Ca/Mg were calculated from the analytical
data for the 4,000 oilfield water samples. Fig.8.1 illustrates a plot of the
Na/(Ca + Mg) ratio versus dissolved solids for samples from the East Texas
Basin. The ratio tends to decrease with increasing dissolved solids concen-
tration. The depletion of sodium with respect to calcium + magnesium can
be attributed to diagenesis of the waters. This correlates with Schoeller's
IBE, indicating that the alkali metals in the water exchange with alkaline
I KEY
Tertiary
Upper Cretaceous
6 Lower Cretaceous
0 Jurassic
x Pennsylvanian
0
DISSOLVED SOLIDS, g / l
Fig. 8.1. Plot of the Na/(Ca + Mg) ratio versus the concentration of dissolved solids in
some formation waters taken from sedimentary rocks in the East Texas Basin.
-
0
7- 0
Fig. 8.2. Plot of the Na/(Ca + Mg) ratio versus the concentration of dissolved solids in
some formation waters taken from the Rodessa formation in the East Texas Basin. The
line is a least squares fit (y = a + bx + ex2 ) for the scattered data.
earth metals on the clays, decreasing the dissolved alkali metals and in-
creasing the dissolved alkaline earth metals.
Fig.8.2 shows the scattered data for the brines taken from the Rodessa
formation in the East Texas Basin. The scattered data were submitted t o
least squares analysis to determine with more certainty how the ratios varied
with depth and with dissolved solids concentrations. The least squares
formula used was y = a + bx + cx2or the approximation to a parabola. If the
best fit is a straight line, the solution will regress to it, and this occurred for
the data shown in Fig. 8.2. Also, if this occurs, the value for c i n the formula
is very small. The curve shown in Fig. 8.2 represents a fit of about 88%of
the sum of the mean, a fit of 100%would be ideal.
Preliminary plots of the scattered data indicated that a better fit could be
obtained with the parabolic least squares approximation than with the least
squares approximation for a straight line. A computer was used t o obtain the
fit.
The least squares curve indicates that the concentration of sodium
decreased with respect t o calcium plus magnesium until the dissolved solids
were about 210,000 mg/l and then increased. This could be attributed to
exchange of sodium in the water for alkaline earths on the minerals until the
concentration of dissolved solids reached 210,000 mg/l, at which point the
solubility products of the alkaline earths are such that they are unable to
stay in solution. This would correlate with Schoeller’s IBE. However, the
IBE does not consider the effect of other ions in solution upon the solubility
product of an ion.
30 - Tertiary
0 Upper Cretaceous
Lower Cretaceous
Jurassic
x Pennsylvanian-
-=2
0
20- 0
:Is 0
10- 0
. . .n
. 0
a
n
n&
n
0
0
315 7b 105 I40 I'l5 2;O

DISSOLVED SOLIDS, g / l
245 280 3;5 z 50
Fig. 8.3. Plot of the Ca/Mg ratio versus the concentration of dissolved solids in some
formation waters taken from sedimentary rocks in the East Texas Basin.
Fig. 8.3 illustrates a plot of. the Ca/Mg ratio versus dissolved solids. Here
the trends appear t o be that the Ca/Mg ratio increases as the dissolved solids
increase. This trend also is related to diagenesis of the waters. The calcium
increases as the magnesium decreases, and this may be related to the forma-
tion of dolomite and chlorite, or to reactions with argillaceous minerals.
Fig. 8.4 is a plot of Ca/Mg versus dissolved solids. The Ca/Mg ratios for
some brines from the Rodessa formation in the East Texas Basin were sub-
mitted t o least squares analysis, and the results were used to plot Fig. 8.4.
The scattered data in Fig. 8.4 did not yield as good a fit to the formula
y = a + bx + cx2 as did the data in Fig. 8.2, because the fit is about 35%of
the mean where 100% is ideal. The line indicates that the ratio increased as
the dissolved solids increased. This is the result of magnesium lost from the
brine by reactions t o form minerals, ion exchange, or shale membrane filtra-
tion. It is not a result of solubility product because most magnesium com-
pounds are more soluble than calcium compounds.
Fig. 8.5 is a plot of Na/(Ca + Mg) against depth from which the samples
were taken. The trend does not indicate a definite relationship for these
samples.
Fig. 8.6 is a plot of the least squares analysis data for Na/(Ca + Mg) versus
depth for some brines from the Rodessa formation in the East Texas Basin.
The plot indicates that the sodium concentration decreases with respect to
calcium plus magnesium until the depth is about 2,350 m and then it in-
creases. The curve in Fig. 8.6 represents a 45% mean fit of the scattered data
where an ideal fit is 100%.
Fig. 8.7 is a plot of Ca/Mg versus depth, and there appears to be an
increase of this ratio with depth which would indicate that magnesium is
depleted more with respect t o calcium in brines taken from deeper strata.
.
7t 0 .
21i
I1 I I 1 I
20 80 I40 200 260 3 10
DISSOLVED SOLIDS, g/l
Fig. 8.4. Plot of the Ca/Mg ratio versus dissolved solids for some formation waters taken
from the Rodessa formation in the East Texas Basin. The line is a least squares fit
(y = a + bx + cx') for the scattered data.
I "
KEY
6C - o Cretaceous gulf
A Cretaceous Comanche
o Jurassic
50 -
x Pennsylvanian
Tertiary
-
-
.. 0
2c -
A
0
0 A 0 0
IC
C
500 1,000 1,500 2,000 2,500 3,000
DEPTH, meters
Fig. 8.5. Plot of the Na/Ca + Mg) ratio versus depth for some formation waters taken
from sedimentary rocks in the East Texas Basin
I I I I I
0 500 1.000 1,500 2,000 2,500 100
DEPTH, meters
Fig. 8.6. Plot of the Na/(Ca + Mg) ratio versus depth for some formation waters taken
from the Rodessa formation in the East Texas Basin. The line is a least squares fit
(y = a + bx + cx2 ) for the scattered data.
KEY
3c - o Cretaceous gulf
A Cretaceous Comanche
. A Jurassic
x Pennsylvanian
0 Tertiary 0
'F 2c
e
I: r" A
0
m
A
A A
10 - A 0
. ..
A P o
8 0
0 I
500
I
If300
.
1,500
I
2,000
I
I
2,500
**
3,( 30
DEPTH, meters
Fig. 8.7. Plot of the Ca/Mg ratio versus depth for some formation waters taken from
sedimentary rocks in the East Texas Basin.
This would indicate that magnesium is taken from the brine by reactions to
form minerals or absorbed by clays.
Fig. 8.8 is a plot of the least squares analysis data for Ca/Mg versus depth
for some brines from the Rodessa formation in the East Texas Basin. The
plot indicates that the concentration of calcium increases with depth with
I I I I I 10 L
0 500 I.000 1.500 2POO 2.500 3pOO
DEPTH, meters
Fig. 8.8. Plot of the Ca/Mg ratio versus depth for some formation waters taken from the
Rodessa formation in the East Texas Basin. The line is a least squares fit (y = a
+ bx + cxz ) for the scattered data.
respect t o magnesium. Reactions occurred between the brines and the

enclosing rocks to deplete the magnesium or increase the calcium in these
brines. The curve in Fig. 8.8 represents a 35% mean fit of the scattered data
where an ideal fit is 100%.
Ternary diagram
Ternary or triangular diagrams are useful in studying the distribution of

the cations and anions in subsurface waters. The ions which often are used in
these diagrams are sodium, calcium, and magnesium or chloride, sulfate, and
bicarbonate. To plot these constituents, the equivalent weights of three
cation constituents or three anion constituents are determined and summed
t o equal 100%.The percentage of the three then are plotted. Dickey (1966)
used these types of plots fo study the composition of some deep subsurface
waters.
Fig. 8.9 is a ternary plot of the normalized values for sodium, calcium,
and magnesium concentrations of some brines taken from oil-productive
sedimentary basins. This diagram indicates that all of these brines contain
more calcium than magnesium. For these particular samples the diagram
seems to indicate that the calcium equivalents tend to increase with the age
of the rocks from which the brines were taken; this does not, however, apply
to all brines.
DISCUSSION 289
Fig. 8.9. Relative concentrations of sodium, calcium, and magnesium in some formation
waters taken from sedimentary rocks in oil-productive basins.
Discussion
Chebotarev (1955) postulated that most subsurface waters are altered by

meteoric water. The three hydrodynamic zones that he considers are: (1)
active exchange, where water is influenced by a relatively high degree of
hydrodynamic movement with flushing-out of most of the salt thus pro-
ducing a low salinity water; (2) delayed exchange, where the hydrodynamic
flow is less rapid, thus leaving a higher salinity water; and (3) stagnant
conditions, where there is little hydrodynamic movement, so that the
salinity accumulates and is higher.
This system was not applied to the 4,000 waters because it evaluates in a
rather complex manner the types and classes of waters from the constituent
compositions in reacting values in percent. The method appears useful in
certain general types of studies.
Palmer’s system does not consider ionic concentrations or possible con-
ditions of saturation of calcium carbonate or calcium sulfate. Some consti-
tuents such as the alkaline earths and chloride and sulfate are lumped togeth-
er. Calcium and magnesium should not be lumped together because even
though they are in the same chemical group in the Periodic Table, they often
behave quite differently in chemical reactions. For example, the solubilities
of many of their salts are considerably different. The same often is true of
the chlorides and sulfates.
The portion of the Palmer system applied t o the 4,000 brines was used
only because it is incorporated in Sulin’s system. The Palmer system by itself
appears to have little value in studying subsurface brines related t o hydrocar-
bon accumulations.
Schoeller separates the waters into six types solely on the basis of the
concentration of the chloride ion and into four additional types using the
concentration of the sulfate ion. Because of this, his classification of oilfield
waters gives considerably more variation in types than the Sulin system, and
thus it is more confusing in interpretation. Schoeller’s formula for deter-
mining relative saturations of sulfates and carbonates in the waters has value
in production problems. For example, knowledge that the sulfates or carbon-
ates may precipitate from a water under certain conditions is useful so that
correct treatment can be applied t o prevent well and/or formation damage.
The index of base exchange also is useful in evaluating diagenetic reactions
that may have occurred to change the water.
Sulin’s system considers some of the ions in determining the type of
water, and the class indicates the predominance of the anion groups. These
two characteristics of waters appear useful in studying waters that are likely
to be associated with hydrocarbon accumulations. For example, several in-
vestigators have determined that the chloride-calcium type in the S1 S2 A2
class of water is the most likely type to be associated with a hydrostatic
environment that promotes the accumulation of hydrocarbons. The next
most likely type is bicarbonate-sodium in the S1 A l A2 class and the least
likely types are chloride-magnesium and sulfate-sodium.
Knowledge of the type and class plus what Sulin describes as the signifi-
cant indicators (direct and indirect) appears useful in hydrocarbon explora-
tion studies. The Sulin system does not consider the degree of saturation of
the water with respect to sulfates and carbonates, which is a disadvantage in
production problems. Also, the system makes no provision for determining
the degree of base exchange which is useful in certain interpretations of
diagenesis.
Bojarski’s modification of the Sulin system appears t o have little value in
evaluating a water that is likely to be associated with petroleum. Many of the
values that were found for the Na/Cl ratio in the 4,000 samples were greater
than 0.85. This value in a chloride-calcium type water according to Bojarski
is not likely t o be found in a water associated with petroleum. It is possible
that the waters that Bojarski evaluated were consistently very concentrated
waters with respect to dissolved solids.
CONCLUSIONS 291
Conclusions
Classification of the subsurface waters aids in interpreting the type of

water that is more likely t o be associated with a hydrocarbon accumulation.
Application of water classification techniques t o 4,000 oilfield brine samples
revealed that the majority of the samples were chloride-calcium type of the
S1 S2 A2 class, had positive IBE values, had a predominant cation sequence
of Na > Ca > Mg, and were very highly concentrated with dissolved chloride.
Many of the 4,000 samples had values greater than 7 for q m
C 0 3)’ (Ca), and some had values greater than 70 for d m , indicating
that many of the waters were saturated or nearly saturated with respect to
carbonate and sulfate.
The Bojarski modification of the Sulin system appears to be of little value.
The ratio Na/(Ca + Mg) generally decreased with increasing dissolved solids
and depth. The ratio Ca/Mg generally increased with increasing dissolved
solids and depth.
Studies of the water characteristics and mapping of the more important
water properties in conjunction with other geological and geophysical data
appear useful in exploration. Maps of certain water characteristics such as
the JCa+’ x SO,,-’ and d ( H C 0 3 - + C03- )’ (Ca+’ ) are useful in solving
production problems.
References
Ambrose, A.W., 1921. Underground conditions in oilfields. US.Bur. Min. Bull., 195, 238
PP.
American Petroleum Institute, 1968. API Recommended Practice for Analysis o f Oilfield
waters. Subcommittee on Analysis of Oilfield Waters, API RP 45, 2nd ed., 49 pp.
Bojarski, L., 1970. Die Anwendung der hydrochemischen Klassifikation bei Sucharbeiten
auf Erdol. 2.Angew. Geol., 16:123-125.
Chebotarev, I.I., 1955. Metamorphism of natural waters in the crust of weathering.
Geochim. Cosmochim. Acta, 8:22-48, 137-170,198-212.
Collom, R.E., 1922. Prospecting and testing for oil and gas. U.S. Bur. Min. Bull., 201, 170
PP.
Dickey, P.A., 1966. Patterns of chemical composition in deep surface waters. Bull. A m .
Assoc. Pet. Geol., 50:2472-2478.
Elliott, Jr., W.C., 1953. Chemical characteristics of waters from Canyon, Strawn, and
Wolfcamp formations in Scurry, Kent, Borden, and Howard Counties, Texas. Pet.
Eng., 25 :B7 7-BB9.
Mills, R. van A., 1925. Protection of oil and gas field equipment against corrosion. U.S.
Bur. Min. Bull., 233, 127 pp.
Ostroff, A.G., 1967. Comparison of some formation water classification systems. Bull.
A m . Assoc. Pet. Geol., 51:404-416.
Palmer, C., 1911. The geochemical interpretation of water analyses. U.S. Geol. Sum.
Bull., 749~5-31.
Reistle, Jr., C.E., 1927. Identification of oilfield waters by chemical analysis. U.S.Bur.
Min. Tech. Paper, No.404, 24 pp.
Reistle, Jr., C.E. and Lane, E.C., 1928. A system of analysis of oilfield waters. U.S.Bur.
M i a Tech. Paper, No.432, 14 pp.
Rogers, G.S., 1917. Chemical relations of the oilfield waters in San Joaquin Valley,
California. U.S. Geol. Sum. Bull., 653:5-116.
Rogers, G.S., 1919. Sunset-Midway oil field, California, 11. Geochemical relations of oil,
gas, and water. U.S. Geol. Sum. Prof. Paper, No.117, 103 pp.
Schoeller, H.,1955. Geochemie des eaux souterraines. Rev. Znst. Fr. Pet., 10:181-213,
219-246, 507-552.
Sulin, V.A., 1946. Waters of Petroleum Formations in the System of Nature Waters.
Gostoptekhizdat, Moscow, 96 pp. (in Russian).
Swigart, T.E. and Schwanenbek, F.X., 1921. Petroleum engineering in the Hewitt
Oilfield, Oklahoma. Cooperative Bulletin published by the US. Bureau of Mines, the
State of Oklahoma, and the Ardmore, Oklahoma, Chamber of Commerce, 61 pp.
U.S.Bureau of Mines, 1965.Analyses of Oilfield Waters. Open-file Report.
Chupter9. SOME EFFECTS OF WATER UPON THE GENERATION,
MIGRATION, ACCUMULATION, AND ALTERATION OF
PETROLEUM
Both water and hydrocarbons are ubiquitous in the crust of the earth and
water is in contact with most if not all chemical and physical reactions. The
water cycle is comprised of mechanisms whereby it precipitates from the
atmosphere t o the crust of the earth and there enters various environments -
e.g., surface and subsurface, chemical and physical, and biogenic and
abiogenic - until it ultimately returns t o the atmosphere. The carbon cycle
comprises the reaction of water, minerals, carbon dioxide, and a catalyst
(chlorophyll) t o form organic plant material composed of carbohydrates,
fats, lignins, proteins, and some hydrocarbons such as alkanes. Production of
additional types of organic material results from the consumption of plants
by animals.
Organic matter from land and marine plants and animals is deposited with
inorganic sediments t o form a biomass from which petroleum and natural gas
originate. Most of the organic matter that will later produce the hydrocar-
bons is deposited in an aqueous environment; therefore, it is evident that
water profoundly influences not only the hydrocarbon precursors but (as
will be demonstrated later) also the generation, migration, accumulation, and
alteration of hydrocarbons.
The environment controls where and how inorganic and organic matter is
deposited. Most of the petroleum source material is marine (autochthonous);
however, Corbett (1955) noted that considerable terrestrial (allochthonous)
derived organic matter is carried by streams and rivers until the bulk of it
often is deposited in deltas. In the marine depositional environment, the
sediment characteristics are controlled by hydraulics and by grain size of the
detrital material. Most coarse-grained sediment will deposit on beaches, bank
tops, and shelves, or where water turbulence is maximum. Finer grained
sediments are deposited in shelf areas, while the finest grained are deposited
in the deep basin (Emery, 1960).
Reducing conditions which aid in preserving organic matter are prevalent in
sediments and bottom waters in many marine environments and in general
tend t o become more reducing with depth below the water-ediment inter-
face. Therefore, most of the oxidation of organic detritus occurs as it falls t o
the sea bottom where the reducing environment will inhibit further
oxidation. Rapid deposition will aid in the preservation of more organic
matter as well as more of the easily oxidized organic compounds.
294 EFFECTS O F WATER ON PETROLEUM
Compaction
Recently deposited sediments have water-filled pore spaces which com-

prise 8Wo or more of their volume, and deposition of a few more hundred
meters of sediments causes this porosity to be reduced to 30% accompanied by
the expulsion of water (Hedberg, 1964). Considerable time is necessary for
water t o be expelled from low-permeability fine clays (Terzaghi and Peck,
1948). In some cases, pressures approaching geostatic are created because
with increasing compaction, the pore water, which cannot easily escape,
carries the full load. The pressure will drop to hydrostatic as the water is
squeezed out, then the water bears the weight of the overlying water column
and the sediments carry the weight of the overlying sediment column minus
the buoyancy factor of the water.
Compaction occurs in most sediments, but it is greater as muds form
shales than when sands form sandstones, and likewise the expulsion of water
from muds is greater. Also carbonate compaction will halt before argil-
laceous mud compaction because carbonate recrystallization causes forma-
tion of a rigid, difficult t o compact, structure.
The clay, montmorillonite, expands when interlayer water is added and
contracts when it is removed. The removal of water from montmorillonite is
a function of pressure and temperature (Burst, 1969). Schmidt (1971) found
that the temperature is about 93-104OC when clay starts releasing intra-
crystalline water and that rate of release increases with higher temperatures.
Powers (1967) also suggested that it is a function of the amount of potas-
sium which is necessary for the transformation of montmorillonite t o illite.
Removal of water from montmorillonite with compaction resulting from
deep burial yields large amounts of free water which can migrate. The
amount of freed interlayer water depends upon the amount of montmoril-
lonite or other clays in an expanded state in the sediment. The conversion of
montmorillonite to chlorite does not require any volume change (Weaver and
Beck, 1969), but the conversion of montmorillonite t o illite reduces the clay
volume about 5096, so it is apparent that this freed water can be equivalent
to 50% of the amount of montmorillonite in the sediment. The dehydration
of montmorillonite occurs in three stages and probably a t three different
depths of burial since the dehydration is a function of depth. Therefore, the
second and third dehydration stages could release water at about 1,500 m
and perhaps below 3,000 m thus allowing new supplies of water t o migrate.
In the Gulf Coast area, the last water is not removed from montmoril-
lonite until the depth is about 4,500 m. The clays in this area are composed
of 50--80% montmorillonite, and the last water out comprises about 20% of
the volume of montmorillonite present so it is evident that considerable
freed water also is available at depths to 4,500 m.
Von Engelhardt and Gaida (1963) noted that compaction of clays appears
to proceed more rapidly if the pore solution is enriched in dissolved electro-
lytes. This may influence the compaction rate as, for example, a nonmarine
GENERATION AND MIGRATION 295
clay may compact more slowly than a marine clay. This could relate t o a
nonmarine source evolving a different type of hydrocarbon.
Generation and migration
Most petroleum and natural gas hydrocarbons are generated from sedi-
ment organic matter by low-temperature chemical reactions (Philippi, 1965).
Philippi also concluded that lipids probably are the major petroleum
precursor and that most petroleum is generated by chemical reactions occur-
ring at temperatures above 115°C. A catalytic effect of the surrounding
sediment, and in particular the shales, influences the rate and type of
hydrocarbon formation.
According t o McIver (1963)the maturation of hydrocarbon is influenced
by depth, and the heavy hydrocarbons in the liquid crude oil decrease with
depth while the lighter hydrocarbons increase with depth. Fig. 9.1 illustrates
how an aromatic type oil might generate and mature to graphite with depth,
and how a paraffinic type crude oil might generate and maturate t o methane
with depth. In reality many crude oils contain both aromatics and paraffins
and the ultimate end product with increasing burial depth would be graphite.
It has been postulated that oil migrates from the fine-grained source rock
to a trap as a discrete oil phase. However, calculations indicate the migration
of oil globules in reservoirs required forces several thousand times greater
than those produced by hydrodynamic gradients in most reservoirs
(Levorsen, 1954).
Because water is present in all sediments that contain organic matter
which transforms to petroleum and natural gas hydrocarbons, it is reasonable
t o assume that water is closely related to petroleum-generation processes. In
essence, water controls the depositional environment of the petroleum
1 r S t a r t i n g material I Starting material
1
P a r a f f i n i c crude oil
1
L i g h t e r hydrocarbons
plus N, S, and 0
compounds
1
I CH 4 1
Fig.9.1. Schematic of how an aromatic crude oil can maturate to graphite with increasing
temperatures associated with increasing depth, and how a paraffinic crude oil can
maturate t o dry methane gas. (After McIver, 1963.)
precursors, while in the subsurface environment, water appears to control

the migration of petroleum hydrocarbons and it is not unlikely that water is
a control in the generation of petroleum.
The generation of hydrocarbon is temperature dependent (Philippi, 1965).
The viscosity of crude oil decreases exponentially with increasing tempera-
ture. The solubilities of petroleum hydrocarbons in water increase with
temperature and decrease as the salinity of the water increases. A tempera-
ture drop from 150’ to 25OC reduces the solubility of petroleum in water by
a factor of 4.5 t o 20.5 (Price, 1973).
Nonhydrocarbon organic acids, esters, and ketones are relatively soluble in
water. Therefore it is possible that these compounds may migrate in water
solution and be reduced t o hydrocarbons in the reservoir. Baker (1959)
demonstrated that artifical solubilizers enhance the ability of saline water to
solubilize and transport hydrocarbons.
Baker (1962) ultrafiltered dilute petroleum acid soap solutions and found
two kinds of colloidal particles present at all finite concentrations. One type
he called a small ionic micelle and the other a large neutral micelle. Accord-
ing to him, “the presence of these colloidal particles in water increases the
water solubility of hydrocarbons by providing hydrocarbonlike regions in
which the hydrocarbon solutes preferentially and selectively dissolve. The
relatively small, spherical, isotropic micelle appears to enhance hydrocarbon
solubility chiefly by incorporating hydrocarbons on its surface. However,
hydrocarbons may be accommodated within the interior of the larger,
cylindrical, anisotropic micelle, as well as on its surface.” He found that
paraffinic, naphthenic, and aromatic hydrocarbons possess different solu-
bilities in the different kinds of micelles. It has not been demonstrated that
these micelles occur in nature, however.
Peake and Hodgson (1966 and 1967) studied the accommodation of
C12-C3,n-alkanes in water. They found that it is possible to accommodate
various hydrocarbons in water in the range of 0.01-100 ppm. Levels this
high are more than adequate t o account for known petroleum reserves. For
example, Hunt (1961) estimated a ratio of 27/1 for the volume of dispersed
hydrocarbons in sediments t o oil in reservoirs, or that about 3.6% of the
dispersed hydrocarbons are trapped in reservoir. Further, it can be shown
that only approximately 1.8 ppm of the source hydrocarbons need t o be
reservoired to account for the known oil reserves, assuming that the source
sediments contain an average of 50 ppm of hydrocarbon. Hodgson et al.
(1964) estimated that it would be necessary t o mobilize only about 1 ppm
of the hydrocarbons from the source sediment to the reservoirs to account
for known reserves.
The odd/even carbon ratio is different in recent sediments, from ancient
sediments and from most crude oils (Bray and Evans, 1961). Baker (1959)
and Peake and Hodgson (1966) concluded that the odd carbon preference
does not persist in the mobilizing water because the accommodation of
hydrocarbon in water is a function of its properties rather than supply.
GENERATION AND MIGRATION 297
Welte (1965) extracted some bituminous rock samples with oilfield brines
which had been cleaned of traces of hydrocarbons. The extracted n-paraffins
had an approximately equal carbon number distribution.
Cooper and Bray (1963) postulated a mechanism whereby the carbon
number distribution may change when fatty acids are dissolved in water.
Each fatty acid can lose COz to form an intermediate product which reacts
to produce an n-paraffin and a fatty acid. The products have one less carbon
atom than the original fatty acids. The oilfield waters that they investigated
contained fatty acids with 14-30 carbon atoms with a balanced distribution
between even and odd carbon numbers.
Kartsev et al. (1959) reported that some oilfield waters contain up t o
5,000 mg/l of sodium naphthenate. Gullikson et al. (1961) found 5,000 t o
8,000 mg/l of organic acid salts in oilfield waters from the Ventura Basin.
Collins (1969) reported that some waters from the Anadarko Basin contain
up t o 3,000 mg/l of organic acid salts. The presence of 50 mg/l of organic
acid salts in waters could transport significant amounts of hydrocarbons out
of source rocks.
It is apparent that petroleum hydrocarbons are generated from organic
matter in sediments, that hydrocarbons are formed primarily at temperatures
above 115'C through abiogenic reactions, and that water serves to transport
the hydrocarbons from the sediments. It is possible that the water also may
transport. petroleum precursors from the sediment. These precursors might
be released from the water phase when the environment is right and then be
transformed into petroleum hydrocarbons.
Fluids in subsurface strata will move from regions of high potential t o a
low-energy region. The permeabilities of the strata influence how quickly
equilibrium can be attained between high-energy and low-energy fluid poten-
tials. The fluids will move toward the strata offering the least resistance and
that have the greatest permeability. Generally the lower energy environment
is straight up, but it may also be straight down or lateral. The source sedi-
ments may overlie a sand structure, in which case the migration is likely t o
be down into the sand structure which may serve as a reservoir.
Energy gradients which cause fluids to move are caused by potentials
resulting from temperature, pressure, elevation, and osmotic forces which
have not been completely described mathematically (Hitchon, 1969). Rapid
tectonic uplift can alter the steady-state equilibrium in a basin.
Primary transit of hydrocarbon or precursors is this migration from the
source sediment and usually is only for a short distance or far enough t o
reach a coarse-grained sediment. After they reach the coarse-grained
sediment, they may or may not be trapped. Secondary migration occurs
after the hydrocarbons or precursors move into the coarser grained sediment,
and it may result in long or short-distance migration. The forces suggested
above affect the migration. Once the petroleum and gas have separated from
the aqueous phase they tend t o move t o higher zones because of gravity
segregation.
Accumulation
Compaction of sediments containing organic matter and water results in

the expulsion of water and solubilized organic matter. Conceivably, the
organic matter and gases can be solubilized or combined with the water
phase as organic acid salts or as other soluble, colloidal, or suspended forms
as a result of increased pressure. Meinschein (1959) outlined a method
whereby ptroleum hydrocarbons may dissolve in compaction-expelled
water, migrate, and later be unloaded to form an accumulation.
Cartmill and Dickey (1970) formed a crude oil in water emulsion and
found that it freely passed through coarse sand. However, it tended t o
coalesce with decreasing grain size. They attributed this to electrochemical
and capillary phenomena. Such a process could operate in a subsurface
environment and form a hydrocarbon accumulation. Related t o this
phenomenon, Kidwell and Hunt (1958) studied the hydrocarbon content
produced from sands. They decided, “that in the lenticular sands hydrocar-
bons are being filtered out of the moving stream of water by capillary
action.”
Jones (1968) gives some data illlustrating the relation of pressure differen-
tials to salinity differences in waters attributed to osmotic flow of water
through a clay-shale membrane. He found that the salinity differential may
exceed 100,000 mg/l between two aquifers separated by a 10-to 50-m layer
of clay-shale. Waters carrying dissolved or solubilized hydrocarbons could
release them because of salting out when contacting highly saline brines
because water will accommodate less hydrocarbon as its dissolved salt con-
centration increases (Baker, 1967). I t also has been speculated that the solu-
bilized or dissolved hydrocarbons may be unloaded from the aqueous phase
because of temperature or presssure changes, filtration, salinity changes, or
the attraction of the organic constituents in the water for a hydrocarbon
accumulation.
According to Baker (1962), “the nature of the oil, whether paraffinic,
naphthenic, or aromatic, probably depends less on the nature of the
hydrocarbons originally available in the sediments for collection than on the
proportions of neutral and ionic micelles that were able t o release crude oil
droplets for subsequent pooling in the reservoir rock. It appears that the
frequency distribution of hydrocarbons in crude oil reflects variations in the
kind and size of the micelles in which the sediment hydrocarbons selectively
dissolve. The subsequent release of solubilized hydrocarbons from aqueous
solution results in the accumulation of oil. The hydrocarbons, after under-
going this solution and release process, would exist in the proportion charac-
teristic of hydrocarbon occurrence in crude oil.”
Welte (1965), in discussing the micelle theory, noted that Neumann
“distilled several crude oils and observed that the surface tension of the
distillate decreased and the dielectric constants increased with increasing
boiling point. This means that the polar surfaceactive molecules which are
ALTERATION 299
necessary for the formation of micelles are enriched in the high molecular-
weight fraction.” The surface-active components were phenols, and it was
found that the interfacial tension between oils and a buffered solution was
strongly pH dependent, suggesting that pH variations in pore solutions are
important for stability of micelles.
Hodgson and Hitchon (1966) theorized that surfactants aid in the mobi-
lization of hydrocarbons in water, allowing them to migrate. During
migration, the surfactants may decompose, leaving the hydrocarbons in an
unstable state in the water phase. If the water is moving, the hydrocarbons
will have a tendency not to agglomerate, but in a hydrostatic quasi-stagnant
area the hydrocarbons will tend to separate by gravity t o form an accumula-
tion.
Hydrocarbons accumulate in areas called traps, and these traps sometimes
are structural traps such as an anticline or stratigraphic traps such as
permeability barrier. Accumulations of hydrocarbons in oil pools apparently
form by the release of hydrocarbons from a water system. The hydrocarbons
accumulate subsequent t o their migration, and the volume of water asso-
ciated with their migration is large; however, tremendous volumes of water
are available in the subsurface, and large amounts of water are freed by dia-
genesis or alteration of clays.
Baker (1967) postulated that the “tar mats” that form near the edges of
sedimentary basins can be attributed t o organic solutes moving in a water
phase. The water merely evaporates at the sediment-air interface with com-
plete release of the heavy oil solute containing compounds of carbon,
hydrogen, nitrogen, oxygen, and sulfur.
Because the solubilities of petroleum hydrocarbons increase with tempera-
ture and decrease with increasing salinity, a mechanism for petroleum
accumulation can be formulated. Upward moving subsurface waters will
decrease in temperature, and as they move upward, the pressure will be
lower, and they may meet more saline waters. Because of these changing
conditions they will release dissolved hydrocarbons.
Alteration
Hydrocarbons are subject t o alteration as soon as they are formed. The

long-chain alkanes may react with other chemicals associated with the water
phase, resulting in bond breaking and the formation of a shorter chain
alkane. Organic acid salts may form which are relatively soluble in water and
they may be further altered if the redox potential or pH of the water
changes. The pH may change because of release of carbon dioxide associated
with a hydrocarbon reaction. The hydrogen proton in the low pH water will
react with dissolved organic acid salts, causing a reversion to the less soluble
organic acid, which will disaccommodate from the aqueous phase.
A positive change in the redox potential of the water could cause oxida-
tion of dissolved or associated hydrocarbons. A negative change in the redox
potential will aid in preserving the hydrocarbon, but if the redox should
become extremely reducing, it could cause hydrogenation of the hydrocar-
bons, this is, however, very unlikely. Reservoired oil becomes lighter with
depth in all productive sedimentary basins; this usually is attributed to in-
creased cracking of the larger hydrocarbon molecules into smaller ones, and
is associated with increased temperatures (Philippi, 1965). This process
requires more hydrogen for the smaller molecules, which may result in con-
densation and polymerization of large molecules. Ultimately, with deeper
burial and higher temperature, the hydrocarbons would be converted t o
methane and graphite (Fig.9.1).
Water washing
Water washing is a simple dissolving of light hydrocarbons by water

moving past a reservoir. The water washing causes the hydrocarbon accumu-
O i l m i g r a t e d l o here
Fig.9.2. Alteration of an oil accumulation caused by tectonism and moving water.

ALTERATION 301
lation t o become more concentrated in heavy hydrocarbons with a lower

API degree gravity because of fractionation.
Water washing occurs when fresh water infiltrates through an outcrop and
flows through the zone containing a structurally trapped hydrocarbon accu-
mulation. Examples of this type are found in the Rocky Mountain area.
Tectonic disturbances can cause the hydraulic potential associated with a
hydrocarbon accumulation t o become dynamic. The resulting hydrodynamic
situation can cause severe alteration of the hydrocarbon accumulation. Fig.
9.2illustrates how oil may migrate to another reservoir if the strata are tilted,
causing oil t o spill from a structural trap. This phenomenon can be observed
in the Rocky Mountain area, where the first row of anticlines at the edge of
a basin often are barren of hydrocarbon accumulation.
Bacterial
Biological degradation of a hydrocarbon accumulation is more complex

than water washing. I t results from the introduction of bacteria into the
reservoir or the reactivation of dormant species.
Alteration of petroleum by bacteria cannot be directly termed alteration
because of water, but water indirectly aids the alteration because water must
be present so that the bacteria can live. Aerobic bacteria can exist and thrive
at the surface or in the subsurface only if several conditions are met, and
these requirements as listed by Beerstecher (1954)and Davis (1967)include
the following:
(1) Water: water must be present t o serve as a medium in which the
bacteria can live.
(2) Bacteria: the bacteria themselves must be present.
(3) Nutrients: certain inorganic trace nutrients must be present.
(4)Temperature: bacteria can exist at temperatures up t o about 90°C but
thrive at temperatures of less than 60°C.
(5)Food: a supply of food such as organic matter or petroleum must be
present.
(6) Toxics: bacterial poisons such as hydrogen sulfide must not be present
in excessive concentrations.
(7)Oxygen: aerobic bacteria require free oxygen dissolved in the water
medium in which they live.
Obviously all of these conditions are most easily met at the surface and
the heavy oils and tars common t o oil seeps are products of fractionation
and aerobic degradation of crude petroleum. Bacterial alteration of petro-
leum can also occur below the surface of the earth if the prerequisite con-
ditions are met. These conditions, especially the presence of dissolved
oxygen, occur only where subsurface waters are in hydrodynamic connec-
tion with the surface and are actively receiving surface-recharge waters. The
presence of free oxygen is critical although it generally exists in very small
concentrations, Kuznetsov (1957)measured the dissolved oxygen content in.
302 EFFECTS OF WATER ON PETROLEUM
waters from wells in the Groznyv oil district and found up t o 10.6 mg/l at a
depth of 249 m. In general, dissolved oxygen was found only in the zone of
active water circulation (surface recharge) and was lacking in the zone of
stagnate water.
Dissolved oxygen in smaller concentrations (less than 1 ppm) sustains
aerobic bacteria and controls the rate at which they consume food (organic
matter and petroleum). Since the oxygen content of subsurface waters
usually is low, alteration and consumption of petroleum in reservoirs may
require thousands or even millions of years to achieve a significant degree of
change.
Kuznetsov (1957) made bacterial counts in produced subsurface waters
and found that the number of bacteria varied directly with the rate of
exchange of water (recharge). The amount of dissolved oxygen varied
directly with the rate of recharge, and the size of the bacterial population is
dependent upon the amount of oxygen available. Aerobic bacteria in
outcropping rocks were carried into the subsurface by recharge waters and
were sustained by dissolved atmospheric oxygen in such waters.
Apparently the rate at which bacteria are able t o move depends only on
the amount of available oxygen and food. All digestible organic matter in the
carrier bed along the recharge path would have t o be removed before the
bacteria could proceed. Biodegradable organic matter in the carrier bed
would be depleted as the oxygen “front” advanced and would permit free
oxygen to reach great distances into the subsurface. The ultimate advance of
bacterial alteration depends principally on time and temperature. If favor-
able conditions are present for sufficient time, bacterial alteration can reach
great distances if the temperature remains below about 6OoC. Although some
forms of bacteria can exist at temperatures up to about 90°C, they appar-
ently are in a dormant state above about 6OoC (Davis, 1967).
Zobell (1949, 1950, 1952) found that all kinds of hydrocarbons appear to
be susceptible t o bacterial action, even though the process is selective. In
general, aliphatic hydrocarbons and aliphatic side chains on cyclic com-
pounds are most susceptible t o attack while naphthenes and aromatics are
least affected. Bacteria prefer the normal alkanes over the branched alkanes
and concentrate on the heavier molecules, rarely touching hydrocarbons as
light as C 3 .
Bacteria consume alkanes at a hydrocarbon fermentation plant at Grange-
mouth, England (Anonymous, 1971). The nearly pure mixture of paraffins is
prepared from crude oil by molecular sieving. The microorganisms assimi-
lated the material with little waste and produced microbial protein which is
used in animal feed.
After consumption of the normal alkanes, branched chain alkanes are
attacked (Winters and Williams, 1971), as are long chains attached t o
aromatic and naphthenic rings (Davis, 1967). Bacteria usually reject ring
compounds and many aromatics are toxic t o bacteria. However, with ideal
conditions, bacteria can break naphthenic rings and consume them. Because
ALTERATION 303
bacterially altered oils contain less paraffinic hydrocarbons than unaltered

oils of the same type, they have higher correlation index values (Smith,
1940) and a lower API gravity.
Winters and Williams (1971) found an increase in nitrogen and optical
activity disproportionate to n-paraffin loss in altered Powder River Basin ofis
in Wyoming and suggested that microbially produced materials are added t o
altered oils. Bailey and Krouse (1970) observed an increase in weight percent
sulfur and asphaltenes and a decrease in saturates (alkanes) in altered
Williston Basin crudes. The relative increase in asphaltenes because of the
selective loss of lighter saturates leads t o increased viscosity of highly altered
oils. The results of bacteria alteration are evidenced by heavy residual oils
common t o oil seeps and breached oil accumulations such as the Athabasca
tar sands.
A bacterially altered paraffinic crude differs from a similarly altered
asphaltic crude. Altered paraffinic crudes contain higher percentages of
asphaltene and naphthene compounds than their unaltered counterparts, and
usually are liquid and mobile. In contrast, severe alteration of asphaltic crude
produces a solid, immobile tar. In the subsurface, the formation of tar seals
halts bacterial alteration and may even act as a trapping mechanism for an
accumulation. An example of oil residue providing a seal occurs in the
Bolivar Coastal fields of the Maracaibo Basin in Venezuela (Brenneman,
1960).
The Maracaibo Basin oils have an asphaltic base, and bacterial alterahion
has produced heavy residues which provides a seal (Rubio, 1959). Appar-
ently, secondary migration of the oils occurred in a carrier-bed system
exposed at the surface. Oil migrated updip without being trapped until it
reached meteoric recharge waters about 5-10 km from the outcrop. Here,
aerobic bacteria reduced the n-paraffin content of the oil, causing an
asphaltic gel t o precipitate which reduced the porosity and permeability,
preventing further migration and allowing oil to accumulate. An example are
the vast reserves of Lagunillas field found in a massive sand downdip from an
asphaltic seal in an area devoid of structural of stratigraphic txappng
mechanisms.
Three types of oil exist in the Bell Creek field, which is a large strati-
graphic trap on the northeast flank of the Powder River Basin. The field
contains common reservoirs, but the oil at the south was not altered and
contained a full spectrum of n-paraffins, while in the middle of the field all
n-paraffins were absent and the oil at the north was deficient in n-paraffins
up to C1,. Selective loss of n-paraffins up t o C1, caused by microbiological
activity occurred (Winters and Williams, 1969).
The Bell Creek altered oils also had lower API gravity, higher nitrogen
content, and higher optical rotation than their unaltered counterparts. The
Bell Creek field is associated with fresh meteoric water in its central and
northern portions but has no oil-water contact at its southern end, the
downdip limit of production in that area being defined by absence of reser-
voir sand. Evidence of aerobic organisms was found in waters produced with
altered Bell Creek oils; oils in the Hackberry field, Louisiana, were bacterial-
ly altered by loss of n-paraffins and singly branched isoparaffins, and in
North Africa an oil was depleted in n-paraffins.'
Conclusions
A suitable mechanism for removing water and organics from source sedi-
ments is compaction and squeezing of the fluids into adjacent beds either
vertically, laterally, or below. Accommodation of hydrocarbons or hydrocar-
bon precursors in the aqueous phase appears t o be a probable means of
primary migration. Variations in salinity, pressure, temperature, pH, redox
potential, capillarity, hydrodynamics, or lithology may cause dissolution of
mobilized hydrocarbons and serve t o begin an accumulation. Finally, the
moving waters can serve t o alter an accumulation.
Alteration of crude oils in the reservoir by water washing and biodegrada-
tion is common t o many producing areas and affects a significant portion of
the world's reserves. Biodegradation is most prevalent in shallow-basin flank
environments where fresh meteoric water reaches into the subsurface and
carries with it aerobic bacteria and the oxygen necessary t o sustain them.
Bacterial alteration selectively removes n-alkanes first, followed by isoal-
kanes and aliphatic side chains on cyclic compounds, leaving an oil relatively
enriched in naphthenes, aromatics, asphaltenes, and sulfur compounds. The
API gravity is lowered, and the nitrogen content and optical activity are
increased. Asphaltic base oils may be reduced t o an immobile tarry residue
that can block porosity and form a seal which may trap oil. Alteration of
paraffinic based crudes produces the so-called naphthenic oils, which,
because of their relatively small asphaltene content, remain mobile even
under advanced alteration conditions.
References
Anonymous, 1971. Food from oil. Nature, 229:79.

Bailey, N.J.L. and Krouse, H.R., 1970. Chemical aspects of crude oil preservation
(abstract). Bull. Am. Assoc. Pet. Geol., 54:834-835.
Baker, E.G., 1959. Origin and migration of oil. Science, 129 (3353):871-874.
Baker, E.G., 1962. Distribution of hydrocarbons in petroleum. Bull. A m . Assoc. Pet.
Geol., 46 (1):76-84.
Baker, E.G., 1967. A geochemical evaluation of petroleum migration and accumulation.
In: B. Nagy and U. Colombo (Editors), Fundamental Aspects of Petroleum Geo-
chemistry. American Elsevier, New York, N.Y., pp.299-329.
Beerstecher, Jr., E., 1954. Petroleum Microbiology - A n Introduction to Microbiological
Engineering. American Elsevier, New York, N.Y., 375 pp.
Bray, E.E. and Evans, E.D., 1961. Distribution of n-paraffins as a clue to recognition of
source beds. Geochim. Cosmochim. Acta, 22:2-15.
Brenneman, M.C., 1960. Chemical study of crudes of the Maracaibo Basin: Venezuela.
Direc. Geol., Bol. Geol., Publ. Espec., No.3, Part 3, pp.1025-1069 (in Spanish).
REFERENCES 305
Burst, J.F., 1969. Diagenesis of Gulf Coast clayey sediments and its possible relation to
petroleum migration. Bull. A m . Assoc. Pet. Geol., 53 (1):73-93.
through porous media. Bull. A m . Assoc. Pet. Geol., 54 (12):2438-2447.
Collins, A.G., 1969.Chemistry of some Anadarko Basin brines containing high concentra-
tions of iodide. Chem. Geol., 4:169-187.
Cooper, J.E. and Bray, E.E., 1963. A postulated role of fatty acids in petroleum forma-
tions (preprints). A m . Chem. SOC.,Div. Pet. Chem., 8 (2):A17-A28.
Corbett, C.S., 1955. In situ origin of McMurray oil of northeastern Alberta and its
relevance to general problem of origin of oil. Bull. A m . Assoc. Pet. Geol., 39
(8):1601-1649.
Davis, J.B., 1967. Petroleum Microbiology. American Elsevier, New York, N.Y., 604 pp.
Emery, K.O., 1960. The Sea Off Southern California: A Modern Habitat of Petroleum.
John Wiley and Sons, New York, N.Y., 366 pp.
Gullikson, D.M., Caraway, W.H. and Gates, G.L., 1961. Chemical analysis and electrical
resistivity of selected California oilfield waters. U.S.Bur. Min. Rep. Invest., No.5736,
21 PP.
Hedberg, H.D., 1964. Geologic aspects of origin of petroleum. Bull. A m . Assoc. Pet.
G e o i , 48 (11):1755-1803.
Hitchon, B., 1969. Fluid flow in the Western Canada Sedimentary Basin, 2. Effect of
geology. Water Resour. Res., 5:460-469.
Hodgson, G.W. and Hitchon, B., 1966.Research trends in petroleum genesis. In: Petroleum
--roc. Eighth Common. Min. Metall. Congr., Aust. N.Z. Paper 34, 59-19.
Hodgson, G.W., Hitchon, B. and Taguchi, K., 1964.The water and hydrocarbon cycles in
the formation of oil accumulations. In: I. Miyake and T. Koyamer (Editors), Recent
Researches in the Fields of Hydrosphere, Atmosphere, and Nuclear Geochemistry.
Maruzen, Tokyo, pp.217-242.
Hunt, J.M., 1961. Distribution of hydrocarbons in sedimentary rocks. Geochim.
Cosmochim. Acta, 22:37-49.
Jones, P.H., 1968. Geochemical hydrodynamics - a possible key to the hydrology of
certain aquifer systems in the northern part of the Gulf of Mexico Basin. Proc. 23rd
Int. Geol. Congr., Prague, 1968, 17:113-125.
Kartsev, A.A., Tabasaranskii, Z.A., Subbotta, M.I. and Mogilevskii, G.A., 1959. Geo-
chemical Methods of Prospecting and Exploration for Petroleum and Natural Gas.
University of California Press, Berkeley, Calif., 349 pp.
Kidwell, A.L. and Hunt, J.M., 1958. Migration of oil in recent sediments in Pedernales,
Venezuela. In: L.C. Weeks (Editor), Habitat o f Oil. American Association of Petro-
leum Geologists, Tulsa, Okla., pp.790-817.
Kuznetsov, S.I., 1957. A review of fundamental research on the microflora of petroleum
deposits. Mikrobiologiya, 26 (6):651-658.
Levorsen, A.I., 1954. Geology of Petroleum. W.H. Freeman, San Francisco, Calif., 703
PP.
"McIver, R.D., 1963. Maturation of oil, an important natural process. Geol. SOC. A m .
Annual Meet., 1963 Paper, 15 pp.
Mehschein, W.G., 1959. Origin of petroleum. Bull. A m . Assoc. Pet. Geol., 43:925-943.
Peake, E. and Hodgson, G.W., 1966.Alkanes in aqueous systems, I. Exploratory investiga-
tions on the accommodation of Czo-C33 n-alkanes in distilled water and occurrence
in natural water systems. J. A m . Oil Chem. SOC.,43 (4):215-222.
Peake, E. and Hodgson, G.W., 1967.Alkanes in aqueous systems, 11. The accommodation
of C, *-C36 'n-alkanes in distilled water. J. A m . Oil Chem. SOC.,44 (12):696-702.
Philippi, G.T., 1965. On the depth, time, and mechanisms of petroleum generation.
Powers, M.C., 1967. Fluid-release mechanisms in compacting marine mudrocks and their
importance in oil exploration. Bull. Am. Assoc. Pet. Geol., 51:(7)1240-1254.
Price, L.C., 1973. Solubility of petroleum in water as function of temperature and
salinity and its significance in primary migration. Bull. Am. Assoc. Pet. Geol., 57:801.
Rubio, F.E., 1959. The conditions of the accumulation of petroleum of the Costaueros
Fields, in the Bolevar District, Lake Maracaibo. Tercer Congr. Geol. Venezolana,
Mem., 3:1009-1023 (in Spanish).
Schmidt, G.W., 1971. Interstitial Water Composition and Geochemistry o f Deep Gulf
Coast Shales and Sands. M.S. Thesis, University of Tulsa, Tulsa, Okla., 121 pp.
Smith, H.M., 1940. Correlation index to aid in interpreting crude-oil analyses. U.S.Bur.
Min. Tech. Rep., No.610, 34pp.
Terzaghi, K. and Peck, R.B., 1948. Soil Mechanics in Engineering. John Wiley and Sons,
New York, N.Y., 566 p9.
Von Engelhardt, W. and Gaida, K.H., 1963. Concentration changes of pore solution
during the compaction of clay sediments. J. Sediment. Petrol., 33 (4):919-930.
Weaver, C.E. and Beck, K.C., 1969. Changes in the clay-water system with depth,
temperature, and time. O f f Water Resour. Res. Project N0.A-008-GA WRC-0769,
Georgia Inst. Technol., Completion Rep., 95 pp.
Welte, D.H., 1965. Relation between petroleum and source rock. Bull. Am. Assoc. Pet.
Geol., 49 (12):2246--2268.
Winters, J.C. and Williams, J.A., 1969. Microbiological alteration of crude oil in the
reservoir (preprints). Am. Chem. SOC.,14(4):E22-E31.
Winters, J.C. and Williams, J.A., 1971. Microbiologic alteration of crude oil in muddy
sandstone and other reservoirs. Bull. Am. Assoc. Pet. Geol., 55:369.
Zobell, C.E., 1949.Action of microorganisms on hydrocarbon. Am. Pet. Inst. Rep. Progr.
1946-1947, pp. 107-132.
Zobell, C.E., 1950. Assimilation of hydrocarbons by microorganisms. Adu. Enzymol.,
10:443--468.
Zobell, C.E., 1952. Part played by bacteria in petroleum formation. J. Sediment. Petrol.,
22:42-49.
Chapter 10. GEOCHEMICAL METHODS OF EXPLORATION FOR PE-
TROLEUM AND NATURAL GAS
The current worldwide concern for energy resources results from the fact
that man eases his burdens and accomplishes enormous amounts of work by
harnessing mineral energy. His material progress is reflected by the amount
of energy his machines consume, and the continued upward thrust of such
progress is dependent upon the supply of mineral fuels. In the United States
this energy has eased the burdens of man t o the extent that every person in
the U.S.A. has available the energy equivalent of the output of 260 men.
About 33% of the total mineral energy demand is supplied by natural gas
and about 42% by crude oil. In 1967 the U S . demand for natural gas was
17.7 trillion cubic feet and the demand for crude oil was 4.5 billion barrels.
The demand for both is expected t o double by the year 2000.
Improved methods of finding petroleum and gas are needed. Many investi-
gators believe that the application of geochemical methods in conjunction
with geological and geophysical methods can markedly improve the
discovery ratio. Data presented in this chapter primarily are in the form of
an annotated survey of the available literature and many of these data were
taken from Petroleum Abstracts with permission of H.O. McLeod (1971).
Introduction
Type of methods
Geochemical methods of exploring for petroleum and gas can be termed

direct and indirect. The direct methods are as follows:
(1) Analysis of soil samples t o determine the amounts of hydrocarbon
gases adsorbed by the soil.
(2) Analysis of free hydrocarbons in soils.
(3) Analysis of soils to determine the amounts of soil wax, paraffin dirt,
and other bitumens.
(4)Analyses of waters t o determine their amounts of dissolved hydrocar-
bons.
These methods are classified as direct because the determined constituents
are present in accumulations of petroleum and gas or derived from them.
The indirect methods are as follows:
(1)Measure the oxidation-reduction potential of soils, rocks, and waters.
(2) Analyze soil samples for various salts such as bromides, chlorides,
carbonates, and sulfates.
308 EXPLORATION FOR PETROLEUM AND GAS
(3) Analyze plants t o determine if their growth is affected by petroleum-

type bitumens.
(4) Analyze soil samples for bacteria such as the types that consume
hydrocarbons.
The indirect methods are so classified because the determined properties
may have originated from something other than petroleum or gas accumula-
tion (Rosaire, 1939; Kartsev et al., 1959). Collected data from both the
direct and indirect methods usually are plotted or contoured to form maps
to identify anomalies and to locate a target area for drilling.
Case histories
A geochemical soil survey led t o the discovery of an oilfield in the Texas

Gulf Coast (Stormont, 1939). Three hundred geochemical surveys were
made in 1939 with a greater than 50%success ratio (Simons, 1940). A gas
discovery at Stroud, Oklahoma, was attributed t o geochemistry (Rosaire et
al., 1940). Pirson (1942) determined the success ratios for the following
exploration methods: random drilling, 5.8%;geology, 8.2%;geophysics,
14.9%;and geochemistry, 57.8%.
Geochemistry predicted the discovery in Oklahoma of the West Edmond
field (Bronston, 1947). The Hardy field in Texas was found by drilling into a
geochemical anomaly where reflection seismographs gave no indication of a
stratigraphic trap (Ransone, 1947). The Soviet Union attributes a 70%
success ratio to wells drilled into geochemical anomalies (Sokolov et al.,
1959). The success of geochemical methods is outstanding when compared
to the overall success ratio for wildcats (Anonymous, 1960b). A company
using an indirect inorganic geochemical technique achieved a 25.5%success
ratio, which is more than double the usual wildcat ratio (Anonymous, 1959).
Success ratios of 65% and 75%, respectively, were obtained using geo-
chemical anomalies in the Texas Gulf Coast and North Texas areas
(Anonymous, 1960a). The Kohav oilfield in Israel was discovered solely
through hydrocarbon geochemistry, and geochemistry successfully predicted
that stepout wells in this field and the Heletz field would be dry (Davidson,
1963).
Analytical techniques, type of hydrocarbon anomalies, and successful
discoveries made by hydrocarbon geochemical studies in Texas were discuss-
ed by Horvitz (1969). According t o him, the anomalies become weak and
tend to disappear after an oil accumulation is produced.
Note: The author has been told by representatives of various companies
that research performed by them has produced evidence indicating that
success ratios claimed for some of these exploration techniques cannot be
met in actual practice. However, the author is at a loss t o cite references to
this effect because there are no documented published data available.
INTRODUCTION 309
Source rocks
There are three basic elements relevant t o hydrocarbon accumulations: (1)

source rock; (2) trap; and (3)reservoir. Before any hydrocarbons can accu-
mulate in commercial quantitites, a source rock must be present. Concepts
of hydrocarbon generation from source rocks have been reviewed by
Hedberg (1964), Erdman (1965), Philippi (1965), Welte (1965), Landes
(1967), Tissot et al. (1971), and Klemme (1972).
Crude oil is liquid in its natural state and is composed primarily of hydro-
carbons often combined with nitrogen, sulfur, and oxygen. Examples of
some hydrocarbons found in petroleum are shown in Fig. 10.1. Natural gas is
a gas in its natural state composed primarily of hydrocarbons often mixed
with carbon dioxide, hydrogen sulfide, and nitrogen. Examples of hydro-
carbons found in natural gas are shown in Fig. 10.2.
Organic matter in rocks usually is divided into the part that is soluble in
common organic solvents and the part that resists these solvents. The non-
soluble organic matter is called kerogen, and the amount of kerogen and
soluble matter can be estimated from their respective carbon contents. The
hydrocarbons in the soluble matter are of primary value in evaluating a
source rock. The evaluation of total organic carbon in rocks can serve as an
ALKANES
c-c-c-c-c-c-c c-c-c-c-c
Porottin I I
C C
Bronchrd Porottin
Monocyclic Bicyclic
Tricyclic
CYCLOPARAFF INS (nophthrnrsl
AROMATICS
Pyrrne
Fig.lO.1. Some of the types of hydrocarbons found in petroleum.

NORMAL PARAFFINS
H H H H H H
I I I I 1 I
H-C-H H-C-C-H H-C-C-C-H
I I I I I I
H H H H H H
Methane Ethone Propane
Fig.lO.2. Some of the hydrocarbons found in natural gas.
indication of a source rock for further estimation of the hydrocarbon poten-

tial of a target area.
Organic matter in all sedimentary rocks is about 2% of the rock mass, and
the amount of trapped oil is about 1.25 x The Clarke for organic
carbon is 1.14% in shales and 0.24% in carbonates (Gehman, 1962). Forsman
and Hunt (1958) found that the ratio of organic material t o organic carbon
in rocks ranges from 1.07 to 1.22. Philippi (1969) found that the ratio
between soluble carbon and total carbon CJCt must be greater than 3 for
shale before it can be construed to be a source rock.
Good source rocks contain more than 130 ppm of petroleumlike hydro-
carbons; fair source rocks, 40-130 ppm; and poor source rocks, less than 40
ppm (Hunt and Meinert, 1954). Philippi (1957) extracted several shale
samples and made a similar conclusion. His ranges were more than 500 ppm
good, 50-500 ppm fair, and less than 50 ppm noncommercial.
Analysis of the gases in cores and cuttings can be used to identify source
rocks (FeugGre and Gkrard, 1970). Hydrocarbons with carbon numbers
below CI4 usually occur in source rocks (Dunton and Hunt, 1962), but the
recent sediments do not contain light hydrocarbons (Erdman, 1967), indi-
cating that long periods of time are necessary in the hydrocarbon generation
process.
Commercial accumulations of oil usually are associated with marine shales
or with organic-rich, fine-grained limestones. A shale commonly contains
0.5% or more organic matter and a limestone 0.2% before oil begins to
occur. The kind and amount of organic matter in a rock indicate its source
potential (Hunt and Meinert, 1954). Less than 5% of the organic matter in
sediments eventually becomes petroleum; therefore, about 95% of it remains
in the source rocks (Hunt and Jamieson, 1956; Philippi, 1957).
Trask and Patnode (1942) studied about 35,000 rock samples and found
that the average amount of organic matter in rocks in close proximity t o
oilfields was about 1.5%. According t o Hunt (1967), more hydrocarbons
appear t o be generated by fine-grained carbonate rocks than by other types
of rocks containing the same amounts of organic matter.
Philippi (1969) concluded that petroleum is generated by chemical
reactions at relatively low temperatures, from source rocks. He further con-
cluded that bacteria alone do not transform organic material into petroleum,
INTRODUCTION 311
and that radioactive bombardment of organic matter contributes little t o the

formation of petroleum. His study of the shales and oil from the Los Angeles
Basin indicated that the Upper Miocene divisions of the D and E shales form
the major source of oil; this was the first documentation of oil source rock
identified by comparing the composition of crude oil and shale hydrocar-
bons. He believes that lipids are the major precursors but that precise data
concerning the chemical conversion of lipids t o petroleum components are
lacking.
Silverman (1964) summarized available data and concluded that petro-
leum is derived from lipids and that the r3C/12C6-values indicate that the
ranges for the lipid fractions of marine and land plants cover the ranges
found in all petroleums. He suggested that polymers are a petroleum hydro-
carbon nrecursor. Organic acids were suggested as hydrocarbon sources by
Cooper and Bray (1963).
Ferguson (1962) derived a quantitative method of determining hydrocar-
bons in sediments, using benzene (rather than mixed solvents), the benzene
is evaporated, and the extract residues are resolved using silica gel chro-
matography. This method was used in an organic geochemistry study of the
Cherokee Group rocks in Kansas and Oklahoma (D.R. Baker, 1962).
A quantitative method using electron spin resonance spectrometry was
used to evaluate the maturation stage of potential source rocks (Pusey,
1973). The method is effective for all types of rocks and all types of
kerogens.
Water and hydrocarbons
Because of the need to locate gas and oil reserves, better hydrogeo-
chemical exploration methods are needed. Water affects the accumulation
and migration of hydrocarbons. Water flowing from compacting sedimentary
rocks moves the oil precursors from source rocks before they concentrate in
a trap area.
Peake and Hodgson (1966, 1967) found that water can accommodate up
to 150 ppm of C,,-C,, hydrocarbons as a fine colloidal suspension. Low-
molecular-weight hydrocarbons such as benzene, methane, ethane, and
propane are somewhat soluble in brines and water; however, the normal
paraffins and cycloparaffins with carbon numbers greater than C8 are rather
insoluble (McAuliffe, 1969), and the insoluble ones are the main con-
stituents of most crude oils. E.G. Baker (1962) postulated that micelles of oil
(colloids) may be transported by water; however, the effects of temperature,
p ~ s s u r e ,and flow through sedimentary rock need further study. Spencer
and Koons (1970) theorized that hydrocarbon precursors may be trans-
ported in water as compounds of nitrogen, oxygen, or sulfur, which later are
reduced t o hydrocarbons. Cartmill and Dickey (1970) stabilized an emulsion
of crude oil in water where the globules averaged 1 p in diameter, and the
emulsion passed through sand but coalesced as the grain size decreased.
According to them, the coalescing process may be an electrochemical rather

than a capillary phenomenon.
Baker (1960) suggested that salts of organic acids form clusters or micelles
in water solution and that these colloidal particles increase the solubility of
hydrocarbons by providing hydrocarbonlike regions within the water. An.
equation was developed which postulates a relation between the abundance
of a hydrocarbon in a crude oil and micellar water solubility.
Fig.lO.3. Illustrations of (a) structural and (b) stratigraphic traps.
The precise manner whereby oil leaves a source rock is not known,but it
apparently is related t o the flow of water from compacting sediments. A
better understanding of the migration process will aid in locating reserves of
oil and especially those in stratigraphic traps or in a combination of strati-
graphic and structural traps. Fig. 10.3 illustrates these types of traps.
HYDROGEOCHEMICAL RESEARCH AND METHODS 313
Hydrogeochemical research and methods
Organic compounds
Saturated hydrocarbons
Davis and Yarbrough (1969)invented a geochemical prospecting method
which involves the analysis of formation waters for the saturated hydrocar-
bons: normal decane, isodecane, butyl cyclohexane, and pentyl cyclohexane.
The presence of these hydrocarbons in the formation water is indicative of
petroleum accumulations.
Buckley et al. (1958)studied the dissolved hydrocarbon gases in waters of
petroleum-bearing strata and determined the “escaping pressure” or the
pressure at which the gas started t o come out of solution. Most of the gas in
solution was methane with lesser quantities of ethane and heavier gases. The
results indicated that the gases generally diffuse into the edgewater rather
short distances from the margins of oil fields.
Kortsenshtein (1965)outlined a general solution for the problem of deter-
mining the oil and gas potential of subsurface strata from data of the gas
saturation of subsurface waters under dephased equilibrium conditions. His
study indicated that water that is saturated or supersaturated with hydrocar-
bon gases can positively be used t o predict whether oil and gas are present in
traps, while waters that are not saturated with hydrocarbon gases cannot be
used to make a positive prediction.
It was observed: (1)that .the direction of increasing saturation indicates
the direction of the accumulation; (2) that a relatively accurate target
location can be obtained by determining where the increasing saturation
pressure intersects by using data from two or more wells; (3) that an increase
in pressure of saturation stratigraphically upward in a well may indicate an
approach t o a gas-water contact; (4)that a constant saturation pressure
laterally presents an unsolvable problem, unless one has data on changes with
depth; and (5)that interpretation of the size of the postulated accumulation
is difficult (Kortsenshtein, 1964).
Continuous investigations made in the oilfields of the Dnepr-Donets
Basin revealed a number of regularities in the composition and degree of gas
saturation of groundwaters. In the presence of oil accumulations, the gas
saturation of waters and the composition of the dissolved gases were deter-
mined with respect to the oil pool in a vertical and horizontal direction. The
regularities of the gas saturation of these waters serve as a basis for oil
exploration (Gutsalo and Krivosheya, 1965).
London et al. (1961)found three classes of gas-water relationships. The
first is represented by formation waters undersaturated with gas in which the
hydrocarbons dominate the chemistry of the water. Commercial gas pools
are not probable under such conditions. Examples of these exist in the
Jurassic and Cretaceous sediments of the West Siberia Lowland. A system in
which predominantly hydrocarbon gases and formation water are in equi-
librium (that is, the gases are at saturation pressure) represents the second
class. The third class is represented by a system in which the pressure of the
dissolved gas is greater than the saturation value, free gas escapes, and the
hydrocarbons are oxidized. Gas pools are likely to be found near the highly
saturated water.
Bond (1962)patented a geochemical exploration method which com-
plements conventional gravimetric, magnetic, seismic, and electrical geologic
survey methods. The technique utilizes the analysis of gas samples, collected
from earth or water samples obtained from locations in the proximity of
hydrocarbon-containing reservoirs. The gas is analyzed t o determine the
isotopic ratio '*C1H4 /13C1H4. If the samples .of oil-gas have not diffused a
considerable distance through the earth, the normal ratio ordinary methane/
heavy methane is in the range of 89.5-93.5. If the gas has diffused up from
a considerable depth, the ratio will be 0.5-2.5 units above normal, which in
turn indicates the presence of petroleum.
Aromatic hydrocarbons
Zarrella et al. (1967)found that the amount of benzene in formation
waters directly reflects the occurrence of a petroleum accumulation in a
formation. They believe that vertical migration of hydrocarbons between
aquifers is restricted and that lateral migration is limited. For example, Fig.
10.4 illustrates how the concentration of a hydrocarbon in a brine may vary
with the distance of the brine from an oil pool.
Zinger and Kravchik (1969)believe that toluene and benzene in water are
hydrochemical indexes which indicate the presence of oil and gas. Accumula-
tions of oil and gas are the principal source of benzene and toluene in oilfield
waters. The ratio benzene/toluene in these waters is greater than 1. Benzene
in waters is considered one of the most important and direct indicators of oil
and gas content (Kortsenshtein, 1968;Kartsev et al., 1969).
If the concentration of a measured aromatic hydrocarbon in a sample of
formation water is equal to a target value, the point from which the sample
was obtained is close to a reservoir of crude oil. Greater differences between
these two values represent greater distance to the crude oil accumulation.
The target value is determined by contacting the type of crude oil expected
in the sampled reservoir with water solutions of salt and measuring the
concentration of the aromatic hydrocarbon in the solutions. This determines
the variation in concentration of this aromatic hydrocarbon as a function of
the salinity of the dissolving water. The salinity of the sample of the water
makes it possible to determine the target value; that is, the value of concen-
tration of this aromatic hydrocarbon existing at the point of contact
between crude oil and formation water (Schmidt, 1970).
daturnted, unsaturated, and aromatic hydrocar'bons

McAuliffe (1969) determined hydrocarbons dissolved in water by gas
chromatographic methods using a number of techniques to separate the
Fig.10.4. Relationship of the concentration of benzene in brines and proximity to an oil

accumulation. (After Zarella, 1969.)
hydrocarbons from water. Separation methods include : (1)direct injection

of the water sample; (2) equilibration of aqueous sample with immiscible
solvent; and (3) equilibration of aqueous sample with gas. Some of the
solubilities found in water were: n-hexane, 10 ppm; n-heptane, 3 ppm;
n-octane, 0.6 ppm; cyclopentane, 156 ppm; benzene, 1,780 ppm; and iso-
propylbenzene, 50 ppm (McAuliffe, 1966). McAuliffe (1967)obtained a
patent for a geochemical method of prospecting for petroleum, which
utilized the concentrations of dissolved hydrocarbons in oilfield brines to
determine the location of a hydrocarbon accumulation.
Fatty acids
Cooper and Kvenvolden (1967)analyzed samples of water from the sub-
terranean formations t o determine the ratios of selected fatty acids. These
ratios indicate the presence of a petroleum reservoir if the ratio of fatty acids
containing an even number of carbon atoms to those containing an odd
number of carbon atoms is not more than 1.6.
Organic acids
Shvets and Shilov (1968) used a quantitative method t o determine water-
soluble organic substances in waters. Waters near oil accumulation deposits
contained up to 3,500mg/l of organic acids. Data on organic carbon, organic
acids, and oil hydrocarbons in waters could be used as indicators of oil
accumulations.
Organic acids dissolved in subsurface waters in the form of salts, or in the
free state, indicate the removal of organic substances from rocks. This
indicates the existence of geochemical processes in which deeply buried
organic substances are being broken by decarboxylation, accompanied by
the formation of hydrocarbons. Thus, organic acids and their salts may be
considered a source for the formation of hydrocarbon accumulations
(Shabarova et al., 1961).
Inorganic and organic compounds
Kolodii (1969) found two types of hydrochemical anomalies in the Pli-

ocene deposits of the South Caspian Basin. The first is associated with struc-
tures in which the red bed part of the producing formation is at a shallow
depth. It manifests itself in sudden increase in dissolved solids in the water
up to 300 g/l. The second anomaly is related t o the lower section and
decreased dissolved solids 5-40 g/l. These waters are often of bicarbonate-
sodium type and were formed by mixing oilfield water with condensate
water which reacted with the host rocks. This hydrochemical inversion is
related to processes accompanying migration of hydrocarbons into the zones
of reduced pressure and temperature. The genetic relationship between
hydrochemical anomalies and the presence of oil (gas) deposits can be used
in oil exploration.
Compared to a general background, methane, heavy hydrocarbons, hydro-
gen sulfide, and carbonic acid concentrations in water increase with the
approach t o a gas and oil accumulation, while the content of sulfates
decreases. The amount of helium in the zone depends upon the difference
between its saturation in the waters and in the accumulation (Savchenko et
al., 1965).
Sudo (1967) studied the major oil- and gasfields of Japan as related to
subsurface water for the exploration of oil and gas deposits. The analytical
data suggested that: (1) the SO,-2 concentration is extremely low, perhaps
resulting from microbiological activity in the initial sedimentary environ-
ment; and (2) dissolved hydrocarbons and naphthenic acid salts in the brine
-
are direct indicators, and I- and HC03 are indirect indicators of oil and
gas accumulations.
Oilfield waters (as well as waters of gasfields and bituminous formations)
are characterized by enrichments of the biophile elements K+, B+3,Br-, and
I-. Alkaline waters with higher concentrations of carbonates and sulfates,
and with naphthenic acids, are associated with naphthenic oils. Alkaline-
HYDROGEOCHEMICAL RESEARCH A N D METHODS 317
earth chloride waters with low-numbered fatty acids are associated with
paraffinic oils ( Krejci-Graf, 1962).
In0 rganic compounds
Serebriako and Tronko (1969) found that the ammonium content of

subsurface waters correlates with hydrocarbon deposits. The amount of
ammonium found was above 80 mg/l in oil- and gas-bearing areas with lesser
amounts in nonproductive areas.
Korobov (1965) determined that the main source of lithium, strontium,
barium, manganese, copper, chromium, and aluminum in oilfield waters is
related to the geochemical and biochemical environments which produced
the oil and gas deposit accumulations. The elements potassium, boron,
barium, iodide, strontium, barium, and gallium are enriched in oilfield waters
(Krejci-Graf, 1962) and are an indication of bituminiferous formations.
The first transition series metals - iron, cobalt, manganese, nickel, vana-
dium, titanium, chromium, and scandium - are soluble in the reduced state
but insoluble in the oxidized state and are useful in determining the proximi-
ty of petroleum deposits according t o a patent issued to Billings (1969). The
method involves analyzing formation waters for one or more of the transi-
tion metals to determine a trend in the amounts of the metal in an area.
According to London (1964), the main indication of the presence of
oil-gas in rocks is the presence of hydrogen sulfide and biogenic nitrogen,
and the degree of sulfate reduction. The author recommends using the
degree of sulfate reduction as a reliable exploratory criterion. The sulfate
concentration decreases in the direction toward an oil accumulation. Sulfates
cannot be regarded as reliable criteria if the water-bearing formation contains
salt and gypsum.
The single most important common ground between hydrodynamics and
log interpretations is the determination of formation resistivity (Rw ), which
is related to the chemical makeup of the formation water. In waters with a
high chloride-ion content, the chloride ion is (for all purposes) the
resistivity-determining ion; but in cases where low-chloride salinity waters
are present, other ions, particularly calcium and magnesium, can make
material changes from the chloride-calculated resistivity. Chemical content
and its variations can be used directly as a tool for finding oil. Variations in
the chemical makeup of formation waters can be used to delineate separate
reservoirs within the same formation, thereby indicating potential strati-
graphic traps. The production engineer also benefits from detailed chemical
analysis (Schwab, 1965).
Radioactive compounds
Filonov (1969) investigated radioactive elements in underground waters of
the Devonian deposits in the Pripvatsky Depression. He found a uniformity
in distribution of uranium and radium and a correlation between the increase
in dissolved solids of the waters and the increase in their radium content. He
concluded that the radium concentration may be used as an indirect in-
dicator of oil.
Gutsalo (1967) studied the geochemical relationship between radium
anomalies and oil and gas deposits. In waters of t h e Dnepr-Donets Basin
associated with hydrocarbons, the amount of Ra concentration depends on
the concentration of dissolved solids and the age of the water-bearing rocks.
Positive hydrogeochemical anomalies exist within the areas of hydrocarbon
influence. The Ra anomalies are related t o the gas component of waters; Ra
concentration is dependent on the partial pressure of hydrocarbons dissolved
in water, which usually amounts t o more than 90%of the total gas pressure.
The Ra concentration in these waters has a direct linear relationship to the
hydrocarbon pressure. The Ra anomalies in the ground waters of the
Dnepr-Donets Basin, near the gas-water or oil-water contacts, appear to
owe their origin t o the hydrocarbon gases, which are genetically related t o
the hydrocarbon deposits.
The formation of radioactive anomalies on the earth's surface, above oil
and gas deposits, is related t o the migration of hydrocarbons, salts, and ions
in water solutions. It is assumed that the radioactive substance is carried
toward the surface by ascending waters and hydrocarbons. The migration of
the water and hydrocarbons is controlled by the permeability of the rock
strata. The infiltration of gaseous hydrocarbons into the surface zone may
increase evaporation, which leads to the increased migration of the water
from surrounding areas. This water may be capable of dissolving radio-
elements, redepositing them in these places of greatest evaporation, which
leads to the increased migration of water from surrounding areas. Liquid
hydrocarbons, because of fractionation, may deposit bituminous material
which is able to extract radioactive material from ground waters. There are
two schools of thought on the matter of formation of surface radioactive
anomalies. They are: (1) the radioelements have a surface origin; and (2) the
radioelements have a deep-seated origin (Sikka, 1963).
The deep-seated origin is substantiated by MacElvain (1963). In fact, he
concluded that the determination of *l0Pband 206 Pb in near-surface samples
should be more rewarding in oil exploration than the beta radiation techniques.
Furthermore, the 210Pb/206Pbratio might indicate the age of the underlying
hydrocarbon accumulation.
The amount of gases in surface soils is a function of the weather and is
influenced by temperature, rain, and wind. Therefore, geochemical explora-
tion techniques using soil samples must be designed so that samples are taken
from a depth that is not influenced by the weather (MacElvain, 1963).
Filonov (1964) studied the radium and uranium contents in different
oilfield waters occurring in the platformal and geosynclinal deposits, in
particular, the variations of the Ra/U ratio. The results revealed: (1) that
oilfield waters contained an increased Ra content and that chloride-calcium
types had a higher Ra concentration than the bicarbonate-sodium waters; (2)
HYDROGEOCHEMICAL RESEARCH A N D METHODS 319
that the water enrichment in Ra occurred .because of leaching from the

reservoir rocks and because of the radioactive decomposition of uranium; (3)
that uranium was either absent or present in insignificant amounts in the
bicarbonate-sodium waters, where it occurred in the form of a complex
anion [(UOzC03)3]*; (4) that the Ra/U ratio in the waters a t the oil-
water contacts was shifted in the direction of increasing R a content; and (5)
that the organic matter in oil had a definite effect on the shift of the Ra/U
ratio because it removed uranium from the salt components dissolved in
water, and supplied additional quantities of Ra because of the radioactive
decomposition of uranium in oil.
A linear relationship between .the radium concentration and dissolved
solids concentration of some water from Paleozoic and Mesozoic age strata
was found (Gutsalo, 1964). Waters in contact with hydrocarbon accumula-
tions were more highly concentrated with radium.
Vilonov (1962) investigated water-oil contact zones in several of the
Soviet producing regions and found that a definite regularity in the distribu-
tion of radioactive elements exist. It was found that, in formation waters of
oil accumulations, there is no definite relationship between the radium and
the uranium content, and that usually uranium and thorium are present only
in small quantities. The influence of oil occurrence on the distribution of
radioactive elements in the formation waters of the oil accumulations
appears most clearly in the zone of the water-oil contact; radium and its
isotopes are concentrated in the water of this zone, and there is a relative
decrease of uranium content, sometimes to zero.
Gutsalo (1969) concluded’ the following:. (1) enrichment of chloride-
calcium brines with helium results from migration through rocks and is
directly proportional t o the amount of radium in solution - radium leached
from the rocks by the underground water determines the amount of helium
lost from the rocks; (2) a positive helium anomaly is formed in underground
water around oil and gas accumulations; (3) the helium anomaly coincides
with a positive radium anomaly; (4) the absolute value of the helium concen-
tration at any point in the anomaly zone is dependent upon the radium
concentration at the point and the time of formation of the accumulation;
and ( 5 ) for any deposit of oil or gas, the areal extent of the positive helium
anomaly in the formation water is dependent upon the value of the ratio of
the partial pressure of the hydrocarbons to the total gas pressure.
Physical properties
Vdovykin (1963) determined the Eh and pH of formation waters and

waters from rivers, lakes, and seas. Fig. 10.5 illustrates the results that he
obtained. Unfiltered formation waters from petroleum producing wells had a
lower Eh and a higher pH than waters taken from rivers, lakes, and seas.
Filtration of the formation waters prior to analysis fof Eh and pH caused
higher Eh and pH readings.
600
SO0 - I
I
I
I
400 - I
I
I
I
f
> 300
200
- 0I
I
---- 1-------
II Q-
(3
100
0
- I I I I I
6 7 8 9 10 I1
PH
Fig.lO.5. Approximate pH and Eh of waters from unfiltered petroleum producing wells

(A); filtered petroleum producing wells ( B ) ;and surface rivers, lakes, and seas (C). (After
Vdovykin, 1963.
Fluid mechanics
Roach (1965) describes how to apply fluid mechanics t o petroleum

exploration. The definition of fluid mechanics as he used it encompasses
complete study of subsurface fluids including physical and chemical charac-
teristics, whether hydrodynamic or hydrostatic, and a complete study of the
characteristics of the reservoir rock.
Maps
Chloride-ion concentrations in water produced from rocks of various ages

and depths were mapped in Lea County, New Mexico, using machine map-
plotting techniques and trend analyses. Anomalously low chloride concen-
trations (1,000-3,000 mg/l) were found along the western margin of the
Central Basin Platform in the San Andres and Capitan Limestone formations
of Permian age. These low chloride-ion concentrations may be caused by
preferential circulation of ground water through the more porous and
permeable rocks (Hiss et al., 1969).
Hanshaw and Hill (1969) studied aquifer systems from: (1) Mississippian
age rocks; (2) Pinkerton Trail Limestone; (3) Paradox member of the Her-
mosa formLtion; (4)Honaker Trail formation; and (5) Permian age rocks.
Recharge in the Paradox Basin occurs on the west flank of the San Juan
Mountains and along the west side of the Uncompahgre Uplift. A series of
potentiometric surface maps were prepared for the five systems studied.
With a few exceptions, most wells in formations above the Pennsylvanian age
strata contain fresh to moderately saline water. Much of the strata below the
Permian age rocks contained waters with dissolved solids concentrations
greater than 35,000 mg/l and some areas favorable for hydrocarbon accumu-
lations. Some of the brines in the Paradox formation contained up to
400,000 mg/l of dissolved solids. Cambrian age strata in much of Colorado is
favorable for the accumulation of hydrocarbons.
Chemical analyses of water from five Cretaceous aquifers were used to
compute ion ratios, which were used in conjunction with structural and
stratigraphic information t o interpret hydrologic conditions in the East
Texas Basin. Ion ratio comparisons made by maps and diagrams show that
the aquifers contain water of distinctive character, and that there are inter-
connections between aquifers, especially near the Mexia-Talco Fault zone
and the Sabine Uplift. A hypothesis is offered that water moves along an
unconformity from the Sabine Uplift eastward toward the East Texas
oilfield where it enters the Woodbine Sandstone. Ion-ratio maps show the
effect of time and of rock composition upon the relative kind and amount of
dissolved solids in the water because- of reactions with minerals and organic
material in the rocks. The hydrodynamic component of the water in the
Woodbine formation from east to west helped form and contain the giant
East Texas oilfield (Parker, 1969).
Karim et al. (1966) studied three exploratory wells drilled on the east
plunge of the Cordillera Isabella, Nicaragua, and all three had gas shows. A
stratigraphic cross section and a localized map showing the relationship of
magnetic highs obtained from an aeromagnetic survey and results of
fluoroanalysis and water-gas surveys are included. The prospect of finding
petroleum in coastal northeastern Nicaragua appears fair.
Water analysis integrated with the existing knowledge of the geologic
framework of an area provides supplementary information t o assist the
exploration geologist in solving geologic problems on both a local and a
regional scale. Isoconcentration maps showing regional variations in the total
solids content of the waters within a given stratigraphic unit are important.
Inorganic water analyses data are useful in the correlation of porous zones,
and benzene analysis is a promising hydrocarbon exploration tool (Noad,
1966).
Maps were prepared delineating: (1) surface outcrop areas of Upper
Cretaceous rocks; (2) axial lines of Upper Cretaceous anticlinal structures;
(3) zones of water types; (4)zones of water groups; and (5) zones of mixed
waters. A diagram of chemical composition of the waters was also prepared.
Using these maps, Galin and Plyushchenko (1963) selected areas in Dagestan
that are favorable for the accumulation of oil and gas.
Reviews
Gerard and Feug&re (1969) concluded that geochemical exploration

techniques are useful in offshore areas. Kroepelin (1967)reviewed geo-
chemical prospecting as applied to petroleum and found that several com-
panies had success ratios up to 59% with it in exploration. Johnson (1970)
states that success ratios of about 35% can be attributed to the use of an
inorganic technique of prospecting for oil and gas. The method is useful in
locating stratigraphic trap accumulations and is based on the postulate that
heavy-metal salts concentrate in the soil profile as a result of vertical migra-
tion of waters above an accumulation of oil.
According to Boyle and Garrett (1970),“geochemical prospecting will
play an ever increasing role in the discovery of hidden ore deposits and
accumulations of hydrocarbons. The methods need no longer be sold since it
has now been generally recognized that they are the only direct approach to
the problems of mineral, oil, and natural gas exploration. The methods are
direct and have generally proved most successful when applied in con-
junction with geological and geophysical exploration techniques.”
Geochemical methods applied t o petroleum prospecting have not reached
their potential, expecially in the United States.
Karaskiewicz (1966)discussed the geomicrobiological and hydrochemical
research done by the Polish Petroleum Institute in 1962-63 in the Lubel-
Nadbuzan region. Specific methods such as the determination of gas dis-
solved in water were applied in this area, and 215 water samples were studied
from the 9,500-km2 area. Anomalies produced by the bacterial activities
which oxidize methane, propane, and butane were determined. Where
methane was present, the bacterial microflora activity was lowest; where the
biological activity was high, methane was absent.
Important aspects of geochemical prospecting are : (1)analytical methods;
(2)transport mechanisms of the hydrocarbons; (3)anomalies associated with
hydrocarbon accumulations; (4) statistical treatment of the data; and (5) the
final result. Significant findings are mapped and interpreted to locate a target
area for drilling (Kroepelin, 1967).
Case study of the Delaware sand (Bell Canyon formation), Texas, by Visher
(1961)
Preliminary work was carried out on the relation of sapropelic material

found in both dark shales and laminated silts, and the oil occurring in reser-
voir rocks. The study indicated that the frequency distribution of the
radicals, - benzene, straight chain, CH3, and CH2 -, was identical in both
the black shales and the residual crude present in oil saturated reservoir
rocks. The entire sequence of Permian rocks from the Bone Springs Lime-
stone through the Lamar Limestone member of the Bell Canyon formation is
composed of dark organic sediments, deposited in a reducing environment.
CASE STUDY OF THE DELAWARE SAND 3 23
The total quantity of the sapropelic material available for the formation of
oil was not determined, but 1%would be a conservative estimate, indicating
that source material in the Delaware Basin is of sufficient quantity t o
produce volumes of oil greater than presently discovered. The oilfields which
have been discovered in the basin primarily are confined t o the upper porous
and permeable sands of the Bell Canyon formation. The present distribution
of oil appears t o be controlled by hydrodynamic conditions. Therefore, the
tracing of the times and paths of migration is dependent upon reconstructing
the paleohydrodynamics.
Potentiometric surface of the upper Delaware sand
A potentiometric surface map of the Bell Canyon formation was made

from a two-dimensional electric analog model of the central and eastern
portions of the Delaware Basin in West Texas (Fig. 10.6). The majority of
pressures were bottom-hole measurements from existing fields. Only a few
shut-in pressures from drill-stem tests were usable because of the short shut-
in time commonly used in this area, and consequently, few tests reached true
formation pressures. A pressure buildup method should have been used.
The total dissolved solids of the formation water range from a low of
90,000 mg/l in the Ford field to over 250,000 mg/l at the South Pyote field.
All pressure readings were corrected for effects of varying total dissolved
solids. The assumption was made that the concentrations of total dissolved
solids are stratified with little mixing. Therefore, the weighted mean is
between 50,000 and 120,000 mg/l.
The potentiometric surface has a hydrodynamic gradient from west to
east with a component of northward flow. In the southeastern portion of the
mapped area, the hydrodynamic gradient is reversed because of the influence
of the eastern flank of the basin.
Stratigraphic traps are formed in areas where linear sand fingers show an
updip decrease in permeability and porosity. The change in permeability and
porosity between the permeability barrier and adjacent reservoir rocks, how-
ever, is not great. The Saber field, for example, has an average porosity of
25% and permeability of 70 md, and the barrier rock an average porosity of
12% and permeability of 3 md. Since in some areas this barrier rock would
be considered a possible reservoir, something in addition to these changes in
porosity and permeability is necessary to prevent the movement of oil into
the barriers.
Under equilibrium conditions water flowing through a formation will have
a greater pressure gradient across a tight zone than a more permeable zone
(see Fig.lO.7). Therefore, the differential pressure in the barrier is greater
than in the reservoir, and varies directly with the decrease in permeability
between the two. When the oil phase reaches the barrier zone, the pressure
gradient increases updip, making i t increasingly more difficult for the oil to
-z-
I
I
CASE STUDY OF THE DELAWARE SAND 325
246 k g A q cm
Fig.10.7. Relationships of hydrodynamic gradients to permeability; I = decreased hydro-

dynamic gradient because of increased permeability; 2 = increased hydrodynamic gradient
because of decreased permeability; 3= average hydrodynamic gradient; 4 = decreased
hydrodynamic gradient.
invade the water-filled rock pores. Finally, the entry pressure of the barrier
rock is greater than the invading force of the oil and migration ceases.
The downdip hydrodynamic flow increases the efficiency of the trap by
reducing the buoyancy effect of the oil. The hydrodynamic enforcement of
stratigraphic traps can increase the oil column many times over what it
would be under hydrostatic conditions and probably accounts for the devel-
opment of commercial stratigraphic oil accumulations in the Delaware Basin.
Formation waters
The initial approach in the study of stratigraphic problems within the Bell
Canyon formation was by the use of formation waters. Over 300 samples of
formation water were collected, analyzed, and processed by computer tech-
niques. Data collected in this manner were posted on maps (Fig.10.8-10)
and contoured. Several aspects of the waters (relative concentration percent-
ages of SO4,Mg, Ca, and total solids) show systematic variations over the
basin. Variation in these parameters is related t o proximity to outcrop and
the degree of transmissibility of the formation. The highest sulfate content is
near the outcrop belt t o the west; the calcium and total dissolved solids
concentrations increase toward more impermeable rocks and areas of low
circulation. In the center of the basin, the waters are characterized by very
high total dissolved solids, high calcium, and low sulfate content, but in the
porous and permeable fingers near the outcrop, salinites are low and sulfate
is high. All gradations between these two extremes exist in the Bell Canyon
formation. In areas where the sand fingers pinch out very rapidly into low-
permeability sediments, the transition between these two extremes of water
composition may take place in a matter of a few well locations. An excellent
3 26 EXPLORATION FOR PETROLEUM AND GAS
example of this is the Saber field (Fig.10.8-10) where a range of waters is

evidenced in one homogeneous, continuous sand body.
The reason for these rapid changes in formation water compositions may
be explained by permeability changes within the Bell Canyon formation. In
areas of low permeability there is less circulation, less dilution, and more
chance for the maintenance of an equilibrium relation between formation
water and sediment. This was substantiated by the distribution of magne-
sium in the waters.
A series of multiple regression analyses was made on the relation of
various dissolved ions in the waters to their total dissolved solids. First, all
the waters were analyzed as a unit to determine the correlation coefficients
and the degree of variability explained by the chosen ions. The second stage
was the breakdown of the waters into three arbitrarily defined groups (based
principally on salinities) t o see if there were any noticeable changes in either
correlation coefficients or degree of explained variance. The only significant
change was in the relation of magnesium t o total solids. In those waters
containing relatively low concentrations of total dissolved solids, there was a
significant positive correlation, but in those with high concentrations of total
dissolved solids, there was a significant negative correlation. This indicates
that the relative concentration of magnesium decreases in waters of high
total dissolved solids. These waters are precisely those that are found in
low-permeability, fine-grained, argillaceous rocks in which magnesium would
most likely be taken out of waters by diagenetic alteration of clay minerals.
The variations found in the formation waters within the Bell Canyon
formation can be used as the basis of an exploration technique. Since the
composition of formation waters is related t o permeability, and permeability
is related t o producibility of reservoir rocks, a workable relation exists
between exploration objectives and water compositions. The refining of the
maps of the distribution of the composition of the waters aids in defining
the distribution of the sand fingers. The updip edges of the permeable sqnd
fingers show increased concentrations of total dissolved solids and decreased
magnesium which are related to the presence of a “barrier” (or trap) updip
from the reservoir sands.
Formation water maps
Maps of the total dissolved solids content (Fig. 10.8), the chloride content
(Fig. 10.9), and the calcium content (Fig. 10.10) of formation waters were
prepared. The inference which may be drawn from these maps is the empiri-
cal association of oilfield occurrence versus the iso-mg/l contours of the
various constituents. This empiricism shows some remarkable alignments and
permits formation-water composition maps to be added to the suite of ex-
ploration tools. Some of the subtleties of constituent composition versus
rock properties do not lend themselves readily t o mapping techniques but
are useful for consideration (i.e., magnesium content versus low permeability
I I
I !
d
W
Y
5
33 0 EXPLORATION FOR PETROLEUM AND GAS
and fine-grained rocks, sulphate concentration variation with relationship to

outcrop, etc.).
It is important to note on each of the constituent maps that the local
variations of the iso-mg/l contours are of greatest importance and not the
precise value of the contour. For example, on the total dissolved solids map
(Fig. 10.8) the concentration in the Ford field is only 50,000 mg/l and
ranges in a re-entrant to about 150,000 mg/l, while in the Wheat field, the
range is from 150,000 mg/l to nearly 250,000 mg/l.
The overall appearance of this map (Fig. 10.8) is a series of fingering
expressions. The various oilfields seem to have an occurrence relationship in
the transition zone from higher to lower concentration. The Mason, Tunstill,
Olds, and Saber fields occur along a transition zone from 250,000 mg/l to
50,000 mg/l. The El Mar and Grice fields are on a transition zone from
250,000 mg/l to 150,000 mg/l. The Two Freds field has a transition zone
from 250,000 mg/l to about 150,000 mg/l. The Wheat and also the Ford
fields occur in a similar transition zone. This relationship of oil occurrence
and differential concentration is of great significance. Even in this limited
area it appears that in the block from longitude 103”30’00” to 103”45‘00‘’
and from latitude 30’45’00” to 32’00‘00” there are several places which,
from this empirical relationship, have some potentialities. The block immedi-
ately south of this, trending southwesterly from the Wheat and also from the
Two Freds fields, needs additional study for better delineation. The area east
of the Two Freds field lacks adequate control but basically shows the
possibility of favorable development.
The chloride map (Fig.10.9) has a configuration similar to the total solids
map. Again it is not the precise iso-mg/l contour which is of prime concern
but the variation in the limited area. This rate of change from higher to
lower concentration appears to be a principal key to occurrence.
The calcium content map (Fig. 10.10) does not show the prominent
fingering, almost pseudodeltaic, effect that the total solids and chloride maps
have. Perhaps this is because of the smaller range of values mapped. Some of
the high to lower concentration effect is present and in other areas, the
Wheat and Two Freds fields, the iso-mg/l closure is developed. This map is
less diagnostic than the others; however, considered in conjunction with the
other two maps, the coexistence of accumulation and the transition zone,
even closure cannot be missed.
Formation water maps of other areas
Fig. 10.11 is a potentiometric surface map of the Arbuckle formation

group in parts of Kansas, Missouri, and Oklahoma drawn by Chenoweth
(1964). As ,noted on the figure, the arrows indicate the theoretical direction
of the water flow which was inferred from the tilted oil-water interfaces.
Detailed pressure data along with reservoir transmissibility data could be
used to construct a similar map for determining detailed flow pattern.
FORMATION WATER MAPS 331
Kilometers
0 55.60
M
I:l,wom
LEGEND
drrpru indicate ItmoretlcoI dimction of
artasion flow. Infarred tmm tilted oil
w01.r i"l,,t.aCe. in B"ter,Ellswrth,
ord Pownee Counties. Kansas.
Arbuckle b n t . " R e W o n amd'or
Gmnite Wash beneath hnnaylvonion.
Ellia.Rush and Barton Countlea.
Kanroa .
Fig.lO.11. Potentiometric surface map of the Arbuckle formation group in South Kansas,
North Oklahoma, and southeast Missouri.
The map shown in Fig.10.12 also was constructed by Chenoweth (1964)

and it is a chloride map of the Arbuckle formation group in Kansas and
Oklahoma. Note the dilute brines near the outcrop areas, which are diluted
by meteoric recharge waters entering the outcrop. As a general rule the
trapped petroleum in this group of formations is found associated with the
more saline brines, and in the transition areas.
Fig. 10.13 is a map that illustrates the variation in salinity at the bottom
of the lower Wilcox formation in portions of Texas, Louisiana, Arkansas,
Mississippi, and Alabama. The most saline brines occur in the deeper basin
areas with the dilute brines nearer outcrop areas. Fig. 10.14 is a similar map
which illustrates the salinity variations in the top of the lower Wilcox forma-
tion in the same area. Fig. 10.15 illustrates the salinity variations at the base
Fig.10.12. Map of the chloride concentrations (mg/l) in the Arbuckle formation waters in
Kansas and Oklahoma
of the upper Wilcox formation, while Fig. 10.16 illustrates the salinity varia-
tions at the top of the upper Wilcox formation. Fig. 10.17 is a salinity map
of the lower Tuscaloosa and Woodbine formations, again most of the trap-
ped petroleum is found in areas where the more saline waters occur and in
transition areas.
Salinity maps are useful as a primary tool in petroleum exploration be-
cause they provide information concerning sand fingering, diagenetic changes
that affect reservoir and source rocks, and stratigraphic traps. Occurrences of
petroleum accumulations often correlate with salinity transition zones, i.e.,
where the salinity ranges from 50,000 mg/l to 100,000 mg/l.
I I
EEL‘ S d V N IZIBLVM NOILVNIZIOd
Fig.10.15. Salinity concentrations in waters taken from the base of the Upper Wilcox
formation in portions of Texas, Louisiana, Arkansas, and Mississippi.
Less than 5,600
Fig.lO.16. Salinity concentrations in waters taken from the top of the Upper Wilcox
formation in portions of Texas, Louisiana, Arkansas, Mississippi, and Florida.
CONCLUDING REMARKS 335
LEGEND
II Greater than im,ooo mg/l (as NOCI) 5.600 lo 70.000 mp/l

Less than 5.600 mg/I
Fig.lO.17. Salinity concentrations in waters taken from the Woodbine (Dexter) and
Lower Tuscaloosa formations in .portions of Texas, Oklahoma, Arkansas, Louisiana,
Mississippi, Georgia, South Carolina, North Carolina, and Florida.
Concluding remarks
Organic acid salts, petroleum hydrocarbons, and other organic compounds

are soluble in water. The ionic composition, the pH, and the Eh of the water
influence the solubilities of the organic compounds. The aqueous solubility
of petroleum hydrocarbons increases with increasing temperature and pres-
sure, and decreases with increasing water salinity. The aqueous solubility of
organic acid salts increases with increasing pH.
A mechanism for the migration of petroleum or petroleum precursors,
therefore, is water. It is known that petroleum hydrocarbons are generated
from organic-rich rocks. The organic material in the petroleum source rocks
is transformed by physicochemical reactions into petroleum precursors
and/or hydrocarbons which are solubilized by water. The water phase moves
the solubilized organics from the source to the reservoir where, because of
temperature, pressure, salinity, pH, filtration, or organic salting-out
phenomena, the organic phase separates from the water.
In the reservoir the petroleum precursors and/or hydkocarbons mature to
crude oil and gas, primarily because of temperature and time. Thermal alter-
ation proceeds both in the fine-grained source rock and in the reservoir at
temperatures above 115OC by abiogenic reactions. With increasing tempera-
ture the quality of the crude oil improves; however, at higher temperatures
the crude oil is destroyed, leaving methane and pyrobitumen.
The primary mechanism in the migration of petroleum involves water,
therefore, it follows that knowledge of certain characteristics of the water is
useful in exploration for oil and gas. The chapters “Classification of oilfield
waters,” and “Some effects of water upon the generation migration, accumu-
lation, and alteration of Petroleum” discuss some of these characteristics.
Fig. 10.18 illustrates some characteristics related to waters that are likely to
indicate an oil or gas accumulation, and some characteristics related t o
waters that are likely to indicate a dry reservoir.
/-2 - C a type water
(a)
Fig.lO.18. Genetic indicators in a water associated with an oil and gas accumulation (a)
compared to indicators in a water associated with a dry reservoir (b).
REFERENCES 337
References
Anonymous, 1959. Is geochemistry worth the effort. Pet. Week, 23-1-1959:20-24.

Anonymous, 1960a. Geochemistry to get full-fledged test as exploration technique. Pet.
Week, 2-12- 1960 :34-35.
Anonymous, 1960b. 10.9% of U.S. wildcats produce in 1959. World Oil, 1 5 0 : l l l .
Baker, D.R. 1962. Organic geochemistry of Cherokee group in southeastern Kansas and
northeastern Oklahoma. Bull. A m . Assoc. Pet. Geol., 46:1621-1642.
Baker, E.G., 1960. A hypothesis concerning the accumulation of sediment hydrocarbons
t o form crude oil. Geochim. Cosmochim. Acta, 29:309-317.
Baker, E.G., 1962. Distribution of hydrocarbons in petroleum. Bull. A m . Assoc. Pet.
Geol., 46 :76-84.
Billings, G.K., 1969. Geochemical petroleum exploration method, U.S. Patent,
No.3,428,431.
Bond, D.C., 1962. Geochemical process. U.S. Patent, No.3,033,287.
Boyle, R.W. and Garrett, R.G., 1970. Geochemical prospecting - a review of its status
and future. Earth-Sci. Rev., 6:51-75.
Bronston, A., 1947. Geochemical data forecast major field. World Oil, 127:112-116.
Buckley, S.E., Hocutt, C.R. and Taggart, M.S., 1958. Distribution of dissolved hydrocar-
bons in subsurface waters. In: L.G. Weeks (Editor), Habitat o f Oil. American Associa-
tion of Petroleum Geologists, Tulsa, Okla., pp.850-882.
through porous media. Bull. A m . Assoc. Pet. Geol., 54~2438-2447.
Chenoweth, P.A., 1964. Potentiometric Surface Map and Salinity Map o f the Arbuckle
Formation Group in South Kansas, North Oklahoma and Southeast Missouri. Sinclair
Oil Company, unpublished.
Cooper, J.E., 1962. Fatty acids in recent-and ancient sediments and petroleum reservoir
waters. Nature, 193:7 44-7 46.
Cooper, J.E. and Bray, E.E., 1963. A postulated role of fatty acids in petroleum forma-
tion. Geochim. Cosmochim. Acta, 27:1113-1127.
Cooper, J.E. and Kvenvolden, K.A., 1967. Method for prospecting for petroleum. US.
Pa ten t , No. 3,30 5,317.
Davidson, M.J., 1963. Geochemistry can help find oil if properly used. World Oil,
157:94-106.
Davis, J.B. and Yarbrough, H.R., 1969. Geochemical exploration. U.S. Patent,
No.3,457,044.
Dunton, M.L. and Hunt, J.M., 1962. Distribution of low-molecular-weight hydrocarbons
in recent and ancient sediments. Bull. A m . Assoc. Pet. Geol., 46:2246-2248.
Erdman, J.G., 1965. Petroleum - its origin in the earth. In: A. Young and J.E. Galley
(Editors, Fluids in Subsurface Environments. American Association of Petroleum
Geologists, Tulsa, Okla., pp.20-52.
Erdman, J.G., 1967. Geochemical origins of the low-molecular-weight-hydrocarbon con-
stituents of petroleum and natural gases. In: Proceedings 7th World Petroleum Con-
gress, Mexico, D.F. Elsevier, London, 2:13-24.
Ferguson, W.S., 1962. Analytical problems in determining hydrocarbons in sediments.
Bull. A m . Assoc. Pet. Geol., 46:1613-1620.
Feugere, G. and Gerard, R.E., 1970. Geochemical logging - a new exploration tool.
World Oil, 170:37-40.
Filonov, V.A., 1964. Influence of organic material in oil on the shift of radioactive
equilibrium in the watersof oil deposits. Sou. Geol., 6:144-146 (in Russian).
Filonov, V.A., 1969. The problem of use of radioactivity of underground waters as an
indirect hydrochemical indicator of oil content. Nauch. Tekh. Sb. Ser. Neftegazov.
Geol. Geofiz., 3:32-35 (in Russian).
Forsman, J.P. and Hunt, J.M., 1958. Insoluble organic matter (kerogen) in sedimentary
rocks of marine origin. In: L.G. Weeks (Editor), Habitat of Oil. American Association
of Petroleum Geologists, Tulsa, Okla., pp.747-778.
Galin, V.L. and Plyushchenko, V.G., 1963. Hydrogeology of the Upper Cretaceous
deposits of Dagestan as related to their oil- and gas-bearing properties. Zzu. Vyssh.
Uchebn. Zaued. Geol. Razued., 4:120-217 (in Russian).
Gehman, Jr., H.M., 1962. Organic matter in limestones. Geochim. Cosmochim. Acta,
26 :885-899.
Gerard, R.E. and Feug&re, G., 1969. Results of an experimental offshore geochemical
prospection study. In: P.A. Schenck and I. Havenaar (Editors), Advances in Organic
Geochemistry. Pergamon Press, New York, N.Y., pp.355-372.
Gutsalo, L.K., 1964. Some regularities of radium distribution in underground waters of
middle part of Dnepr-Pon Depression. Geokhimiya, 12:1305-1312 (in Russian).
Gutsalo, L.K., 1967. Geochemical relation of radium anomalies in groundwaters to oil
and gas deposits. Dokl. Akad. Nauk S.S.S.R., 172:1174-1176 (in Russian).
Gutsalo, L.K., 1969. The nature and regular features of distribution of helium anomalies
in underground waters adjoining oil and gas deposits, Sou. Geol., 8:112-123 (in
Russian ).
Gutsalo, L.K. and Krivosheya, V.A., 1965. Certain regularities in the gas-saturation of
groundwaters associated with oil-gas-bearing structures in the central DDB (Dnepr-
Donets Basin) and their significance for oil-gas prospecting. Geol. Nefti Gaza, 3:51-53
(in Russian).
Hanshaw, B.B. and Hill, G.A., 1969. Geochemistry and hydrodynamics of the Paradox
Basin region, Utah, Colorado, and New Mexico. Chem. Geol. 4:263-294.
Hedberg, H.D., 1964. Geologic aspects of the origin of petroleum. Bull. A m . Assoc. Pet.
Geol. 48:1755-1803.
Hiss, W.L., Peterson, J.B. and Ramsey, T.R., 1969. Saline water in southeastern New
Mexico. Chem. Geol., 4:341-360.
Horvitz, L., 1969. Hydrocarbon gqochemical prospecting after thirty years. In: W.B.
Heroy (Editor), Unconventional Methods in Exploration f o r Petroleum and Natural
Gas. Southern Methodist University, Dallas, Texas, pp. 205-21 8.
Hunt, J.M., 1967. The origin of petroleum in carbonate rocks. In: G.V. Chilingar, H.M.
Bissel and R.W. Fairbridge (Editors), Developments in Sedimentology, 9B. Carbonate
Rocks -Physical and Chemical Aspects. Elsevier, Amsterdam, pp.225-251.
Hunt, J.M. and Jamieson, G.W., 1956. Oil and organic matter in source rocks of petro-
leum. Bull. A m . Assoc. Pet., Geol., 40:477-488.
Hunt, J.M. and Meinert, R.N., 1954. Petroleum prospecting. U.S. Patent, No.2,854,396.
Johnson, A.C., 1970. How to hunt oil and gas using the inorganic surface-geochemical
method. Oil Gas J., 68:llO-112.
Karaskiewicz, J., 1966. Forecasting of oil- and gasbearing capacity in the light of geo-
microbiological and geochemical investigations. Nafta, 22 (12):19-20 (in Polish).
Karim, M, Chilingar, G.V. and Hoylman, H.W., 1966.Northeast Nicaragua has gas and oil
indications. World Oil,162:84,86, 91,92,94,96.
Kartsev, A.A., Dudova, M. Ya. and Diterikhs, O.D., 1969. Benzene homologs in under-
ground waters and their relation to oil. Geol. Nefti Gaza, 7:41-45 (in Russian).
Kartsev, A.A., Tabasaranskii, Z.A., Subbota, M.I. and Moglevskii, G.A., 1959. Geo-
chemical Methods of Prospecting and Explomtion f o r Petroleum and Natural Gas
(English transl. edited by P.A. Witherspoon and W.D. Romey) University of California
Press, Berkely, Calif., 349 pp.
Klemme, H.D., 1972. Geothermal gradients. Oil Gas J . , 69:136,141-144; 70:76-78.
Kolodii, V.V., 1969. Origin of some hydrochemical anomalies in petroleum provinces.
Nauch. Tekh. Sb. Ser. Neftegazou. Geol. Geofiz., 8:37-40 (in Russian).
REFERENCES 339
Korobov, D.S., 1965. Distribution of trace elements in water and rock of oil deposits in
the Saratov-Volgograd region of the Volga and its significance in petroleum explora-
tion. In: Soviet Advances in Nuclear Geophysics (English transl. of Yadernaya
Geofizika). Consultants Bureau, New York, N.Y., pp.171-178.
Kortsenshtein, V.N., 1964. The estimation of the possible oil-gas presence according to
the groundwater analyses and the evaluation of the forecast oil-gas reserves. Dokl.
Akad. Nauk S.S.S.R., 158:856-859 (in Russian).
Kortsenshtein, V.N., 1965. Estimating the oil and gas potential from data on gas satura-
tion of groundwater under dephased equilibrium conditions. Dokl. Akad. Nauk
S.S.S.R., 150:635-638 (in Russian).
Kortsenshtein, V.N., 1968. A comparative description of the benzene content in stratal
waters of the Mesozoic complexes of South Mangyshlak and of East Caucasia. Dokl.
Akad. Nauk S.S.S.R., 180:697-699 (in Russian).
Kravchinskii, Z.Ya. 1960. Comparison of chemical characteristics of waters in productive
red-colored formations. Geol. NeftiGpza, 12:42-44 (in Russian)..
Krejci-Graf, K., 1962. Oilfield watek. Erdol Kohle Petrochem., 15:102-109 (in
German).
Kroepelin, H., 1967. Geochemical prospecting. In: Latest Developments within the Oil
Industry. Proceedings 7 t h World Petroleum Congress. American Elsevier, New York,
N.Y., 1B:37-57.
Landes, K.K., 1967. Eometamorphism and oil and gas in time and space. Bull. A m . Assoc.
Pet. Geol., 51 :828-841.
London, E.E., 1964. The degree of groundwater saturation with dissolved hydrocarbons
and sulfates as a criterion for the evaluation of oil-gas prospects. Geol. Nefti Gaza,
11:41-47 (in Russian).
London, E.E., Zor’kin, L.M. and Vasil’ev; V.G., 1961. Location of gas reserves based on
dissolved gas content of formation waters. Geol. Nefti Gaza, 3:35-40 (in Russian).
MacElvain, R.G., 1963. What d o near-surface signs really mean in oil finding. Oil Gas J.,
61(7):133-136; 61(8):139-146.
McAuliffe, C.D., 1966. Solubility in water of paraffin, cycloparaffin, olefin, acetylene,
cycloolefin, and aromatic hydrocarbons. J. Phys. Chem., 70:1267-1275.
McAuliffe, C.D., 1967. Geochemical method of prospecting for petroleum. U.S. Patent,
No.3,345,137.
McAuliffe, C., 1969. Determination of dissolved hydrocarbons in subsurface brines.
Chem. GeoL, 4:225-233.
McLeod, H.O., 19711 Bibliogmphy Related to Research and Geochemical Methods for
Petroleum and Gas Exploration. Petroleum Abstracts, 200 pp., unpublished.
Noad, D.F., 1966. Water analysis: a key to exploration. Can Petrol., 7:12-14; 7:16-18.
Parker, J.W., 1969. Water history of Cretaceous aquifers, East Texas Basin. Chem. Geol.,
4:111-133.
Peake, E. and Hodgson, G.W., 1966, 1967. Alkanes in aqueous systems, I and 11. J. Am.
Oil Chem. SOC.,I, 43:215-222; 11, 44:696-702.
Philippi, G.T., 1957. Identification of oil source beds by chemical means. Proc. 20th Int.
Geol. Congr., Mexico City, 1957, Sect. III, pp.25-40.
Philippi, G.T., 1965. On depth, time, and mechanism of petroleum generation. Geochim.
Cosmochim. Acta, 29:1021-1049.
Philippi, G.T., 1969. Essentials of the petroleum formation process are organic source
material and a subsurface temperature controlled chemical reaction mechanism. In:
P.A. Schenck and I. Havenaar (Editors), Advances in Organic Geochemistry. Pergamon
Press, New York, N.Y., pp.25-46.
Pirson, S.J., 1942. Theoretical and economic significance of geodynamic prospecting.
World Petrol., 13:38-42.
Pusey, 111, W.D., 1973. How to evaluate potential gas and oil source rocks. World Oil,
176:7 1-75.
Ransone, W.R., 1947. Geochemical history of the Hardy oilfield, Jones County, Texas.
Geophysics, 12: 384-392.
Roach, J.W., 1965. How to apply fluid mechanics to petroleum exploration. World Oil,
160:7 1-7 5.
Rosaire, E.E., 1939. The Handbook of Geochemical Prospecting. Rosaire, Houston,
Texas, 120 pp.
Rosaire, E.E., McBermott, E. and Fash, R.H., 1940. Discussion of geochemical explora-
tion. Bull. Am, Assoc. Pet. GeoL, 24:1434-1463.
Savchenko, V.P., Vinogradov, V.L. and Yakoylev, Yu. I., 1965. Front- and back-wall
effect and its importance in prospecting. Geol. Nefti Gaza, 7:36-40 (in Russian).
Schmidt, G.W., 1970. Geochemical prospecting method. US.Patent, No.3,524,346.
Schwab, R., 1965. Logging important aspect of hydrodynamic studies. Oilweek,
16:36-37.
Serebriako, 0.1. and Tronko, I.V., 1969. Ammonium content in groundwaters of the
northwestern Caspian region as an indication of oil and gas. GeoL Nefti Gaza, 9:57-60
(in Russian).
Shabarova, N.T., Tunyak, A.P. and Nektarova, M.B., 1961. Study of organic acids in
subsurface waters. Geol. Ne f t i Gaza, 11:50-5 1(in Russian).
Shvets, V.M. and Shilov, I.K., 1968. On organic matter in underground waters of the
southwestern part of the Azov-Kuban Artesian Basin. GeoL Nefti Gaza, 8:46-49 (in
Russian).
Sikka, D.B., 1963. Mechanisms explaining the formation of radiometric anomalies. Izv.
Akad. Nauk S.S.S.R., Ser. GeoL, 6:73-87 (in Russian).
Silverman, S.R., 1964. Investigations of petroleum origin and evaluation mechanisms by
carbon-isotope studies. In: H. Craig, S.L. Miller and G.J. Wasserburg (Editors), Isotopic
and Cosmic Chemistry. North-Holland, Amsterdam, pp.92-102.
Simons, H.F., 1940. Scope of soil analysis increased during year. Oil Gas J., 38:54.
Sokolov, V.A., Alexeyev, F.A., Bars, E.A., Geodekyan, A.A., Mogilevskii, G.A.,
Yurovskii, Yu.M. and Yasenev, B.P., 1959. Investigations into direct oil detection
methods. Proc. 5th World Petrol Congr., Sect. I , Paper, No.36, pp.667-687.
Spencer, D.W. and Koons, C.B., 1970. Studies on origin of crude oil data. Presented at
SPE Symp., 44th Annual Meet., Calgary, A l t a , June 22-24, 1970.
Stormont, D.H., 1939. Operation of a Gulf Coast field on soil survey information. OilGas
J., 38:28-29.
Sudo, Y., 1967. Geochemical study of brine from oil and gas fields in Japan. J. Japan
Assoc. Pet. Technol., 32: 286-296 (in Japanese).
Tissot, B., Califet-Debyser, Y., Deroo, G. and Oudin, J.L., 1971. Origin and evolution of
hydrocarbons in Early Toarcian shales, Paris Basin, France. Bull. Am. Assoc. Pet.
GeoL, 55:2177-2193.
Trask, P.D. and Patnode, H.W., 1942. Source Beds o f Petroleum. American Association of
Petroleum Geologists, Tulsa, Okla., 566 pp.
Vdovykin, G.P., 1963. Oxidation-reduction potential of formation waters of the
northwest Cis-Caucasus and of some surface waters. Pet. GeoL, 7:286--290.
Vilonov, V.A., 1962. A feature of the distribution of radioactive elements in the water-
oil contact zone. Geokhim. Nefti Neftyanykh Mestorzhdenii, Akad. Nauk S.S.S.R.,
Moscow, pp.199-206 (in Russian).
Visher, G.S., 1961. Petrologic study of the Delaware sand (Bell Canyon formation),
Texas. Rep. Sinclair Oil Company, unpublished.
Welte, D.H., 1965. Relation between petroleum and source rock. Bull. Am. Assoc. Pet.
Geol., 49:2246-2268.
Zarrella, W.M., 1969. Applications of geochemistry to petroleum exploration. In: W.B.
Heroy (Editor), Unconventional Methods in Exploration for Petroleum and Natuml
Gas. Southern Methodist University, Dallas, Texas, pp.29-41.
REFERENCES 341
Zarrella, W.M., Mousseau, R.J., Coggeshall, N.D., Norris, M.S. and Schrayer, G.J., 1967.
A h , 31 :1155-1166.
Zinger, A S . and Kravchik, T.E., 1969. On direct hydrochemical indices in the presence of
oil and gas and the role of micellar solubility in the case of hydrocarbon migration in
aqueous solutions. DokL Akad. Nauk S.S.S.R., 189:180-184 (in Russian).
Chapter 11. GEOPRESSURED RESERVOIRS
The composition of the waters in normally pressured reservoirs often

differs from the composition of the waters in geopressured or abnormally
high-pressured reservoirs. There are several theories concerning the cause of
the geopressured zones; many papers have been written about their occur-
rence and causes (Burst, 1969; Dickey et al., 1968, 1972; Fowler, 1970;
Harkins and Baugher, 1969; Hottmann and Johnson, 1965; Jones, 1969;
Powers, 1967;Schmidt, 1973;Wallace, 1969). Knowledge of how to locate
geopressured zones is important in drilling operations, because if such a zone
is drilled into without adequate preparation, the well may blow out, perhaps
causing a fire, loss of the well, loss of the drilling rig, or even loss of life. The
usual precaution, if the driller knows of a high-pressure zone, is to increase
the weight of the drilling mud; however, the continual use of heavyweight
mud is much more expensive than drilling with a lighter weight mud. Drilling
rig time is worth about $2,000 per day, and it costs about $44,000 per
kilometer to drill a well on land. A drilling barge in the bay can cost from
$4,000 to $lO,O.OOper day while a drilling ship plus a full crew costs about
$25,000 per day. Considering.the foregoing costs plus the cost for a special
crew t o extinguish a fire at an ignited blowing well can be very expensive
because the initial fee for the fire extinguishing personnel is about $25,000.
Steps, therefore, are taken by the drilling company to assure that an
adequate drilling rig is used, that the optimum size borehole is drilled, that
the correct weight drilling mud is pumped down, that strong enough casing is
inserted into the well, and that blow-out preventers are operative.
Geopressure
Dickinson (1953)defined abnormally high pressure (geopressure) as any

pressure exceeding the hydrostatic head of a column of water (extending
from the subsurface tapped stratum to the land surface) containing 80,000
mg/l of dissolved solids. Formations with equal or less pressures are con-
sidered normal or subnormal. In the Gulf Coast area the normal pressure
gradient is about 0.107 kg m-l, or about equal to 0.21 g cm-3 of
drilling mud (Harkins and Baugher, 1969). Normal pressure in the Rocky
Mountain region has a gradient of 0.100 kg m-', although excep
tions occur in western Montana, the Denver Basin, the Powder River Basin,
and the San Juan Basin, mostly in Cretaceous rocks (Finch, 1969). (A
gradient of 0.118 kg m-l is normal in the Williston Basin in North
344 GEOPRESSURED RESERVOIRS
Dakota.) Normal pressure is that which is normal for the particular area
involved and is related t o the salinity of the reservoir water, rock types, and
geologic setting, but in general, it is that pressure exerted by a column of
water from the surface t o the observed subsurface formation, which is equal
to and will balance the subsurface formation pressure. Abnormally high
pressures are those which exceed this normal hydrostatic head.
Geostatic ratio
Abnormal pressures can be expressed in terms of a “geostatic ratio,”

which is the ratio of the observed fluid pressure in a subsurface formation t o
the overburden pressure of the overlying sediments. This load at a given
-
depth is approximately 0.231 kg m down t o depths of more than
6,100 m, because the density of rocks changes slowly with depth (Penne-
baker, 1968). Any abnormal pressure will therefore have a geostatic ratio
between 0.0327 and 0.0703 kg in the Gulf Coast area and between
0.0304 and 0.0703 kg in the Rocky Mountain area.
Compaction model
The compaction concept was demonstrated using a model consisting of

perforated metal plate separated by metal springs in water and enclosed in a
cylindrical tube (Terzaghi and Peck, 1948). The springs were used to
simulate communication between deposited particles and with the initial
pressure upon the upper plate, the springs d o not move because all of the
pressure is supported by the water, assuming that water does not escape
from the system. Relating the fluid pressure (FP) t o the total pressure (TP)
one can derive an equation X=FP/TP to record various formation pressures
(X)t o determine the geostatic ratio using the model.
Origin of abnormal pressures
Abnormally high pressures in a formation can be caused by compaction.
Factors which may cause them are, according to Hottmann and Johnson
(1965),“the ratio of shale t o sand thickness, the mean formation permeabili-
ty, the elapsed time since deposition, the rate of deposition, and the amount
of overburden.” These parameters are interrelated in compaction, which is
the controlling factor in fluid pressures within subsurface sedimentary
environments (Harkins and Baugher, 1969). Dickinson (1953)reports that
the fluid pressures within sediments are predominately controlled by two
factors; namely: (1)the compression as a result of compaction; and (2)the
resistance t o expulsion of water.
Compaction begins with sedimentation and deposition of soft muds com-
posed of up to 90% water (Wallace, 1969). In an environment where depo-
sition continues, gradual compaction occurs whereby the muds become clay
ORIGIN OF ABNORMAL PRESSURES 345
minerals and shales. The shales are primarily clay minerals with flat or
tabular grain shapes; with additional overburden, the pressure packs the
grains closer together, with a resultant expulsion of water from the inter-
vening spaces. In the early stages of compaction, the shale possesses high
porosity and permeability, and the expelled water always flows to areas of
least resistance and pressure (often porous sand). As the overburden in-
creases, the porosity and permeability of the shale decrease until equilibrium
is approached and the pressure in all directions is equal. A t this point,
expulsion of additional water is limited. Tectonics, of course, could alter the
subsurface environment.
Deposition and sedimentation of sand are somewhat different because the
sand grains are in contact in the first stage and sand compaction is about
complete with deposition. However, reduction of porosity can occur by: (1)
solution of the sand grains at contact points; and (2) rearrangement of the
grains because of very high pressures.
Clay beds separating aquifers are often referred to as semipermeable mem-
branes. Such beds can separate aquifers containing waters of different salin-
ities, causing a hydrostatic head in the direction of the more saline water.
Fig. 11.1. Sand dikes in the Simpson Sand formed by the actiqn of highly pressured
subsurface waters forcing the lighter colored sand intrusively into the primary sandstone.
The primary sand was formed from white beach sands during Ordovician time.
Osmotic pressure can develop, which is dependent upon osmotic efficiency

of the clay bed and the differences in salinities of the two aquifers (Young
and Low, 1965).According to Jones (1969),stepwise increments of osmotic
pressure may develop wiih depth through a series of bedded sands and clays
acting as a multistage pump, thus producing the high reservoir pressures in
the northern Gulf of Mexico basin.
Fertl and Timko (1972)discuss 17 possible causes of abnormally high
pressures. They are rate of sedimentation, tectonic activities, potentiometric
surface levels, reservoir structures, areal salt deposition, shallow-reservoir
repressuring, paleopressures, mud volcanoes, secondary precipitation of
cementation constituents, diagenesis of volcanic ash, rehydration of
anhydrite, diagenesis of clays, osmosis, permafrost, earthquakes, chemical,
thermal chemical, and biochemical effects, and tidal disturbances.
Fig. 11.1 illustrates one type of action that results from high pressures,
where sand dikes formed by the action of highly pressured subsurface waters
forcing the lighter colored sand intrusively into the primary sandstone. The
primary sandstone was formed from white beach sands during Ordovician
time.
Abnormal pressures in the Gulf Coast area
In the Gulf Coast area, the abnormal pressure seems t o be related t o rapid
deposition of sediments and low regional transmissibility. Fluid pressures are
near hydrostatic where there is continuity with normally pressured aquifers
and where the sands are sufficiently permeable to dissipate the expelled
water from the compacting fine-grained rock.
In some of the deep oil and gas wells of the Gulf Coast, the pressure of the
interstitial fluids (oil, gas, or water) in kilograms per square centimeter is
normally the depth in meters multiplied by 0.107.This is slightly more than
the pressure required t o sustain a column of water to the surface. At great
depths where the geological section is mostly shale, fluids at abnormally high
pressures are found. Sometimes the pressures are very high, approaching 0.2
kg cme2 m-l. Often the increase in fluid pressure is abrupt, taking place in
a vertical interval of 30 m or less. In other areas, the increase in pressure is
more gradual, extending over 300 m of vertical section. The depth at which
the pressure starts t o increase ranges over a wide interval. Abnormal pres-
sures are found at depths as shallow as 1,000m in some offshore fields, and
wells in some areas have been drilled deeper than 7,000 m without encoun-
tering abnormal pressures.
Forty-one formation water samples from gasfields in southwestern
Louisiana were obtained and analyzed to determine the relationships of the
chemical composition of the waters to normal and abnormally pressured
geologic zones (Dickey et al., 1972).The concentration of dissolved solids in
the waters from the overpressured zones is generally less than in the normal
pressure zones, and this knowledge is significant in electric log interpreta-
tion.
ABNORMAL PRESSURES IN THE GULF COAST AREA 347
Fig. 11.2. Slash lines showing the general area in Louisiana where the samples were
obtained.
Previous work in the area suggested that the abnormally fresh waters were
found in the same part of the section as were the abnormally high pressures
(Dickey et al., 1968). The general locations of the wells are shown in Fig.
11.2. They were from the South Lewisburg, Church Point, Branch, South
BOSCO,North Duson, Duson, Ridge, and Andrew fields, all in Acadia, and
Lafayette Parishes, Louisiana. The water samples were analyzed chemically
by using the procedures published by the American Petroleum Institute
(1968). The analytical data are summarized in Table 11.1.
A subsurface cross section, Fig. 11.3, was constructed in a general north-
south direction showing the stratigraphy and structure across seven oilfields
in the area of study (Fajardo, 1968). The initial pressures of the shallower
reservoirs are normal. However, below 2,450 m many reservoirs contain
fluids with abnormally high pressures. The 4.9-m amplified normal curve was
used to recognize the first appearance of abnormal pressures in the shale
section. The fluid pressure gradients were estimated following the method
described by Hottmann and Johnson (1965). Shale resistivity and fluid pres-
sure gradient versus depth were plotted for 50 wells in different fields of the
study area, and of these, 22 are included in the cross section.
All of the 41 waters belong t o the chloride-calcium class of Sulin (1946),
and none has the composition of meteoric water. The principal cation is
sodium, although the concentration of calcium is always high. In some of the
more concentrated brines, the calcium concentration is nearly 40,000 mg/l
and constitutes over half the reacting value of the sodium. Magnesium is
variable in amount, and in two samples it is absent. Chloride is the predomi-
nant anion, amounting always t o more than 49.5% of the total reacting
values. Sulfate usually is absent and never is present in concentrations greater
than 0.5% of the total reacting value.
In Fig. 11.3, the top of the section is 2,100 m below sea level. The electric
logs indicate the lithology, which is quite sandy down t o a depth of 2,700 m
TABLE 11.I
Formation-waters sample locations, constituents found in the waters, shale resistivity (SR),and fluid pressure gradients (FPG)
Sample Location o f Depth Zone Specific Concentration (mg/l) SR at FPG
number of well (m) gravity spl. depth (kg cm-2
(S-Twp-R) CI
(60°/600F) HCO3 SO, B Br 1 Na Ca Mg K Li Sr Ba NH4 organic 6’)
acid as
acetic
1*I 73-105-34E 4,66+4,664 Bolivina-mex 1.035 39,000 387 0 49 35 18 17,800 1,070 78 518 10 ND ND 246 84 0.77 0.185
2’1 2l-lOS-03E 3,643-3,645 U. Camerina 1.062 55,600 541 407 62 61 22 32,300 2,210 369 247 7 ND ND 295 48 0.59 0.159
3*’ 2*lOSFo3E 3,625-3,627 L. Camerina 1.057 56,600 826 234 62 52 21 34,200 1,380 213 200 6 ND ND 238 96 0.62 0.157
4’1 20-10S-aE 3.613-3,615 U. Camerina 1.059 50,000 630 38 37 57 21 29,500 1,380 194 204 6 ND YD 650 72 ND ND
5*1 17-10S43E 4,060-4,063 Bolivina-mex 1.085 72.800 448 tr. 43 81 19 41,100 3,850 583 267 6 ND ND 178 96 0.38 0.195
6 O8-lOS43E 3,047-3.049 Discorbis 1.045 33,300 503 50 32 37 16 19.200 1,390 224 85 3 P:D ND 200 120 1.1 0.107
7 25-07S43E 3,325-3,327 U. Tweedel 1.060 51,700 180 0 18 21 15 34,800 2,730 544 208 4 ND ND 538 24 ND ND
8 25-07603E 3.293-3.296 L. Tweedel 1.069 58,000 363 tr. 26 41 18 34,400 2,020 194 230 5 ND ND 301 76 ND ND
9 U. Nodosaria 1.083 50,900 507 33 35 70 23 27,500 3,050 719 162 6 ND ND 362 24 1.02 0.107
10 Daigle 1.051 45,300 545 0 28 38 18 26,000 2,310 167 134 4 ND ND 279 96 0.83 0.107
11 U. Nodosaria 1.065 45,600 574 tr. 29 35 22 24,900 2,950 447 262 4 ND ND 210 72 0.97 0.107
12 Tweedel 1.069 57,900 586 tr. 26 56 25 33,300 2,660 389 in1 4 ND ND 254 96 ND ND
13 Tweedel 1.062 52,600 579 0 34 45 20 31,700 1,570 0 192 5 ND ND 364 120 ND ND
14 Struma 1.128 116,000 322 0 38 128 26 49,600 21,600 2.180 427 9 ND ND 218 144 0.90 0.107
15 Frio 1.061 50,900 330 67 23 43 18 29,600 1,890 408 31 5 9 ND ND 230 48 ND ND
16 Nodosaria 1.149 135,000 92 223 52 154 24 66,800 15,200 1,270 813 9 ND ND 250 24 ND ND
17** Klumpp D 1.090 79,300 334 0 48 169 74 46,400 2,950 447 427 10 0 50 202 144 0.60 0.157
18 Frio 1.058 46,600 741 60 46 40 23 24,700 2,660 1,010 172 6 0 5 377 120 0.84 0.107
19 Frio 1.058 47,900 788 72 48 52 21 26,600 2,180 303 166 5 0 5 206 96 ND ND
20 Frio 1.057 45,800 694 ND 44 47 22 ND ND ND 157 4 0 8 96 72 1.o 0.107
21 Frio 1.144 125,000 135 0 47 64 23 61,900 15,700 159 830 15 0 19 243 48 0.94 0.107
22*’ U. Texana 1.090 84.500 419 122 40 20 22 45,800 7,390 565 324 7 ND ND 110 96 0.35 0.191
23 U.Texana 1.092 80,300 363 tr. 45 62 20 45,800 3,300 972 376 9 0 33 142 144 0.70 0.131
24*’ Frio 1.220 201,000 0 352 75 213 18 80,600 38,800 2,140 782 17 ND ND 294 48 0.50 0.191
25” Frio 1.202 184,000 0 tr. 67 204 19 68,900 33,200 5,770 640 17 ND ND 282 120 0.35 0.203
26 Frio l.lS3 111,000 112 ti. 42 94 24 53,600 14,300 428 798 12 ND ND 295 48 1.4 0.107
27 Nodosaria 1.139 119,000 76 0 52 117 28 52,700 18,400 1,200 771 18 ND ND 279 24 1.o 0.107
28 Marg howei 1.070 61,600 550 0 67 14 5 35,200 3.610 17 137 5 ND ND 180 168 0.38 0.193
29 Homeseeker 1.144 100,000 66 0 42 201 21 40,600 18,300 1,090 1,150 17 N D l ND 222 72 1 .o 0.107
30 Nodosaria A 1.145 109,000 73 0 39 71 21 51,800 14,400 700 631 15 ND ND 368 96 ND ND
31 Horn-eker D‘-4 1.088 80,000 270 77 34 110 34 47,800 2,760 35 392 10 0 110 349 24 ND ND
32 Klumpp E 1.055 44,400 244 130 36 70 30 25,800 1,510 447 166 5 ND ND 195 12 0.83 0.107
33*’ Brookshire 1.089 77,500 171 0 18 81 21 44,200 3,530 836 236 2 140 97 219 96 ND ND
3483 Brookshire 1.089 78,000 206 0 18 82 19 44,400 3,270 972 235 2 128 109 214 624 ND ND
35*3 Brookshire 1.082 72,300 234 0 33 58 18 41,200 3,270 564 294 3 265 41 258 144 ND ND
36” Brookshire 1.089 75,400 203 0 18 162 19 42,600 3,370 894 232 2 171 102 306 48 ND ND
37 Nodosaria 1.140 129,000 80 0 41 174 24 77,800 4.560 0 ND ND ND ND 152 ND 0.95 0.107
38 Homeseeker 2-D 1.120 120,500 240 0 43 134 38 68,800 5,610 564 375 5 ND ND 295 96 0.92 0.107
39*3 Marg tex 1.069 55,500 539 102 52 40 26 32,500 3,210 136 176 5 0 7 179 192 0.65 0.152
40 U. Moicene 1.082 74,400 249 0 26 79 18 42,800 2,950 855 264 2 195 85 167 432 ND ND
41*3 Marginulina 1.050 49,700 482 8 8 43 60 35 29,600 1,780 141 71 2 0 4 160 312 0.57 0.16X
*’ Abnormal pressure.
*’ Abnormal pressure. but normal chemically.
* 3 Samples 33-36 are from the Abheville field or the south and of the area sampled and appear to be in a different chemical family,
S-Twp-R = section-township-range; ND = not determined.
?6 KEY
*33a34 X AbnzDressure
Normal pressure
4.5
il
30
I
40
I
50
1
60
I
70
I
80
CHLORIDE, g / l
I
90
I
100
1
110
I
120 I D
Fig. 11.4. Plot of the depth of the wells versus concentrations of chloride in the forma-
tion waters.
-‘-I
2.5 c 33.
3p
034
/
KEY
,
X Abnormal pressure/
?‘Normal presauro
0 0.5
BICARBONATE, p/l
Fig. 11.5. Plot of the depth of the wells versus concentrations of bicarbonate in the
formation waters.
in the north to 3,050 m in the south. Below this depth, the sands become
less abundant and less widespread.
The first abnormal pressure as calculated from shale resistivity is indicated
by an arrow. The location of a producing horizon from which a water sample
was taken is shown by the sample number in a circle. When the water sample
was taken from a nearby well, not shown on the section, it was projected
onto the section and shown as the sample number inside a square in Fig.
11.3.
There is a general tendency for the dissolved salt concentration of the
water samples t o increase with depth. This is shown in Fig. 11.4, which
shows chloride plotted against depth. Since chloride is the predominant
I KEY
)< Abnormal pressure
0 Normal pressure
25
K
27
X
\ 24
\
I 10 I
CALCIUM, g/l
Fig. 11.6. Plot of the depth of the wells versus concentrations of calcium in the formation
waters.
anion, it serves as an indication of the degree of concentration. The samples

of water from abnormally pressured sands are shown as circles in x'es. All of
them except 17, 22, 24, and 25 fall below the average concentration line,
that is, they are less concentrated than they should be for their depth of
burial. Sample 1 especially is much too weak.
Bicarbonate, while occurring in much smaller quantities, shows the reverse
relation, decreasing in amount with depth, as shown in Fig. 11.5. The waters
from horizons with abnormal pressures have more bicarbonate than they
should, considering their depth of burial.
Calcium increases with depth, as shown in Fig. 11.6. It would be more
correct t o say that there are two types of water. Type 1includes waters with
less than 5,000 mg/l calcium, all of which are shallower than 3,800 m; type 2
is water with more than 5,000 mg/l calcium, most of which is deeper than
3,500 m. The only minor constituent that indicated a significant change with
depth was potassium, and it appears to increase relative to sodium. The
abnormally pressured waters seem deficient in potassium for their depth.
Normal pressure
x Abnormal pressu&e-Solution
o N o r m a l pressure- A l t e r e d relict bittern
100 -
O\&o
0%
'0
0
a0 - \
0 "\
" \
O \
\
-
\
0
60- "\
.\O
A
0
0
0
SODIUM, g/l
Fig. 11.7. Comparison of some brines of a bittern type from the Michigan Basin with
some brines from some normal and abnormally pressured reservoirs in Louisiana.
36 0 GEOPRESSURED RESERVOIRS
I50
Normal pressure
x Abnormal pressure
I25
I00
.
-
\
2
5-
75
/
0 X
0
In
X
50 - x
25
X*
Sodium’=mg/l Na + 40 mg/l Ca
0 I I I I
0.05 010 0.15 0.20 !5
BROMIDE, g/l
Fig. 11.8. Plot of Na’ versus Br from some brines from normal and abnormal pressured
reservoirs in Louisiana.
Four of the waters from high-pressure sands (17, 22, 24, 25) have normal
concentrations of dissolved solids for their depth. The other waters from
high-pressure sands (1-5, 13, 28, 39, 41) have lower concentrations than
normal. They also have less calcium, more bicarbonate, and a higher Cl/K
ratio.
About 80% of the material in the Gulf Coast shale is clay. Assuming that
the waters have reacted with montmorillonite, there should be a direct rela-
tionship of calcium t o sodium. Plotting the calcium and sodium data in
Table 11.1 plus some data for some brines from the Michigan Basin (as
shown in Fig. 11.7) indicate that a relationship of calcium t o sodium does
exist in the Gulf Coast waters and that they probably have reacted with
montmorillonite. Fig. 11.7 also indicates that the Gulf Coast waters are not
an altered relict bittern as are the Michigan Basin brines.
In an ion exchange reaction with montmorillonite, 2 moles of sodium are
exchanged for 1 mole of calcium, therefore, if salt is redissolved the bromide
content in solution should be proportional to the original redissolved solu-
ABNORMAL PRESSURES IN THE GULF COAST AREA 36 1
tion. However, because of the exchange reaction the sodium in solution

should be Na + 46/40 Ca or Na'. Fig. 11.8 is a plot of Na' versus Br for the 41
samples. The data scatter to some extent but this can be expected if biogenic
derived bromide is present and the presence of iodide indicates that such is
the case. Fig. 11.8 indicates that re-solution of salt is a control in these
samples.
Fig. 11.9 shows further evidence that the Louisiana brines were formed by
re-solution of salt. For example, the dashed line in the left portion of Fig.
11.9 is a plot of Na' versus Br of salt dissolved in distilled water, and the
solid line just t o the right is a replot of Na' versus Br for the Louisiana
brines. The next dashed line t o the right is Na' versus Br for evaporating sea
water, and the curved dashed line is Na' versus Br for relict brines from the
Michigan Basin.
Notable differences in the waters found in the normally and abnormally
pressured rocks are evident (Schmidt, 1973). The dissolved solids in the
-Re-solution solt
in pure water
Southwestern Louisiano brines
I
\*
I *\
I \*
I \
\ t
9vaporoting h
25 seo water *\ \
I f *\
I .3:
Sodium I=mg/l sodium +%mg/l calcium m>
I 2 3 4
BROMIDE, g/l
Fig. 11.9. Replot of Na' versus Br of the Louisiana brines (Fig. 11.8);plus data for relict
Michigan brines, evaporating sea water, and resolution of salt. Resolution of salt is an
important control for the Louisiana brines
normally pressured sandstones range from 600 to 180,000 mg/l, while in the
geopressured sandstones the range is from 16,000 to 26,000 mg/l. The dis-
solvedsolids in the water in the pores of the shales adjacent to normally
pressured sandstones are lower than the dissolved solids in the water in the
sandstones, but the dissolved solids concentrations are similar in the waters
of the adjacent high-pressure sandstones and shales. The concentration order
in shale pore water is > HC03- > Cl-, and in normally pressured
sandstone water it is C1- > HC03- > S 0 4 - 2 .
The temperature gradient in the geopressured zone is about O.8l0C/25 m,
while in the normally pressured zone it is about 0.44OC/25 m. This change in
temperature gradient is believed t o be related t o the porosity, where a
greater porosity causes a decreased thermal conductivity (Schmidt, 1973).
The clay mineral composition in the geopressured zone is predominantly a
nonexpandable type, while in the normally pressured zone montmorillonite,
an expandable type, frequently occurs. This change is believed to be related
t o the temperature, and the heat allows the release of water from the clays at
temperatures of about 93-104OC. This released water will dilute the pore
water and cause the dissolved solids to decrease.
The total amount of water released by Gulf Coast shales in geopressured
zones is about 13%of the total in the system (Schmidt, 1973). This can be a
cause of the lower salinity of the waters found in the geopressured zones.
Fowler (1970) studied the Chocolate Bayou field in Texas and evaluated
the relationships between geopressure and the migration and accumulation
of hydrocarbons. He concluded that faults tend t o act as barriers separating
fluid systems in the area; however, cross-formational flow occurs with
geopressure causing shale ultrafiltration of the waters. The ultrafiltration
produces salinity variations in the waters. Hydrocarbon accumulation in the
area is controlled by the hydrodynamic flow.
According to Fowler (1970), hydrocarbons are trapped in the upper sands
because of slight pressure differentials across fault traps in the West Choco-
late Bayou field. However, in deeper strata, abnormal pressures have caused
hydrodynamic flow and pressures greater than the displacement pressure in
the fault, resulting in no trapped hydrocarbons.
In essence then, sands with pressure gradients greater than 0.20 kgcm-'
m-' in the Chocolate Bayou field do not contain commercial amounts of
hydrocarbons. It also appears that the size of the accumulation may decrease
with increasing pressure gradients up t o about 0.16 kg cm-* m-'. The
accumulation size may increase with pressure gradients in the range of
0.16-0.19 kg cm-2 m-' and then decreases.
Detection of abnormal pressures
Estimation of formation pressures from electrical surveys is related to the

following assumptions, concerning the origin of abnormal pressures (Foster
and Whalen, 1966):
DETECTION OF ABNORMAL PRESSURES 363
(1) Shale porosity is a function of net overburden pressure and normally

decreases with an increase in depth.
(2) Shales with abnormal pressure will have a higher porosity than nor-
mally pressured shales at the same depth, because of the greater amounts of
interstitial fluids.
(3) Sand bodies (confined by lensing, faulting, etc.) surrounded by shale
will have a pressure similar t o those in the shales.
Data from acoustic and resistivity logs can be used to establish a shale
transit time or shale resistivity versus depth of normal hydrostatically pres-
sured formations. Deviation from the derived curve is used t o determine
abnormal pressures (Hottmann and Johnson, 1965).
The acoustic log is a function of porosity and lithology; therefore, in any
given shale sequence it is primarily a measure of porosity. The acoustic
response in normally pressured shales decreases in travel time (velocity in-
creases) with increasing depth. This is the “normal compacted trend”, and
the pressures in the shale are normal, or hydrostatic. Deviation from the
“normal compaction trend” indicates an abnormally pressured zone.
Relating the difference in the travel time of the observed formation pressure
( A T , ) t o a normal formation pressure (AT,) t o the formation pressure
gradient (calculated from known depths and pressures of wells in the area), a
pressure gradient (AT, -AT,) can be determined. The reservoir pressure can
be found by multiplying this gradient by the depth.
Fertl and Timko (1970) discuss several methods, using the theory of
“departure from the normal” t o detect abnormally pressured zones. Methods
they discuss are as follows:
(1) Bulk density -this is a measurement of the intensity of back-scattered
electrons produced by gamma-ray bombardment; this intensity varies with
the bulk density of the rocks surrounding the borehole.
(2) Conductivity measurements - measure of an induction log. Electro-
motive forces set up a current, which is detected by a receiver and recorded.
Overpressured shales are noted by greater-than-normal conductivity reading
resulting from higher-than-normal water content and porosity.
(3) Borehole temperature - geopressured as usually associated with an
increase in temperature.
(4)Presence of gas in mud - this is not always a good detector, for gas
can evolve from formation cuttings, as they come to the surface.
One of the best means of obtaining subsurface information, other than
drilling, is the use of the reflection seismograph. This geophysical tool is a
measure of time between the earth’s surface and various subsurface reflecting
horizons. The differences in interval velocities between these different
horizons (formations) can be used to obtain a plot of average interval travel
time, which varies exponentially with depth.
The degree of departure from a “normal” plot of this travel time versus
depth is related t o abnormally pressured reservoirs in the Gulf Coast area
(Pennebaker, 1968). This departure is noted as an increase in the normally
decreasing travel time with depth, because of the undercompacted forma-

tions. To measure the formation pore pressure, plots of equal pore pressure
gradients are compared, by an overlay, to the abnormally pressured interval
travel time depth plots.
Forgotson (1969),by experience with wells in the Gulf of Mexico, noted
that the presence of high background gas and high trip gas, together with
lower than normal shale density, does not necessarily indicate the proximity
of an abnormally pressured reservoir. He believes that a minimum of 200%
increase in the shale penetration rate when drilling is the best available means
to predict abnormal pressures.
A recent series of papers explains how downhole temperatures and pres-
sures can affect drilling (Fertl and Timko, 1972;Timko and Fertl, 1972).
Methods of detecting abnormal pressures, compensating for them, and evalu-
ating the hydrocarbon potential of geopressured strata are discussed.
References
American Petroleum Institute, 1968. API Recommended Practice f o r Analysis of Oilfield

Waters. Subcommittee on Analysis of Oilfield Waters, API, RP 45, 2nd ed., 49 pp.
Burst, J.F., 1969. Diagenesis of Gulf Coast clayey sediments and its possible relation t o
petroleum migration. Bull. A m . Assoc. Pet. GeoL, 53:73-93.
Dickey, P.A., Collins, A.G. and Fajardo, I., 1972. Chemical composition of deep forma-
tion waters in southwestern Louisiana. Bull. Am. Assoc. Pet. GeoL, 56:1530-1533.
Dickey, P.A., Shiram, C.R. and Paine, W.R., 1968. Abnormal pressures in deep wells of
southwestern Louisiana. Science, 160:609-615.
Dickinson, G., 1953. Geological aspects of abnormal reservoir pressures in Gulf Coast
Louisiana, Bull. A m . Assoc. Pet. GeoL, 37:410-432.
Fajardo, I., 1968. A Study of the Connate Waters and Clay Mineralogy. M.S. Thesis,
University of Tulsa, Tulsa, Okla., 50 pp.
Fertl, W.H. and Timko, D.J., 1970. Overpressured formations, 2. How abnormal
pressure-detection techniques are applied. Oil Gas J., 68:62-71.
Fertl, W.H. and Timko, D.J., 1972. How downhole temperatures, pressures affect drilling.
World Oil, 174(7):67-70; 175(1):47-49; 175(2):36-39, 66;175(4):45-50; 176(2):
47-50.
Finch, W.D., 1969. Abnormal pressure in the Antelope field, North Dakota. J. Pet.
Technol., 21:821-835.
Forgotson, J.M., 1969. Indication of proximity of high pressure fluid reservoir, Louisiana
and Texas Gulf Coast. Bull Am. Assoc. Pet. GeoL, 53:171-173.
Foster, J.B. and Whalen, H.E., 1966. Estimation of formation pressures from electrical
surveys - offshore Louisiana. J. Pet. TechnoL, 18:165-171.
Fowler, Jr., A.W., 1970. Pressures, hydrocarbon accumulation, and salinities - Chocolate
Bayou field, Brazoria County, Texas. J. Pet. TechnoL, 22:411-423.
Harkins, K.S. and Baugher, 111, J.W., 1969.Geological significance of abnormal formation
pressures. J. Pet. TechnoL, 21:961-966.
Hottmann, C.E. and Johnson, R.K., 1965. Estimation of formation pressures from log-
derived shale properties J. Pet. TechnoL, 17:717-721.
Jones, P.H., 1969. Hydrodynamics of geopressure in the North Gulf of Mexico Basin. J.
Pet. TechnoL, 21:803-810.
Pennebaker, E.S., 1968. Detection of abnormal pressure formations from seismic field
records. Presented at API Southern Dist. Meet., San Antonio, Texas, March 6-8,
1968, API Paper, No. 926-13C.
REFERENCES 365
Perry, D.R., 1969. A Correlation of Reserves and Drive Mechanisms with Reservoir Pres-
sure Gradients on Geopressured Gas Reservoirs in Southwest Louisiana. M.S. Thesis,
Southwest Louisiana University, Lafayette, La., 54 pp.
Powers, M.C., 1967. Fluid-release mechanisms in compacting marine mudrocks and their
importance in oil exploration. Bull. Am. Assoc. Pet. GeoL, 51:1240-1254.
Schmidt, G.W., 1973. Interstitial water composition and geochemistry of deep Gulf Coast
shales and sandstones. Bull. A m . Assoc. Pet. Geol., 57:321-377.
Sulin, V.A., 1946. Waters of Petroleum Formation in the System o f Natural Waters.
Gostoptekhizdat, Moscow, 96 pp.
Terzaghi, K. and Peck, R.R., 1948. Soil Mechanics in Engineering Practice. John Wiley
and Sons, New York, N.Y., 56 pp.
Timko, D.J. and Fertl, W.H., 1972. How downhole temperatures, pressures affect drilling.
\ World Oil, 175(5):73-81; 175(6):79-82; 175(7):5*62; 176(1):45-48; 176(4):.
62-65.
Wallace, W.E., 1969. Water production from abnormally pressured gas reservoirs in South
Louisiana. J. Pet. Technol., 21 :969-982.
Young, A. and Low, P.F., 1965. Osmosis in argillaceous rocks. Bull Am. Assoc. Pet.
Geol., 49:1004-1008.
Chapter 12. COMPATIBILITY OF OILFIELD WATERS
Waters used for the secondary recovery of oil by waterflooding usually

contain a number of inorganic salts and sometimes organic salts in solution.
It is common practice t o test the compatibility of the injection water and
water in the formation before starting a waterflood operation. Often this test
is performed by mixing the injection water with the formation water in a
glass container and observing t o determine if a precipitate forms. The precip
itate or scale can be analyzed to determine its composition.
Waters are compatible if they can be mixed without producing chemical
reactions between the dissolved solids in the waters and precipitating insolu-
ble compounds. The precipitated insoluble compounds are undesirable be-
cause they can reduce the permeability of a porous petroleum-productive
rock formation, plug input wells in waterflood systems, and cause scale
formation in water pumps and lines.
Some of the more common ions that frequently occur in oilfield waters
and that cause precipitation in incompatible waters are: Ca+2,S P 2 , Ba+2,
Fe+?,HC03-, and
Common reactions are:
CaC1, + Na2S04 += 2NaC1+ CaSO, 3-

CaC1, +MgS04 += MgClz +CaS04 3-
c a w 0 3 )2 -b C 0 2 + H 2 0 + CaCO, J-
CaC12 + 2NaHC03 += 2NaC1+ C 0 2 + H 2 0 + CaCO 3-
SrC1, + NaS04 += 2NaC1+ SrSO, 3-
SrClz +MgSO, += MgC12 + S r S 0 4 J-
BaC12 +NaS04 += 2NaC1+ BaSO, -1
BaCl2 +MgSO, += MgC12 +BaS04 3-
Fe + H2S -b H2 + FeS 3-
Fez03 + 6H2S += 6H,O + 2Fe2 S3 5.
A relatively insoluble compound CA where C is the cation and A is the
anion will precipitate from an aqueous solution if:
where ac = the cation activity, a A = the anion activity in the solution, and
SCA = the solubility product of the compound CA. When two salts with a
common cation (CAI and CA2) are in equilibrium in a solution, the follow-
ing will hold:
36 8 COMPATIBILITY OF OILFIELD WATERS
a~~ I ~ A , = &!Al PCA,

If a A l l a A , > SCA /SCA,,CA, will precipitate, and CA2 will dissolve if
a~~ I ~ A , <SCA,ISCA,.
Deposition of scale in both primary and secondary recovery producing
wells and formations is a very costly problem in the petroleum industry. The
scale not only restricts production but also causes inefficiency and produc-
tion equipment failure. Scale deposits are caused by mixing incompatible
waters and by environmental changes during the production of well fluids.
For example, as production begins, the pressure drops in the vicinity of the
wellbore, allowing dissolved gases to escape from solution. The loss of C 0 2
can cause calcium carbonate t o precipitate.
The decrease in pressure also can cause the vapor pressure of the brine to
increase. The temperature of the brine will decrease because heat energy is
required to vaporize the water, causing calcium sulfate t o precipitate.
Wellbore and formation damage
In several case studies Vetter and Phillips (1970) found that calcium sul-
fate deposits form in both primary and secondary petroleum production
operations. The scale forms within the formation and causes production loss
and permanent damage. In many cases damage to the formation cannot be
corrected even by fracturing. Research has indicated that sodium carbonate
can cause the metathesis of anhydrite and gypsum to calcium carbonate.
This might work in a formation that is partially plugged. For example, a
water containing sodium carbonate could be injected into the formation and
allowed t o react with the scale. An acidified water then could be injected
into the formation to remove the carbonate and hopefully clean the forma-
tion, allowing recovery of more oil.
Potential scale deposition should be predicted as soon as the well begins
production, and the correct inhibitor should be added immediately rather
than following the common practice which is to pull the tubing after a
production decline and find scale on the metal surface. Scaling can occur
within the formation and never show up on the tubing. Pressure drops are
the primary cause of calcium sulfate scaling within a formation, when the
formation brine is saturated with calcium sulfate.
Important variables related t o scaling are:
(1)Temperature of the formation in relation to solubility of the possible
scale former in the fluids passing through it. CaS04 becomes less soluble
WELLBORE AND FORMATION DAMAGE 369
with increasing temperature (Blount and Dickson, 1969), while BaS04

becomes more soluble (Templeton, 1960).
(2) Subsurface pressures change for any system, with the highest pressure
found while the fluid flows through the formation. The greatest pressure
change is at the sand face of the producing well (Vetter and Phillips, 1970),
which causes this area t o be where solubility changes are the greatest. Depo-
sition of scale at this point is the most damaging to oil production and the
most difficult t o discover or to remedy. Very few data are available on
pressuresolubility relations of most scale forming compounds, but CaSO,
has been shown to decrease in solubility with decrease in pressure at NaCl
concentrations t o 10%(Fulford, 1968).
(3) Brine concentration, exclusive of precipitating compounds, also in-
fluences scale formation. Most electrolytes in ionic form cause an increase in
the solubility of compounds which form scales. The solubility normally
increases with increasing electrolyte concentration unless some other solubil-
ity equilibrium is reached. This can occur, for example, when BaS04 satura-
tion level is reduced because of increasing amounts of Ca+2 ion in the
solution (Davis and Collins, 1971). Other properties of brine known to in-
fluence the solubility levels of scale formers are gases in solution, hydrogen-
ion concentration, ion pairs, and dissolved organic chelates (Weintritt and
Cowan, 1967).
Waterflooding of petroleum reservoirs has been successfully carried out
for many years. Large quantities of petroleum are produced through second-
ary recovery by forcing water (usually a brine) into an oil sand which has
become unproductive by primary production methods. However, the
efficiency of the operation is often low, and the amount of petroleum
remaining in the sand after waterflooding can be as high as 50% of the
original accumulation (Shaffer, 1967). One of the reasons such a high pro-
portion of the oil remains unrecovered is because the injection pressures
become economically too large to continue forcing water through the sand
t o displace the oil. The gradual deposition of solid material precipitating
from the water closes the permeable channels and slows the flow at the
producing well.
Dilution of the water injected into a formation often occurs, and addition-
al makeup water is necessary. The slow mixing of connate (interstitial)
waters of the formation or the introduction of water from associated
aquifers, both underground and on the surface, contributes to the instability
of the injection water.
The deposition of scale in wellbores, sand faces, and piping has reduced oil
production in many fields. Removal of scale is difficult, often impossible,
and methods t o avoid its formation need additional development. Scaling
results from the precipitation of a solid from a formation water or from
injection water in waterflood operations. The most common causes of scale
are: (1) temperature and/or pressure changes t o which the formation water is
subjected; (2) dilution with makeup water (in secondary recovery opera-
370 COMPATIBILITY O F 0ILF IELD WATERS
tions) or mixing with other formation water containing incompatible ions;

(3) evaporation causing increased concentrations of dissolved solids allowing
saturation t o be reached; (4) supersaturation caused by formation water
flowing through and dissolving slightly soluble solids.
When the composition and temperature of a brine, saturated with CaSO,,
remain constant, precipitation will occur if the pressure drops. Scaling is not
likely with increasing pressure under the same conditions. However, these
conclusions must be modified if the brine is flowing through beds conkaining
soluble compounds of calcium or sulfate or if another water source is
altering the brine concentration. Because of moderate rise in brine tempera-
ture as it travels betwsen the injection wellhead and the bottomhole and the
rapid rise in pressure, scaling of CaS0, is not likely in the injection well.
The formation of BaS0, scale is worthy of special attention. Most barium
compounds are relatively insoluble, and large volumes of brine often are
necessary t o cause heavy BaS0, scale. The most unique characteristic of this
compound is its crystal growth (Weintritt and Cowan, 1967). It will remain
in a supersaturated solution for an unpredictable time and then will precipi-
tate slowly and in a crystal form which has not been duplicated in the
laboratory. Some observers attribute this phenomenon t o the requirement of
a unique solid crystal acting as a seed to promote BaSO, precipitation.
Furthermore, the forming crystal adheres to other larger solids suspended in
the solution or attached to the associated solid phase. This causes the scale
to occur in larger quantities than if it were pure barite.
In the Raleigh field, Smith County, Mississippi, a scale consisting of con-
centric rings of prismatic barite commonly occurs. The barite prisms are
about 0.5 mm in length and contain up t o 1%strontium and lead.
The pumping equipment in the Pisgah field, Rankin County, Mississippi,
often is plagued with a scale composed of metallic lead containing small
fragments of steel. The steel is from the pumping mechanism but the lead
must be from the formation water because the amount of dissolved lead
ranges up t o 100 mg/l. Maintenance of the wells to remove the lead scale
occurs as often as every 10 days.
Knowledge of the solubilities of BaS0, and SrS0, in solutions containing
NaCl, CaCl,, and NaHCO, needs to be increased t o better understand
various precipitation reactions that occur when waters containing these salts
mix. Information concerning the effects of heat and pressure upon these
reactions is lacking.
Solubility of calcium compounds in various salt solutions
Frear and Johnston (1929) measured the solubility of calcite in water

saturated with carbon dioxide and obtained an activity of 4.8 x at
25OC. Ellis (1963) determined that the solubility of calcite was significantly
less in the laboratory salt solutions than in hydrothermal solutions with
similar ionic strength.
SOLUBILITY OF CALCIUM COMPOUNDS 371
Stiff (1952) developed a graphic method of predicting the tendency of

oilfield waters to deposit calcium sulfate. Diagrams can be used to find the
maximum solubility of a salt in waters of similar composition. This informa-
tion is useful in predicting that a given brine has a scale forming tendency.
However, better pressure and temperature data in respect t o their effect on
scale formation are needed.
Akin and Lagerwerff (1965) studied the solubility of calcite in relation t o
ionic strength. The soluble salts used were NaC1, NaHCO,, and CaC12; the
ionic strength of the solutions ranged up t o about 0.09 and their data agreed
well with the Debye-Huckel theory. They also studied the effect of Mg"
and S04-2 and found that the solubility of calcite was enhanced by these
ions relative to theoretical values.
Ostroff and Metler (1966) determined the solubility of calcium sulfate
dihydrate in the system NaC1-MgC12-H20 in 5.50 molal NaCl and 0.340
molal MgC12 admixtures at 28", 38', 50°, 70°, and 90'C. Their results
indicate that the solubility increases in the presence of small amounts of
MgC12 in NaCl solutions up t o about 2.5 molal NaC1. The MgC12 effect
decreases in higher molalities of NaCl until at about 4 mold NaCl a plateau is
reached.
Shaffer (1967) studied the solubility of gypsum in sea water and sea-water
concentrates. He found that the solubility product of gypsum is greater in
the highly concentrated brines and also that in these brines it increased with
increasing temperature.
Glater et al. (1967) developed a method t o measure calcium sulfate scaling
thresholds in saline water samples at 100'C. They found a correlation of
ionic strength with calcium sulfate solubility, and used a gSaphical method to
relate scaling threshold to the concentration of calcium and sulfate ions in
saline water.
Pytkowicz et al. (1967) measured in situ the pressure coefficient of the
aragonitic oolites with pH electrodes, Their results indicate that the pressure
coefficient or the apparent solubility must be known to obtain accurate
solubility data at high pressures.
Fulford (1968) found that the solubility of gypsum or anhydrite increases
with pressure because of a small decrease in total volume as the scale dis-
solves. Subsequently with a pressure drop a supersaturated solution forms
and gypsum precipitates. In very concentrated brines this does not occur
because the solubility of gypsum in very concentrated brines is less depen-
dent upon pressure. He presented several equations t o calculate anhydrite
and gypsum solubilities.
Blount and Dickson (1969) determined the solubility of anhydrite in NaCl
solutions at 100'-450'C and 1-1,000 bars. They found that anhydrite
solubility increased with temperature and NaCl concentrations.
Glew and Hames (1970) determined the solubilities cf gypsum, disodium
pentacalcium sulfate, and anhydrite in sodium chlor,de solutions. Their
results indicated that the solubilities of these comr#ounds decreased in
chloride solutions with molalities greater than 3.5.
372 COMPATIBILITY OF OILFIELD WATERS
Vetter and Phillips (1970) included the effects of complicated downhole

phase equilibria to develop an improved thermodynamic method to predict
deposition of calcium sulfate. According to them the calculated solubility is
as accurate as the experimentally determined solubility; however, additional
data are needed concerning the solubility of gypsum in brines at high pres-
sures. These data are needed to determine which CaS04 compounds are
formed under high pressure in brines. Knowledge of pressure drops either at
the wellhead or within the reservoir is important to determine where scale
deposition occurs.
Solubilities of the sulfates of barium and strontium in saline solutions
Neuman (1933) published results of studies of BaS0, solubility in

aqueous solutions of potassium, magnesium, and lanthanum as chlorides and
nitrates. His data show that BaS04 solubility increases with the increasing
complexity of the major solute, and in the order (3, -1) > (2, -1) > (1,-1)
of equal molality solutions.
Gates and Caraway (1965) analyzed California oil-well scale and found in
a BaS0,-type scale significant amounts of strontium along with iron, calci-
um, magnesium, and some carbonate. Weintritt and Cowan (1967) studied
the unique characteristics of BaS04-scale deposition and concluded, “the
presence of strontium in barium sulfate scales deposited from oilfield waters
appears to be common.” All of the sulfate deposits analyzed contained
strontium sulfate in concentrations ranging from 1.2 to 15.9%, and barium
sulfate in concentrations ranging from 63.7 t o 97.5%.
Templeton (1960) studied the solubility of BaS04 in solutions at 25OC
and at sodium chloride molalities between 0.1 and 5.0. He found that at
constant ionic strength the solubility of BaS0, increases with increasing
temperature, and observed that calcium sulfate exhibits an inverse reaction
with increasing temperature.
Experimental determination of some solubilities of the sulfates of barium

and strontium
A radioisotope-tagged solution of Na2SO4 was prepared from which

aliquots were taken (Davis and Collins, 1971). The radioactive isotope was
35S.One aliquot was used to precipitate BaS04 by addition of an excess of
BaCl,; a second portion was used to precipitate SrS04 by addition of
appropriate equivalents of SrC12.These suspensions were stirred and allowed
t o settle. Following a 2 4 t o 48-hour settling period, the precipitates were
washed onto a 0.45-pm pore size filtering medium, and the washings were
continued until the sulfate ion in the filtrates could not be further reduced.
The tagged precipitate was removed from the filter, dried in an oven at
105OC, and transferred t o storage vials. Standard samples of the sulfates were
prepared by chelation in a 0.W solution of EDTA. Various strengths of
RESULTS AND DISCUSSION 373
5-80 mg/l of BaSO, and 50-800 mg/l of SrS04 were made and used as
reference counting samples for all of the sulfate determinations.
A nonionic detergent (Triton X-100) and toluene emulsion (Patterson and
Greene, 1965) were prepared, whereby l-cm3 sulfate samples in brine could
be counted with greater than 20%efficiency. The emulsion forms a clear gel
and permits a homogenous dispersion of the aqueous phase in the fluor with
no salting out.
Solutions of various salts, such as those usually found in formation waters,
were made up in strengths of 0.005-1.77 molal, and tagged solid barium or
strontium sulfate was added. The chlorides of sodium, calcium, magnesium,
and potassium were prepared, as were solutions of sodium bicarbonate,
sodium borate, and potassium bromide. All of the solutions were stored in
plastic bottles. To determine sulfate solubility, a 20-cm3 portion of one of
the prepared salt solutions was transferred to a small plastic stoppered vial,
and 0.1 g of the solid, tagged sulfate was added. This suspension was shaken
in a wrist-action type shaker for 72 hours and then allowed t o settle a
minimum of 24 hours without opening the vial.
The samples were prepared in duplicate to assure equilibrium, and the
operation was repeated when better precision was needed. The temperature
of the suspension was raised briefly above the stabilized room temperature
(25OC k l 0 C ) with a heat lamp during the shaking period, but no change was
permitted during the last 24 hours nor during the settling period. When the
sample container was opened, it was quickly filtered through a double
Whatman No.42 filter paper, and 1 ml was transferred t o a counting vial
which contained 12 ml of the Triton emulsion and 7 ml of deionized water.
The sample then was counted in a liquid scintillation counter for 50 minutes.
The chelated sulfate standards were counted in the same time period. By this
method, the correction for radioactive decay could be omitted and the
soluble sulfate values determined from a graph of the chelated standards (in
mg/l) versus the counts per minute. Barium was analyzed by emission spec-
troscopy, but adequate precision at levels of 1 mg/l and less was difficult to
achieve in the presence of ionic-strength salts encountered in some solutions.
Results and discussion of the experimental investigation
The values obtained from solubility measurements are shown in Table 12.1.
The amounts of the alkaline sulfates which dissolve in other electrolyte
solutions are tabulated alongside the total ionic strength of each solution.
Ionic strength is the most useful concept yet developed t o include the com-
bined effects of the activities of several ionic species in a solution. Lewis and
Randall (1923) state, “in dilute solutions, the activity coefficient of a given
strong electrolyte is the same in all solutions of the same ionic strength.” It
is defined as s = H m 1 2, ’, where m 1 = the ionic molality, and 2 , = the
charge of the ion in solution, the summation being taken over all ions,
positive and negative. By definition, the activity of the dissolved species
approaches the concentration value (molality) a t infinite dilution.
Since the thermodynamic solubility product Ku = x mso, x y2 and

since y equals unity at zero ionic strength, a plot of log mso, versus the
ionic strength function would extrapolate to zero concentration where log
KuS = log rnso,. Fig. 12.1-3 give plots representing six electrolytes and the
values of Ku'h are determined graphically. The value for the sulfate solu-
bilities in pure water was determined experimentally and agreed with values
in the literature.
In Fig. 12.1, the plot of BaS04 solubility versusds for the six electrolytes
is almost identical at low ionic strength, a phenomenon to be expected from
the statements above. The extrapolated KuS values of all systems are 1.05 x
loe5 (within experimental limits). This fact must be correlated with the
nature of the equation defining ionic strength. The square of the ionic
valence gives the Mg+' and Ca+' ions four times the numerical weight of the
Na+ and K+ ions. Molality values would indicate that the bivalent ions cause
increased solubility effects.
Because borates are present in many oilfield waters, sodium borate was
included t o find differences in sulfate solubility in electrolytes containing a
complex ion. As shown in Fig. 12.1 and 3, the solubility deviated from that
of monatomic electrolytes, and the relationship described does not hold at
higher solubilities of electrolytes containing complex ions.
Another ion commonly found in mineral waters is bicarbonate. Many
water-bearing zones contain limestone and dolomite which slowly erode in
water of low pH. The water carries away carbonates and bicarbonates. In this
study, NaHCO, solutions of 0.005-1.0 molal were saturated with tagged
10.0 -
0.0-
- -
KEY -
-
6.0- 0 CoC12 __
- MgC12 -
4.0 - A No2 84 07 -
0 - A KBr -
I 0 NaCl
0
x 2.0 - X KCI -
0
0
v)
0
m
lI.
0
1.0
0.8
0+ SO4 Mo Io I i t ies
1
0.6
k
_I
a
J
0.4
0
H
0.2
0.I
0 0.5 I .o I .5 2.0 2.5
4 l O N l C STRENGTH
Fig. 12.1. Concentration of saturated B a s 0 4 in strong electrolyte solution.
TABLE 12.1
Solubility of Bas04 and SrS04 in electrolyte solution
Major solute Bas04 5r504

(molality) major solute system major solute system
Bas04 total ionic SrS04 total ionic

(mg/l) strength (mg/l) strength
ca Cl2
0.010 (5.0)* 6.2 0.03016 214 0.0347
0.015 247 0.0504
0.020 7.6 0.06013
0.025 295 0.0815
0.045 260 0.1403
0.050 11.5 0.15020
0.090 508 0.2819
0.100 15.5 0.30027
0.136 590 0.4197
0.200 (17.3)* 17.7 0.60030
0.226 757 0.6959
0.300 16.2 0.90028
0.400 16.3 1.20028
0.456 1,152 1.3947
0.500 16.6 1.50028
0.934 1,942 1.8438
1.000 (16.3)* 11.3 3.00019
2.000 (10.8)* 2.5 6.00004
M m 2
0.005 172 0.0188
0.010 (2.9)* 5.4 0.02979 203 0.0344
0.015 233 0.0499
0.020 6.9 0.05982
0.025 295 0.0805
0.049 394 0.1571
0.050 9.8 0.14927
0.074 422 0.2324
0.099 13.3 0.29693
0.125 530 0.2866
0.196 18.0 0.58831
0.254 731 0.7782
0.474 25.9 1.42244
0.525 1,063 1.5993
0.902 32.0 2.70655
1.637 (44.5)* 33.2 4.93257
-
* See footnote at end of Table.
TABLE 12.1 (continued)


Na Cl
0.010 (5.3)* 3.6 0.01006 134 0.0129
0.015 149 0.0182
0.020 (5.6)* 4.2 0.02007
0.025 172 0.0288
0.050 5.4 0.05009 199 0.0543
0.086 265 0.0914
0.100 (7.3)* 7.1 0.10012
0.172 332 0.1756
0.200 (11.3)* 10.0 0.20017
0.257 420 0.2667
0.431 525 0.4423
0.500 14.8 0.50025
0.869 699 0.8840
1 .ooo (22.3)* 20.2 1.00035
1.771 760 1.7875
2.000 (35.7)* 27.2 2.00047
KCl
0.010 (3.7)* 4.2 0.01007 144 0.0131
0.015 169 0.0185
0.020 4.9 0.02008
0.025 167 0.0286
0.050 6.3 0.05011
0.067 375 0.0754
0.100 8.6 0.10015
0.200 11.2 0.20019
0.202 396 0.2109
0.338 502 0.3492
0.500 16.8 0.50029
0.684 742 0.7001
1.000 21.6 1.00037
1.396 802 1.4139
2.000 (25.8)* 27.2 2.00047
* See footnote at end of Table.
BaS04. However, only trace amounts of barium were found in solution,

though the sulfate content increased with the amount of NaHCO, in solu-
tion. This apparent anomaly can be reconciled by the ionization of the
HC03- ion into CO,-*, which in appreciable concentration would reduce
the Ba+2 ion concentration according to the solubility product K,, = M B x~
MC03.
TABLE 12.1 (continued)


XBr
0.010 (3.6)* 4.1 0.01007 152 0.0133
0.015 163 0.0186
0.020 4.7 0.02008
0.042 215 0.0467
0.050 6.3 0.05011
0.084 262 0.0900
0.100 8.2 0.10014
0.126 320 0.1335
0.200 11.0 0.20019
0.211 420 0.2207
0.426 509 0.4378
0.500 16.2 0.50028
0.866 669 0.8812
1.000 21.8 1.00037
2.000 (23.9)* 26.7 2.00046
Na2 B4 0 1
0.010 6.0 0.03010
0.013 320 0.0462
0.020 7.8 0.06014
0.039 600 0.1307
0.050 12.9 0.15022
0.065 690 0.2100
0.100 (23.7)* 21.0 0.30036
0.200 (33.9)* 34.8 0.60060
Pure Water 2.5 0.00004 114 0.0025
* Parentheses indicate barium ion and sulfate. ion determinations made separately.
The effect of high concentration of CaClz on BaS04 solubility is in-

dicated by the solid curve of Fig. 12.1. At concentrations of 2 molal, the
BaS04 solubility has dropped t o values close to that of the compound in
pure Hz0. The maximum value is reached between 0.2 and 0.4 molal where
the decline begins. The accompanying broken line of Hg. 12.1, which is a
plot of Ba+2 + SO4-' ions determined separately, shows the reduced solu-
378 COMPATIBILITY OF 0IL F IE LD WATERS
-1 IONIC STRENGTH
Fig. 12.2. Concentrations of saturated SrS04 in strong electrolyte solutions of NaCl, KCl,
and KBr.
-
- K a = 2.4 x
I I I I I I I
4 IONIC STRENGTH
Fig. 12.3. Concentrations of saturated SrSO4 in strong electrolyte solutions of MgClz,

CaClz , and Na2 B4 0,.
bility of the Ba ion caused by the equilibrium Ca+’ + S04-2 *

CaSO, ( K s p
< 1.95 10-4).
The effect of the ions of strong electrolyte solutions on SrS04 solubility
is similar t o that observed when BaS04 solubility was studied. The Na+, K+,
C1-, and Br- ions have approximately equal effect, and all determined
values fall on a common curve (Fig. 12.2). The increase in sulfate solubility
is marked in dilute solutions but reaches a maximum a t concentrations with
ionic strength near 1. This is the average value calculated for sea water. When
bivalent ions Mg+’ and Ca+2 are used in the strong electrolyte (Fig. 12.3),
the SrS04 solubility remains of the same relation t o the total ionic strength
as for the monovalent ions.
A study of the system SrS04-NaHC03 -H2 0 was limited by the insolu-
bility of SrCO, . The ionization of the bicarbonate to H+ and COSw2would
result in the precipitation of any Sr+’ which dissolves and leaves the S04-2
in solution. This relationship is similar t o the BaS04-NaHC03 -H2 0 system
and is worthy of special note. That is, when carbonate or bicarbonate waters
are diluted or intermixed with waters containing barium or strontium, an
unstable solution is formed.
Experimental data indicate that maximum sulfate solubility in strong
electrolytes begins at an ionic strength of approximately 1.When the princi-
pal cation in solution is the Ca+2 ion, sulfate solubility decreases after the
ionic strength exceeds unity. Blount (1965), when measuring solubility of
CaSO, in the system CaS04-NaClLH2 0, and Lucchesi and Whitney (1962),
TABLE12.11
Sulfate solubilities in synthetic brines

~ ~ ~~~
Concentration (molal)
brine 1 brine 2 brine 3
Na+ 1.2179 1.7399 2.4359

Ca+’ 0.0250 0.0374 0.0499
Mg+’ 0.0206 0.0823 0.0411
K+ 0.0051 0.0051 0.0193
c1- 1.3019 1.9650 2.6113
Br- 0.0125 0.0188 0.0250
1- 0.0001 0.0000 0.0001
Barium sulfate solubility

Bas04 (mg/l) 60 63 66
Bas04 (molality) 2.57 x D4 2.70 x D4 2.83 x o4
Ionic strength (s) 1.3600 2.1038 3.0278
Strontium sulfate solubility

SrS04 (mg/l) 813 922 , 958
SrS04 (molality) 44.26 x lo4 50.19 x lo4 52.18 x lo4
Ionic strength (s) 1.3777 2.1239 3.0487
when measuring SrS04-NaC1-H2 0 solubility equilibria, found similar

maximums. By using ionic-strength calculations in place of weight per unit
volume, the predictions of mineral water stability become more accurate and
dilutions more feasible.
Three synthetic brines were made with salts concentrations in the range of
many formation waters and containing the major salts found in these waters.
Table 12.11 gives these concentrations and results of a BaS04 and a SrS04
solubility determination. The values found when plotted against the ionic-
strength function of the brine fall on the same curve as the barium salt in
Fig. 12.1 and the strontium salt in Fig. 12.2. No carbonates were added t o
these synthetic brines.
Brine stabilization
Efforts to stabilize the brines used in petroleum production have been

extensive and successful in many cases, but the complexity of the problem in
other cases is reported. Water treating units are considered necessary in
waterflooding operations, but none fully satisfy the operator’s apprehension
that there may be plugging within the reservoir. Addition of solubilizing,
chelating, and clarifying agents to the brine has helped, but economics limit
the quantities used. Tests for compatibility of the fluids as they exist in the
formation and in the wellbore give erroneous results because the subsurface
environment cannot be fully duplicated a t the surface.
To aid brine stabilization programs, several studies of the solubilities of
various relatively insoluble compounds have been made as previously discus-
sed. Usually the results of these studies are reported as solubility products of
various pure compounds (CaS04, BaS04, CaCO, , etc.) in the presence of
other ions, dissolved gases, ion pairs, and various sized crystals of the com-
pound under study. Some efforts have been made using mixed cations and
anions in solution with the compound under study, including limited study
of sulfates in sea water or synthetic sea water (Shaffer, 1967). Various easily
measured parameters such as percent chlorides, total solids, and ionic
strength have been plotted against solubility product of the potential scale
former. Very little correlation suitable for direct field application has been
found. For example, the author has measured BaS04 solubility in CaC12,
MgC12, NaC1, and other salt solutions using ionic strength as the common
property. However, a synthetic sea water containing these compounds and
having comparable ionic strength will dissolve double the weight of BaS04 in
milligrams per liter with respect to any solution containing a single salt.
Fulford (1968) and Vetter and Phillips (1970) proposed useful formulas
and graphs t o use in predicting scaling from calcium sulfate. Fig. 12.443 are
included for possible use in predicting potential scale problems from calcium
sulfate, strontium sulfate, and barium sulfate. The figures are plots of molal
solubility versus ionic strength. The advantage of this plot is that the ionic
strength of any given water can be calculated from its chemical composition,
BRINE STABILIZATION 381
IONIC S T R E N G T H
Fig. 12.4. Solubility o f CaSO4 versus ionic strength of aqueous solutions (Ostroff and
Metler, 1966).
t
y 12- KEY
1
IONIC S T R E N G T H
Fig. 12.5. Solubility of SrS04 versus ionic strength of aqueous solutions containing
CaClz, MgClz, NaCl, KCl, and KBr (Davis and Collins, 1971).
and the solubility of a given compound is a direct function of the ionic

strength of the solution. Therefore, a very good approximation of the solu-
bility of a given compound in a given water solution can be made. For
example, if a water with an ionic strength of 0.1 contains 0.001 molal of
strontium sulfate, it can be assumed that the water is undersaturated with
respect to strontium sulfate as illustrated in Fig. 12.5. However, if the water
contains 0.003 molal of strontium sulfate it is oversaturated and some treat-
ment should be made if the water is to be reinjected.
01 I a I I ,111
I
I , I I I IIII I , I , 1 , 1 ,
0.01 0.02 0.040.0601 a2 04 060.0 1.0 2 4 6 8 3
IONIC STRENGTH
Fig. 12.6. Solubility of B a s 0 4 versus ionic strength of aqueous solutions containing

CaClz, MgClz, NaCI, KCl, and KBr (Davis and Collins, 1971).
Similar curves can be made using appropriate solubility data for calcium
carbonate and for iron compounds. However, it should be noted that appli-
cation of this technique only gives an estimation of the maximum solubility
of a compound in waters of similar ionic compositions. Better data on pres-
sure and temperature and how they affect the solubilities are needed before
adequate prediction equations can be developed.
Mixing of subsurface waters
Mixing of surface and subsurface waters results in solutions which are

either saturated or undersaturated with relatively insoluble compounds such
as calcium carbonate, calcium sulfate, strontium sulfate, and barium sulfate.
These compounds are considered because they often are found in scales
formed because of mixing of formation waters.
Hydrodynamic potentials caused by differences in elevation, weight of the
overlying fluids and rocks, secondary cementation of rock pores (Levorsen,
1967), temperature differences, osmotic pressures, and chemical and
physical reactions cause subsurface waters t o move (Hubbert, 1953).Popov
and Goldshteyn (1957) described a large hydrodynamic system of de-
scending fresh water and ascending saline water which could mix to form a
fresh-saline water mixture. Henningsen (1962)found that recharge waters
into Trinity aquifers were two types of water from strata of different
lithology and with basinward movement of the waters they mixed to form a
third type of water. Mixing of fresh waters with encroaching sea water
occurs according t o Kohout (1960),Columbus (1965),and Upson (1966).
MIXING OF SUBSURFACE WATERS 383
Estimating strontium sulfate saturation in waterflood makeup brines (Biles,

19 72)
The data in Tables 12.1 and I1 were used by Biles (1972)to estimate the
saturation point of strontium sulfate in waterflood makeup waters. Ac-
cording to him, brines used as makeup water for waterflood operations often
are more concentrated in dissolved solids than are the single solute samples
shown in Table 12.1. However, considering that the sodium concentrations
are 96, 93,and 97 mole %, respectively, in the synthetic brines 1, 2, and 3
shown in Table 12.11, it appears reasonable in lieu of experimental data to
extend the NaCl data to the higher.concentration range with these data.
Fig. 12.7 is a plot of the milligrams per liter of strontium sulfate in
solution as a function of the total ionic strength of the solution. The data
were taken from Table 12.1 and 11. A smooth curve can be plotted for Fig.
12.7 if the strontium sulfate value at 1.7875 total ionic strength (Table 12.1)
is ignored. This curve can be extrapolated for use in estimating the amount
of strontium sulfate in milligrams per liter that is likely t o be soluble in more
concentrated brines. A similar curve could be plotted for the solubility of
barium sulfate.
Consider a brine that does not contain the stoichiometric combining
weight ratio of strontium and sulfate as shown in Table 12.111.To compare
the amount of strontium sulfate apparently at equilibrium in this brine with
the solubility data in Tables 12.1 and 11, it is necessary to use another
approach. The solubility product of a solute A, Bm is determined by the
molalities of the ions composing the solute and their activity coefficients:
"--"I L
z
0 1.200-
I-
4
a
I I I
I 1) 2.0 3.0 3
TOTAL IONIC STRENGTH
Fig. 12.7. Solubility of strontium sulfate versus ionic strength of the solution (J. Biles,
written communication, Cities Service Oil Company, Tulsa, Okla., 1972).
TABLE 12.111
Composition of a brine that does not contain a stoichiometric combining weight ratio
of strontium and sulfate*
Ion mg/l me/l Molalit y
Na+ 49,000 2,130 2.24

K+ 220 6 0.00589
ca+' 11,500 574 0.303
Mg+' 2,400 197 0.104
Ba+' 25 < 1 0.00018
sr+' 1,000 23 0.012
Fe" 101 4 0.00189
c1- 106,140 2,990 3.15
s04-' 170 4 0.00189
Total 170,556
* Total ionic strength = 3.54;density at 22OC = 1,120 g/l; grams H20/1= 950.
The activity coefficients are determined primarily by the total ionic

strength of the solution, and in a solution saturated with the solute A, B, :
If the total ionic strength is unchanged, ~ [K,,/(YA~ Y B ~ ) are

Y B and ]
constant. Therefore, the concentration of A in equilibnum with a given
concentration of B in a saturated solution of A, B, is defined:
Plotting the data in Tables 12.1 and I1 for the solubility of SrS04 in
sodium chloride and synthetic brine solutions as the product of the molal-
ities of strontium and sulfate versus total ionic strength, as shown in Fig.
12.8,indicates that the brine in Table 12.111 is undersaturated in SrS04 by
d(280 x lo-') - (227 x lo-') molal. This method is in error to the extent
that the SrS04 solubility is affected differently by the ions in the brine in
Table 12.111 than by the ions in the brines shown in Table 12.11. Neverthe-
less, this approach is valuable in that a reasonable estimate can be made of
the degree of undersaturation of SrS04.
Now consider the advisability of mixing the brine shown in Table 12.111
with another brine which contains less dissolved solids and a comparable
percentage of cations as sodium and about 1,850 mg/l of sulfate. To deter-
mine the solubility of SrS04 in various mixtures of waters, the product of
the weighted average molalities of strontium and sulfate was determined for
MIXING OF SUBSURFACE WATERS 385
-
240 -
200 -
KEY
0 In NaCl solutions
A In synthetic brines
3
TOTAL IONIC STRENGTH
Fig. 12.8. Solubility of strontium sulfate as a product of the molalities of strontium and
sulfate versus the ionic strength of the solution (J. Biles, written communication, Cities
Service Oil Company, Tulsa, Okla., 1972). Filled square shows the product of the
molalities of (Sr )(SO4 ) in Table 12.111 brine.
6
t
J
9
E
a
aw.
z 600 -514 g / kiloliter
0
I-
a 500
a
3
I-
3
a
2
0
* 200
??
100
m
100
TABLE 12.X BRINE, percent
Fig. 12.9. Plot of the supersaturation of a mixture of the brine shown in Table 12.111 with
a brine containing 1,850 mg/l sulfate versus the Table 12.111 brine in percent (J. Biles,
written communication, Cities Service Oil Company, Tulsa, Okla., 1972).
each mixture and compared with the comparable values in Tables 12.1 and I1
and Fig. 12.8. It was determined that a maximum supersaturation of 514
g/kl occurred when the mixture contained 60% of the brine shown in Table
12.111 as illustrated in Fig. 12.9. Mixtures containing less than 9%and more
than 97% of the brine shown in Table 12.111 were undersaturated in &SO4
when mixed with a brine containing 1,850 mg/l of sulfate. The same error
mentioned in the above paragraph will be present, but the correction would
not greatly affect the value obtained using solubility data from Table 12.11.
References
Akin, G.W. and Lagerwerff, J.V., 1965. Calcium carbonate equilibria in aqueous solutions
open t o the air, I. The solubility of calcite in relation to ionic strength; 11. Enhanced
solubility of CaC03 in the presence of Mg” and SO4’-. Geochim. Cosmochim. Acta,
29: 343-360.
Blount, C.W., 1965. The Solubility o f Anhydrite in the Systems C a S 0 4 - H z 0 and
CaS04-NaCl-H2 0 and Its Geologic Significance. Ph.D. Dissertation, University of
California, Riverside, Calif., 179 pp.
Blount, C.W. and Dickson, F.W., 1969. The Solubility of anhydrite (CaS04) in NaCl-H’O
from 100 t o 45OoC and 1 to 1000 bars, Geochim. Cosmochim. Acta, 33:227-245.
Columbus, N., 1965. Viscous model study of sea water intrusion in water table aquifers.
Water Resour. Res., 1:318-323.
Davis, J.W. and Collins, A.G., 1971. Solubility of barium and strontium sulfates in strong
electrolyte solutions. Environ. Sci TechnoL , 5:1039-1043.
Ellis, A.J., 1963. The solubility of calcite in sodium chloride solutions at high tempera-
tures. A m . J. S c i , 261:259-267.
Frear, G.L. and Johnstonb J., 1929. Solubility of calcium carbonate (calcite) in certain
.
aqueous solutions a t 25 J. A m . Chem. SOC.,51:2082-2093.
Fulford, R.S., 1968. Effects of brine concentration and pressure drop o n gypsum scaling
in oil wells. J. Pet. Technol., 20:559-564.
Gates, G.L. and Caraway, W.H., 1965. Oil well scale formation in waterflood operations
using ocean brines, Wilmington, Calif. US.Bur. Min. Rep. Invest., No.6658, 28 pp.
Glater, J., Ssutu, L. and McCutchan, J.W., 1967. Laboratory method for predicting
calcium sulfate scaling thresholds, Environ. Sci Technol, 1:41-52.
Glew, D.N. and Hames, D.A., 1970. Gypsum, disodium pentacalcium sulfate, and
anhydrite solubilities in concentrated sodium chloride solutions, Can. J. Chem.,
48:3734-3738.
Henningsen, E.R., 1962. Water diagenesis in Lower Cretaceous Trinity aquifers of Central
Texas. Baylor Univ. Geol. Studies, Bull., 3~38.
Hubbert, M.K., 1953. Entrapment of petroleum under hydrodynamic conditions, Bull.
A m . Assoc. Pet. Geol., 37:1954-2026.
Kohout, F.A., 1960. Cyclic flow of salt water in the Biscayne aquifer of southeastern
Florida, J. Geophys. Res., 65:2133-2141.
Levorsen, A.I., 1967. Geology o f Petroleum (revised by F.A.F. Berry). W.H. Freeman,
San Francisco, Calif., 724 pp.
Lewis, G.N. and Randall, H.M., 1923. Thermodynamics. McGraw-Hill, New York, N.Y.,
723 pp.
Lucchesi, P.J. and Whitney, E.D., 1962. Solubility of strontium sulfate in water and
aqueous solution of hydrogen chloride, sodium chloride, sulfuric acid and sodium
sulfate by the radiotracer method. J. AppL Chem. (London), 12:277-279.
Neuman, E.W., 1933. Solubility relations of barium sulfate in aqueous solutions of strong
electrolytes. J. Am. Chem. SOC.,55:879-884.
REFERENCES 387
Ostroff, A.G. and Metler, A.V., 196:. Soltbility of calcium sulfate dihydrate in the
system NaCl-MgC12-H20 from 28 to 70 C. J. Chem. Eng. Data, 11:346-350.
Patterson, M.S. and Greene, R.C., 1965. Measurement of low energy beta-emitters in
aqueous solution by liquid scintillation counting of emulsions. Anal. Chem.,
37 :85 4-85 7.
Popov, A.I. and Goldshteyn, R.I., 1967. Hydrologic zoning of hydrostatic systems as a
mineralizing factor in the stratal cover of Central Asia. Dokl. Akad. Nauk S.S.S.R.,
Earth Sci Sect., 17:118-120 (transl.).
Pytkowicz, R.M., Disteche, A. and Disteche, S., 1967. Calcium carbonate in sea water at
in situ pressures. Earth Planet. S c i Lett., 2:430-432.
Shaffer, L.H., 1967. Solubility of gypsum in sea water and sea water concentrates at
temperatures from ambient t o 6 5 C. J. Chem. Eng. Data, 12:183-188.
Stiff, H.A., 1952. A method for predicting the tendency of oilfield waters to deposit
calcium sulfate. AIME, Pet. Trans., 195:25-28.
Templeton, C.C., 1960. Solubility of barium sulfate in sodium chloride solutions from
25' to 95'C. J. Chem. Eng. Data, 5:514-516.
Upson, J.E., 1966. Relationships of fresh and salty groundwater in the Northern Atlantic
Coastal Plain of the United States. U.S. Geol. Surv. Prof. Paper, No.550-C,
pp. 2 35-2 4 3.
Vetter, O.J.G. and Phillips, R.C., 1970. Prediction of deposition of calcium sulfate scale
under down-hole conditions. J. Pet. TechnoL, 22:1299-1308.
Weintritt, D.J. and Cowan, J.C., 1967. Unique characteristics of barium sulfate scale
deposition. J. Pet. TechnoL, 19:1381-1394.
Chapter 13. VALUABLE MINERALS IN OILFIELD WATERS
In the early days of the oil industry, oilfield brines were allowed t o flow
by natural drainage into streams until it was noted that some of the once
good fishing streams contained less fish. Fur-bearing animals had disappeared
in these areas and dead trees and barren soils now bordered these same
streams that once had luxurious vegetation. A few years prior t o 1935,
litigation pertaining t o pollution of fresh water was taking a heavy toll from
oil operators. In certair, older oil producing areas, extensive plots of ground
still are barren, with no living vegetation. The litigations against oil operators
combined with legislation for fresh-water protection to force better disposal
techniques.
At first, evaporation ponds were employed; however, usually more brine
drained into fresh-water aquifers than evaporated. Until recently a widely
employed practice for disposal was the dumping of oil brines into salt-water
bodies when they existed nearby. This disposal method was practiced along
the Gulf of Mexico and in California. Authorities in these areas insisted that
oil separation be highly efficient to prevent damage to fish and oyster popu-
lations. Recently the State pollution boards have ruled that oilfield brines
can no longer be dumped into surface salt-water bodies. In California excess
oilfield waters are being injected into porous subsurface formations as
rapidly as the injection systems can be constructed.
The Plains States are not only situated in a hard water beit, but seldom
have they had an overabundance of usable or surface ground waters. For this
reason, State legislatures passed laws for the protection of fresh-water sup-
plies, allowing the return of oilfield brines t o subsurface formations and
allowing the repressuring or waterflood of oil properties with salt water.
Subsurface brine disposal has since become the common practice. Since the
laws were passed to allow subsurface disposal, more legislation has both
forced such disposal and set up tight controls for it. A survey of cost data on
subsurface injection in 1968 showed that subsurface disposal costs ranged
from 6.6 to 19.8 cents per m3. These figures were based on operating costs
plus 5-year amortization.
Costs vary with the amount of treatment necessary before injection, the
number of production wells per injection well, and the costs of drilling
injection wells or the depth of the injection formation. The depths of
disposal wells normally encountered required no injection pressure. The
brines flow readily into the receiving formations under the gravity head
alone.
390 VALUABLE MINERALS IN OILFIELD WATERS
Most of the 1.23 billion m3 of saline water that is produced yearly with
petroleum is an expense t o oil producers even though some of these waters
contain salts yielding valuable elements which might be economically recov-
ered (Angino, 1967). Elements found in some brines in economic concen-
trations are magnesium, calcium, potassium, lithium, boron, bromine, and
iodine. Many of them are recovered by chemical companies from sea water,
salt lakes, and subsurface saline waters (Collins, 1966; Brennan, 1966).
The recovery of minerals from saline waters dates back to the first time
that someone precipitated a compound from a salt solution. Precipitation is
the most used separation process employed in separating minerals from sea
water or subsurface brines. Research continued on the separation methods
which show economic promise in mineral separation from saline waters.
The Office of Saline Water, U S . Department of the Interior, supports
research aimed at mineral recovery processes to be integrated with fresh-
water plants. The object of this research is to reduce the cost of the
produced fresh water by selling the extracted minerals at a profit. Now
consider mineral recovery as a means of reducing the cost of oilfield brine
disposal. There are additional advantages t o mineral removal other than
profits from the sale of the mineral. For instance, magnesium in sea water
causes great expense because of scale formation in fresh-water plants.
Recovery of iodine and bromine from oilfield brines
Iodine
Iodine consumption in the United States exceeds domestic production.

The Dow Chemical Company is the sole domestic producer of iodine. 75% of
our domestic consumption is imported from Japan and Chile (Miller, 1965).
Chilean nitrate deposits furnish most of the world’s supply of iodine. The
United States and Japan obtain iodine from subsurface brines. In Michigan,
Dow Chemical liberates iodine from brines by chlorination and blows t h e
iodine out with air. Japan recovers iodine from brine by the cuprous iodine,
electrolytic, or active carbon methods.
In the United States, iodine was discovered in an oilfield brine by C.W.
Jones in Louisiana in 1926. The Dow Chemical Company and Jones com-
bined t o produce iodine from a brine well in Louisiana in 1928. At that
time, iodine sold at a price between $9 and $11per kg. In 1929 General Salt
Company began extracting iodine from oilfield brines in California. General
Salt halted operations when Chile cut the iodine price to $3.30 per kg. In
1931 Deepwater Chemical Company began to produce iodine from oilfield
brines in California. Deepwater Chemical halted recovery of iodine from
brines in the late 1950’s.
The Dow Chemical Company moved its iodine recovery operation from
Louisiana to the California oilfield brines in 1932. The move was made for
two reasons. The first reason was that California brines contained 60 ppm
RECOVERY OF IODINE AND BROMINE 391
iodine as compared to Louisiana’s 35 ppm. Secondly, the Dow Chemical

Company was producing the brine in Louisiana from its own brine wells. In
California the brine was produced by oil producers, because older wells
produced 1 0 m3 of brine for every cubic meter of oil.
The Dow Chemical Company used two methods to obtain brine in Cali-
fornia. The first was by paying royalties t o oil producers for brines of high
quality which were delivered at one pick-up point. The second was from an
extensive brine gathering system which Dow built t o collect the brines from
independent producing companies. The second method of disposal was done
for the producers in lieu of royalties. At one time, Dow operated three
iodine recovery plants in California. Only one of the plants utilized a com-
plete iodine recovery process.
In 1961 Dow began iodine recovery from Michigan brines at Midland,
Michigan. These brines are not oilfield brines and although the Michigan
brines contain only 35 ppm, compared to California’s 60 ppm, Dow found
the Michigan operation less costly. Oilfield brines of California have two
disadvantages. First, the brine source near DOW’Soperation dwindled, and
secondly, production costs in California rose.
Several economic advantages were available in the Michigan operation. For
example, the iodine recovery process was integrated with processes for the
recovery of calcium chloride, magnesium hydroxide, magnesium sulfate,
bromine, potassium chloride, and magnesium chloride. The Midland
operation boosted iodine recovery by using brines which were heated t o
91°C for other extraction processes. The absence of oil in the Michigan
brines negated the cost of oil removal. In California, oil removal is necessary
t o prevent interference with the oxidation step in the recovery process. The
brine feed for the Midland operation is composed of brines produced from
various strata in order t o obtain the desired feed for the most economical
products.
Bromine
Bromine is another element that is recovered from oilfield brines. One

plant that is located in Arkansas recovers bromine from the Smackover
formation in the Catesville field. The bromine recovery project was originally
included as part of the plans to unitize the field in 1956.
TABLE 13.1
Bromide recovery economics at Catesville
Minimum economical production 900,000 kg/year

Designed production 1,800,000kglyear
Designed brine feed 1,400 m3/day
Plant cost $ 1,000,000
Plant payout period 6 years
TABLE 13.11
Profitability of Catesville bromide project
Return o n investment 16.7%

Profit as sales percent 14.3%
Profit per m3 processed $ 0.346
Investment $ 1,100,000
Profit per year $ 180,000
Location of a bromine plant at Catesville offered several important advan-

tages such as high bromide content (up to 6,000 mg/l) of the brine, field
operation under a single company (unitization), excellent rail and road
facilities, low-cost fuel, and regional market outlets. Production of
Smackover brine in 1956 was approximately 190 m3 daily from four oil
wells. This quantity of brine was not quite economical for a bromine recov-
ery plant. Additional pumping equipment was installed in some of the wells
in order t o provide 795 m3 of brine daily for the bromine recovery project.
Depleted oil wells later were employed for brine production to raise the
plant feed to 1,430 m3 daily. Table 13.1 shows the initial economics associ-
ated with the bromine project as reported by Kincaid (1956).
The economic data presented in Table 13.1 are based on the bromine
prices of 1956. In 1956 the price of bromine was 66 cents per kg. The price
fluctuates with supply and demand. The data shown in Table 13.11 were
calculated by assuming that all of the bromine is sold at 70 cents per kg, that
the total investment is not more than $1.1million, and that the payout time
is 6 years (Cox, 1967).
Minerals recovered from saline waters
Sodium chloride
Minerals are recovered from practically every type of saline water. By far
the largest recovery is that of sodium chloride in solar evaporation processes.
From the point of view of oilfield brine disposal, where solar evaporation is
possible, the cost of disposal is small. The salts recovered, if any, would
probably pay for the construction of evaporation pits.
Lithium
Lithium is produced from brines by Foote Mineral Company at Silver

Peak, Nevada, and by American Potash and Chemical Corporation at Trona,
California. American Potash and Chemical Corporation recovers a coproduct
lithium sodium phosphate from Searles Lake, California, brines. However,
the largest lithium production is from lithium ore mined in North Carolina
MINERALS RECOVERED FROM SALINE WATERS 393
by Lithium Corporation of America. Domestic production of lithium has not

been reported since the mid-l950’s, because individual companies do not
want t o disclose confidential data. In 1954 about 36,000 metric tons were
produced in the United States. The staff of the U.S. Bureau of Mines (1968)
reports that both the lithium industry and the government are hampered by
restrictions on publishing statistical data on the production and consumption
of lithium metal, alloys, and compounds. These restrictions inhibit the
determination of requirements, the evaluation of market potentialities, and
the planning of future action.
Potassium
Tallmadge et al. (1964) report that the commercial recovery of potassium

from brines only has been attempted on a pilot plant scale. Precipitation
appears the most promising either by the addition of a selective agent
specific t o potassium, or by fractional crystallization of saturated brines.
Potassium compounds occur in many rocks and minerals, but the com-
mercial sources are limited t o soluble salts in bedded salt deposits and brines.
The major deposits of potassium salts in the United States are part of the
Permian Salt Basin that underlies parts of Colorado, Kansas, Oklahoma,
Texas, and New Mexico, and the Paradox Basin of southwestern Colorado
and southeastern Utah. However, commercial beds of potassium minerals
have been found only in New Mexico. Commercial operations have been
limited t o about 1 4 0 km2 east of Carlsbad, New Mexico. These deposits
were discovered by oil well drillers. Commercial recoveries on a limited scale
are made from the brines of Searles Lake, California, and Bonneville Flats,
Utah.
Rubidium
The rubidium-producing industry is very small. During 1958 rubidium

production in the United States was only about 100 kg annually, and during
that year some new technical-grade rubidium compounds were prepared
from alkali carbonate residues of lithium operations. As with many other
minerals found in oilfield brines, the production of rubidium is not
published because it is withheld as confidential company data. However,
with current accumulated stocks and a very small consumption, it is doubt-
ful that the recovery of rubidium from brines would be economical even at
$935 per kilogram.
Cesium
Cesium, both as a metal and as an industry, is similar to rubidium. The

demand for both is small, and the known uses are few. Both cesium and
rubidium are obtained commercially from lepidolite, a lithium mineral.
Cesium and rubidium are byproducts of the lithium industry, and both are
recovered from the residues of the lithium production process are
precipitation from solution. The high concentrations of cesium and rubidium
in the residues and the fact that the amount therein greatly exceeds demand
virtually preclude their removal from oilfield brines on a competitive basis.
Magnesium
Magnesium comprised one-third of the value which Collins (1966) attrib-

uted to the minerals wasted by oilfield brine disposal, and the price used
was that of magnesium metal. In the primary.meta1 form, magnesium com-
mands its highest price. When magnesium is sold as contained in other com-
pounds, its value is less than 2 cents per kg as compared to 77 cents per kg
for primary magnesium. Magnesium and magnesium compounds are
produced from the following four raw material sources: (1) sea water; (2)
dolomite; (3) ores other than dolomite; and (4) evaporite deposits and lake
and well brines. In 1963, well brines, bitterns, and sea water combined with
calcined dolomite or lime accounted for more than half of the domestic
production of magnesium compounds used as chemicals, filters or bases in
many industrial products including basic refractories.
Magnesium and magnesium compounds are produced and recovered by
several companies in the United States. The Dow Chemical Company pro-
duces magnesium chloride crystals, magnesium chloride fluxes, and magne-
sium hydroxide from well brines and calcined dolomite at Ludington,
Michigan. The Michigan Chemical Company produces precipitated magne-
sium carbonate, magnesium hydroxide, and magnesium oxide from well
brines and calcined dolomite at St. Louis, Missouri. The Dow Chemical
Company produces magnesium chloride, caustic-calcined magnesia, and
magnesium hydroxide from sea water and oyster shells at Freeport, Texas.
Magnesium compounds are recovered from solution by precipitation of
magnesium hydroxide. This method is so economical that a large part of the
production of magnesium and magnesium compounds in the United States is
derived from sea water. The Dow Chemical Company is the major source of
primary magnesium and in 1963 it had a capacity at Freeport, Texas, of
50,000 metric tons per year; at Velasco, Texas, the capacity was 34,000
metric tons. In 1963 the U S . production of primary magnesium was 69,000
metric tons.
The Dow process for magnesium recovery from sea water first precipitates
magnesium hydroxide. The hydroxide source is calcium hydroxide made
from oyster shells. After settling and thickening, a slurry of 17%magnesium
hydroxide is attained and neutralized with hydrochloric acid to form a 15%
solution of magnesium chloride. After evaporation and dehydration, the
resultant 48% magnesium chloride solution is mixed with dried magnesium
chloride t o form a paste. The paste is dried t o granules which consist of 74%
magnesium chloride, and the granular material is the feed to electrolytic cells
which produce magnesium metal and chlorine. The chlorine is then con-
verted t o hydrochloric acid which is used in the neutralization step. Shreve
(1956) lists three economic factors of importance in the precipitation of
magnesium hydroxide. They are: (1) the source of the hydroxide; (2)the
dewaterirg procedures used for removal of the magnesium hydroxide from
the dilute solution; and (3) the purification of precipitates.
The source of the hydroxide is the major economic deterrent factor
against the increased use of well brines. Sea water provides the magnesium,
and the sea also furnishes the oyster shells for calcium hydroxide produc-
tion. For well brine feed, dolomite often is employed, and a large source of
dolomite must be economically available. Tallmadge et al. (1964) report that
waste sodium hydroxide has been tested in Japan, but in most areas, calcium
salts are the least expensive sources of the hydroxide. Thus the choice of a
raw material must be belanced in cost against plant size and market. While
Michigan brines contain four t o five times the magnesium concentration of
sea water, the reduced size in necessary equipment for processing the brine
does not completely overcome the cost of producing and disposing of the
brine. This would appear t o make oilfield brines more attractive than other
subsurface brines if a hydroxide source is available at an equivalent expense.
Tallmadge et al. (1964) report methods for extracting magnesium from
brines by methods other than hydroxide precipitation. However, none
appear economically attractive when compared to precipitation unless com-
bined with other processes or products. Among those studied are solar
evaporation to produce chloride, use of ion-exchange resins with lime and
carbon dioxide or waste liquor from the ammonia-soda process, and elec-
trolysis.
Calcium
Calcium production from brines does not appear economical when com-
pared to the source of the world’s calcium consumption. The largest amount
of calcium is produced by the mining of mineral deposits (notably gypsum)
found extensively throughout the world. Proposals for methods t o recover
calcium from brines have been made and are under study, but to compete
commercially beyond extremely small, local demands, considerable research
is needed.
Mixed salts
Mixed salts are precipitated by evaporation of sea water and brines, pro-
ducing crude separations. The costs of these separations are low compared t o
those of highly purified compounds or metals. There are several drawbacks
which prevent greater use of this type of recovery. The product does not
command a high price, the plant must be at the brine sburce, there must be
solar evaporation conditions, and a local market must exist for the majority
of the mixed salts. Uses which have been suggested include heat-treating salt
baths in the steel industry, raw materials for refractory or catalyst manufac-
ture, and fertilizer components.
Precipitation other than by solar evaporation is accomplished by cooling
or adding chemical agents. Simple cooling may be all that is necessary for
more concentrated brines, but fractional crystallization is necessary for
dilute brines such as sea water. Again local markets dictate whether cooling
or freezing processes will yield the correct products for a particular area.
Adding chemicals t o precipitate a specific product is the most fruitful of the
nonsolar evaporation processes. Most of the processes have been aimed at the
production of fertilizer. Potassium and magnesium are the minerals in sea
water that are most valuable for use in fertilizers. Salutsky and Dunseth
(1962) report that metal ammonium phosphates (MAP) containing magne-
sium, calcium, iron, manganese, copper, and many other trace metals com-
prise a high-analysis fertilizer.
The production of metal ammonium phosphates (MAP) in the United
States was started by W.R. Grace and Company in 1960 on a semicom-
mercial scale. The method which Grace used to produce MAP was not
disclosed until 1962 after it was patented. The fertilizers are nonburning,
long-lasting sources of nitrogen, phosphorus, and various trace metals. Be-
cause of their low solubility, MAP’s will not cause salt injury t o seeds or
plants.
In magnesium ammonium phosphate, practically all of the P z 0 5 is avail-
able, and the size of the MAP granules applied to plants determines how long
the nutrients will be available. Thus, availability of nutrients can be control-
led by granulation and, since growing time varies from crop t o crop, MAP’S
can be tailored to a specific crop (Anonymous, 1961). Therefore, fewer
applications are necessary with MAP’S than with fertilizers of higher solu-
bility and high nitrification rates.
W.R.Grace and Company developed the MAP process for two purposes.
First, it is useful t o remove scale-forming materials from sea water before
desalination. Secondly, it would yield the valuable, high-analysis fertilizer,
magnesium ammonium phosphate. In 1962, W.R. Grace and Company
(Anonymous, 1962) reported that the process was ready for the pilot plant.
The process is based on phosphate precipitation. To descale sea water and
produce high-analysis fertilizer at the same time, wet-process phosphoric acid
and anhydrous ammonia are added continuously t o raw sea water. This
precipitates the scale-forming elements - calcium, magnesium, iron, and
other metals - as metal ammonium phosphates and other phosphates. The
precipitated solids are removed by settling, and the descaled sea water is
pumped to the saline water conversion plant. The descaled water holds only
1%of the original magnesium and 5% of the original calcium. The slurry of
MAP’s is dewatered t o about 35--40% solids by continuous centrifuges and
thin it is heated t o 90°C. This converts MAP hexahydrate t o monohydrate.
The slurry is filtered, washed, mixed with recycle fines, and granulated. Fig.
I I
S e t t l i n g ond Descoled sea w o t e k
thickeninq
t
1
Dehydration
1
Filtrotion ond
Wash
wos hing
Gronulotion
Drying Crushing
I k
Undersize
1
Screeninq
Oversize I
Finished product
to storoge
Fig. 13.1. Diagramatic flowsheet for producing descaled sea water and fertilizer.
13.1 shows a process flowsheet for producing descaled sea water and fertil-
izer.
Several questions surround the economics of the process. For a plant
descaling 3,800 m3 of sea water per day (output: about 10,000 metric tons
per year of fertilizer), the fertilizer would have to command a price higher
than that of conventional farm fertilizers. The estimate assumes 1962 market
prices for raw materials (phosphoric acid and ammonia) and does not take
credit for the increased value of the descaled sea water. The cost is just about
the same for Grace’s present method of producing MAP’S. Because of its
premium quality, MAP can go t o the market as a specialty product.
In the phosphoric acid-ammonia process, 2 moles of ammonia per mole of
MAP are lost. to ammonium chloride in neutralizing the phosphoric acid.
Using disodium phosphate in place of the acid loses no ammonia, and using
monosodium phosphate only loses ‘1mole of ammonia. If a cheap method
were developed for producing the sodium phosphates, ammonia waste would
be reduced. The simplest method for producing sodium phosphates involves
the neutralization of phosphoric acid with either dilute sodium hydroxide or

soda ash. Caustic soda and chlorine can both be produced from sodium
chloride brines.
Chlorine
Chlorine is the most abundant element in oilfield brines, and the removal
of all chlorides from oilfield brines would virtually desalt the brine. How-
ever, present commercial methods for the extraction of chlorine from brines
requires evaporation to acquire a saturated brine. Solar evaporation first
produces the saturated brine, then electrolytic processes are employed to
generate chlorine gas. The largest production of chlorides from brines is from
solar evaporation of sea water and salt lake brines. Many oilfield brines are
very close t o saturation with sodium chloride at surface temperatures. Sub-
surface brines are employed as raw materials for chlorine production, but the
amount produced is not significant when compared t o surface brine produc-
tion.
Shreve (1956)describes the methods still in use for chlorine production.
Caustic soda and chlorine are coproducts in the electrolytic process. Purifica-
tion of the brine is necessary to produce a purer caustic soda and lessen
clogging of the cell diaphragm with consequent increased voltage demand.
Calcium, iron, magnesium, and sulfate must be removed. The higher concen-
trations of magnesium and calcium in oilfield brines cause greater expense in
brine purification unless the removed compounds can be sold. Hydrogen,
caustic soda, and chlorine are the products of this type of recovery. The
hydrogen presents a disposal problem and is frequently made into other
compounds such as hydrochloric acid or ammonia or is employed for
hydrogenation of organic compounds.
Iodine and bromine
Iodine and bromine are two minerals which have closely related processes
for recovery from brines. Both are displaced from ions to elements in solu-
tion by chlorination and then stripped from solution by air. When bromine is
liberated by chlorination, iodine is oxidized to the iodate ion. After bromine
is stripped from solution, the iodate ion can be reduced t o free iodine by
treatment with ferrous chloride and then stripped by air as was the bromine.
The greater portion of bromine production in the United States is from
well brines. Slightly over 50% of the domestic production is from well
brines, 35% comes from sea water, and the remainder comes from oil well
brines and saline lake brines. Substantial expansion recently completed by
two producers of bromine from brines should give the industry sufficient
capacity for several years t o supply the expected increase in markets (Miller,
1965). The 2-million-kilogram-per-year plant designed for oilfield brines in
Arkansas by Michigan Chemical represented 50% of the domestic production
at that time. In 1963 the domestic production of bromine was about 12

million kg.
The recovery of bromine from brines and sea water is accomplished by
displacement of the bromide ion with chlorine. The resulting free bromine
dissolved in water is then stripped from the solution with air and recovered.
A large number of modifications t o the basic process (developed by the Dow
Chemical Company) have been proposed and patented from time to time.
Tallmadge et al. (1964) report that studies are being carried out on the effect
of pH, temperature, organic impurities, chlorine concentration, and foreign
ion concentration on the displacement reaction between the bromide ion
and chlorine gas. The use of chlorine water rather than gas for the displace-
ment step has been suggested t o reduce the adverse effects of magnesium and
calcium interference where more concentrated brines are used. Activated
carbon with adsorbed chlorine also has been proposed as a means of carrying
out the displacement step.
Iodine production in the United States uses oilfield brines and sub-surface
brines exclusively. Dow Chemical is the sole domestic producer of crude
iodine. Dow extracts iodine from oilfield brines in California and from deep-
well brines in Michigan. Roughly 1.1 million kg of crude iodine were impor-
ted from Chile in 1963 and about 0.5 million kg were imported from Japan.
Japanese production of iodine is almost exclusively from deepwell brines,
while Chilean production is from nitrate deposits containing the minerals
lautarite and dietzite. Miller (1965) reports that Chilean reserves are in
excess of 1 billion tons as a byproduct of the nitrate minerals industry. It
was price cuts by Chile that forced all domestic producers except Dow out
of iodine production; however, the recent nationalization of the Chilean
mines has changed the picture and pushed the price of iodine to about $5.06
per kg.
Multiproduct production
Tallmadge et al. (1964) report that it is very probable that the most
economic system for removal of minerals from sea water may involve two or
more recovery steps in some integrated fashion. Only a few multiproduct
processes are operated on a commercial scale. The Dow plant in Midland,
Michigan, is such a plant. The first step in studying multiproduct processes is
t o determine how much of each product can be sold. The second step is to
determine the engineering design and production costs for such a plant.
Angino (1967) has pointed out that several processes exist that can be set
up t o recover elements from petroleum-associated waters. These existent
methods and new methods should be utilized and developed to conserve the
natural resources dissolved in brines and to aid in the abatement of soil and
fresh water pollution. The recovery methods used may include desalination
(Christensen et al., 1967), ion exchange (Klein et a1.,'1968), and ion ex-
change plus other methods (George et al., 1967; Waters and Salutsky, 1968).
Fig. 13.2. How gas, oil, and brine are separated after production from subsurface strata.
Fig. 13.2 illustrates a wellhead through which gas, oil, and water are
produced from a subsurface formation. Often they are produced as a mix-
ture and it is necessary t o separate them in a tank such as that illustrated and
sometimes referred to as a gunbarrel. In this tank the water or brine will
settle to the bottom with the oil interface forming over the brine, and the
gas will rise to the top. The gas is drawn off the top, the oil is pumped off
the top of the water and stored in an oil tank, and the water is siphoned
from the bottom into a skimming tank for further oil-water separation. The
water is siphoned from the bottom of the skimming tank into a settling pond
where additional oil-water separation occurs. The brine or water could be
pumped from the settling pond t o a chemical plant for recovery of valuable
elements or to an injection well.
Fig. 13.3 illustrates a possible scheme for recovery of some elements plus
fresh water from the brine. For example, the raw brine could be concen-
trated by a desalination process which would also yield fresh water. Sulfate
then could be taken from the concentrated brine and used t o produce sulfur
or sulfur compounds. Next, iodine could be recovered, then bromine, fol-
lowed by calcium, sodium chloride, and magnesium as suggested in the
figure. The remaining sludge could be dried and disposed of as a solid or it
could be recycled for additional recovery of elements.
FRESH-WATER PRODUCTION 401
RAW B R I N E
1
I DESALINATION
PROCESS I-* PRODUCT
F R E S H WATER
CONCENTRATED
BRINE
PRODUCT
I SULFATE
PRECIPITATION I-* SULFUR AND
SULFUR COMPOUNDS
PRODUCT
I IODINE
RECOVERY
I O D I N E AND
I O D I N E COMPOUNDS
I
t.
BROHINE
RECOVERY
PRODUCT
BROMINE AND
BROMINE COMPOUNDS
PRODUCT
CALCIUM coMp0uM)s
PRODUCT
SODIUM CHLORIDE
SODIUM CHLORIDE OR
RECOVERY SODA ASH
AND
CHLORINE
PRODUCT
MAGNESIUM AND
MAGNESIUM COMPOUNDS
---l---
+
SLUDGE CONTAINING CALCIUM, STRONTIUM,
BARIUM, MAGNESIUM AND OTHER ELEMENT
COHPOUNDS. THE SLUDGE CAN BE D I S P O S E D
A S A S O L I D OR RECYCLED FOR CHEMICAL
RECOVERY
Fig. 13.3. Diagramatic flowsheet for producing fresh water and valuable elements from
brines.
Fresh-water production
Dwindling fresh water supplies and polluted supplies have increased re-
search on how t o best obtain fresh water from saline water. Several plants
throughout the world produce fresh water from sea water. The price of
water for municipal purposes is a highly specific thing. The availability of
fresh water and costs of obtaining it vary from place to place. Conventional
water supplies range in cost from a few dollars per 1,000 m3 to over $260
per 1,000 m3.
The average cost of conventional water supplies in the United States was
$100 per 1,000 m3 in 1952. This was chosen as the goal for saline water
conversion costs. Several authors have estimated ultimate costs of saline
water conversion based on thermodynamic considerations. Dodge and
Eshaya (1960) have examined the minimum expected costs for saline water
conversion. Prior t o their calculations, other authors reported sea water con-
version costs to be ultimately less than $79 per 1,000 m3. Dodge and Eshaya
expanded earlier work t o look at departure from isothermal operation, finite
product recovery, differential as opposed to single stage operation, and salt
concentration in the feed. They found that $90 dollars per 1,000 m3 is the
smallest cost for desalination of sea water.
Consider the case for converting brackish water with 5,000 ppm sodium
chloride. For converting 50% of the feed t o fresh water, 187 kWhr per 1,000
m3 was the power requirement. For 35,000 pprn sea-water conversion, the
power requirement was 1,530 kWhr per 1,000 m3. Both calculations were
for 50% recovery of fresh water from feed, where the average power costs
used in determining conversion costs are 1.5 cents per kWhr. At this rate, the
difference in power costs for sea water over brackish water is $20 dollars per
1,000 m3.
Oilfield brines contain up to seven times the concentrations of dissolved
salts compared with sea water. Would the power be seven times again as
expensive per 1,000 m3? At over $132 for power and $92 for other costs,
the cost of obtaining fresh water from oilfield brines probably would be
prohibitive when consideration is given to the other sources for feed t o a
conversion plant in the same area. An additional factor is that most oilfield
brines with their high concentrations are nearly saturated. Removing 50% of
the water would in essence leave a precipitated salt. Therefore, since no
conversion processes under study deal with saturated brine effluents, it is not
technologically feasible to completely desalt oilfield brines at this time.
Preliminary economic evaluation
The “brine refinery” concept (Collins, 1966) yields a processing plant the
size of a large petroleum refinery. The market prices used were for the
recovery and sale of the pure elements. The $3 billion in sales from 0.95
billion m3 of brine is the highest sales income possible that would result
from recovering and selling the minerals in the form that gives the highest
unit price.
Consider what is probably the best case of a “brine refinery”, a system
that would gather 22 million m3 per year. The cost of gathering and
disposing of this brine would be approximately 9.4 cents per m3. The
question is whether or not minerals could be sold at a profit such that the
disposal expense would be negated or a profit made. First, the minerals t o be
PRELIMINARY ECONOMIC EVALUATION 403
sold must be determined. At 7 ppm lithium, 163 metric tons per year could
be produced. This is a large fraction of present consumption and probably
would depress the sale price. The same holds true for most other elements of
such a refinery.
The assumptions lead to a brine refinery that would process 22 million m3
per year and sell $35 million of minerals. Assuming a 15% return on in-
vestment and a profit of 15% of sales, the plant would require $35 million-
investment and yield 23.6 cents per m3 of brine processed. The original
disposal operation without mineral recovery was such that only $6 million
was invested. The “brine refinery” would turn brine disposal into a profit.
But the new investment is six times that for disposal only. It is doubtful that
any large oil producer would be interested in a 15%return on investment,
and small ones would never gather the cash.
Would a chemical company be interested in such an operation since they
operate at about a 15% return? Companies that currently remove minerals
from brines use brines that are more concentrated in the minerals desired. It
is doubtful that a process could combine several less economical operations
into a more economical one, and this would probably be true even if the
brine was supplied t o a chemical company free of charge. Only in the special
case where an oilfield brine contained a concentration very near t o a brine
that would be the most economical for separation would the oilfield brine be
a best alternate. Therefore, a tax incentive for pollution abatement or some
other economic incentive such as price increase of recovered chemicals is
necessary.
Other economic factors
Table 13.111 illustrates the approximate amount of valuable chemicals per

1 million kg of brine produced from a given depth should contain before it
can be considered of economic value at present market conditions. The
values shown in Table 13.111 should allow a profit if conventional or better
recovery operations are utilized. The marketed end product will influence
the selection of the recovery operation as well as the delivered price. The
TABLE 13.111
Dollar value of dissolved chemicals a brine should contain per million kg of brine produced
from a given depth
Value of dissolved chemicals Depth of well

(m)
$ 462 760
$ 968 2,130
$ 1,430 3,050
TABLE 13.W
Amount of element necessary in 1 million kg of brine to produce a chemical worth $ 550

at the market
Element in the brine Concentration of element Market product

(ppm/106 kg of brine)
Sodium 50,000 sodium chloride

Potassium 14,000 potassium chloride
Lithium 170 lithium chloride
Magnesium 8,000 magnesium chloride
Calcium 11,000 strontium chloride
Strontium 4,000 strontium chloride
Boron 1,400 sodium borate
Bromide 1,700 bromine
Iodide 250 iodine
Sulfur 5,300 sodium sulfate
price information used to make the approximations was taken from the
U.S. Bureau of Mines (1968).
Factors that must be considered in evaluating a saline water as an econom-
ic ore are the cost of bringing it t o the factory, the cost of the recovery
process, and the cost of transporting the recovered products to market.
Assuming that a brine is produced only for the purpose of recovering its
dissolved chemicals, a prime factor is the cost of pumping the brine. It will
cost less to produce the brine from a shallow well than from a deep well.
Therefore, neglecting other factors, a brine must contain a certain amount of
recoverable chemicals before it can be considered economically valuable, and
the farther it must be pumped, the more chemicals it must contain.
Today the possibility of recovering elements from brines that are pumped
t o the surface is increasingly important because the brines present a pollu-
tion hazard if their disposal is improper. Consider the fact that 1 m3 of brine
containing 100,000 ppm of chloride is capable of polluting 400 m3 of fresh
water so that they are unfit for human consumption.
Table 13.IV illustrates the value that chemicals recovered from brines have
at the market; however, because the market fluctuates, these values are
approximate. The column on the left indicates the elements that are found
in petroleum-associated brines, and the second column indicates the concen-
tration that a given brine must contain before it can be used to produce a
given amount of chemical. For example, a brine containing 50,000 ppm of
sodium will contain sufficient sodium in 1 million kg of brine to produce
sodium chloride worth about $550.
The data in Table 13.IV indicate that some petroleum-associated waters
contain sufficient sodium to establish them as economic for the production
of sodium chloride. This is not necessarily true, because factors such as
market demand, ease of recovery, and proximity to market may be discour-

aging in certain geographic areas. Such factors must be fully considered
before startup of a chemical from brine recovery operation. One important
goal that should not be discounted nor overlooked is developing a means of
ultimately disposing of these brines so that they are not a pollution hazard.
Coupling of this goal with the fact that many of these brines contain
economic concentrations of several elements should make such recovery
operations more attractive. Additionally, several important chemicals can be
produced from these elements instead of those shown in the market product
column in Table 13.IV. An example is soda ash, which is a basic chemical in
many manufacturing processes. Furthermore, the figures shown in Table
13.111 are applicable only if the brine is produced solely for the recovery of
its dissolved chemicals. If the brines are pooled from several petroleum
production operations, the cost of pumping the brine becomes less, and the
necessary amounts of chemicals dissolved in 454,000 kg of brine become less.
At the present time, many petroleum-associated brines are injected into
subsurface strata, and it is assumed that they are thus disposed of perma-
nently (Crouch, 1964). However, this method of disposal appears subject to
question, because in some instances, fresh waters apparently have been
polluted by disposal of brines. Subsurface disposal operations are suspected
in certain areas as possibly contributing to increased earthquakes and ground
tremors (Evans, 1966; Bardwell, 1966).
The storage of brines in earthen pits is known to cause pollution of nearby
soils and streams. Such ponds which have been abandoned for 10 years still
contribute to soil pollution (Bryson et al., 1966). Sound conservation should
favor the recovery of valuable elements from brines, and with proper plan-
ning, the recovery processes should aid in the ultimate disposal of unwanted
brines. Conservation of this type not only will develop new resources, but
will benefit the oil producer and the national economy and will aid in
abating pollution of soils, potable waters, and streams.
Work necessary f o r an exact preliminary evaluation
Aries (1954) spells out the marketing research techniques employed in the
chemical industry. There are ways to quickly determine where a market for a
product is. Usually these places are currently served by some producer or
another. If the competition is located far from the market, then an evalua-
tion of a closer area source is readily made. To find product users, the
following methods and approaches are utilized: advertising, company
analysis, product analysis, industry analysis, use analysis, and other miscel-
laneous methods. If new markets must be found, the following types of
work are utilized: personal interview, questionnaire, trade analysis, company
records, and published sources. Before an economic analysis can be made for
a given area, probably several man-months of the listed methods would be
required. The product of this type of market study would be a list of

elements and compounds that could be sold from a given place. The quanti-
ties and prices obtgined would then allow an economic calculation of the
production costs.
With the quantities, prices, and production costs in hand, it is still not a
simple matter t o determine what type of plant t o operate. Regardless of
whom the investor might be, he will want to know what return on invest-
ment he will get, what risk is involved, and what payout period exists for the
project. Depending on the investor, he may want t o limit the plant size by
the amount of money he can invest. This does not simply scale down the
plant. It may rearrange various ratios of certain products produced in order
to give the investor the combination of profits, return on investment, risk,
and actual size of investment that he desires.
Locations o f valuable brines
Table 13.V lists the approximate geographic locations where subsurface

saline waters containing valuable elements are found. The numbers in the left
column of the table correspond to the numbered arrow locations on Fig.
13.4. The second column in the Table indicates the age of the geologic strata
from which the waters were obtained. These waters are in or near oil-
productive sedimentary basins. Concentrations of various elements present in
the waters are given in columns 3 through 11 of Table 13.V. These concen-
trations are representative of one or more subsurface waters from each loca-
tion; however, the concentrations should not be considered typical of all
subsurface waters in an area or stratum. For example, some waters near
location 1 from Mississippian age strata may contain 1,000 ppm of bromine,
while other waters 80 km away, but from the same geologic strata, may
contain 3,600 ppm of bromine. The elemental composition of ocean water is
consistent; the composition of subsurface saline waters is inconsistent.
Fig. 13.5 is a map showing some areas in the United States where brines
containing high concentrations of sodium are found. The solid circles on the
figure represent areas where brines containing 75,000-80,000 mg/l of
sodium can be found, the open circle represents brines containing
80,000-95,000 mg/l, and the triangle represents brines containing more
than 95,000 mg/l.
Fig. 13.6 is a map showing some areas in the United States where brines
containing high concentrations of calcium are found. On this figure the solid
circle represents brines containing 20,000-30,000 mg/l of calcium, the open
circle 30,000-50,000 mg/l, and the triangle more than 50,000 mg/l.
Fig. 13.7 is a map illustrating some of the areas in the United States where
high concentrations of magnesium in brines are found. On this figure the
solid circle indicates brines containing 5,000-10,000 mg/l of magnesium,
the open circle 10,000-30,000 mg/l, and the triangle more than 30,000
mg/l.
PRELIMINARY ECONOMIC EVALUATION
l l c
TABLE 13.V
Geographic location, geologic age of saline water-bearing strata, and concentration of some of the elements found in the brine's
Location* Age of subsurface Concentration (ppm)
strata lithium sodium potassium magnesium calcium strontium boron sulfur chloride bromide iodide
1 Mississippian 10 28,000 40 10,000 60,000 3,000 40 400 179,000 3,200 40 c

2 Permian 30 100 1,000 25,000 100 5 ND 20,000 9,000 ND ND
3 Permian 40 55,000 2,500 9,000 30,000 ND 90 400 166,000 1,200 25
4 Devonian 100 66,000 10,000 5,000 40,000 2.000 ND 100 198,000 700 20
5 Cambro-Ordovician ND 8,000 ND 11,000 20,000 ND ND 350 79,000 ND ND
6 Pennsylvanian 5 51,000 100 600 10,000 1,000 10 30 98,000 600 1,000
7 Jurassic 100 68,000 4,000 5,000 30,000 ND ND 60 175,000 5,000 10
8 Miocene 15 73,000 600 6,000 30,000 ND 60 60 ia4,ooo 200 20
9 Devonian 25 74,000 700 5,000 30,000 900 ND 600 182,000 2,000 40
10 Devonian 60 14,000 8,000 15,000 70,000 1,500 300 400 200,000 2,500 40
11 Mississippian * 16,000 9,000 11,000 14,000 800 ND 20 90,000 1,500 40
12 Devonian 10 58,000 3,000 5,000 20,000 1,000 ND 60 143,000 1,300 30
13 Devonian 90 72,000 2,000 4,000 35,000 ND ND 4 186,000 1,800 20
* See arrows in Fig. 13.4. ND = not determined.

Fig. 13.5. Approximate geographic locations of brines containing high concentrations of

sodium.

calcium.
' 0
A
LEGEND
5,000-10,000mg/l
10,000-30,000
> 30,000

magnesium.

bromine.
DISPOSAL BRINES 411
Fig. 13.8 is a map illustrating some of the areas in the United States where
high concentrations of bromide in brines are located. On this figure the solid
circle represents brines containing 1,500-2,000 mg/l of bromide, the open
circle 2,000-3,000 mg/l, and the triangle more than 3,000 mg/l.
Disposal brines
In an attempt to acquire as much information as possible about salt water

disposal facilities in the various States, the literature was surveyed and State
and Federal agencies and oil companies were contacted. The acquisition of a
true compilation of the exact number of disposal facilities, disposal wells,
and total number of barrels of brine injected was an impossible task without
a large surveillance force; however, the data in Table 13.VI are reasonably
representative.
To determine the value of the minerals in the brines flowing into these
disposal systems, samples were obtained from 40 systems. The samples were
analyzed for concentrations of lithium, sodium, potassium, magnesium, cal-
cium, boron, ammonium, sulfate, bicarbonate, chloride, bromide, and
iodide.
Table 13.VII lists the state, county, subsurface formation, and dissolved
solids (DS) from which the brine samples were obtained. The specific gravity
of each sample plus the ionic values determined in the laboratory also are
given in Table 13.VII. Sample 1is sea water, sample 2 is a brine that contains
a high concentration of iodide, and sample 3 is a brine from which bromine
currently is extracted.
Worth and value estimates
The estimates of the value of a brine are related to the market and the
recovery process. The market is dependent upon demand; however, in the
following estimates the demand was not considered. According to Chris-
tensen et al. (1967):
Brine value = market value of products - operating costs exclusive of

brines value - fixed costs - profit
The maximum value of a brine can be found by letting the “return on

investment” equal zero, and the brine worth is:
Brine worth = brine value + profit = market value of products - oper-

ating costs - fixed costs
The brine value is less than the brine worth by the amount of profit
expected. Although the information necessary to obtahi an accurate calcula-
TABLE 13.VI
States where oilfield brines are disposed, total number of subsurface salt water disposal wells (SWDW) and largest disposal facilities
State Total number of Largest SWDW SWDW SWDW Remarks

SWDW ( m3 /day) 2 4700 m3/day 2 1590 m3 /day
Ah. 14* 2,385

Alaska no data
Ariz. 4 191 0 0
Ark. 421 9,221 5 14 + bromine plant effluent
Calif. 216 25,120
Colo. 25* 2,544
Fla. 5 954 0 0
Ill. no data
Ind. 325 159 0 0
Kans. 3,150 3,180 2**
KY. 60** 111 0 0
La. 1,304 average SWDW is 21 7 m3 /day
Mich. 52 2 318 0 0 average SWDW is 32 m3 /day
Miss. 370 1,113 0 0
Mont. 20 0 239 0 0
Nebr. 3 135 0 0
Nev. - - 0 0 only 7 producing wells for State
N. M. 300 2,703 0 4
N. Y. - - 0 0
N. D. 62 318 0 0
Ohio 28* 318 0 0 2
Okla. 4,900
- - 2P
Pa. 0 0 most of this is by ponding
S. D.
Texas
-
7,173**
- 0 0 ponding used
number of SWDW permitted from
2
_.
P
U
1950-1971
8
Utah 3** 143 0 0
W.Va.
wyo.
95**
19*
159 0 0 5
2--
-
* Salt water disposal Systems that may have more than one salt water disposal well per system.
* * Approximate.
TABLE 13.VII
Analyses of some disposal brines, seawater, and a proven economic brine*
Brine State County Formation Sp. gr. Constituents (mg/l)
Na Ca Mg K Li B NH4 CI Br I SO4 HCOi DS

1 Seawater - - -
c
1.025 10,500 400 1,350 380 0.17 4.6 19.000 65 0.06 3,468 140 35,308
2 Okla. Kingfisher Oswego 1.124 56,250 8,300 260 180 14 18 300 98,300 1,500 1,300 180 50 166,652
3 Ark. Columbia Smackover 1.230 74,000 44,440 4,340 4,410 370 200 - 202,050 5,725 15 220 95 335,865
4 Kans Pawnee Arbuckle 1.036 14,430 2,480 700 260 20 10 20 32,850 60 10 2,000 450 53,290
5 Kans. Barton Arbuckle 1.025 9,850 1,450 490 75 3 10 30 19,460 50 5 2,350 350 34,123
6 Kans. Butler Hunton 1.012 5,990 760 260 70 3 0 0 10,380 20 2 1,400 60 18,855
7 Kans. Ellis Arbuckle 1.034 16,800 2,630 690 190 10 5 0 30,500 50 2 2,880 315 54,072
8 Kans Pratt LKC-Arb 1.020 9,400 1,200 320 105 5 3 0 17,YOO 50 3 1,100 250 30,332
9 Kans. Barton LKC-Arb 1.050 23,300 4,300 1,300 160 5 12 30 45,100 150 10 2,270 260 76,895
10 Ark. Palm Graves 1.046 19.900 3,500 900 200 5 10 45 42,200 500 10 0 170 67,439
11 Ark. Ouachita L. Graves 1.048 21,100 3,800 1,030 160 5 16 40 43,100 400 10 0 160 69,819
12 Ark. Ouachita L. Graves 1.046 20,500 3,800 930 140 5 12 40 42,400 600 10 0 60 68,495
13 Ark. Union Smackover 1.192 64,200 34,500 3,950 1,845 160 140 320 178,100 2,450 5 650 100 286,420
17 Ark. Columbia Smackover 1.162 54,500 27,600 1,315 3,500 230 160 50 150,000 3,500 5 190 200 241,250
18 N.M. Lea San Andres 1.020 9,150 1,500 500 245 5 10 70 17,800 50 3 2,000 1,000 32,329
19 N.M. Lea Devonian 1.036 18.200 1.850 500 370 10 0 60 32,650 30 0 2,260 500 56,428
20 N.M. Lea Devonian 1.028 i3:goo i$oo 340 20 2 0 0 24,300 25 0 2,000 600 42,687
21 N.M. Lea Penn 1.043 21,600 2,840 770 150 10 40 90 42,600 210 5 360 380 69,055
22 N.M. Lea Devonian 1.039 19,350 2,400 410 560 10 5 0 37,240 40 2 1,630 490 62,137
23 Texas Gaines Devonian 1.025 12,380 1,970 365 400 5 5 100 22,400 40 0 610 590 38,865
24 Texas Cherokee Woodbine 1.056 30,000 8,650 345 105 2 40 50 49,100 270 30 240 400 89,232
25 Texas Rusk Petit 1.153 58,700 10,320 1,130 790 35 10 50 135,500 210 30 270 0 207,045
26 Texas Cherokee Woodbine 1.070 36.200 3.300 690 860 2 0 40 61,300 400 35 30 300 103.1 57
27 Texas Wood Woodbine 1.065 34;OOO 10;530 110 550 5 5 10 57,100 370 30 90 400 103,200
28 Texas Wood SubClarkville 1.037 21,200 840 205 360 2 10 25 31,800 180 35 0 450 55,107
29 Texas Wood Paluxey 1.076 34,600 6,750 970 250 10 20 40 68,700 100 25 420 300 112.185
30 Texas Hopkins Paluxey 1.010 5,640 630 40 50 2 10 0 8,350 70 5 120 500 15,417
31 Ala. Mobile Rodessa 1.031 12,180 5,630 480 400 5 0 90 29,400 40 10 710 190 49,135
32 Ala. Mobile Rodessa 1.039 14,500 6.750 550 320 10 12 80 37,000 50 10 300 160 59,742
33 Ala. Mobile Rodessa 1.052 18,400 8,860 680 460 10 30 90 47,540 30 12 400 140 76,652
34 La. LaSalle Wilcox 1.064 35,600 1,650 600 310 5 25 250 62,500 70 20 0 380 101,410
35 La. Calcasieu - 1.084 44,800 3,960 230 300 5 10 25 74,600 25 20 0 140 124,115
36 La. Cameron Miocene 1.076 42.600 2,335 135 200 5 12 12 68,900 30 20 0 300 114,549
37 Calif. Fresno - 1.026 13,600 1,855 780 200 1 5 0 28,870 15 20 0 240 45,616
38 Calif. Kern Kern 1.000 210 70 5 15 1 0 0 330 2 0 40 130 803
39 Ariz. Apache L.Hermosa 1.012 7,760 1,520 50 20 1 0 12 11,600 20 12 0 170 21,165
40 Okla. Woods Hunton 1.123 57,600 10,120 1,640 1.000 10 40 260 115,500 326 150 350 120 187,096
41 Okla. Oklahoma Wilcox 1.151 68,750 13,270 2,460 980 10 0 140 138,600 540 10 520 85 225,365
42 Miss. Wayne Wilcox 1.003 2,880 50 10 10 0 0 0 3,861 3 0 410 255 7,479
* Sp. gr. =specific gravity; DS = dissolved solids,

TABLE 13.VIII
Formulas for calculating maximum worth, brine worth, and brine value
____
Maximum worth = ( X i ) (market value of compound i)*

Brine worth = M.W. - (market cost + fixed charges)
Assume:
brine worth = M . W . 4 . 7 5 (M.W.)
Also assume:
brine value = M.W. x 0.1
* X = amount of compound, and i = number of compounds.

tion of brine value can be gained only by detailed market research, an
estimate can be made by assuming which products will be recovered from
the brine, calculating their market values, and relating the total values (or a
single product value) back to the brine value after assuming that the brine
worth equals a fraction of the total value.
Table 13.VIII presents formulas for calculating the maximum worth, brine
worth, and brine value. In this study, the maximum worth was calculated
TABLE 13.IX
Value of assumed recoverable compounds used in calculating brine value
Cation Compound Cation ($/ton*) Compound ($/ton)

Na NaCl (rock) 18.09 7.11
Mg MgClz (99%) 259.38 66.14
MgS04 346.99 69.22
Li LiCl (technical) 11,515.10 1,87 3.91
Sr SrC13 518.89 286.60
K KCl 58.88 30.86
Ba BaClz (technical) 284.29 187.39
Ca CaClz 120.64 43.54
NH4 NH4 C1 412.26 138.89
Anion Compound Anion ($/ton) Compound ($/ton)

B NazB40, * lOHzO 519.72 55.39
c1 NaCl (rock) 11.72 7.11
so4 NazS04 (salt cake) 45.64 30.86
MgS04 86.75 69.22
Br NaBr 1,135.69 881.84
I NaI (U.S.P.) 9,114.61 7,716.10
HC03 NaHC03 81.24 59.03
(303 CaC03 26.20 15.71
* Metric tons.
WORTH AND VALUE ESTIMATES 41 5
TABLE 13.X
Value of brine constituents*
Assumed brine composition (kg/rn3 of brine):

Calcium 23.36
Magnesium 2.25
Potassium 5.27
Lithium 0.34
Boron 0.31
Sodium 57.65
Bromide 1.95
Iodide 0.04
Sulfate 0.11
Bicarbonate 145.60
Assumed products (kg):
NaCl 146.01 at $ 7.ll/ton** = 1.04
CaCl2 64.73 at $ 43.54lton = 2.82
MgCl2 8.75 at$ 66.14/ton = 0.58
KCl 10.06 at $ 30.86/ton = 0.31
LiCl 2.05 a t $ 1,873.91/ton = 3.84
Na2 B4 0 7 10H20 2.97 at $ 55.39lton = 0.16
NaBr 2.52 at $ 881.841ton = 2.22
NaI 0.04 at $ 7,716.10/ton = 0.31
MgS04 0.13 at $ 69.221ton = 0.01
NaHCO 0.10 at $ 59.031ton = 0.01
Maximum worth = $ 11.30

Brine worth = $ 11.30-314 (11.30) = $ 2.82/m3
Brine value = $ 11.30 x 0.1 = $ 1.13
* Assuming: 75% of market cost is operating and fixed charges.

** Metric tons.
from the market value of compounds that can be derived from the ions
found in the brine. This is not necessarily the m d i m u m worth because the
bnnes contain some ions other than those determined. For example, the
brines probably contain strontium, barium, rubidium, manganese, etc.
Table 13.IX gives the values of chemicals (Anonymous, 1971; U.S. Bureau
of Mines, 1969) that were used to calculate the brine worth of a sample, as
shown in Table 13.X. Note that the brine worth in Table 13.X depends upon
the products that are assumed t o be recovered, and that 75% of the market
cost is operating and fixed charges.
Table 13.XI illustrates the values that were found for brine worth and
brine value for each of the brines collected for the study. Also shown in
Table 13.XI is a ratio of brine value for a commercial brine (brine 3) over the
disposal brine. Only brines 2, 13, 14,15,16, and 17 hadratios of less than 2,
TABLE 13.XI
Brine worth, brine value, and ratio commercial brine value/disposal brine value
Brine Brine worth Brine value Ratio State

($/m3) ($/m3)
~ _ _ ~ _ _ ~
1 0.19 0.08 19.38 Sea water
2 3.69 1.45 1.07 Okla.
3 3.86 1.55 1.00 Ark.
4 0.28 0.P1 14.09 Kans.
5 0.17 0.07 22.14 Kans.
6 0.10 0.04 38.75 Kans.
7 0.28 0.12 12.92 Kans.
8 0.14 0.06 25.83 Kans.
9 0.39 0.16 9.69 Kans.
10 0.41 0.17 9.12 Ark.
11 0.41 0.17 9.12 Ark.
12 0.82 0.33 4.70 Ark.
13 2.38 0.95 1.63 Ark.
14 2.50 1.00 1.55 Ark.
15 2.99 1.19 1.30 Ark.
16 2.61 1.04 1.49 Ark.
17 2.48 0.99 1.57 Ark.
18 0.18 0.07 22.14 N.M.
19 0.22 0.09 17.22 N.M.
20 0.16 0.07 22.14 N.M.
21 0.33 0.13 11.92 N.M.
22 0.25 0.10 15.50 N.M.
23 0.19 0.08 19.37 Texas
24 0.52 0.21 7.38 Texas
25 1.11 0.33 4.70 Texas
26 0.48 0.19 8.16 Texas
27 0.64 0.25 6.20 Texas
28 0.26 0.11 14.09 Texas
29 0.51 0.05 31.00 Texas
30 0.10 0.04 38.75 Texas
31 0.32 0.13 11.92 Ala.
32 0.37 0.15 10.33 Ala.
33 0.53 0.21 7.38 Ala.
34 0.34 0.14 11.07 La.
35 0.38 0.15 10.33 La.
36 0.33 0.13 11.92 La.
37 0.23 0.09 17.22 Calif.
38 0.01 0.01 155.00 Calif.
39 0.17 0.07 22.14 Ark
40 1.06 0.42 3.69 Okla.
41 0.96 0.39 3.97 Okla.
42 0.02 0.01 155.00 Miss.
indicating that the majority of the brines probably do not contain enough
valuable minerals t o be considered commercial by themselves. However,
these six brines may warrant further investigation, it sufficient brine is avail-
able.
CONCLUSIONS 417
Brine 1 is sea water, and some chemicals are recovered from sea water.
Therefore any brine that is disposed of in large volumes and has a ratio of
less than 20 may warrant investigation as a source of minerals because these
brines may be considered as polished ores. Brine 40 along with brine 2 may
contain commercial amounts of iodine. Although the current market for
iodine is attractive, this market may change, depending upon the interna-
tional political atmosphere. Nevertheless the market for iodine once estab-
lished should be fairly stable.
Brines 38 and 42 cannot be considered brines; in fact, brine 38 is almost
potable and with little treatment would be potable. Brine 42 could be used
for irrigation and as drinking water for certain types of livestock.
Conclusions
Some brines contain valuable minerals that if recovered would help pay
for part or all of their disposal costs. Recovery of certain minerals and
potable water should lower the potential of the disposed brine as a pollutant.
Brine value and brine worth formulas should be applied to disposal waters to
determine the relative value of their recoverable minerals.
References
Angino, E.E., 1967. Dissolved salts in oilfield brines - a wasted resource? In: E.E.
Angino and R. Hardy (Editors), Proceedings 3rd Forum o n Geology of Industrial
Minerals -State GeoL Sum. Kansas, Spec. Distrib. PubL, No.34, pp.120-125.
Anonymous, 1961. Grace moves ahead with new fertilizer. Chem. Eng. News, 39:83-84.
Anonymous, 1962. Descaling: route to MAP. Chem. Eng. News, 40:52--53.
Anonymous, 1971. Current prices of chemicals and related materials. Oi4 Paint, Drug
Rep., 200:24-37.
Aries, R.S., 1954. Marketing Research in the Chemical Industry. Chemonomics, New
York, N.Y., 220 pp.
Bardwell, G.E., 1966. Some statistical features of the relationship between Rocky Moun-
tain arsenal waste disposal and frequency of earthquakes. Mountain Geol., 3: 37-42.
Brennan, P.J., 1966. Nevada brine supports a big new lithium plant. Chem. Eng.,
7 6 :86-8 8.
Bryson, W.R., Schmidt, G.W. and O’Connor, R.E., 1966. Residual salt of brine affected
soil and shale, Potwin area - Butler County, Kansas. Kansas State Dep. Health, Bull.,
3( 1):28 pp.
Christensen, J.J., McIlhenney, W.F., Muehlberg, P.E., Hunter, J.A., Heintz, J.A., Jebens,
R.H. and Bacher, A.A., 1967. A feasibility study on the utilization of waste brines
from desalination plants, I. U.S. Off. Saline Water Res. Dew. Progr. Rep., No.245, 359
PP.
Collins, A.G., 1966. Here’s how producers can turn brine disposal into profit. Oil Gas J.,
64: 112-1 13.
Cox, R.L., 1967. An examination of the feasibility of mineral recovery from oilfield
brines. Dep. Chem. Pet. Eng., Univ. Kansas, 41 pp., unpublished.
Crouch, R.L., 1964. Investigation of alleged groundwater contamination Tri-Rue and
Ride oilfields, Scurry County, Texas Texas Water Comm. Rep., No.LD-O464MR, 16
PP.
Dodge, B.F. and Eshaya, A.M., 1960. Thermodynamics of Some Desalting Processes.
Advanced Chemistry Series, No. 27. American Chemical Society, Washington, D.C.,
246 pp.
Evans, D.M., 1966. The Denver area earthquakes and the Rocky Mountain arsenal
disposal well. Mountain Geol., 3’:23-26.
George, D.R., Riley, J.M. and Crocker, L., 1967. Preliminary process development studies
for desulfating Great Salt Lake brines and sea water. U S . Bur. Min. Rep. Invest.,
No.6928, 34 pp.
Kincaid, E.E., 1956. Two moves pay off at Catesville. Oil Gas J., 54:96-98.
Klein, G., Cherney, S., Ruddick, E.L. and Vermeulen, T., 1968. Calcium removal from
sea water by fixed-bed ion exchange. Desalination, 4:158-166.
Miller, W.C., 1965. Bromine. US. Bur. Min. Bull., 630:159-164.
Salutsky, M.L. and D u m t h , M.G., 1962. Recovery of Minerals from Sea Water by
Phosphate Precipitation. Advanced Chemistry. Series, No. 38. American Chemical
Society, Washington, D.C., 199 pp.
Shreve, R.N., 1956. The Chemical Process Industries. McGraw-Hill, New York, N.Y., 2nd
ed., 1004 pp.
Tallmadge, J.A., Butt, J.B. and Solomon, H.J., 1964. Minerals from sea salt. Ind Eng.
Chem., 56:44-56.
U.S. Bureau of Mines, 1968. U.S. BuMines Minerals Yearbook, Metals, Minerals, and
Fuels. Washington, D.C., Vol. 1-11, 1208 pp.
U.S. Bureau of Mines, 1969. U S . BuMines Minerals Yearbook, Metals, Minerals, and
Fuels. Washington, D.C., Vol. 1-11, 1194 pp.
Waters, Jr., O.B. and Salutsky, M.L., 1968. Separating potassium and sodium sulfate from
brines and bitterns. U.S. Bur. Min., W.R. Grace Company, U.S. Patent, No.3,402,018.
Chapter 14. SUBSURFACE DISPOSAL
Because many oilfield waters contain appreciable quantities of dissolved

solids that are capable of polluting fresh waters and lands, they must be
disposed of in some manner. One of the most generally used methods of
disposal is injection into a subsurface aquifer. Sedimentary rocks that were
deposited in an ancient marine environment are the most likely type to
possess the necessary geologic characteristics for injection sites. The technol-
ogy for the installation of disposal wells for oilfield wastes is highly devel-
oped, and a similar technique has been applied to disposal of other types of
wastes (Donaldson, 1964). It is believed that about 75,000 wells have been
drilled in the United States for salt-water injection and disposal wells by oil
companies (Smith, 1970).
History of brine disposal operations
Wells for the production of salt water were in operation in the United
States as early as 1800, 59 years before Col. Drake brought in the first oil
well. Salt well operators were not happy with the discovery of oil in some of
their operations and not knowing what t o do with the oil, they often moved
t o new locations to avoid the “messy nuisance” which spoiled their opera-
tion. Today salt water is the messy nuisance that oil-well operators must
handle.
Early oil operators simply allowed the produced brines t o run off into
streams or drain into fresh-water aquifers. Landowners were so ecstatic over
the royalty payments that the oil producers could do whatever they desired
with the salt water. Landowners simply took their checks and moved into
new locations. Then population density increased, fresh water was being
polluted, farm land was damaged, and land was more valuable because of its
increasing scarcity compared to the population. Therefore, controls on
oilfield brine disposal became necessary.
The salt-water brine pond was used as an early method of keeping the
brines from fresh-water drainage; however, earthen ponds often leak, and
this method polluted fresh-water supplies. Sometimes these brine ponds were
called evaporation ponds. However, it was found that in the majority of
cases, evaporation and rainfall compensated each other, with the brine vol-
ume continually increasing (Jones, 1945). In west Texas, evaporation ponds
are successful as a means of brine disposal. In the colder climates, winter
usually caused a continuous gain in brine pond gage height. These gains were
420 SUBSURFACE DISPOSAL
not overcome in the summer months. In cases where evaporation ponds are
well-lined to prevent drainage, the problem of how to handle the salt
deposits begins to mount up. Where solar evaporation conditions are favor-
able, very little salt is needed for the roads when it snows because it seldom
snows there.
The expense of oilfield brine disposal is the least where salbwater bodies
are nearby, but even here some care must be taken. The oil content must be
less than 30 pprn for disposal into the oceans in coastal areas, otherwise, the
oil collects on the shore and becomes a hazard to oyster and fish life. Along
the Gulf of Mexico strict controls by wildlife authorities and oyster and fish
industries monitor the brines disposed in the Gulf. A small quantity of oil
gives both fish and oysters a bad taste.
Subsurface injection
Injection may not be the proper word since pressure is not always neces-
sary. The Plains States are in a hard water belt and have the strictest controls
on subsurface disposal. Whether or not subsurface disposal expenses should
always be considered the 'most expensive means of disposal is debatable.
Pressure maintenance by injected brine assists oil production rates. Oilfield
brines are also used for waterflooding; therefore, not injecting the brine may
decrease production and result in a more expensive method of disposal.
Shallow well disposal was one of the first subsurface disposals employed
where a shallow well is defined as one using a horizon less than 305 m in
depth. A shallow well takes considerable input brine for a time, but in-
creasing pressures become necessary as time goes by and this continually
increases disposal costs. Shallow wells are also more likely to allow the
injected brine t o reach fresh-water supplies in some areas.
When a deep-seated bed is known to be available, it is less costly in the
long run to make use of the deeper formation. Deep disposal wells accept
immense volumes of brine by gravity, and eliminating the necessity of injec-
tion pressures reduces disposal expenses. It often was discovered that clog-
ging occurred near the bore of the injection well and treatment was neces-
sary to reduce clogging. Today treatment plants compose a large fraction of
the subsurface disposal operations.
Early in the history of subsurface brine disposal, legislation simply
allowed any means of disposal. Since then legislation has for all practical
purposes forced subsurface disposal and set up tight controls and safeguards
for protection of fresh water. In the 1930's, the Railroad Commission of
Texas allowed the increase of oil allowable by 1 m3 for every 50 m3 of brine
returned to subformations. The increase was fixed at a maximum of 0.8 m3
(Laurence and Leusler, 1958) and this incentive program did a great deal to
encourage subsurface disposal of brines in the East Texas field.
PRESENT-DAY TECHNOLOGY 421
Present-day technology in subsurface disposal
Several areas of research in the handling of brines were opened because of

the subsurface disposal of oilfield brines. The most notable problems
encountered were scaling in the salt-water lines, corrosiveness of brines, in-
efficient separation of oil from the water, and poor well cements. Today,
methods to correct most of these problems are available.
The scaling problem is a culprit which causes large expenses in brine-
handling operations. Morris (1959) describes the method which the East
Texas Salt-Water Disposal Company (1953) uses for handling scale formation.
Asbestos-cement pipe is used almost exclusively in their salt-water gathering
system. This pipe does not deteriorate from salt-water use and has been
found quite practical to prevent corrosion. However, accumulation of scale
often occurs t o such an extent that the inside of the pipe must be cleaned
regularly to maintain injection capacity. Sections of 8-inch lines often are
reduced by scale t o 2-inch lines over a period of years.
The East Texas Salt-Water Disposal Company (1953) had the maintenance
responsibility for 611 km of such pipeline. To accomplish this they had four
pipeline-cleaning crews at work 5 days a week cleaning pipelines exclusively.
The most efficient method for scale removal was scrapers. The scraper is
forced through the pipeline by pump pressure applied against a rubber
cementing plug. The plug is followed by a wire brush, and a second type of
scraper is used where high temperatures are used to treat water-oil emul-
sions. The scale deposits in this case are too hard t o be removed by a wire
brush, and the scraper used to remove this scale is mechanically operated by
a flexible cable. Table 14.1 compares the costs of pipeline cleaning methods
(Morris, 1959).
Treating oilfield brines before injection is necessary to remove suspended
solids which clog the formation. When a brine is brought to the surface in
East Texas, the temperature changes from 64OC t o atmospheric. The pres-
sure reduces from an average of 74 kg/cm2 to atmospheric. Carbon dioxide
and petroleum gases, which were dissolved in the salt water, are allowed to
escape. Oxygen and other elements of the air mix with the elements of the
salt water creating many precipitates.
TABLE 14.1
Comparative costs of pipeline cleaning methods
Pipe size Costs ($/m)

(cm) cleaning machine acidizing replacement
10.2 0.249 0.958 4.04
15.2 0.174 1.919 5.77
20.3 1.050 4.259 7.64
25.4 1.234 5.899 9.61
Scale-forming precipitates are removed by settling tanks. Elements that

will precipitate in the well are chemically precipitated and removed, and oil
is skimmed from the settling tanks. Smaller oil particles are removed by
using baffles in the skimming process. Aeration is used t o oxidize ferrous
compounds and cause them to precipitate for filtration. A high aeration
efficiency tends t o reduce the chemical treatment costs.
Chlorination is used as an oxidizing agent t o control algae and bacterial
growth because algae. and bacterial growths cause plugging problems in the
treatment plant as well as in the injection wells and disposal formations.
Chemical treatment with hydrated lime and alum removes iron compounds,
calcium compounds, and small amounts of hydrocarbon products. Small floc
particles which remain in suspension after chemical treatment to cause pre-
cipitation are removed by filtration.
Where a closed system can be operated, treatment costs are decreased.
Exclusion of oxygen prevents oxidation of iron, maintains a low corrosion
rate and prevents the growth of aerobic bacteria. Laurence and Leuszler
(1958)point out that the maintenance of pressure on the system will hold
carbon dioxide in solution and reduce the precipitation of calcium and mag-
nesium carbonates. In some cases a closed system will maintain the stability
of the water for reinjection into the producing formation with little or no
treatment. Usually a closed system will require only oil skimming and filtra-
tion, but in some cases chemicals and bactericides are necessary t o control
bacterial growth in a closed system.
Economics and oilfield brine disposal
Investments and operating costs for oilfield brine disposal systems are
difficult to obtain and compare because of differences in accounting systems
used by various oil operations. Some operations have brine disposal costs
incorporated into oil production costs, others only separate costs of treating
and report it as brine disposal costs.
Elliston and Davis (1944)reported a survey taken in the early 1940’s.
They report investments for 256 systems totalling $4.2 million and operating
costs totalling $1.2 million for 86 systems. Disposal costs then range from a
few cents per m3 t o 63 cents per m3. The main variables controlling the
costs were the amount of treatment necessary, the size of a system, and the
well depth necessary t o dispose of the brine. East Texas Woodbine formation
disposal averaged approximately 12 cents per m3 injected. Brine from the El
Dorado field in Kansas averaged about 6.3 cents per m3. Operations where
one disposal well served only eight production wells yielded a cost of 44
cents per m3. A case where one disposal well served 15 production wells gave
a disposal cost of only 5 cents per m3. In general, the more brine disposed
into one well, the smaller the cost.
A small operator or even a major company is not economically justified in
installing a deep disposal well if the development limit of his lease or field is
ECON OM ICS 423
only one or two producing wells. Some deep disposal wells show a potential
capacity of 1,600-3,200 m3 of brine intake per day under actual test.
Therefore, it is evident that it is possible for several operators to use the
same disposal well.
The largest oilfield brine disposal association is the East Texas Salt-Water
Disposal Company. This company serves the East Texas oilfield located in
northeast Texas, which is located in parts of five counties, and the company
handles approximately 90%of the brine produced in this field. In 1942, with
disposal costs averaging in excess of 12.6 cents per m3 and the amount of
produced brine increasing, the disposal company was formed by 250 large
and small operators. During the second year of operation, costs of disposal
dropped t o about 10.7 cents per m3. Table 14.11 shows the history of
disposal costs for the East Texas Salt-Water Disposal Company (1953).
Table 14.11 shows that although there was a large difference between the
costs of labor, equipment, materials, etc., from 1944 through 1958, the
company’s cost per m3 of disposed brine changed very little. This probably
indicates that in this case, increasing costs have been balanced by technologi-
cal improvements. The amount of brine disposed per year roughly follows
inverse variations t o disposal cost variations. In the data gathered by Elliston
and Davis (1944) on disposal system costs, investment amortization was
roughly 63%of the total disposal costs. With greater fixed costs than variable
operating costs, the unit costs should increase as the amount of brine in-
jected decreases.
TABLE 14.11
East Texas Salt-Water Disposal Company’s costs
Year Brine injected Total cost before

(m3) taxes ($/m3
__
1942 88,600 0.450
1943 6,454,000 0.108
1944 12,606,000 0.088
1945 14,034,000 0.090
1946 17,546,000 0.089
1947 21,666,000 0.086
1948 22,932,000 0.087
1949 22,715,000 0.091
1950 21,535,000 0.091
1951 19,458,000 0.096
1952 19,310,000 0.089
1953 18,320,000 0.086
1954 18,514,000 0.092
1955 20,O27,000 0.084
1956 20,979,000 0.084
1957 21,488,000 0.086
1958 21,969,000 0.087
The East Texas Salt-Water Disposal Company is considered a public utility

under the Statutes of Texas, which simply gives the company a monopoly on
salt-water disposal in the East Texas field. The company has in excess of $6
billion invested in over 630 km of pipeline& 35 treating plants, 64 collection
centers, 212 centrifugal pumps, and 60 injection wells. All of this investment
is not in one integrated gathering system, but in several small systems for
different producing areas.
One small system which services a producing area is made up of three
collection centers, three treatment centers, and five injection wells. In the
design of a disposal system for a producing area, the location of treatment
plants, collection centers, and injection wells must be optimized. Pumping
costs and pipeline costs dictate the economic size and location of collection
centers and treatment plants are highly automated, requiring little attention.
Pipelines for gathering brines are laid t o obtain gravity flow where possible.
The unit disposal costs for a particular field are determined by several
factors. The more brine produced in an area, the lower the unit disposal
costs. An open disposal system requires more treating than a closed one,
hence higher costs. An area where gravity flow is attained for most of the
brine will yield lower costs. With the costs of brine disposal per barrel
remaining constant, oil producers still realize higher disposal costs on a total
basis. As an oilfield is produced, the produced brine per m3 of oil increases.
While the disposal costs in subsurface brine disposal are to be minimized,
the expenditure is not without benefit t o the oil producer. Some produced
brines are combined with additional brine-well production for waterflood
purposes. Here the brine is injected into the oil-producing formation in order
to displace more oil toward the oil wells. Another value of the brine injec-
tion is in pressure maintenance. The pressure in an oil reservoir decreases as
the oil and brine are produced. This increases the costs of pumping the oil to
the surface. Returning the brine to the reservoir formation helps to maintain
the pressure. In the East Texas field the pressure dropped from 114 kg/cm2
to 70 kg/cm2 before the subsurface brine disposal program was inaugurated.
The brine disposal halted the decreasing reservoir pressure. Now additional
brine from other formations is added to the oilfield brine and the reservoir
pressure has increased. As Morris (1960) points out, it is estimated that brine
disposal in the East Texas field is responsible for the availability of an
additional 95 million m3 of oil. An oilfield with 16 million m3 of oil
recovery is considered a major field in the oil industry.
Injection well versus disposal well
An injection well in an oilfield waterflooding operation is a well into

which water or brine is injected to sweep in-place oil out of the formation
and into an oil-production well. A primary production oil well producing
from a water-drive reservoir, i.e., already under a natural flood, usually is not
subjected t o a waterflood operation except possibly peripheral injection.
ACCEPTABLE GEOLOGIC AREAS 425
Water is injected into an oil reservoir that does not possess a natural water
drive; this injection water is obtained from water supply wells, and in many
cases some of it is recycled.
In oilfield waterflooding operations, an injection well is used to introduce
water (often brine) into the strata t o be flooded. In such an operation the
oil-water interface is kept as uniform as possible to clearly sweep in-place
oil out of the formation and to a production well, i.e., to obtain maximum
sweep efficiency. The injection pressure and injection rate usually are low
and slow during the beginning of a waterflood, and the input well usually is
not fractured because channeling is thus less likely.
Disposal wells are used in oil-production operations t o dispose of the
waters that are produced with petroleum from a natural water-driven reser-
voir, and some petroleum reservoirs produce large quantities of such brine
water. Disposal wells usually are fractured so that the subsurface formation
will accept large quantities of fluids at little or no injection pressure; many
of these wells initially operate on a vacuum. In a disposal operation, a
uniform flood front and absehce of channeling is not required. The primary
consideration normally is to put large amounts of fluid into a reservoir at the
least possible cost.
Acceptable geologic areas
Most of the major synclinal basins contain favorable locations for d e e p

well injection (Warner, 1967). The most acceptable areas are in porous sedi-
mentary rock strata, e.g., sandstone, limestone, or dolomite. Such strata are
found under about 50% of the land area in the United States, mainly in the
Central Plains States and southwest coastal areas.
Some areas such as the west coast are underlain by complex geologic
strata, which has not been satisfactorily studied. Areas where volcanic rocks
are present at the surface usually are not acceptable for disposal wells.
McCann et al. (1968) reported on possible disposal sites in the New York
area of the Appalachian Basin. They found some suitable and some un-
suitable areas. Hardaway (1968) reported on the possibilities of waste
disposal in a structural syncline in Pennsylvania and found that disposal into
certain horizons appeared promising but that additional seismic and well-test
data are needed. Similar studies were made by Briggs (1968) for the
Michigan Basin; Edmund and Goebel (1968) for the Salina Basin; Garbarini
and Veal (1968) for the Denver Basin; and Peterson et al. (1968) for the San
Juan Basin.
In summary, the most likely acceptable areas are found in the Puget
Willimette Valley and Great Valley of California, in the Mid-Continent and
Great Plains, on the Gulf coast, in the central and eastern Great Lakes area,
and in much of the Mississippi River drainage basin. Areas that are likely to
be unacceptable are the Western Mountain ranges; the Ozark, Wichita,
Arbuckle Llano Uplift area; the Mexia fault area; the Atlantic Coast area; the
4 26 SUBSURFACE DISPOSAL
Appalachian Mountain area; most of the New England States; and certain
areas near western Lake Superior and Lake Michigan.
Geologic maps
To conduct a feasibility study of a project in depth, it is desirable to have

a suite of maps that show the following (L.R. Reeder, written communica-
tion, 1972): (1)surface geology; (2) subcrop at various horizons; (3)struc-
ture; (4) tectonics; ( 5 ) convergence; (6)isopachs of various reservoirs; (7)
potentiometric gradient of various reservoirs; (8) depth to base of fresh
water zones; (9) fresh-water wells; (10)all wells other than fresh-water wells,
showing total depths.
Suitable disposal zones
The development of subsurface disposal operations by the oil industry

indicates that almost all types of rocks possess large enough porosities and
permeabilities to accept large amounts of fluid under favorable conditions.
The injected wastes must be confined t o the disposal formation so that fresh
water and other valuable natural resources are protected. Some of the
characteristics that are required of an acceptable disposal zone are as
follows:
(1)The rocks in the disposal strata should have large porosity, permeabili-
ty, and thickness so that a significantly large volume is available for fluid
injection at relatively high rates and at reasonably low pressures.
(2) The disposal reservoir should be of large area extent suitable for in-
jection of large quantities of fluid.
(3) The reservoir rocks should be uniform and not too heterogeneous to
allow calculations concerning the behavior of injection fluids, injection pres-
sures, and possible fluid rock reactions.
(4) The injection zone should contain brackish or salty water (a
salaquifer). Waters containing more than 1,000 mg/l of dissolved solids are
used for domestic, irrigation, and industrial water in some areas (Warner,
1968).
(5)The proposed injection zone must be separated from fresh-water zones
both laterally and vertically. Such a zone should be vertically below the level
of fresh-water circulation and confined vertically by strata that are imperme-
able. A rule of thumb is the depth at which a confined salaquifer is present;
however, this is not always applicable because in some areas salaquifers
overlie fresh-water aquifers.
( 6 ) There should be no unplugged or improperly plugged wells penetrating
the proposed zone in the vicinity of the disposal well.
(7) The fluids t o be injected should be compatible with the rocks in the
injection strata and with the fluids in the strata. If they are not, precipitates
will form and plug the well. Wastes incompatible with the native fluids can
EVALUATION OF THE DISPOSAL ZONE 427
be injected behind a buffer zone; however, it is difficult t o buffer the rock

strata to prevent incompatible fluid and rock reaction (Gabarini and Veal,
1968).
(8) The injection zone should have a low internal hydraulic pressure to
allow a sufficient margin for injection of fluids without causing hydraulic
fracturing of the surrounding strata and to assure a long operating life of the
disposal well.
(9) The potential injection zone should be surrounded above, below, and
laterally by impermeable strata or aquicludes. Many potential zones are sur-
rounded above and below by such strata, and the lateral movement can be
monitored in the injection zone. Good seals to prevent fluid movement are
provided by anhydrite, clay, gypsum, marl, salt, slate, and unfractured shale.
(10) The hydrodynamic gradient, if any, of the proposed disposal forma-
tion should be determined so that the path of movement can be calculated.
Evaluation of the disposal zone
To evaluate a disposal zone requires a detailed study of the geology of the

area and maps showing the surface geology, tectonics, surface and subsurface
structure and stratigraphy, salinity, potentiometric gradient, and presence of
all wells in the area. The information t o construct the maps can be obtained
from studies outlined in Table 14.111 (Ross, 1968).
Drill cutting samples and cores are taken during drilling. The various logs
and drill-stem tests can be run after the entire hole, or a portion of it, has
been drilled. Pumping and injectivity tests can be performed through the
drill pipe and an open-hole packer before the well is completed or through
casing or tubing after the well is completed.
TABLE 14.111
Method of obtaining data to evaluate a disposal zone
Necessary information Methods
Fresh-water zones State agencies, Federal agencies, drill

exploitation well, and log
Porosity electric log, sonic log, radioactive log,
core samples
Permeability drawdown test, pressure buildup test,
micrologs, core samples
Minerals in strata core samples
Fluid pressure in strata drill-stem tests
Subsurface formations and electric log, sonic log, drilling log, core
thickness samples, radioactive log
Temperature of strata temperature log
Injectivity profile injection test, differential temperature
log, spinner log, radiotracer log
Reservoir transmissibility
The reservoir transmissibility can be calculated using knowledge obtained

from the following first two items, but it is influenced and may change
during disposal operation because of phenomena associated with items
shown in (3)through (8).
(1)Core analysis data.
(2)Reservoir transient tests (Matthews and Russell, 1967).
(3)Behavior of fractured reservoirs;
opening of natural fractures during pumping (Snow, 1968)
(4)Artificially induced fractures (DeLaguna, 1966).
(5)Dissolution of rock by injected fluid.
(6)Accidentally induced fractures.
(7)Plugging from suspended solids;
bacteria;
corrosion products.
(8)Plugging by clay swelling.
(9)Plugging by incompatibility of fluids (Ostroff, 1964).
Compressibility of rock and water
Both the reservoir rock and interstitial fluid are compressible to a very
small degree. It is the compression factor that provides the space needed t o
inject extraneous fluids into an otherwise full reservoir. Waters and rock in
the salaquifer are compressed by the injected waste liquids in an ever-
expanding cylinder away from the wellbore. Since the rock and water com-
pressibility is of small magnitude, the salaquifer must be of large areal extent
t o distribute the pressure buildup. If the formation is confined by faulting,
sand pinch-out, or restricted permeability in the region of the disposal well, a
very limited area will be available t o compress the formation rock and water,
and pressure will build up rapidly or injection rates will decline to a point
where the operation becomes impractical.
Water compressibility
The compressibility of pure water is known to be dependent upon the

pressure, temperature, and gas in solution in the water. Note that there is a
wide range of compressibilities and that increasing pressures reduce the
value, whereas increasing temperatures enlarge it. The compressibility of
pure water at 408 atm and 93.3"C is approximately 4.2 x lo-" cm2/dyne.
Since with increasing depth higher pressures and temperatures are encoun-
tered, it is expected that compressibility will increase, but the magnitude will
depend upon the relative increases in pressure and temperqture.
A t a given pressure and temperature, the effect of gas in solution in pure
water is t o increase the compressibility over that of pure water at the same
pressure and temperature. A graphical method of correction for gas solu-

bility indicates that the effect of gas solubility on the compressibility of
water is great, and a reservoir water containing 3.56 m3 of natural gas per
m3 will have a compressibility approximately 18%greater than that of pure
water at the same temperature and pressure (Amyx et al., 1960).
Rock compressibility
The porosity of sedimentary rocks has been shown by Krumbein and Sloss
(1963) t o be a function of a degree of compaction of the rock. The com-
pacting forces are a function of the maximum depth of burial of the rock.
Sediments which have been buried deeply, even if subsequently uplifted,
exhibit lower porosity values than sediments which have not been buried a
great depth.
Apart from the effect of compaction on grain arrangement, rock minerals
are also compressible. Three kinds of compressibility must be distinguished
in rocks: (1) rock matrix compressibility; (2) rock bulk compressibility; and
(3) bore compressibility (Amyx et al., 1960). The compressibility of each
parameter above is the fractional change in volume of that parameter with a
unit change in pressure. Data correlated with “net overburden pressure’’
indicate that the pore compressibility is a function of pressure. In summary,
pore volume compressibilities of consolidated sandstones are in the order of
7 x lo-” t o 14 x lo-’’ cm2/dyne.
Critical pressures of confining beds
Impermeability of overlying and underlying beds is essential. So that the

possibility of breakthrough from pressure of injection and pressure from
evolved C 0 2 or thermal expansion does not take place, all operating pres-
sures should be below the critical pressures needed to fracture the forma-
tions. The value of the critical pressure usually ranges from 0.11 t o 0.33
atm/m of well depth. To design adequate surface and pumping equipment
and t o accurately evaluate the hydrologic properties of the disposal forma-
tion, injectivity tests should be made. If possible, the test should be made at
the critical input pressure; i.e., the point at which the formation begins to
fracture. This point will determine the maximum safe injection pressure
(McLean, 1969).
Natural and artificial escape routes
Joints, faults, fractures from excessive pumping pressures, formation oub

crops, and unplugged or poorly plugged wells all represent potential escape
routes for injected waste fluids. Many operations can be conducted within
the restrictive limits established by the factors above, and some of the
factors can be corrected. However, each deserves serious consideration .in
conducting a feasibility study for deep-well disposal.
Art i ficial fracturing t o increase permeability
Permeability may be increased by artificially fracturing the formation and

propping the resulting fractures with silica sand, glass beads, aluminum
pellets, or other such agents. Accidentally induced fractures may be pro-
duced when critical pressures are exceeded in pumping. Unless these frac-
tures are propped, they will usually close and subsequently heal when pres-
sure is relieved. In any case, the natural tendency will be for fractures to be
horizontal when they are developed at depths of less than 305 m and t o be
vertical when developed at depths greater than 305 m. However, several
sophisticated methods are being tried to direct the fractures below 305 m
into the horizontal plane (L.R. Reeder, written communication, 1972). The
generally accepted theory is that the attitude of the induced fracture is
related t o the regional stress and is oriented in a direction perpendicular t o
the least principal stress.
Pressure-dista n c e t ime relationship
An important but little understood consideration is the pressure effects at

various distances from the wellbore for given times and volumes of injected
fluids. Where legal situations may develop, or where disposal is conducted in
the vicinity of potentially valuable mineral deposits, this factor becomes very
important. This information is useful in predicting long-range reservoir per-
formance and design of injection equipment and the effect on unplugged
wells in the vicinity. Equations have been developed which describe fluid
withdrawal from water wells. These equations can be used t o describe the
converse condition; that is, fluid injection into a subsurface formation. The
rate at which the pressure increases in a formation and the distance that this
higher pressure moves radially out from the injection well can be computed
for a specific injection rate from non-equilibrium equations. If equilibrium
flow conditions are approached, however, their equations are no longer
applicable. As water is continuously pumped into a homogeneous uniform
aquifer of infinite areal extent, the pressure radius will increase but at a
decreasing rate because of the expanding storage area available (Davis and
Dewiest, 1967).
Potentiometric levels and gradients should be determined for disposal res-
ervoirs to help analyze and anticipate fluid movement and monitoring meth-
ods needed. Depleted oil or gas reservoirs often make ideal reservoirs for the
disposal of oilfield brines or other types of liquid waste, because of the
volume available as a result of the production of oil and gas.
Semipermeable beds
Shales and beds of clay at one time were considered impermeable to fluids
but it now is postulated that ground waters are transported across these
beds. This postulate assumes that the beds act as semipermeable membranes
where the membranes separate waters of different salinity. Transport of
water across a shale can result in lower pressure on one side of the shale
versus higher pressure on the other side. Assuming that the shale acts as a
membrane, the lower pressured side will be the effluent side and will contain
filtered or fresher water while the high pressured side will contain the more
salty water which will become even more salty as the filtration process
proceeds.
Because hydrodynamic conditions exist in many ground-water aquifers, it
should be a mandatory requirement that the hydrodynamic conditions of
the proposed sedimentary disposal aquifer be thoroughly determinzd. Water
flow in aquifers usually is determined by use of contour maps cf water
elevation in wells plus aquifer permeability and aquifer thickness. This
method will not give a true calculation if much water is transported through
semipermeable shale or clay beds. Pressure maps are necessary in establishing
such transport and the low-pressure aquifer should be used for a disposal site
rather than the high-pressure aquifer.
t Injection
Fig.14.1. Cross section of disposal well.

Considerations during drilling and well completion
Fig. 14.1 is a cross section of a disposal well. Note that it is cased through
the entire fresh-water zone and cemented to a competent horizon below the
fresh-water zone. The annulus of the well is filled with an inhibitor fluid
under pressure, and the well is equipped with a continuous monitor to detect
casing or tubing failure. The well shown in Fig. 14.1 is completely lined with
cement to a bottom competent zone. The materials used for tubing, casing,
or valves can be carbon steel, plastic, fiberglass, stainless steel, etc., depen-
ding upon requirements.
Core samples taken during drilling operations should be reacted with the
proposed liquid waste to determine what reaction might occur and how to
prevent or inhibit the reactions if they are likely to damage the well; for
example, to determine what precipitates form and what gases evolve to give
pressure increases.
Treatment facilities, such as filtration, pH adjustment, and additives,
probably will be necessary. For example, the quantity of solids in the fluid
and their plugging characteristics with the disposal zone must be determined.
If the quantity of suspended solids is excessive, their concentration must be
reduced by filtration, settling, decantation, or gas flotation (Amstutz and
Reynolds, 1968). A reduction in permeability of the injection horizon and
resulting increase in injection pressure or decrease in injection rate can occur
as a result of plugging of the pores. Plugging can be caused by suspended
solids or entrained gas in the injected fluid, reactions between injected and
interstitial fluids, autoreactivity of the waste at aquifer temperature, and
pressure and reactions between injected fluids and aquifer minerals. Plugging
at or near the wellbore can also be caused by bacteria, mold, and fungi.
Selm and Hulse (1959) state that the chemical reactions between injected
waste and interstitial water which can cause plugging precipitates to form are
as follows:
(1)Precipitation of alkaline earth metals, such as calcium, barium, stron-
tium, and magnesium, as relatively insoluble carbonates, sulfates, ortho-
phosphates, fluorides, and hydroxides.
(2) Precipitation of heavy metals, such as iron, aluminum, cadmium, zinc,
manganese, and chromium, as insoluble carbonates, bicarbonates, hy-
droxides, orthophosphates, and sulfides.
(3) Precipitation of oxidation-reduction reaction products.
Additional causes of formation plugging in disposal wells are as follows:
(a) partial decomposition or dispersion of salaquifer minerals yielding solid
matter in suspension; (b) viscosity increases with an increase of pH; (c)
coalescence of gel films at constrictions in pores; and (d) complete gelation
of the entire advancing front. These reactions impair the flow the greatest
when they occur near the wellbore and the least when they are far removed
from the wellbore.
Clay minerals occur in sedimentary rocks and are known t o reduce the
FLUID TRAVEL 433
permeability of sandstones to water as compared to their permeability to air.

The water permeability of a clay-bearing sandstone decreases with decreasing
water salinity, decreasing valence of the cations in solution, and increasing
pH of the water.
Quartz, feldspars, carbonates, micas and clays, and iron oxides generally
constitute the main components in sandstone aquifers. Limestone and
dolomite are primarily carbonates but, if impure, may contain as much as 50
percent noncarbonate minerals such as quartz and clay minerals. Quartz,
feldspars, and micas are nonreactive except in highly alkaline or acid solu-
tions. Carbonate minerals are soluble in acids. The reaction of carbonate
minerals with acid wastes can be beneficial, if no undesirable precipitates
form and if the generation of CO, gas does not cause excessive pressure
buildup or plugging of the injection zone.
Surface treatment or injection of a compatible liquid (buffer liquid) t o
move the indigenous brine away from the wellbore before the injection of
incompatible waste liquid are two methods of preventing precipitation. If
the incompatible fluids do mix later and precipitate solids, they form only at
the mixing front and are at such a distance from the wellbore that pore
volume is more than ample to contain solids without detectable restriction
of injection.
Waste waters that are stable on the surface can become unstable at aquifer
temperature and pressure. This instability can lead t o polymerization of
resin-like materials as suggested by Selm and Hulse (1959).Other reactions,
such as the precipitation of calcium carbonate, can occur because of the
decreased solubility of dissolved gas at high temperature (Ross, 1968).Case
(1970)presents several methods that are useful in handling waters and ana-
lyzing problem scales.
Organic growths causing operational problems are rare in deep-well
disposal projects. Most injected liquids contain sufficient heavy metals and
toxic organic or inorganic solutions to provide unfavorable environments for
any type of bacterial life. Amstutz and Reynolds (1968)note that fungi can
exist and proliferate under a wider range of environmental conditions than
can slimes, algae, or bacteria and are more likely to be encountered than the
other three types of organisms. He cites an operational problem where fungi
growth was encountered in a system disposing of spent sulfuric acid with a
pH of 2.8. Problems with organic growths are most likely to appear where
liquids are stored or passed through open ponds.
Fluid travel
The pressure radius increases but at a decreasing rate when a liquid is

pumped into a homogeneous uniform aquifer of infinite areal extent because
of the available expanding storage area. Nonequilibrium equations can be
used t o compute the distance that the higher pressures move out radially
from the injection wellbore and the rate at which the pressures increase
(Wright, 1969; Davis and Dewiest, 1967; Ferris e t al., 1962). Hydraulic
gradients and potentiometric levels should be determined for disposal zones
to calculate and monitor the injected fluid movement.
Hazards of underground waste disposal
Contamination of shallow aquifers
Unplugged or poorly plugged wells that penetrate zones into which waste
is being injected provide escape routes by which the waste liquid can reach
and contaminate shal!ow fresh-water aquifers. This has been common in
oilfield experience and is a factor to consider whenever an operation is
conducted in the vicinity of old wells.
Vertical fracturing caused accidentally by excessive injection pressures or
during the process of hydraulic fracturing acts in a manner similar to unplug-
ged wells if the fractures breach the impermeable horizons isolating the
injection zone. Surface contamination may occur if there is some malfunc-
tion or material failure of surface or well equipment.
Earthquakes
The most notable example of a hazard attributed t o deep-well disposal

thus far is the Rocky Mountain arsenal well located about 16 km northeast
of Denver 1,581 m above sea level, completed September 11, 1961.The well
was drilled t o a total depth of 3,671 m, and injection was made into a zone
of fractured gneiss from 3,650t o 3,671 m. From March 1962 until February
1966, a volume of 0.625 million m3 was injected at a maximum rate of 1.95.
m3/minute and 75 atm (average rate 0.76 m3/minute and 34 atm). The
seventh week after injection began, an earthquake of magnitude 1.5 was
recorded (April 24, 1962). From April 24, 1962, through August 1967,
1,514 earthquakes were recorded with magnitudes ranging from 0.5 t o 5.3;
all were relatively shallow in origin and from an area about midway between
central Denver and the arsenal well (Hollister and Weimer, 1968).
Present consensus is that the earthquakes are products of a regional stress
field of tectonic origin, triggered by the local incremental strain from injec-
tion into the arsenal well. In several aspects, however, the stress-strain rela-
tionship in the vicinity of the arsenal well seems not t o have been resolved
fully. Injection-triggered earthquakes were tentatively identified.
State regulations and tax incentives
Regulations
Regulations concerning construction and operation of disposal wells are

not standard and vary from State to State (L.R. Reeder, written communica-
STATE REGULATIONS AND TAX INCENTIVES 435
tion, 1972). Missouri, Ohio, Texas, and West Virginia have regulations
dealing specifically with disposal wells. Apparently no States have regula-
tions which specifically prohibit disposal wells; however, Ohio permits dis-
posal only into the Mountain Simon sand. Texas probably has the most
specific and perhaps the most equitable regulations, which are obtained from
the Texas Water Development Board. Some of their regulations shown on
their Form GW-14 are as follows:
“A preliminary report is required before application can be processed.
This report should include but not necessarily be limited to the following
information:
(a) An accurate plat showing location of proposed injection well.
(b) A map indicating location of water wells and all artificial penetrations
(oil and gas wells, exploratory tests, etc.) of the proposed injection inter-
val(s) in the general area of the proposed injection well. Reasonable diligence
shall be used to locate such penetrations. Well and abandonment records for
all exploratory oil and gas tests located within the area owned and operated
by application should accompany map. (Details within 5 km radius generally
acceptable.)
(c) Description of local topography and geology pertinent t o injection
program. Depth of deepest strata containing fresh water or water of suitable
quality for potential beneficial development as determined by well develop-
ment and/or electrical logs. (Generally required minimum of 91 m of shale
between injection zone and base of fresh water.)
(d) A detailed description of the chemical, physical, and biological charac-
teristics of the waste to be injected. Complete chemical analyses of all inor-
ganic constituents should be reported in ppm or mg/l. If organic fractions are
present, all such constituents should be reported in ppm, mg/l, as individual
percentages by weight, or in other appropriate terms.
(e) The anticipated average and maximum rate of injection in gallons per
minute or barrels per day. Estimated yearly volume of injected waste and
anticipated life of project. (Semiannual reports of monthly volumes, injec-
tion rates, pressures, cumulative volume, workovers.)
(f) Data on completion and operation of proposed injection well:
(1)Total depth of well.
(2) Casing size, grade, type, weight, and setting depth of all strings; size
and type of tubing; name, model, and depth of tubing packer setting.
(3) Cement-type and volume of cement t o be used on each casing string
and calculated top of cement behind each string. Describe and give percent
of all cement additives. Run a cement bond log.
(4) Proposed injection interval(s) and perforations. This should include
the interval(s) t o be utilized initially and the entire zone requested for future
development.
(5) Diagramatic sketch of proposed well.
(6) Anticipated maximum and average wellhead injection pressures.
(7) Description of possible hydraulic fracturing and/or acidizing pro-
grams, if anticipated.
(8) Description of proposed injectivity tests. (Logs must be run and

submitted. Cement Bond Logs should be specifically required but are not
(L.R. Reeder, written communication, 1972).
(g) Characteristics of injection interval(s).
(1) Lithology, porosity, permeability, temperature.
(2) Natural reservoir fluid pressure and equivalent hydrostatic head;
fluid saturation and chemical characteristics; and fracture gradient or critical
injection pressure.
(h) Compatibility of injected waste and formation fluids.
(i) Calculated rate of fluid displacement by injected waste and directions
of dispersion.
(j) Description of program to monitor water quality in fresh-water
aquifers.
(k) Surface installations.
(1) Detailed description of pretreatment process and facilities to be
used (include flow diagram if available).
(2) Description of type of materials to be used in pretreatment facilities
and transmission lines.
(3) Description and location of all waste retention ponds, if such are to
be used in conjunction with the injection well.
In the event an existing well is to be converted to an injection well,
applicant should submit a complete electric log, all other logs or surveys
performed on the well, and complete casing and cementing data.”
Tax incentives
Some states give preferential tax advantages to enterprises utilizing subsur-

face disposal for water pollution abatement. The following examples were
taken from Wright (1969).
(1) Connecticut. Exempts pollution control facilities from property tax.
(Conn. General Stats. Anno., Sec. 12(81)(51), (Supp. 1966).)
(2) Florida. Very limited incentive. Provides lower valuation on control
facility for ad valorem tax purposes; no tax on sale of facility, etc. (14A Fla.
Stat. Anno., Sec. 403.241.)
(3) Georgia. Exempts control facilities from ad valorem tax. (Georgia
Code Anno., Sec. 2-5405.)
(4) Idaho. Exempts facilities from ad valorem tax. (Idaho Code Anno.,
Sec. 63-105T.)
( 5 ) Illinois. Provides limited incentive, much like Florida, exempts portion
of value of facility from tax on sale of property. (Ill. Anno., Stats., Ch. 120,
Sec. 502.)
(6) Indiana. Exempts personal property from ad valorem tax when the
same is used for pollution abatement. (Ind. Stats., Anno., Sec. 64-241.)
(7) Massachusetts. Exempts control facilities from ad valorem tax. Also
gives credit against corporate or income tax for portion of cost of facility.
(Mass. General Laws Anno., Ch. 59, Sec. 5 (Supp. 1968).)
COSTS OF DISPOSAL SYSTEMS 437
( 8 ) Michigan. Exempts data certified facilities from personal property

taxes and taxes on sales of fixtures. (MCLA, Sec. 323.356.)
(9) Minnesota. Exempts facilities from ad valorem tax. (Minn. Stats.
Anno., Sec. 272.02.)
(10) New Hampshire. Provides limited exemption by reduced assessment
for property taxes purposes. (Rev. Stats. Anno., Sec. 149: 5-A.)
(11) New Jersey. Exempts water or air pollution abatement equipment of
devices from ad valorem taxes. (NJSA 54:4-3.56.)
(12) New Yorh. Exempts facilities from ad valorem tax. (N.Y. Real
Property Tax Law, Sec. 477.) Also provides deduction for expenses asso-
ciated with pollution control. (N.Y. Tax Law, Sec. 208( 9)(g), 612(h),
706( 9).)
(13) North Carolina. Exempts facilities from ad valorem tax. Also gives
credit against corporate or income tax. (N.C. General Stats. Sec. 105-296)
(Supp. 1967).)
(14) Ohio. Exempts facilities from personal property taxes, franchise
taxes and sales and use taxes. (Ohio Genl. Code, Sec. 6111.31, Ohio Water
Pollution Control Act 1967.)
(15) Oklahoma. Provides credit against income tax liability for cost of
facility. (82 Okla. Stats., 923.)
(16) Oregon. Provides credit against corporate or income tax. (Ore.
Revised Stats., Sec. 314.250.)
(17) South Carolina. Exempts facilities and equipment from ad valorem
tax. (S.C. Stats., Sec. 65-1522(50).)
(18) Washington. Provides that owner of facilities may have exemption
from ad valorem tax or a credit in a like amount against use of business
occupation tax. (RWS 82.04.20.)
(19) Wisconsin. Provides deduction for expenses associated with pollution
control (Wis. Stats. Anno., Sec. 71.04(26), 71.05(1)(b)5, and 71.05(2B));
exempts facilities from ad valorem tax (Wis. Stats. Anno., Sec. 70.11(21);
also provides accelerated depreciation of control works. (Wis. Stats. Anno.,
Sec. 71.04( ZB).)
Costs of disposal systems
Rice (1968) gave some investment costs of disposal systems for disposal of
oil-associated brines, and at that time he estimated that where large volumes
of water from several oil or gas wells are to be disposed of the average cost
per well amounted t o $7,900. Pretreatment o f t h e brines such as oil removal
by gravity separation, flotation, or filtration adds to the cost of disposal and
will vary with the type of operation (Wright and Davies, 1966). Treatment to
insure compatibility of the injected waters t o prevent deposition in the
injection well and plugging adds t o the cost (Ostroff, 1963). Bleakley (1970)
described Shell Oil Company’s salt water disposal operation in its Southern
Region Onshore Division. The cost of the total initial installation was $6
million, and they disposed 58,800 m3 of brine per day from 28 fields. Brine
disposal costs can represent 2% of net oil sales and up to 25%of total lifting
costs (Smith, 1970).
Conclusions
The storage capacity of potential salaquifers, although considerable in

certain areas, is nevertheless limited if long-term continuous disposal is con-
cerned, so that space should be treated as a natural resource. Deep-well
disposal has been successful in many cases, and the evidence for questionable
operations is inconclusive. However, a high success ratio does not preclude
caution in the use of the method. Neither does it relieve the operators of the
responsibility of close monitoring of the injected fluids and the reservoir,
and the meticulous maintenance of the well facilities.
Additional research is needed to develop better methods of evaluating
potential disposal reservoirs, of confining and monitoring the lateral move-
ment of fluids in disposal zones, and of determining the rate of mixing of
injected fluids with native fluids. Regulations and laws concerning disposal
operation should be standardized.
References
Amstutz, R.W. and Reynolds, L.C., 1968. Is the earth’s crust going to waste, 11. Types of
fluids injected and treating procedures Presented at Natl. Pet. Refiners Assoc., Mid-
Continent Regional Meet., Wichita, Kansas, June 12-1 3, 1968.
Amyx, J.W., Bass, Jr., D.M. and Whiting, R.L., 1960. Petroleum Reservoir Engineering.
McGraw-Hill, New York, N.Y., 610 pp.
Bleakley, W.B., 1970. Shell’s SWD meets pollution standards. Oil Gas J., 68:144-146.
Briggs, Jr., L.I., 1968. Geology of subsurface waste disposal in Michigan Basin. In: J.E.
Galley (Editor), Subsurface Disposal in Geologic Basins: A Study o f Reservoir Strata -
A m . Assoc. Pet. Geol., Mem. 10, pp.128-153.
Case, L.C., 1970. Water Problems in Oil Production. The Petroleum Publishing Company,
Tulsa, Okla., 133 pp.
Davis, S.N. and Dewiest, R.J.M., 1967. Hydrogeology. John Wiley and Sons, New York,
N.Y., 463 pp.
DeLaguna, W., 1966. Disposal of radioactive wastes by hydraulic fracturing. NucL Eng.
Design, 3:338-352, 432-438.
Donaldson, E.C., 1964. Subsurface disposal of industrial wastes in the United States. U.S.
Bur. Min. Inform. Circ., No.8212, 34 pp.
East Texas Salt-Water Disposal Company, 1953. Salt-Water Disposal East Texas Field.
Petroleum Extension Service, Austin, Texas, 116 pp.
Edmund, R.W. and Goebel, E.D., 1968. Subsurface wastedisposal potential in Salina
Basin in Kansas. In: J.E. Galley (Editor), Subsurface Disposal in Geologic Basins: A
Study o f Reservoir Strata - A m . Assoc. Pet. GeoL, Mem. 10, pp.154-164.
Elliston, H.H. and Davis, W.D., 1944. A method of handling salt-water disposal including
treatment of water. Presented at API Meet., Tulsa, Okla.,May, 1944, API Paper,
N0.851- 18F.
Ferris, J.G., Knowles, D.B., Brown, R.H. and Stallman, R.W., 1962. Theory of aquifer
tests, ground-water hydraulics. US. Geol. Sum. Water Supply Paper, No.l536-E, 174
PP.
REFERENCES 439
Garbarini, G.S. and Veal, H.K., 1968. Potential of Denver Basin for disposal of liquid
wastes. In: J.E. Galley (Editor), Subsurface Disposal in Geologic Basins: A Study o f
Reservoir Stmta -Am. Assoc. Pet. G e o l , Mem. 10, pp.165-185.
Hardaway, J.E., 1968. Possibilities for subsurface waste disposal in a structural syncline in
Pennsylvania. In: J.E. Galley (Editor), Subsurface Disposal in Geologic Basins: A
S t u d y of Reservoir Stmta - A m . Assoc. Pet. GeoL, Mem. 10, pp.93-125.
Hollister, J.C. and Weimer, J.C., 1968. Geophysical and geological studies of the relation-
ships between the Denver earthquakes and the Rocky Mountain arsenal well. Colo.
School Min. Q., 63(1):1-251.
Jones, O.S.,1945. Disposition o f oilfield brines. University of Kansas Press, Lawrence,
Kansas, 45 pp.
Krumbein, W.C. and Sloss, L.L., 1963. Stratigraphy and Sedimentation. W.H. Freeman,
San Francisco, Calif., 2nd ed., 660 pp.
Laurence, L.L. and Leuszler, W.E., 1958. ABC's of treating and handling injection water.
Pet. Eng., 30:B52-B54, B59.
Matthews, C.S. and Russell, D.G., 1967. Pressure Build-up and Flow Tests in Wells.
Society of Petroleum Engineers, AIME, 167 pp.
McCann, T.P., Privasky, N.C., Stead, F.L. and Wilson, J.E., 1968. Possibilities for disposal
of industrial wastes in subsurface rocks on north flank of Appalachian Basin in New
York. In: J.E. Galley (Editor), Subsurface Disposal in Geologic Basins: A Study o f
Reservoir Strata - Am. Assoc. Pet. G e o l , Mem. 1 0 , pp.43-92.
McLean, D.D., 1969. Subsurface disposal - precautionary measures. Ind. Waste Eng.,
August 1969: 20-22.
Morris, W.S., 1959. Cleaning asbestos-cement pipelines in salt-water disposal service. Pet.
Eng., 31:B46-B49.
Morris, W.S., 1960. Subsurface disposal of salt Water from oil wells. Water Pollut. Control
Fed. J., 32:l-20.
Ostroff, A.G., 1963. Compatibility of waters for secondary recovery. Prod. Monthly,
27 :2-4-9.
Ostroff, A.G., 1964. Introduction to Oilfield Water Technology. PrenticeHall, Engle-
wood Cliffs, N.J., 412 pp.
Peterson, J.A., Loleit, A.J., Spencer, C.W. and Ullrich, R.A., 1968. Sedimentary history
and economic geology of San Juan Basin, New Mexico and Colorado. In: J.E. Galley
(Editor), Subsurface Disposal in Geologic Basins: A S t u d y of Reservoir Strata - A m .
Assoc. Pet. G e o l , Mem. 1 0 , pp.186-231.
Rice, I.M., 1968. Salt-water disposal in the Permian Basin. Prod. Monthly, 32(3):28-30.
Ross, R.D., 1968. Industrial Waste Disposal (Reinhold Environmental Engineering Series).
Reinhold, New York, N.Y., 340 pp.
Selm, R.P. and Hulse, B.T., 1959. Deep-well disposal of industrial wastes. Proc. 14th Ind.
Waste C o n f , Purdue Univ. Eng., Ext. Ser., No.104, pp.566-586.
Smith, W.W., 1970. Salt-water disposal: sense and dollars. Pet. Eng., 42:64-65.
Snow, D.T., 1968. Fracture deformation and change of permeability and storage upon
change of fluid pressure. Colo. School Min. Q., 63:201-244.
Warner, D.L., 1967. Deep wells for industrial waste injection in the United States -
summary of data. US. Dep. Inter., Fed. Water Pollut. Control Adm., Water Pollut.
Control Res. Ser. PubL, No.WP 20-10, 45 pp.
Warner, D.L., 1968. Subsurface disposal of liquid industrial wastes. In: J.E. Galley
(Editor), Subsurface Disposal in Geologic Basins: A Study o f Resenroir Stmta -Am.
Assoc. Pet. GeoL, Mem. 1 0 , pp.11-20.
Wright, C.C. and Davies, D.W., 1966. The disposal of oilfield waste water. Prod. Monthly,
30:14-17; 22-24.
Wright, J.L., 1969. Disposal wells - a worthwhile risk. Presented at 98th Annual Meet.,
AIME, Washington, D.C., February 16-20 1969, Reprint, 1 5 pp.
Chapter 15. SOLUBILITIES OF SOME SILICATE MINERALS IN SALINE
WATERS
In a petroleum reservoir the subsurface saline waters are continually in

contact with minerals that contain silica. Knowledge of the solubilities of
some of these silicate minerals is needed in geochemical studies of sedimen-
tary rocks. To increase this knowledge, a study was made of the solubility of
five clay minerals in saline solutions. The solubilities of kaolinite, montmoril-
lonite, nontronite, illite, and serpentine were determined as a function of
time in saline solutions at 25OC and 1atm.
Hydrothermal solution equipment capable of operating at 25"-600°C and
at 1-2,100 kg/cm2 was designed and constructed to study the solubility of a
serpentine (verde antique) in saline solutions at elevated temperatures and
pressures. Most of the expei-iments were done between atmospheric pressure
and the vapor pressure of water at 15OoC(Collins, 1969).
Composition and structure of minerals studied
Table 15.1 summarizes the approximate chemical compositions of the

illite, kaolinite, montmorillonite, nontronite, and serpentine used in the ex-
periments. All samples were natural ones and were obtained from commer-
cial suppliers.
TABLE 15.1
Chemical composition of silicate minerals
Constituent Composition (wt.%)
illite kaolinite montmorillonite nontronite serpentine

-
SiOz 56.91 44.82 49.91 39.92 44.16

A12 0 3 18.50 37.20 17.20 5.37 0.90
Fez03 4.99 0.41 2.17 29.46 0.27
FeO 0.26 0.07 0.26 0.28 2.10
MgO 2.07 0.25 3.45 0.93 40.07
CaO 1.59 0.58 2.31 2.46 0.02
Naz 0 0.43 0.40 0.14 - -
K2 0 5.10 0.43 0.28 - -
Hz O+ 5.98 12.92 7.70 7.00 7.15
Hz 0- 2.86 1.76 15.77 14.38 5.01
Ti02 0.81 1.26 0.24 0.08 0.01
MnO - - 0.04 - 0.02
442 SOLUBILITIES O F SILICATE MINERALS
Illite consists of three-layer sheets made up of two layers of silica in

tetrahedral coordination, one layer of aluminum in octahedral coordination,
and an intersheet of adsorbed potassium. Kaolinite is a double sheet with
one layer of aluminum and silica in octahedral coordination and one layer of
silica with the silicon in tetrahedral coordination. The chemical structure of
montmorillonite consists of two layers of silica in tetrahedral coordination
and one layer of magnesium in octahedral coordination with oxygen and
hydroxyl in the anion positions. Nontronite belongs to the montmorillonite
group and is characterized by its abundance of iron in tetrahedral and
octahedral positions; all members of this group swell in water because of
introduction of interlayer water in the direction of the C-axis. Serpentine is
monoclinic and composed of four molecules of the formula
Mg, Si2O5(OH), ;the structure can be chain- or sheetlike.
In all silicates, the silicon-oxygen relation is the same; a silicon atom
always occurs in the center of four oxygen atoms. This tetrahedron appar-
ently is the fundamental invariable unit of silicate structure. Silicate types
differ by the relationship of the tetrahedra in a structure to each other.
A serpentine includes at least two distinct minerals, antigorite and
chrysotile. Most asbestos are chrysotile. Verde antique is antigorite.
The sheet structure of the serpentine is the disilicate type. Such a struc-
ture occurs with tetrahedra all in one plane, with each tetrahedron joined to
other tetrahedra by three atoms lying in the common plane. Extension of
this linkage gives a hexagonal network in the plane.
The serpentine minerals are composed of hydrated magnesium silicate
layers. These layers may have ordered or disordered stacking arrangements.
Oxygen atoms usually are the largest atoms in the structure. They are chiefly
responsible, therefore, for the unit cell size.
Serpentine minerals can be transformed thermally. Transformation occurs
at about 6OO0C and probably proceeds as:
2Mg3Si05(OH), + 3MgzSi04 + SiOz + Hz0
Serpentines are the magnesium analogs of kaolin. Their basal spacing is

7.2-7.3 8, compared with kaolinite at 7.15 A.
Analytical procedure for dissolved silica
1ml of a 4% ammonium molybdate solution in 0 . N sulfuric acid solution

was added to a portion of the aqueous phase; 15 ml of 4.5N sulfuric acid was
added, and the mixture was transferred to a separatory funnel. The mixture
was agitated for 1minute in the funnel with ethyl acetate. Then the ester was
extracted and transferred to a spectrophotometer cell. The absorbance of the
ester was determined at 3350 A.
SILICATE SOLUBILITIES AT 25OC AND 1ATM 443
TABLE 15.11
Composition of aqueous solutions used in ambient conditions study
Solut io n Molality of added salt
Hz 0 -
HzO + CaClz 0.23
Hz 0 + CaClz 0.46
Hz 0 + MgClz 0.13
Hz 0 + MgClz 0.25
HzO + NaCl 0.42
Hz 0 + NaCl 0.88
Hz 0 + NaHC03 0.30
H 2 0 + NaHC03 0.61
Silicate solubilities at 25OC and 1 atm
Samples of the minerals'(Tab1e 15.1) were ground to 200-mesh size, por-

tions were placed in polyethylene bottles, and each mineral was allowed to
react 6 months with the aqueous phases shown in Table 15.11. N o precau-
tions were taken t o exclude CO,; therefore, each phase was presumably
saturated with CO, from the atmosphere. Portions of the aqueous phases
were removed periodically from the bottles and analyzed for dissolved silica
and pH. Fig. 15.1-10 illustrate smoothed solubility curves for dissolved sili-
con versus time. In general, it appears that as the concentration of dissolved
salts in solution is increased, the solubility of the silicate minerals decreases
at ambient temperature and pressure.
. 0 1 2 1
-
-
.03I
,040
-0 ,046
t- ,044
0
0
-I
Z .042
.04 I
loo 1,000 10 xx)
HOURS
Fig. 15.1. Silicon concentration as a function of tim: for illite-Hz O,-H2 O-CaClz,
-Hz 0-NaHC03 ,-Hz 0-NaCI, and -H2O-MgCl2 at 25 C. a = Hz 0; b = CaC12,0.23M;
c = NaHC03, 0.30M; d = NaCl, 0.42M; e = MgClz, 0.13M.
444 SOLUBILITIES O F SILICATE MINERALS
.03I I- "
--
a ,048
0 .046
E"
- ,044
0
0
2
.z ,042
.04 I
I00 1,000 I o.oO0
HOURS
Fig, 15.2. Silicon concentration as a function of tim% for illite-Hz O,-HzO-CaCI2,

-Hz 0-NaHC03 ,-Hz 0-NaCI, and -I& 0-MgC12 a t 25 C. a = Hz 0; b = CaClz, 0.46M;
c = NaHC03, 0.61M; d = NaCl, 0.88M;e = MgClZ, 0.25M.
HOURS
Fig. , 15.3. Silicon concentration as a function of time foro kaolinite-H2 O,-Hz 0-

NaHC03 ,-HzO-NaCI,-Hz 0-CaC12, and -Hz O-MgCl2 a t 25 C. a = Hz 0; b = CaClz,
0.23M; c = NaHC03, 0.30M; d = NaCl, 0.42M; e = MgClz, 0.13M.
.04 I I I I I I Ill
100 1.000 IC 00
HOURS
Fig. 15.4. Silicon concentration as a function of time forokaolinite-H2 0,-Hz 0-

NaHC03 ,-HZO-NaCl,-Hz 0-CaC12, and -Hz 0-MgClz at 25 C. a = Hz0; b = CaClz,
0.46M; c = NaHC03, 0.61M; d = NaCl, 0.88M; e = MgClz, 0.25M.
SILICATE SOLUBILITIES AT 25'C AND 1 ATM 445
5 . 0 3 4 1 i
-
!
In
.03I
100
. o 3
I.000
HOURS
2 2
I ,, I(
10.000
Fig. 15.5. Silicon concentration as a function of time for montmorillonite-H2 O,-H2O-

CaC12 ,-H2 0-NaHC03 ,-Hz 0-NaCl, and -Hz 0-MgC12 a t 25'C. a = Hz 0; b = CaClZ,
0.23M; c = NaHC03, 0.30M; d = NaCl, 0.42M; e = MgC12, 0.13M.
.
.04 I 0 4 2 ~
I00 1,000 10,000
HOURS
Fig. 15.6. Silicon concentration as a function of time for montmorillonite-H2 O,-Hz 0-

CaClz ,-Hz 0-NaHC03 ,-HZ 0-NaCl, and -Hz 0-MgC12 at 25'C. a = H2 0; b = CaCl2,
0.46M;c = NaHC03, 0.61M; d = NaCl, 0.88M;e = MgCl2, 0.25M.
100 1,000 10 100

HOURS
Fig. 15.7. Silicon concentration a s a function of time for no~tronite-H2O,-H2

O-CaCl~,-H~0-NaHC03,-HzO-NaCl, and -H20-MgCl2 a t 25 C. a = HzO; b =
CaCl2, 0.23M; c = NaHC03, 0.30M; d = NaCl, 0.42M;e = MgClz, 0.13M.
446 SOLUBILITIES OF SILICATE MINERALS
1,000 i00
HOURS
Fig. 15.8. Silicon concentration a s . a function of time for nontronite-HzO,-Hz

O-CaC1~,-H~O-NaHCO~,-H~O-NaCI,and -HzO-MgC12 a t 25'C. a = HzO; b =
CaClz , 0.46M; c = NaHC03, 0.61M; d = NaCI, 0.88M; e = MgClz, 0.25M.
.032
.031 -
f ,048-
l.000 IC
HOURS
Fig.' 15.9. Silicon concentration as a function of time for serpentine-Hz 0,-HzO-

CaClz ,-Hz 0-NaHC03 ,-HZ 0-NaCI, and -Hz 0-MgC12 a t 25'C. a = Hz 0; b = CaClz,
0.23M; c = NaHC03, 0.30M; d = NaCl, 0.42M; e = MgC12, 0.13M.
,032
-
- ,031
.048 -
I I I I I Ill I 1 1 1 Ill1
100 1.000 10
noum
Fig. 15.10. Silicon concentration as a function of time for serpentine-Hz 0,-Hz 0-
CaClz,-H? 0-NaHC03,-HzO-NaCI, and -HzO-MgCIz a t 25'C. a = H z O ; b = CaC12,
0.46M; c = NaHC03, 0.61M; d = NaCI, 0.88M;e = MgClz, 0.25M.
EXPERIMENTAL EQUIPMENT 447
Experimental equipment
Because the silicates, which cause scale deposits in desalination equip-

ment, react and equilibrate slowly with aqueous solutions, pressurized ther-
mal equipment was used t o approach equilibrium more rapidly. No appropri-
ate equipment was commercially available; therefore, the U.S. Bureau of
Mines designed and built the equipment shown in Fig. 15.11-14, based on
the type used by Dickson et al. (1963).
The pressure vessel was machined from sonic-tested stainless steel, and the
sample container was machined from pressure-formed Teflon 7. A Teflon
filter disk 1.429 cm in 0.d. by 0.318 cm thick (item 7 in Fig. 15.14) was
used t o filter the aqueous phase when a sample was collected for analysis.
Fig. 15.11. Hydrothermal equipment; A P pressure gage, 30,000 psi; B = temperature

recorder; C = pressure recorder; D = temperature controller; E = rheostat; F = high-
pressure valves; G = reactor oven; H = reactor rocking mechanism; Z = hydraulic pump; J =
hoist; K = Teflon sample container; L = closure piece; and M = sampling valve.
44 8 SOLUBILITIES OF SILICATE MINERALS
EXPERIMENTAL METHOD 44 9
The equipment was capable of operating at 30"-600°C and at ambient to

2,100 kg/cm2. It was designed to agitate the sample continuously by rocking
the muffle reactor forward and backward t o 70" from the horizontal. To
prolong a run, aqueous phase was added t o the sample container with the
recharging assembly shown in Fig. 15.12.
Experimental method
The solubility of a serpentine (verde antique) was determined at temper-

atures up to 200°C and at pressures up to 1,055 kg/cm2 in H20-NaC1 and
H20-CaC12 phases. Serpentine ground t o 200 mesh was placed in the
Teflon container along with several small chunks of the same material. Then
the aqueous phase was added, and the container was placed in the reactor.
Aqueous phases were prepared with freshly distilled water and 99.99%
purity salts. Equilibrium was approached by heating the sample t o 200°C and
adjusting the pressure to 1,055 kg/cm2. The sample was equilibrated at these
conditions for 48 hours before a portion of the aqueous phase was with-
drawn for analysis. The temperature then was dropped 25" or 50°C sequen-
tially while the pressure was held constant, and portions of the aqueous
phase were withdrawn every 24 hours and analyzed for pH and dissolved
silica.
Fundamental equations
Fundamental equations that denote thermodynamic relations for the

effect of variation of pressure, temperature, and composition of solution
upon the solubility relations of chemical compounds emerge from the con-
dition for equilibrium in multicomponent systems derived by Gibbs (1928).
At saturation (a condition of equilibrium between solid serpentine and solu-
tion), the molar free energy, e, of the Mg3Si205(OH), in the solid must
equal the partial mold free energy, G, of the Mg3Si2O5(OH), in solution.
A J ~infinitestimal change maintaining equilibrium dG, therefore, must equal
dG as shown by:
where u = the molar volume of serpentine; s = the molar entropy of serpen-

tine;v, = the partial modal volume of serpentine in solution at saturation; S,
= the partial molal entropy of serpentine in solution at saturation; m = the
total molal concentration of H4Si04 in solution; T = the absolute temper-
ature, and P = the pressure, n
i- bars.
The molar free energy, G, is a function of temperature, pressure, and
concentration. At saturation, the concentration, m, is also a function of
temperature and pressure.
As related to the quantities v,v!, and (aG/am)p,T, the effect of pressure
! d 4'
Fig. 15.13. Muffle furnace with pressure vessel, tubes, and thermocouples in place; 1 = thermocouple; 2 = sampling tube,lined with
Teflon spaghetti tubing; 3 = pressure control tube; 4 = insulation; 5 = heating elements; 6 =closure piece; 7 = steel piece for hoist lift;
and 8 = front view.
P ? ? B
I'
Fig. 15.14. Pressure vessel with Teflon sample container in place; 1 = pressure vessel; 2 = thermocouple well; 3 = Teflon seal; 4 =
sample tube lined with Teflon spaghetti tubing; 5 = high-pressure tubing; 6 = closure piece; 7 = Teflon filter; 8 = Teflon sample
container; and 9 = sample bottle closure pieces.
FUNDAMENTAL EQUATIONS 45 1
on the solubility of serpentine at constant temperature is given by:
where the change in molality with pressure at saturation and constant tem-
perature is (am/aT)T. The effect of temperature on-the solubility of ser-
pentine at constant pressure and the quantities s, S,, and (aG/am)p,T is
shown by:
where (am/aT)pequals the change in molality with temperatures at satura-

tion and constant pressure. At saturation and constant composition, the
relationship between the effect of pressure variation on the equilibrium tem-
perature and the quantities u,v,,s, and & is given by:
where (aT/aP), = the change in temperature with pressure necessary t o

maintain constant composition of the solution at saturation.
In equations 2, 3, and 4, u,v,, &, and (aG/am),,~are unknown quanti-
ties. Although values for s are reported in scientific literature, only two of
these equations are independent; therefore, a third relationship must be
derived t o calculate values of these quantities.
The differential molal entropy of solution at saturation (s - K) or ASis
the change in entropy that occurs when 1mole of serpentine is dissolved in a
large volume of solution saturated with H4Si04. Multiplying AS, by the
absolute-temperature gives the differential heat of solution at saturation
fi,.If S, is more than s, the entropy of H4Si04 enlarged when serpentine is
dissolved.
The value of (aT/aP]min equation 4 may bedetermined wjth an equation
-_-
empirically derived from constant molality if AS,, AH,,or AV, is known. To
obtain V,, S,, or (aG/am)p,T,however, a value is needed h terms of activity
coefficients as shown by:
where u = the number of ions formed by the dissociation of one H4SiOq

molecule, R = the gas constant, T = the absolute temperature in degrees
Kelvin, rn = the molal concentration of H4Si04, and y+ = the mean molal
activity coefficient of H4 Si04 in solution.
Equations is obtained by combining equations 2 and 5 and can be used

t o calculate V , or V , because the value of v is known:
Values for (aln rn/aP)T can be obtained by differentiating empirical

equations for the solubility of serpentine as a function of temperature at a
series of pressures (see Tables 15.111-X). Evaluation of equation 6 , however,
requires a method of determining the mean molal activity coefficient for
changes in molality at constant temperature and constant pressure and at the
saturation molal concentration.
Although activity coefficients can be evaluated in several ways, most
methods are based on dilute-solution treatments derived from Debye-Huckel
theory. This theory, even when applying the limiting law (Lewis and
Randall, 1961) shown in equation 7, is applicable only t o solutes in infin-
itely dilute solutions:
In yf = A 1 z+z- 1
6 (7)
where A = a theoretical parameter and is a function of T and P ; 1z+z-I = t h e
absolute of the product of the charges on the solute ions (for H2Si04); and s
= the ionic strength of the solution (s equals one-half the summation of the
products formed by multiplying the molal concentration mi of each individ-
ual ion multiplied by the charge of that ion zi2 and s = '/zZjrnizi2).
Equation 8 is a more complete form of the Debye-Huckel expression
because it accounts for the average effective diameters of the solution ions
and is used to calculate activity coefficients at concentrations t o 0.01 molal:
where B = a theoretical parameter and is a function of T and P , and a = the

average effective diameter of the solute ions.
Differentiation of equation 8 yields:
Most of these parameters, except y+, are reported in the literature (Klotz,
1950); the parameter yk for H4Si04 in solutions more concentrated than
0.01 molal is not reported. These values can be obtained graphically (Klotz,
1950) by plotting the logarithm of the molal concentration of H4Si04 in
solution versus the square root of the ionic strength of the solution, drawing
a smooth curve through the experimental points, and extrapolating to zero
EXPERIMENTAL DATA 453
ionic strength to obtain the logarithm of the square root of the equilibrium
constant K,. The y? can be calculated from:
yk = Ka'h/m (11)
Because of limited data, the activity coefficients were not calculated.
Experimental data and empirical equations
Table 15.111 presents the smoothed data for the amounts of silicon that
went into solution when reacting serpentine with aqueous 0.025M calcium
chloride solutions at various temperatures and pressures.
The coefficients for a, b, c, and d , and the standard deviations u given in
Table 15.IV were derived from a least squares fit of the silicon solubilities of
serpentine in aqueous 0.025M calcium chloride solutions for a first-degree
equation S = a+ bt, a second-degree equation S = a + b t + c t ' , and a
third-degree equation S = a + bt + ct2 + d t 3 .
Table 15.V gives the smoothed data for the amounts of silicon that went
into solution when reacting serpentine with 0.05M calcium chloride at
TABLE 15.111
Smoothed molal silicon solubilities from serpentine in 0.025molal calcium chloride solu-
tions at various temperatures and pressures
Temperature Pressure
("C)
176 kg/cm2 352 kg/cm2 703 kg/cma 1,055kg/cm2
30 0.1045 x 0.1735 x 0.2045 x 0.2585 x

50 0.1120 10-~ 0.1810 x 0.2110 0.2660 x
76 0.1150x 0.1795 x 0.2120 0.2670x lo-'
100 0.1075 x 0.1705 x 0.2050 x 0.2605 x 10"
125 0.9650x lo4 0.1535 x 0.1895 x 0.2455 x
150 0.7200x lo4 0.1270 x 0.1655 x 0.2210
175 0.4200 x lo4 0.90oo lo4 0.1350 x 0.1905 x
200 0.1150 x lo4 0.5500 x lo4 0.10oo 0.1590 x lo-'
various temperatures and pressures. Table 15.VI contains the coefficients for
first, second, and third degree equations derived from a least squares fit of
the data in Table 15.V. Table 15.VII presents the smoothed data for the
amounts of silicon that went into solution when reacting serpentine with
aqueous 0.05M sodium chloride solutions at various temperatures and pres-
sures. Table 15.VIII contains the coefficient for first, second, and third
degree equations derived from a least squares fit of the data in Table 15.VII.
TABLE 15.IV
Coefficients for empirical equations and resulting standard deviations (a) derived from the
data in Table 15.111
Pressure
176 kg/cm2 352 kg/cm2 703 kg/cm2 1,055 kg/cm2
S=a+bt
a 0.14567 x 0.22177 x 0.24774 x 0.30126 x
b -0.55728 x 10” -0.71175 x -0.61810 x 10” -0.59899 x
U 0.16388 x lo4 0.17652 x 0.16566 x lo4 0.16282 x lo4
S = a + b t + c t2
a 0.86221 x lo4 0.15754 x 0.18723 x 0.24202 x
b 0.81657 x lod6 0.77264 x 0.78028 x 0.77015 x 10”
c -0.60087 x lo-’ -0.64922 x lo-’ -0.61160 x lo-* -5.59881 x lo-’
U 0.21557 x lo-’ 0.18634 x lo-’ 0.99021 x 0.17255 x lo-’
S = a + b t + ct2 + d t 3
a 0.77886 x lo4 0.14803 x 0.18125 x loW3 0.23157 x
b 0.11362 x lo-’ 0.11370 x lo-’ 0.10095 x lo-’ 0.11707 x
c -0.92296 x lo-’ -0.10164 x lo-’ -0.84257 x lo-’ -0.10025 x lo-’
d 0.93284 x lo-’ ’ 0.10636 x lo-’’ 0.66894 x lo-’ ’ 0.11693 x lo-’’
U 0.18318 x lo-’ 0.13391 x lo--’ 0.56229 x 0.97345 x 10”
TABLE 15.V
Smoothed ‘molal silicon solubilities from serpentine in 0.05 molal calcium chloride solu-
(“C)
176 kg/cm2 352 kg/cm2 703 kg/cm2
30 o.1010 10-~ 0 . 2 0 4 5 ~lov3 0.3165 x

50 0.1030 x 0 . 2 1 0 5 ~lov3 0.3355 x
75 0.9850 x loW4 0.2025 x 0.3175 x
100 0.9300 x 0.1955 x 0.3170 x
125 0.8650 x lo4 0.1850 x 0.2975 x
150 0.7250 x 0.1720 x 0.2840 x
175 0.5650 x lo4 0.1540 x 0.2735 x
200 0.3650 x 0.1325 x 0.2525 x
EXPERIMENTAL DATA 455
TABLE 15.VI
data in Table 15.V
-.
Pressure
176 kg/cm2 352 kg /cm 703 kg/cm2
S = a + bt
a 0.12328 x lo-’ 0.23102 x loV3 0.34775 x
b -0.37430 x -0.43229 x 10” -0.42874 x
(3 0.76194 x 10- 0.80191 x lo-’ 0.93876 x lo-’
S = a + bt + ct2
a 0.95751 x lo4 0.20276 x 0.32185 x lo-’
b 0.26192 x lov6 0.21998 x 0.16988 x 10”
C -0.27826 x lo-’ --0.28487 x lo-’ -0.26181 x lo-’
U 0.12056 x lo-’ 0.20391 x lo-’ 0.61658 x lo-’
S = a + bt + ct2 + d t 3
a 0.98227 x lo4 0.19892 x 0.29920 x
b 0.16698 x 10- 0.36685.~10” 0.10385 x lo-’
C -0.18257 x lo-’ -0.43253 x lo-’ -0.11373 x lo-’
d 0.27713 x lo-’ ’ 0.42652 x lo-” 0.25355 x lo-’’
U 0.11573 x lo-’ 0.19706 x lo-’ 0.53361 x lo-’
TABLE 15.VII
Smoothed molal silicon solubilities from serpentine in 0.05 molal sodium chloride solu-
(“C)
176 kg/cm2 352 kg/cm2 703 kglcm’ 1,055 kglcm’
30 0.1700 x lo4 0.3700 x 10- 0.5300 x lo4 0.7400 x 10-

50 0.1950 x lo4 0.3950 x lo4 0.5500 x lo4 0.7700 x lo4
75 0.2100 10- 0.4100 x lo4 0.5600 x lo4 0.7750 x lo4
100 0.2200 lo4 0.4300 x 0.5700 x lo4 0.7900 x lo4
125 0.2200 lo4 0.4200 x lo4 0.5600 x lo4 0.7650 x lo4
150 0.2100 0;4100 x lo4 0.5500 x lo4 0.7600 x lo4
175 0.1950 x 10- 0.4000 x lo4 0.5250 x lo4 0.7400 x lo4
200 0.1700 x lo4 0.3700 x lo4 0.5050 x 10- 0.7150 x lo4
TABLE 15.VIII
Coefficients for empirical equations and resulting standard deviations ((7) derived from the
data in Table 15.VII
Pressure
176 kg/cm2 352 kg/cm2 703 kg/cm2 1,055 kglcm’
S=a+bt
a 0.19939 x lo4 0.40047 x lo4 0.56299 x lo4 0.77924 x lo4
b -0.56601 x -0.13533 x 10- -0.17005 x -0.19774 x loU7
U 0.18996 x lo-’ 0.2Q454 x lo-’ 0.18211 x lo-’ 0.19618 x lo-’
S = a + bt + ct2
a 0.13066 x lo4 0.32714 x lo4 0.49784 x lo4 0.71234 x lo4
b 0.15828 x 0.16960 x lov6 0.13354 x 0.13485 x low6
C -0.69472 x 10*-0.74119 x -0.65845 x lov9 -0.67625 x low9
U 0.28312 x 0.40924 x low6 0.38334 x 0.71101 x
S = a + bt + ct2 + dt3
a 0.12438 x lo4 0.31912 x lo4 0.48420 x lo4 0.68299 x lo4
b 0.18236 x 10” 0.20036 x 0.18585 x 10” 0.24740 x
C -0.93739 x -0.10512 x -0.11856 x -0.18106 x
d 0.70283 x lo-’’ 0.89777 x lo-’ 0.15266 x lo-’ ’ 0.32852 x lo-’ ’
U 0.26986 x 10” 0.39434 x 10- 0.33519 x 0.58769 x 10-
TABLE 15.IX
Smoothed molal silicon solubilities from serpentine in 0.1 molal sodium chloride solutions
at various temperatures and pressures
(“C)
176 kg/cm2 35 2 kg/cm2 703 kg/cm’ 1,055 kglcm’
30 0.2350 x lo4 0.3200 x lo4 0.5700 x 0.1355 x

50 0.2800 x lo4 0.3650 x 0.6100 x lo4 0.1435 x lo4
75 0.3100 x lo4 0.4000 x lo4 0.6500 x lo4 0.1475 x lo4
100 0.3300 x lo4 0.4150 x 0.6600 x lo4 0.1495 x lo4
125 0.3200 x 0.4350 x lo4 0.6450 x lo4 0.1480 x lo4
150 0.2900 x lo4 0.4050 x 0.6150 x lo4 0.1450 x low4
175 0.2700 x lo4 0.3600 x lo4 0.5700 x lo4 0.1415 x lo4
200 0.1750 x lo4 0.2500 x lo4 0.4950 x lo4 0.1325 x lo4
SUMMARY AND CONCLUSIONS 457
TABLE 15.X
data in Table 15.IX
Pressure
176 kg/cm2 352 kg/cm2 703 kg/cm2 1,055 kg/cm2
S=a+bt
a 0.30724 x lo4 0.39656 x lo4 0.64963 x lo4 0.14527 x
b -0.27390 x lo4 -0.24585 x -0.42217 x -0.21139 x
a 0.45114 x lo-' 0.54718 x lo-' 0.46091 x lo-' 0.56072 x lo-'
S = a + bt + ct2
a 0.14511 x 10- 0.20125 x lo4 0.48267 x lo4 0.12509 x
b 0.34728 x 0.42680 x low6 0.34365 x 0.44518 x
c -0.16387 x lo-' -0.19742 x lo-' -0.16876 x lo-' -0.20395 x lo-'
a 0.84998 x 0.12037 x lo-' 0.65103 x lod 0.10164 x lo-'
S = a + b t + ct2 + d t 3
a 0.16424 x 104 0.27529 x 0.47371 x lo4 0.12176 x
b 0.27393 x low6 0.14290 x 0.37802 x lov6 0.57294 x
c -0.89940 x lov9 -0.88707 x -0.20340 x lo-' -0.33271 x lo-'
d -0.21411 x lo-' '-0.82867 x lo-' '0.10033 x lo-'' 0.37291 x lo-' '
a 0.80897 x 0.65538 x lod 0.63945 x 0.90915 x 10"
Table 15.IX contains smoothed data for the amounts of silicon that went
into solution when reacting serpentine with aqueous 0.W sodium chloride
solutions at various temperatures and pressures. Table 15.X contains the
coefficients for first, second, and third degree equations derived from a least
squares fit of the data in Table 15.IX. Plots of the silicon solubilities versus
temperature are shown in Fig. 15.15.
Summary and conclusions
Solubility determinations were made of five clay minerals in aqueous

saline solutions. In general, the dissociation of silicon as silica from the clay
minerals decreased with increasing concentrations of dissolved salts at
ambient temperatures and pressures. Solubility determinations of a serpen-
tine mineral in aqueous saline solutions at elevated temperatures and pres-
sures were determined in specially designed hydrothermal equipment. The
standard deviations of the experimental data were acceptable within the
limits of the equilibria time, and the smoothed data yielded acceptable
empirical equations. The equipment proved to be of excellent design and
construction.
The solubilities of silicon from a serpentine in aqueous salt solutions at
various temperatures and pressures can be calculated with the coefficients
- -
,031 - -
,040
1 :-
,046- -
/ - =
- -
-
-
0
2 .034.- 0.025 Molal calcium chloride - 0.05 Molol calcium chloride

-
i
x - x y
,031
.048 -
.O 46
.O 44 KEY
1,055 Kg/Cm2
0 703 Kg/Cm2
.042- 352 KQ/Cm2
x 176 Kg/Cm2
.04 I
20 40 60 00100 200 20 40 60 00100 200
TEMPERATURE, "C
Fig. 15.15. Molal silicon solubilities from serpentine in aqueous chloride solutions at
various temperatures and pressures.
for a first-degree equation S = a + bt, a second-degree equation S = a + bt +

c t 2 , and a third-degree equation S = a + bt + ct2 + d t 3 given in Tables 15.IV,
VI, VIII, and X. The first-degree equation can be used to make a rapid
calculation with fairly good accuracy; the third-degree equation can be used
to calculate a more accurate value consistent with the experimental data.
Although equation 11 can be used to obtain approximate activity coeffi-
cients, considerably more data are needed for calculating accurate activity
coefficient values.
The solubility values obtained in a study of five clay minerals in aqueous
saline solutions indicated that in general the silicon solubilities decreased
REFERENCES 459
with increasing concentrations of dissolved salts at ambient conditions. To

make more specific conclusions, a more detailed study would be necessary.
References
Collins, A.G., 1969. Solubilities of some silicate minerals in saline waters. U.S. offSaline
Water Res. Dew. Progr. Rep., No.472, 27 pp.
Dickson, F.W., Blount, C.W. and Tunell, G., 1963. Use of hydrottermal y l u t i o n
equipment to determine the solubility of anhydrite in water from 1 0 0 C to 275 C and
from 1 bar t o 1,000 bars pressure. A m . J. S c i , 261:61-78.
Gibbs, J.W., 1928. The Collected Works of J. Willard Gibbs, 1. Thermodynamics. Long-
mans Green, New York, N.Y., 353 pp.
Klotz, I.M.,1950. Chemical Thermodynamics. Prentice Hall, Englewood Cliffs, N.J., 369
PP.
Lewis, G.N. and Randall, M., 1961. Thermodynamics. McGraw-Hill, New York, N.Y., 2nd
ed., 723 pp.
Chapter 16. ENVIRONMENTAL IMPACT OF OIL- AND GAS-WELLDRIL-
LING, PRODUCTION, AND ASSOCIATED WASTE DIS-
POSAL PRACTICES*
No detailed studies have been made about how drilling fluids, drilling
muds, well cuttings, and well-treatment chemicals may contribute to pollu-
tion. Studies of well blowouts and possible development of communication
between a fresh-water aquifer and an oil-bearing sand have been made
(Vedder et al., 1969) as have studies of possible pollution related to poor
production practices (Schmidt and Wilhelm, 1938). The fact that the brines
produced with oil and gas can contribute t o pollution is well known
(Crouch, 1964; Grandone and Schmidt, 1943; Taylor et al., 1940; Wilhelm
and Schmidt, 1935), but no universally satisfactory method of their disposal
is available. Disposal of brine by solar evaporation in evaporating ponds has
been investigated (Gunaji and Keyes, 1968), but final disposal of the residue
salts needs further development. Some brines contain valuable minerals
which are economically recoverable, and treatment or disposal of such brines
should be coordinated with mineral-recovery processes whenever possible
(Collins, 1966).
Several publications are available about oilfield brine disposal by subsur-
face injection into porous and permeable strata (Morris, 1956; Payne, 1966;
Rice, 1968); the staff of the East Texas Salt-Water Disposal Company (1953)
has prepared a comprehensive report that describes gathering systems,
pumps, treatment methods, and injection wells. Subsurface injection of
oilfield wastes provides a good method for disposal of potential water pollu-
tants, but the results are not always satisfactory (Donaldson, 1964; Watkins
et al., 1960). This disposal method has been blamed as the possible cause of
earthquakes, and if a natural disaster, such as an earthquake, occurs, new
faults or fractures in subsurface strata may provide communications between
the strata containing the waste and fresh-water aquifer (Bardwell, 1966;
Evans, 1966; Warner, 1966).
Drilling
Drilling fluids and muds
The most modern drilling method is the rotary system which requires
circulation of drilling fluid for removal of drilled cuttings from the bottom
* Reprinted with permission from Journal Water Pollution Control Federation,
43~2383-2393 (1972).
462 ENVIRONMENTAL IMPACT OF PETROLEUM PRODUCTION
of the hole t o keep the drill bit and the bottom of the hole clean. The
drilling fluids are pumped from ground surface through a drill pipe and bit to
the bottom of the hole and returned t o the surface through the annulus
between the hole and the drill pipe. The flow of formation gas, oil, and brine
into the drill hole is blocked by a fluid-mud column which produces a
hydrostatic pressure that counterbalances or exceeds the formation pres-
sures.
In certain geological environments, abnormally high-fluid pressures are
encountered, i.e., the hydrostatic pressure is greater than 0.107 kg cm-*
m-' of depth. When oil or gas wells are drilled into such an environment,
there always is the possibility of a blowout unless elaborate precautions are
taken and correct drilling muds are used. A situation can develop in un-
cemented or poorly cemented environments if degradation or sloughing
around the casing in a high-pressure zone occurs, allowing the pressured
hydrocarbons t o flow along the outside of the casing to a zone of lower
pressure; Fig. 16.1 shows how this can happen. Drilling fluids may include
gases, liquids, foams, and solids suspended in liquids. Liquid drilling fluids
include crude oil, fresh water, and salt water. Most of the solids suspended in
Conductor casing
Surface casing
Possible pollution
Oil string casing Siough o f f area
Fig. 16.1. Manner in which heaving shales or incompetent zones slough off and permit
communication of a lower zone with an upper zone.
DRILLING 463
TABLE 16.1
Some constituents used in drilling fluids and muds
Quebracho extract
Lignosulfonates, calcium and chrome derivatives
Acrylonitrites (such as hydrolyzed polyacrylonitrite)
Sodium salts of meta and pyrphosphoric acid
Natural gums
Tannins
Molecularly dehydrated phosphates
Subbituminous products
Protocatechuic acid
Barite
Lignins (such as humic acids)
Bentonite
Sugar cane fibers
Lime
Granular material, such as ground nutshells
Corn starch
Salt water
Soluble caustic/lignin product
Carboxy methyl cellulose
Crude oil
Sulfonated crude oil
Oil emulsions
Sodium chromate
Anionic and nonionic surfactants
Organophylic clay
Soaps of long-chain fatty acids
Phospholipids (e.g., lecithin)
Asbestos
liquids are called drilling muds and include the following: suspensiuns of
clays and other solids in water (water-base muds); suspension of solids in oil
(oil-base muds); oil-in-water emulsions (oil-emulsion muds); and water-in-oil
clay emulsion (inverted emulsion muds). Table 16.1 lists some of the com-
pounds in drilling muds (Caraway, 1953; Simpson et al., 1961).
Sulfonated drilling muds are prepared by: (1) sulfonating asphaltic crude
oil with sulfuric acid, followed by neutralization with sodium silicate and ion
exchanging with hydrated lime; or (2) absorbing concentrating sulfuric acid
on a porous carrier, e.g., diatomaceous earth, and then sulfonating asphaltic
crude oil with acid carrier, followed by partial neutralization with sodium
hydroxide and ion exchanging with hydrated lime.
The usual asphaltic crude oils that are used yield a 5- t o 7-wt.% carbon
residue and have an API gravity in the range of 26'-31'. Some blends may
contain an 18' API asphaltic crude oil with a 12-wt.% carbon residue blend-
ed with a paraffinic 42' API crude oil with a 0.5-wt.% carbon residue. These
muds are usually mixed with oil a t the drilling site and used in the drilling
operation. As the cuttings plus the used drilling mud are recovered from the
well, the drilling mud is usually separated from the cuttings and reused.
Some, of course, will be lost because it adheres to the cutting; therefore,
some will present a possible water or land pollution hazard (Messenger,
1963).
The use of quebracho, starch, and carboxy methyl cellulose in formulating
drilling muds has decreased in the last decade, whereas the use of chrome
lignosulfonates has increased. The use of lime-treated mud systems has also
decreased, whereas the use of low-solid muds, invert emulsions, and chrome
lignosulfonate systems has increased.
Considerable money is invested in drilling muds, especially in the heavier
muds; consequently, they are recovered for reuse. Such muds are primarily
used for emergencies, such as lost circulation and high-pressure kicks from
both gas and salt water. Many of the used muds are treated with high
concentrations of lignosulfonates t o produce a stable mud with specific
properties.
Possible sources of pollution from drilling fluids and muds are the fluids
and muds that may be spilled during drilling, those that may escape into a
subsurface fresh-water aquifer, those that cling to the drill cuttings, and
those that are not reused. The data in Table 16.1 indicate that several con-
stituents in drilling fluid and mud are capable of polluting water and land.
Chemical treatment of wells
Wells are treated with acids t o increase the permeability of the reservoir
rocks. This increases fluid flow and increases the recovery of oil and gas; it
also improves fluid injection in secondary oil recovery and disposal
operations. Hydrochloric, nitric, sulfuric, hydrofluoric, formic, and acetic
acids are used. Such treatments produce soluble compounds including cal-
cium chloride, sodium sulfate, sodium fluoride, etc., and in addition, may
leave partially spent acids in solution.
The volume of acid used t o acidize a well may range from 1.9 t o 1 2 m3,
depending upon the amount of acid-soluble strata, the thickness of the
horizon being treated, and the calculated productivity of the well (Hurst,
1970). Table 16.11 lists the approximate amounts of hydrochloric, formic,
TABLE 16.11
Volume of acids used for oil- and gas-well treatment
Acid Volume (m3 /year)
Hydrochloric 3.3 lo5

Formic 7.6 x lo2
Acetic 3.8 x lo2
DRILLING 46 5
and acetic acid used in the United S

treatment.
Other pollution problems can develop when the salt-enriched solution plus
any unspent acid are withdrawn from the well, because subsequent disposal
of these solutions is complicated by their tendency t o form precipitates and
their low pH. It also is difficult t o inhibit (Harris et al., 1966) the acid-
treatment solution t o prevent corrosion, and when corrosion does occur, the
acid solutions and other fluids will escape at the point of pipe failure and
pollute the adjacent zone, which may be a fresh-water aquifer.
Corrosion inhibitors
According to Hurst (1970), a universal “super” inhibitor has evaded the

researcher for 40 years. Such an inhibitor would be useful to prevent steel
casing and tubing from corroding as a result of acid treatment of a well. The
best available high-temperature inhibitor is a combination of sodium arsenite
with an alkyl phenolethylene oxide surfactant, and arsenic-type inhibitors
have been used for both low- and high-temperature applications since the
1930’s. Table 16.111 lists some of the inhibitors used in the United States
(Cowan, 1970).
TABLE 16.111
Types and amounts of inhibitors used in oil- and gas-well treatment

~~
Inhibitor
Sodium msenite 4.54 lo5

Imidazoline 5.68 lo5
Abiethylamine 3.18 lo5
Coal tar derivatives 1.14 lo5
Acetylenic alcohol-alkyl pyridine 1.36 lo5
~~
TABLE 16.IV
Types and amounts of other additives used in oil- and gas-well treatment
Additive
Lactic acid (44%) 2.61 x lo5

Citric acid 9.08 x lo3
Alkylaryl sulfonic acid 2.27 x lo5
Zirconium oxychloride (20%) 1.14 lo5
Quaternary ammonium derivatives 9.08 lo4
Polymers 4.54 lo4
Gum gum 2.61 x lo6
Fluid loss agents 8.17 lo5
Emulsion preventers 2.04 lo5
Other additives
To reduce friction, reduce loss, sustain permeability, prevent emulsions

from forming, and prevent precipitation, additives are added to the oil- or
gas-well systems. Table 16.IV lists some of the compounds used for these
purposes and the approximate amounts used in 1year.
Possible pollution from petroleum
An opening or cylindrical hole from the ground surface to a subsurface

oil-t o gas-bearing formation is a well. Such an opening usually is lined with a
metal pipe cemented in place, and production equipment is fastened t o the
cased hole to regulate and control oil or gas withdrawal rates. Before drilling
Possible break
Conductor casing
Pass i b Ie po I Iu t ton
Surface casin
Fig. 16.2. Probable manner whereby a well blowout can develop communication between
an upper sand and a lower sand.
PRODUCTION 467
a well, some knowledge of the geologic formations to be penetrated is useful,

as is knowledge of the approximate depth of the target petroleum-bearing
zone. This information is needed so that the appropriate diameter, length,
and type of tubular goods can be selected in planning the well.
Most States have laws requiring the setting of surface casing to protect the
fresh-water subsurface sands from invasion by brines and hydrocarbons from
deeper horizons. Therefore, a minimum of two strings of casing - the sur-
face casing and the oil-string casing - will almost always be required. Addi-
tional strings of casing may be required if heaving shales are found while
drilling, if abnormal pressures are encountered, or if a zone of lost circula-
tion is found. Each additional string of casing requires more capital and
increases the cost of the well.
If appropriate precautions are not taken in planning, drilling, and com-
pleting an oil or gas well, disastrous consequences can occur. For example,
during drilling operations or when pulling the drill pipe, a well may blow out
if adequate mud pressure is not maintained. Such a situation may develop if
the mud line is accidentally broken or if the well casing is not properly
cemented to competent zones. Fig.16.2 illustrates what might occur if fluid
from a high-pressure well escapes into an incompetent zone and develops
communication of a lower hydrocarbon-bearing horizon with an upper sand.
Production
Possible pollution from petroleum
Crude oil in excessive amounts is detrimental to vegetation; oily wastes on

surface waters can cause a fire hazard, can be deleterious to fish life, and
gradually will combine with particulate matter, sink, and thus pollute the
bottom of the stream or lake. Further, crude oil has destructive effects on
fowl that may swim in the polluted water and may damage the surrounding
flora and the surrounding beaches. Mercury concentrations in excess of 20
ppm are present in some crude oils. In essence then, it can be assumed that
excessive amounts of produced crude oil that finds its way to surface lands
or waters will cause deleterious pollution.
The composition of the crude oil that pollutes the water or land will
determine the extent and type of pollution. For example, some heavy crude
oils possess a specific gravity of about 1, contain about 5-wt.% sulfur, and
have an overall minimum boiling point of about 27OoC. Conversely, some
light crude oils contain virtually no elements other than carbon and
hydrogen, have 0.8 or less specific gravity, and distill below 27OoC. The
major nonhydrocarbons in crude oils are basic and nonbasic nitrogen and
sulfur compounds and acidic and nonacidic oxygen compounds. Usually the
nonhydrocarbons are more highly concentrated in the heavier portions of
the crude oils. In an overall classification, most crude oils can be classified as
naphthenic paraffinic, or intermediate; the naphthenic type usually is the
heaviest, the paraffinic the lightest.
46 8 ENVIRONMENTAL IMPACT OF PETROLEUM PRODUCTION
Once the crude oils escape upon land or water, they are subjected to
evaporation, oxidation, solution, dispersion, and utilization by micro-
organisms. The lighter crude oils will evaporate more readily than will the
heavy ones. The lower hydrocarbons, e.g., methane and benzene, though
relatively insoluble in water, will be more soluble than the higher molecular
weight hydrocarbons; the crude oils containing sulfur compounds probably
will oxidize less rapidly than will those containing metallo compounds.
Crude oils, when spread on salt water, such as the sea, will quite rapidly form
highly stable water-in-oil emulsions, as was exhibited in the Torrey Canyon
disaster.
This type of emulsfon forms thick blobs of oil which are fairly resistant t o
dispersal, oxidation, and bacterial reactions. The reason that this type of
emulsion forms with salt water has not been clearly established. A means of
readily reverting such emulsions t o an oil-in-water type would be desirable
for quick dispersal (Dean, 1968).
Emulsions of petroleum and brine or mixtures of crude oils and sand that
are difficult to break can be found on surface disposal ponds. Should these
ponds overflow, the surrounding land or surface streams will be polluted.
Crude oil also may escape from leaky connections, improperly plugged wells,
improperly cased and cemented wells, holes in lines or storage tanks, or as a
result of an accident. Burning of the petroleum or emulsions, or both, that
enter brine ponds can contribute to air pollution, and all of the petroleum
will not be completely consumed by the fire.
Oil production may produce pollution in onshore or offshore areas from
blowouts of the wells, dumping of oil-based drilling muds and oil-soaked
cuttings, or losses of oil or brine in production, storage, and transportation.
Over 320,000 km of pipelines operating at pressures to 70 kg/cm2 are used
throughout the country and in offshore areas. Any rupture or accidental
puncture of any of these lines results in pollution.
Possible pollution from natural gas
Blowouts of natural gas wells will contribute to pollution, especially if the

natural gas contains appreciable quantities of hydrogen sulfide. Many gas
wells contain enough hydrogen sulfide t o pollute any fresh water they may
contact. Such contact may develop if a well is faulty and communication
between the gas zone and an upper fresh-water zone occurs. Brines associ-
ated with hydrogen sulfide-bearing gas zones also will contain appreciable
quantities of the sulfide.
Possible pollution from oilfield brines
Waters associated with petroleum in subsurface formations usually contain

many dissolved ions. Those most commonly present in greater than trace
amounts are sodium (Na+), calcium (Ca+2), magnesium .(Mg+*), potassium
PRODUCTION 469
(K+), barium (Ba+’ ), strontium (Sr+’), ferrous iron (Fe+’ ), ferric iron
(Fe+3), chloride (Cl-), sulfate (SO4-’ ), sulfide (S-’ ), bromide (Br-’ ),
bicarbonate (HC03-), and dissolved gases, such as carbon dioxide (CO’),
hydrogen sulfide (H,S), and methane (CH,). The stability of petroleum-
associated brine is related to the constituents dissolved in it, the chemical
composition of the surrounding rocks and minerals, the temperature, the
pressure, and the composition of any gases in contact with the brine
(Fulford, 1968).
Scale inhibitors are added to waters and brines t o prevent the precipitation
reactions. Some of the chemicals used in these inhibitors are listed in Table
16.V.
TABLE 16.V
Chemicals used in scale inhibitors
Ethylenediamine tetraacetic acid salts

Nitrilotriacetic acid salts
Sodium hexametaphosphate
Sodium tripolyphosphate
Sodium carboxymethyl cellulose
Aminotrimethylene phosphate
Knowledge of the oxidation state of dissolved iron in brines is important

in compatibility studies. Brines in contact with the air will dissolve oxygen,
and their Eh generally will be from 0.35 t o 0.50 mV. Brines in contact with
petroleum in the formation normally will have an Eh lower than 0.35 mV, as
will waters in contact with reducible hydrocarbons (Hem, 1961). Any
change in the oxidation state of brine containing dissolved iron may result in
the deposition of dissolved iron compounds.
The sediments or precipitate formed from brines can cause environmental
pollution directly or indirectly. For example, if the produced brines are
stored in a pond, the sediments may cause soil pollution; if the brines are
injected into a disposal well, the sediments may plug the face of the disposal
formation, resulting in the necessity to increase injection pressures which
may rupture the input system.
The amount of salt water or brine produced from oil wells varies consider-
ably with different wells and is dependent upon the producing formation
and the location, construction, and age of the well. Some oil wells produce
little or no brine when first produced, but as they are produced, they gradu-
ally produce more and more brine. As some wells become older, the pro-
duced fluids may be more than 95% brine; or for each cubic meter of oil
coming t o the surface, 100 m3 or more of brine also is produced. The
produced brines differ in concentration but usually consist primarily of
sodium chloride in concentrations ranging from 5,000 t o more than 200,000

ppm; the average probably is about 40,000 ppm. For comparison, note that
sea water contains about 20,000 ppm of chlorides. 1 m3 of brine containing
100,000 ppm of chloride will raise the chloride content of 400 m3 of fresh
water above the maximum recommended for drinking water. Petroleum-
associated brines may escape and contact fresh water or soil in different
ways. For example, t o protect the upper fresh waters from the deeper
mineralized waters that might rise in the drilled well, the upper portion of
the well is sealed by a string of cemented surface casing. If a well has
insufficient surface casing, an avenue may be provided for the escape of
brines if they are under sufficient hydrostatic head t o cause them to rise in
the hole to the surface or to the level of fresh water sands.
Handling the tremendous volume of brine produced simultaneously with
petroleum is hazardous. Basically, the problem is to handle and dispose of
the brine in such a manner that it does not contact soil or fresh water and
cause detrimental pollution.
Currently, some produced brines are being discharged into approved sur-
face ponds, whereas most brines are returned underground for disposal or to
repressure secondary oil or gas recovery wells. The discharge of brines to any
surface drainage is strictly prohibited in most States. Potential water and soil
pollution problems are associated with both disposal methods. For example,
if the surface pond is faulty, the brine will contact the soil and various
chemical reactions will occur between the soil and the brine. Sometimes the
brine will pass through the soil, reappear at the surface, and produce scar
areas; sometimes it will pollute the soil and leaching will pollute surface
streams or shallow subsurface aquifers.
Residual salt concentrations beneath or near abandoned unsealed disposal

ponds
Unsealed surface ponds used for the disposal of oilfield brines have
polluted fresh surface waters, potable groundwaters, and fertile land. Be-
cause of chemical and physical phenomena and dispersion, the movement of
soluble pollutants from these pqnds is complex. For example, the soluble
pollutants move slowly in relation to the soil-water flow rate, and dispersion
effects a displacement which causes the contaminated zone to grow.
The Kansas State Department of Health studied the soils beneath and near
and old unsealed brine disposal pond that had been abandoned for 10 years.
During its use, the pond received more than 29,000 metric tons of salts, and
most of those soluble salts probably escaped by soil leaching and down-
drainage and penetrated below the underlying limestone formation. Eleven
test holes were drilled into the soil and shale beneath and adjacent t o the
pond, both above and below the natural drainage slope. Chemical analysis of
the test hole core samples indicated that more than 430 tons (about 1.4%of
the original) of soluble residual salt still remained t o be leached out of the
DISPOSAL 471
soil and shale in the pond area. This amount of soluble or leachable salt
remaining in the area indicates that the return of the subsurface water and
soil t o their prepollution level is a very slow process and may take several
decades. Network pollution zones appear to form where formation fracture
conjugates occur. Leaching appears to be entirely dependent upon the
flushing mechanism provided by meteoric water.
The cation concentrations in the clay minerals were evaluated by X-ray
diffraction techniques to trace cation transportation rates. Chloride analysis
was selected as the most useful single means of detecting the presence of
oilfield brine pollution, but the associated cation concentration should also
be determined t o formulate a more complete picture. Cation adsorption
studies are apparently useful in differentiating brine-polluted soil and shale,
clay mineral studies provide the information on the environmental charac-
teristics of the pollution media, and cation exchange information aids in
explaining the apparent differential transportation rates of ions in brine
seepage solutions (Bryson et al., 1966; Siever, 1968).
Disposal
Subsurface disposal
A problem associated with subsurface brine disposal is casing leaks in the

disposal well, which could allow the brine to enter fresh-water aquifers. Fig.
16.3 shows how an improperly designed disposal well and a leaky oil well
can pollute a fresh-water aquifer. Erroneous geologic knowledge of the sub-
surface formation into which the brine is being pumped presents another
problem. Brine usually is pumped into a subsurface formation that contains
similar brine; however, exact knowledge of the faulting and fracturing of
such a subsurface formation is difficult t o discern. Because the brine is
pumped into the formation, bottomhole pressure must not exceed 0.23 kg
cm-’ m-l of overburden, or the hydraulic pressure may cause fracturing
and in time, the wastes may migrate t o a fresh-water zone.
Petroleum-associated brines from two different formations may form
precipitates if they are mixed. For example, with a well used for disposal of
brines produced from several producing oil wells, it is imperative that precau-
tions be taken in mixing and treating the brines before injection. If the
brines are incompatible and inappropriate precautions are taken, there is a
possibility that deposits will form and filter out on the face of the injection
formation, thus reducing the permeability. The quantity of deposits formed
from incompatible brines depends on ions present. The more common
deposits resulting from reactions of incompatible brines are gypsum
(CaS04 * 2 H’O), anhydrite (CaS04), aragonite (CaCO,), calcite (CaC03),
celestite (SrS04), barite (BaS04), troilite (FeS), and siderite (FeCO, ).
Subsurface brine disposal can be categorized as confinement or contain-
ment; confinement is the placement of brines in a horizon where any move-
Fig. 16.3. Routes by which salt water can enter fresh water wells from faulty oil or
disposal wells.
ment can be controlled or monitored, while containment is the placement

that precludes the movement of the brines out of a formation or zone. Note
that containment cannot be used for an unlimited supply of brine, but that
confinement necessitates the monitoring of the migration of the brines. The
necessary knowledge to define the hydrodynamics of brines injected into
subsurface environments is expensive to obtain, and much of the necessary
fundamental knowledge of subsurface formations is not available. Forma-
tions into which brines are often pumped for disposal are called salaquifers,
and these zones consist of permeable sedimentary rock. Some information
needed before such a zone can be used for disposal operations is: How big is
the zone? If the brine migrates in the zone, might it reappear in another
zone or perhaps migrate to the surface? What mechanisms control move-
ment in a given salaquifer or perhaps out of it? What steps are necessary t o
assure containment or confinement of the brine within the salaquifer? It is
difficult, if not impossible, to develop adequate knowledge concerning how
or where escape channels may occur from a salaquifer. Test drilling is the
only known method that can provide such knowledge, and the drilling is
expensive, as is the subsequent evaluation (Drescher, 1965).
DISPOSAL 473
Joint ownership of disposal systems by several companies helps minimize

installation and maintenance costs. Brines can be gathered through common
lines and accumulated at a central location, so that one disposal well serves
many producing wells. Investment costs primarily depend upon the brine
characteristics from the producing formations, the receptivity of the disposal
formation, and the condition of the surface soil for gathering-line instal-
lation. Where the brine production is relatively small, a complete system can
be installed for less than $500 per producing well, but if large volumes of
brine are produced, the disposal well may be able t o service only a few
producing wells, and the cost may be $8,000 or more per well. In 1968 the
operating costs in representative fields in the Permian Basin amounted t o
0.42 mill/m3 for 6.04 million m3 of brines (Research Committee, Interstate
Oil Compact Commission, 1968; Rice, 1968).
In 1967 in Texas, there were 41,000 active oil wells, and about 6,900
active gas wells from which more than 0.8 million m3/day of brine was
produced. That amount of brine contains approximately 66 x lo6 kg of
salts, and that amount of daily produced salt can pollute 98 million m3 of
fresh water to the point that it would not be acceptable as drinking water.
Waterflooding of oil sands was begun in Bradford field, Pennsylvania, in
1907, and was developed into a systematic operation after 1934. Con-
siderable care must be exercised in using this method to recover oil, e.g.,
there is a danger that the reinjected brine will migrate to fresh-water streams.
Subsurface disposal of oilfield brines, as well as industrial wastes, is being
increasingly used to replace surface disposal (Enright, 1963; Research Com-
mittee, Interstate Oil Compact Commission, 1960). The ideal conditions for
formations used for such disposal are large areal extent, high permeability
and porosity, overlying and underlying aquicludes, low internal pressure,
salaquifer, compatible fluids, no unplugged wells open t o an outcrop, and
uniformity. The reservoir used for disposal must be tremendous in size, and
even though the amount of fluid that can be injected is large it is ultimately
limited.
Many things are not known about what happens within a formation used
for disposing of wastes. For example, many wastes are low in pH and ap-
parently no studies have been made of how the pH changes with time within
the formation. Conceivable the acid can react with the rock and perhaps
break out. It is known that most accidental fractures of the formation or the
overlying aquiclude will be horizontal if the well is no deeper than 300 m.
However, if the disposal well is deeper than 450 m, the fracture orientation
is likely to be vertical, and vertical fractures can, if large enough, cause
communication with an upper zone.
Salt water under pressure will attempt to escape from any type of con-
finement. The salt water may escape through fractures becausa of a
mechanical failure within the individual well system, through an old drill
hole that penetrates the injection zone, or through a natural fault system
caused by a recent earthquake.
Recovery o f valuable elements before disposal
Elements found in some brines in economic concentrations are magne-

sium, calcium, potassium, lithium, boron, bromine, and iodine. Many of
them are recovered by chemical companies from sea water, salt lakes, and
subsurface saline waters (Brennan, 1966; Collins, 1970).
Factors which must be considered in evaluating a saline water as an
economic resource are the cost of bringing it to the factory, the cost of the
recovery process, and the cost of transporting the recovered products t o
market. Assuming that a brine is produced only for the purpose of recov-
ering its dissolved chemicals, a prime factor is the cost of pumping the brine.
It will cost less t o produce the brine from a shallow well than from a deep
well. Therefore, disregarding other factors, a brine must not only contain a
certain amount of recoverable chemicals, but it must be available in large
quantity before it can be considered economically valuable, and the farther
it must be pumped, the more chemicals it must contain.
References
Bardwell, G.E., 1966. Some statistical features of the relationships between Rocky
Mountain arsenal waste disposal and frequency of earthquakes. Mountain Geol., 3:
37-42.
Brennan, P.J., 1966. Nevada brine supports a big new lithium plant. Chem. Eng., 73:
86-88.
Bryson, W.R., Schmidt, G.W. and O’Connors, R.E., 1966. Residual salt of brine affected
soil and shale, Patiwin areas Buller Co., Kansas. Kansas State Dep. Health Bull.,
3( 1): 28 pp.
Caraway, W.H., 1953. Quebraco in oil well drilling fluids. Petrol. Eng., 25:B81-83, B86,
B88, B89, B92.
Collins, A.G., 1966. Here’s how producers can turn brine disposal into profit. Oil Gas J . ,
64: 112-1 13.
Collins, A.G., 1970. Finding profits in oil well waste waters. Chem. Eng., 77: 165-168.
Cowan, J.C., 1970. Some secondary properties of chemicals used for mineral scale inhibi-
tion. Div. Pet. Chem., A m . Chem. SOC.Meet., Houston, Texas, February 22-27, 1970,
Preprints, pp. F47-F57.
Crouch, R.L., 1964. Investigations of alleged groundwater contamination, Tri-Rue and
Ride oilfields, Scurry County, Texas. Texas Water Comm. Rep., No. L.D.-0464-MR,
1 6 PP.
Dean, R.A., 1968. The chemistry of crude oils in relation to their spillage on the sea. In:
J.D. Carthy (Editor), Proceedings Symposium Field Studies Council, Biol. Eff. Oil
Pollut. Luttoral Communities, London, pp. 1-6.
Donaldson, E.C., 1964. Subsurface disposal of industrial wastes in the United States. U.S.
Bur. Min. Inform. Circ., No. 8212, 34 pp.
Drescher, W.J., 1965. Hydrology of deep-well disposal of radioactive liquid wastes. In: A.
Young and J.E. Gallup (Editors), Fluids in Subsurface Environments - A m . Assoc.
PetroL GeoL, Mem. 4 , pp.399-406.
East Texas Salt-Water Disposal Company, 1953. Salt- Water Disposal East Texas Field.
Petroleum Extension Service, Austin, Texas, 116 pp..
Enright, R.J., 1963. Oil field pollution. Oil Gas. J., 61:76-87.
REFERENCES 475
Evans, D.M., 1966. The Denver area earthquakes and the Rocky Mountain arsenal
disposal well. Mountain GeoL, 3:23-26.
Fulford, R.S., 1968. Effects of brine concentration and pressure drop on gypsum scaling
in oil wells. J. Pet. TechnoL, 20:559-564.
Grandone, P. and Schmidt, L., 1943. Survey of subsurface brine-disposal systems in
western Kansas oilfields. U S . Bur. Min. Rep. Invest., No.3719, 20 pp.
Gunaji, N.N. and Keyes, Jr., C.G., 1968. Disposal of brines by solar evaporation. U.S. O f f .
Saline Water Res. Dev. Progr. Rep., No.351, 213 pp.
Harris, O.E., Henrickson, A.R. and Coulter, A.W., 1966. High-concentration hydrochloric
acid aids stimulation results in carbonate formations. J. Pet. TechnoL 18:1291-1296.
Hem, J.D., 1961. Stability field diagrams as aids in iron chemistry studies. J, Am. Water
Works Assoc., 53:211-232.
Hurst, R.E., 1970. Market for completion and stimulation chemicals. Div. Pet. Chem.,
A m . Chem. Soc., Meet., Houston, Texas, February 22-27, 1970, p.l5(12)F9
(abstract).
Messenger, J.U., 1963. Composition, properties and field performance of a sulfcnated
oil-base mud. J. Pet. TechnoL, 15:259-263.
Morris, W.S., 1956. Salt waters disposal from the engineering viewpoint. Presented to the
Res. Comm., Interstate Oil Compact Comm., Dallas, Texas, May 31, 1956.
Payne, R.D., 1966. Salt water pollution problems in Texas. J. Pet. Technol., 18:
1401-1407.
Research Committee, Interstate Oil Compact Commission, 1960. Production and Disposal
of Oilfield Brine in the United States and Canada. The Interstate Oil Compact Com-
mission, Oklahoma City, Okla., 95 pp.
Research Committee, Interstate Oil Compact Commission, 1968. Subsurface Disposal of
Industrial Wastes. The Interstate Oil Compact Commission, Oklahoma City, Okla., 109
PP.
Rice, I.M., 1968. Salt water disposal in the Permian Basin. Prod. Monthly, 32:28-30.
Schmidt, L. and Wilhelm, C.J., 1938. Disposal of petroleum wastes on oil producing
properties. U.S. Bur. Min. Rep. Invest., No.3394, 36 pp.
Siever, R., 1968. Establishment of equilibrium between clays and sea water. Earth Planet.
Sci. Lett., 5:106-110.
Simpson, J., Cowan, J.C. and Beasley, Jr., A.E., 1961. Some recent advances in oil-mud
technology. Presented at 36th Annual Meet., AIME, Dallas, Texas, October 8-1 1 ,
1961, SOC.Pet. Eng. Paper, No. 150, 16 pp.
Taylor, S.S., Holliman, W.C. and Wilhelm, C.J., 1940. Study of brine disposal systems in
Illinois oilfields. U.S. Bur. Min. Rep. Invest., No.3534, 20 pp.
Vedder, J.G., Wagner, H.C. and Schollhomer, J.E., 1969. Geologic framework of the
Santa Barbara channel retion. U.S. GeoL Sum. Prof. Paper, No.679, pp.1-11.
Warner, D.L., 1966. Subsurface injection of liquid wastes. In: N.E. Grosvenor, J.D. Haun
and D.T. Snow (Editors), Natural Gas, Coal, Groundwater: Exploring New Methods
and Techniques in Resources Research. University of Colorado Press, Boulder, Colo.,
pp. 1 07-1 2 5.
Watkins, J.W., Armstrong, F.E. and Heemstra, R.J., 1960. Feasibility of radioactive waste
disposal in shallow sedimentary formations. NucL Sci Eng., 7:133-143.
Wilhelm, C.J. and Schmidt, L., 1935. Preliminary report on the disposal of oilfield brines
in the Ritz-Canton field, McPherson Co., Kansas. U.S. Bur. Min. Rep. Invest.,
No.3297, 20 pp.
REFERENCE INDEX*
Agaev, A.A., 185 Bentor, Y.K., 225

Ahrens, L.H., 133, 136 Bernard, J.L., 157, 216, 217
Akin, G.W., 371 Berner, R.A., 207
Alexeyev, F.A., 308 Berry, F.A.F., 182
Alkemade, C.T., 59 Berry, J.W., 186, 244
Al’tovskii, M.E., 211 Biles, J., 383
Ambrose, A.W., 272 Billings, G.K., 65,244, 317
American Petroleum Institute, 19,25, 27, Birch, F., 212
47,105,254, 272,347 Bischoff, J.L., 223
Amstutz, R.W., 432,433 Bixler, H.J., 241
Amyx, J.W., 429 Black, A.P., 186
Andrews, L.J., 179 Bleakley, W.B., 437
Angino, E.E., 65,132,226, 390, 399 Blount, C.W., 369, 371, 379, 447
Anonymous, 179,194,302,308,396, Blyth, C.R., 243
415 Bogomolov, G.V., 43
Aries, R.S., 405 Bohon, R.L., 179
Armstrong, F.E., 461 Bojarski, L., 254,260, 282
Attaway, D.H., 108, 109, 110 Bond, D.C., 244, 314
Atwater, G.I., 209 Bonoli, L., 182
Ault, W.U., 208 Booth, R.L., 25
Aumeras, M., 161 Borchert, H., 228
Bordovskii, O.K., 165,184
Baar, C.A., 163 Boyle, R.W., 322
Baas Becking, L.G.M., 170,208 Braitsch, O., 163,227
Bacher, A.A., 399,411 Brannock, W.W., 107
Bailey, N.J.L., 303 Brasted, R.C., 183
Baker, D.R., 311, 312 Braus, H., 183
Baker, E.G., 211, 296,298,299,311 Bray, E.E., 296, 297, 311
Ballinger, D.G., 25,43 Bredehoeft, J.D., 243
Bardwell, G.E., 405,461 Brennan, P.J., 390,474
Barnes, H.L., 224 Brenneman, M.C., 303
Bars, E.A., 429 Brey, M.E., 65,207
Bass, Jr., D.M., 429 Briggs, Jr., L.I., 425
Baugher,III, J.W., 343, 344 Brod, I.O.,211
Bazilevich, Z.A., 162 Broecker, W.S., 202
Beasley, Jr., A.E., 463 Brongersma-Sanders, M., 204
Beck, K.C., 207,294 Bronston, A., 308
Beckman, H.F., 105 Brooks, R.R., 82
Beerstecher, Jr., E., 301 Brown, R.H., 434
Bennett, J.H., 227 Bruderer, W., 210
*Only page references to text pages are made in this index. References to pages
containing bibliographic details have been omitted. These details are given at the end
of each chapter.
478 REFERENCE INDEX
Bryson, W.R., 405, 471 Dall'Aglio, M., 1 5 2

Buckley,S.E., 12, 169, 181, 313 Dapples, E.C., 208
Buckman, H.D., 138 Davidson, M.J., 308
Burges, A., 186 Davies, D.W., 437
Burmistov, D.F., 162 Davis, J.B., 185, 301, 302, 313
Burnam, C.W., 185 Davis, J.W., 145, 147, 369, 372
Burriel-Marti, F., 54 Davis, S.N., 430, 434
Burst, J.F., 140, 209, 240, 294, 343 Davis, W.D., 422, 423
Bush, P.R., 224 Dean, J.A., 54, 80
Butler, G.P., 205 Dean, R.A., 468
Butt, J.B., 393, 395, 399 Deffeyes, K.S., 203, 204
Degens, E.T., 181, 182, 207, 210
Califet, Y., 182 DeLaguna, W., 428
Caraway, W.H., 207, 297, 372, 463 Deroo, G., 309
Carpelan, L.H., 203 DeSitter, L.Y., 242
Carpenter, A.B., 226 Dewiest, R.J.M., 430, 435
Cartmill, J.C., 210, 298, 311 Dickey, P.A., 1 , 205, 210, 226, 274, 288,
Case, L.C., 195, 433 298, 311, 343, 346
Castagno, J.L., '43 Dickinson, G., 343, 344
Chapman, G., 182 Dickson, F.W., 369, 371, 447
Chave, K.E., 207, 240 Diehl, H., 96
Chebotarev, I.I., 226, 262, 265, 267, 289 Dingman, R.J., 132, 226
Chenoweth, P.A., 330, 331 Disteche, A., 371
Cherney, S., 399 Disteche, S., 371
Chilingar, G.V., 207, 245, 321 Diterikhs, O.D., 308, 314
Christ, C.L., 50, 167, 198 Dodge, B.F., 402
Christensen, J.J., 399, 411 Donaldson, E.C., 419, 461
Christman, R.F., 186 Drescher, W.J., 472
Clark, S.P., 194 Dressman, R.C., 25
Clarke, F.W., 197 Drong, H.J., 209
Claussen, W.F., 179 Dudova, M.Ya., 308, 314
Clayton, R.H., 243 Duffy, J.R., 180
Cloke, P.L., 168, 208 Dunham, K.C., 224
Coggeshall, N.D., 12, 181, 314 Dunlap, H.F., 32, 3 5
Collins, A.G., 15, 27, 29, 37, 43, 54, 60, Dunseth, M.G., 396
61, 63, 83, 94, 96, 107, 108, 109, 110, Dunton, M.L., 310
111, 135, 137, 140, 143, 145, 156, Dutoit, M.M.S., 1 8 6
1 5 7 , 1 6 9 , 1 7 1 , 204, 219, 226, 227, Duursma, E.K., 178
229, 232, 233, 238, 297, 346, 369,
372, 3 9 0 , 3 9 4 , 4 0 2 , 4 4 1 , 4 6 1 , 4 7 4 East Texas Salt-Water Disposal Company,
Collom, R.E., 272 421,423,461
Columbus, N., 382 Ebrey, T.G., 94, 9 6
Conolly, J.F., 180 Eckhardt, F.J., 238
Cooper, J.E., 184, 297, 311, 315 Edmund, R.W., 1 7 7 , 4 2 5
Corbett, C.S., 293 Egleson, G.D., 226, 234
Coulter, A.W., 465 Eichelberger, J.W., 25
Cowan, J.C., 369, 370, 372, 463, 465 Eley, D.D., 177
Cox, D.L., 392 Elliott, Jr., W.C., 256
Craig, H., 91 Ellis, A.J., 227, 370
Crocker, L., 399 Elliston, H.H., 422, 423
Crouch, R.L., 4 0 5 , 4 6 1 Emery, E.M., 184
Czamanske, G.K., 224 Emery, K.O., 201, 207, 293
Engelbrecht, R.S., 1 8 3
Enright, R.J., 473
REFERENCE INDEX 479
Epstein, S., 91 Glew, D.N., 371

Erdman, J.G., 309,310 Goebel, E.D., 425
Eshaya, A.M., 402 Goldberg, E.D., 149
Ettre, L.S., 185 Goldschmidt, V.M., 133,136,138, 140,
Evans, D.M., 405,461 141,145,147,150,151,155,158
Evans, E.D., 296 Goldshteyn, R.I., 382
Evans, G., 205 Goncharov, Yu.,152
Evans, M.E., 177 Gordon, W.C., 3
Ewing, G.C., 203 Gorgy, S., 108
Grabau, A.W., 205
Fabricand, B.P., 65,207 Grabowski, R.J., 157,216, 217
Fajardo, I., 346, 347 Graf, D.L., 243
Fash, R.H., 308 Grandone, 461
Ferguson, W.S., 311 Grauer, A., 211
Ferris, J.G., 434 Graves, Jr., R.W., 201
Fertl, W.H., 346, 363,364 Greene, R.C., 373
Fettke, C.R., 2 Grim, R.E., 230, 232
Feugere, G., 310, 322 Griswold, W.T., 1
Filonov, V.A., 317,318 Gullikson, D.M., 297
Finch, W.D., 343 Gunaji, W.N., 461
Fisher, F.L., 107 Gutsalo, L.K., 313, 318,319
Fleischer, M., 133, 136, 141, 143, 145,
147,149,150,151,152,153,156, Haddenhorst, H.G., 209
158,159,161,162,164 Halliburton Company, 74,118
Forgotson, J.M., 364 Ham, W.E., 201
Forsman, J.P., 310 Hames, D.A., 371
Foster, J.B., 362 Hanshaw, B.B., 224, 320
Fowler, Jr., W.A., 245,343, 362 Hanson, W.E., 181
Frank, H.S., 177 :Hardaway, J.E., 425
Frear, G.L., 370 Harkins, K.S., 343, 344
Friedman, G.M., 202,207,234 Harris, O.E., 465
Friedman, I., 91,243, 244 Hastings, W.W., 107
Fuchtbauer, H., 209 Hawthorne, R.R., 32, 35
Fulford, R.S., 369, 371, 380, 469 Hays, J., 245
Fulton, Jr., R.B., 154, 216,217 Hedberg, H.D., 178, 309
Furman, N.H., 44 Heemstra, R.J., 461
Heintz, J.A., 399,411
Gaida, K.H., 294 Helgeson, H.C., 224
Galin, V.L., 321 Hem, J.D., 133,135,147,155,159,161,
Garbarini, G.S., 425,427 168,170,469
Garrels, R.M., 50,167,198 Hemley, J.J., 152
Garrett, R.G., 322 Henningsen, E.R., 382
Garrison, A.D., 2 Henrickson, A.R., 465
Gates, G.L., 207,297,372 Herrmann, R., 59, 227,228
Gautier, A., 108 Hill, G.A., 224, 320
Gehman, Jr., H.M., 310 Hiltabrand, R.R., 239
Geodekyan, A.A., 308 Hiss, W.L., 320
George,’D.R., 399 Hitchon, B., 244, 245, 296, 297, 299
George, W.O., 107 Ho, A., 186
Gerard, R.E., 310, 322 Hocutt, C.R., 12,169,181,357
Gevirtz, J.L., 202 Hodges, Jr., R.M., 241
Gibbs, J.W., 449 Hodgman, C.D., 34
Ginsburg, R.N., 202 Hodgson, G.W., 180,210,296,299,311
Glater, J., 371 Hoerr, C.W., 183
480 REFERENCE INDEX
Holliman, W.C., 461 Klotz, I.M., 352

Hollister, J.C., 434 Knopf, A., 155
Holser, W.T., 163, 205 Knowles, D.B., 434
Hood, D.W., 183,184 Knudsen, M., 172
Horvitz, L., 308 Kwrner, W.E., 184
Hottmann, C.E., 343, 344, 347, 363 Kohout, F.A., 382
Howell, J.V., 1 Kolodii, V.V., 316
Hoylman, H.W., 321 Koons, C.B., 311
Hubbert, M.K., 382 Kopp, J.F., 25
Hulse, B.T., 432,433 Korobov, D.S., 317
Hunt, J.M., 178,205, 210, 296, 298, 310 Kortsenshtein, V.N., 313,314
Hunter, J.A., 399,413 Kovieheva, I.S., 210
Hurst, R.E., 464, 465 Koyama, T., 165
Kozin, A.N., 233
Ibert, E.R., 107 Kozlov, M.F., 43
Illing, L.V., 202 Kramer, J.R., 225, 226, 240
Imbimbo, E.S., 65,207 Krause, H.R., 183
Krauskopf, K.B., 107
Jamieson, G.W., 205, 310 Kravchik, T.E., 314
Jankowsky, W.J., 209 Krejci-Graf, K., 232, 317
Jebens, R.H., 399,413 Krivosheya, V.A., 313
Jeffery, L.M., 183, 184 Kroepelin, H., 322
Jeltes, R., 181 Kroner, R.C., 25
Jobelius, H., 211 Krouse, H.R., 303
John, L.M., 183 Krumbein, W.C., 204,205,429
Johnson, A.C., 322 Kudelskii, A.V., 43
Johnson, R.K., 343, 344 Kuznetsov, S.I., 301, 302
Johnston, J., 370 Kumetsova, Z.I., 211
Jones, B.F., 223, 225 Kvenvolden, K.A., 205, 315
Jones, O.S.,419
Jones, P.H., 245,298,343,346,347,363 Lagerwerff, J.V., 371
Jones, P.J., 32 Lamontagne, R.A., 182
Landes, K.K., 309
Kabot, F.J., 185 Lane, A.C., 3
Kaley, M.E., 201 Lane, E.C., 272
Kaplan, I.R., 82, 170, 208 Larson, T.E., 150
Karasik, M.A., 152 Larson, T.J., 241
Karaskiewicz, J., 322 Latimer, W.M., 29, 167
Karim, M., 321 Laurent, P., 183
Kartsev, A.A., 297,308,314 Laurence, L.L., 420,422
Kawai, K., 165,179 Leobourg, M., 11
Kazmina, T.I., 153 Leuszler, W.E., 420,422
Keefter, R.M., 179 Levorsen, A.I., 197, 295,382
Kelley, W.P., 230 Lewis, G.N., 373,452
Kellog, M.W., and Company, 172,173 Lichtenberg, J.J., 25
Keyes, Jr., C.G., 461 Linnenbom, V.J., 181
Khitarov, N.I.,140, 232, 240 Litchfield, C.D., 183
KidwelI, A.L., 178,298 Lochte, H.L., 185
Kimura, S.,241 Loeb, S., 241
Kincaid, E.E., 392 Loleit, A.J., 425
King, R.M., 150 London, E.E., 313, 317
Kinsman, D.D.J., 205 Long, F.A., 180
Klein, G., 399 Longbottom, J.E., 25
Klemme, H.D., 309 Lotze, F., 204
REFERENCE INDEX 481
Louis, M., 182 Mills, R. van A., 2, 272

Low, P.F., 242, 346 Milne, I.H., 242
Lowenstam, H.A., 202 Mitgarts, B.B., 153
Lucchesi, P.J., 379 Moeller, T.,133, 135,136, 141
Lucia, F.J., 203,204 Mogilevskii, G.A., 297
Lutz, F.B., 108 Moore, C.A., 202
Lyon, T.L., 138 Moore, D., 170, 208
Morgan, C.O., 130
Mandl, I., 211 Morgan, J.J., 199
Manheim, F.T., 223 Morris, R.C., 205
Manjikian, S., 241 Morris, W.S., 421,424,461
Manuel, O.K., 227 Mousseau, R.J., 12,324
Marcy, V.M., 43, 101,122 Muehlberg, P.E., 399, 411
Margoshes, M., 83 Muir, R.O., 228
Marsden, S.S., 165,179 Mun, A.I., 162
Marsh, G.H., 47 Munn, M.J., 1, 2
Martell, A.E., 145 Murata, K.J., 107
Mason, B., 138, 143, 144, 149,150, 151, Myagkov, V.F., 162
153,158,159,161,163,165
Matthews, C.S., 428 Nagy, B., 180
Mayeda, T.K., 91, 243 Nalco Chemical Company, 186
Maxey, G.B., 243 Namoit, A.Y., 177
McAlister, J.A., 11 Natural Gasoline Association of America,
McAuliffe, C.D., 180, 181,210, 311,314, 186
315 Naumor, G.B., 168
McBain, J.W., 183 Nektarova, M.B., 185, 316
McBermott, E., 308 Nemethy, G., 177
McCann, T.P., 425 Neuberg, C., 211
McCutchan, J.W., 371 Neruchev, S.G., 210
McDevit, W.F., 180 Neuman, E.W., 372
McElvain, R.G., 318 Neumann, H.J., 211
McIlhenney, W.F., 399,413 Noad, D.F., 11, 321
McIver, R.D., 295 Nordby, H.E., 186
McKelvey, J.G., 242 Norris, M.S., 12, 314
McLean, D.D., 429 Nutter, B.P., 11
McLeod, H.O., 307
McNellis, J.M., 132 O’Conner, J.T., 183,474
Meents, W.F., 243 O’Conner, R.E., 405
Meinert, R.N., 310 Oden, S., 185
Meinschein, W.G., 298 Odum, H.T., 145
Mellon, M.G., 35 Ostroff, A.G., 169,256,274,371, 428,
Merritt, Jr., L.L., 92 437
Messenger, J.U., 464 Oudin, J.L., 309
Metcalfe, L.D., 184 Ovchinnikou, N.V., 165
Metler, A.V., 371
Meyer, H.W.H., 185 Packman, R.F., 186
Meyerhoff, A.A., 212 Page, H.J., 186
Michaels, A.S., 241 Paine, W.R., 343, 347
Middleton, F.M., 183 Palmer, C., 254,256, 257
Midgett, M.R., 25 Parker, J.W., 224, 321
Miholic, S., 150 Pate, B.D., 108
Miller, E.E., 209 Patnode, H.W., 310
Miller, J.C., 226 Patterson, M.S., 373
Miller, W.C., 390, 399 Pavolva, G.A., 165
482 REFERENCE INDEX
Payne, R.D., 461 Rice, I.M., 437,461,473

Peake, E., 180,296,311 Rich, J.L., 2
Pearson,C.A., 108,109,110,111 Rieke, 111, H.H., 245
Peck, R.B., 206, 294, 344 Riley, G.A., 206
Pennebaker, E.S., 344, 363 Riley, J.M., 399
Peterson, J.A., 320,425 Ringwood, A.E., 194
Peterson, J.B., 425 Risley, G.A., 201
Pettijohn, F.J., 197 Rittenberg, S.C., 207
Philipp, W., 209 Rittenhouse, G., 157,216, 217, 229
Philippi, G.T., 295, 296, 300, 309,310 Roach, J.W., 320
Phillips, R.C., 368,369, 372, 380 Robinson, J.W., 65
Phleger, F.B., 203 Robinson, L.R., 183
Piper, A.M., 132 Rogers, G.S., 256
Piper, T.J., 183 Rogers, W.B., 1
Pirson, S.J., 169, 208, 308 Rosaire, E.E., 308
Pirsson, L.V., 155,168 Rosenqvist, I.T., 207
Platte, J.A., 101 Rosin, J., 23
Plumley, W.J., 201 Ross, C.S., 197
Pollard, T.A., 207 Ross, R.D., 426,427,433
Popov, A.I., 382 Rubio, F.E., 303
Posner, A.M., 183 Ruddick, E.L., 399
Potter, E.C., 27 Russell, D.G., 428
Pourbaix, M.J.N., 30, 167 Russell, W.L., 242
Powers, M.C., 294,343
Prescott, J.M., 183 Sahama, T.G., 197
Presley, B.J., 82 Salutsky, M.L., 396, 399
Price, L.C., 178, 296 Sandell, E.B., 98,99
Privasky, N.C., 425 Sanders, J.E., 207,234
Pugin, V.A., 140, 232, 240 Saraf, D.N., 180
Pusey, 111, W.D., 311 Savchenko, V.P., 316
Pytkowicz, R.M., 371 Scheraga, A.A., 177
Pyushchenko, V.G., 321 Schilthuis, R.J., 1, 3
Schmidt, G.W., 294, 314,343,361, 362,
Quaide, W., 205 405
Querio, C.W., 226,234 Schmidt, L., 461,471
Schoeller, H., 232, 254, 267,271
Rae, A.C., 182 Schollhomer, J.E., 461
Rainwater, F.H., 31 Schrayer, G.T., 12, 314
Rakestraw, N.W., 108 Schrink, D.R., 107
Ralston, A.W., 183 Schwab, R., 317
Ramirez-Munoz, J., 54, 67 Schwanenbek, F.X., 272
Ramsey, T.R., 320 Scribner, B.F., 83
Randall, M.H., 373,452 Selby, S.M., 34
Rankama, K., 197 Selm, R.P., 432,433
Ransone, W.R., 308 Serebriako, O.I., 317
Reed, W.E., 210 Sestini, F., 186
Reeder, L.R., 426,430,434,436 Shaborova, N.T., 185, 316
Reichertz, P.O., 207 Shaffer, L.H., 369,371,380
Reistle, Jr., C.E., 131,272 Shankland, R.S., 34
Research Committee, Interstate Oil Shaw, D.R., 166,244
Compact Commission, 473 Shaw, T.I., 166
Rettig, S.L., 223 Shearman, D.J., 205
Reuter, J.H., 181,182,210 Shilov, I.K., 316
Reynolds, L.C., 432,433 Shimp, N.F., 243
REFERENCE INDEX 483
Shiram, C.R., 343, 347 Timko, D.J., 346, 363, 364

Shiskina, O.V., 165 Tissot, B., 309
Shreve, R.N., 395 Tollin, G., 186
Shvets, V.M., 211, 316 Tooms, J.S., 227
Siegel, A., 182 Torrey, P.D., 1, 2
Siever, R., 207,471 Trask, P.D., 310
Sikka, D.B., 318 Trelease, S.F., 160
Sillen, L.G., 145, 166 Tronko, I.V., 317
Silverman, S.R., 311 Truesdell, A.H., 223, 225
Simons, H.F., 308 Trump, R.P., 242
Simpson, J., 463 Tunnell, G., 447
Skinner, B.J., 193 Tunyak, A.P., 185, 316
Sloss, L.L., 429
Slowey, J.F., 184 Ulrich, R.A., 425
Smales, A.A., 108 Upson, J.E., 382
Smith, G.F., 96 U.S. Bureau of Mines, 254, 272, 393, 404,
Smith, H.M., 303 41 5
Smith, N.O., 180 Usiglio, J., 203
Smith, P.V., 211
Smith, W.W., 419,438 Valyashko, M.G., 162,227
Snow, D.T., 438 Vandenburgh, A.S., 223
Sokolov, V.A., 308 Van Everdingen, R.O., 223
Solomon, H.J., 393, 395, 399 Van Nostrand Press, 211
Souriragan, S.,241 Vasil’ev, V.G., 313
Spencer, C.W., 425 Vasileuskaya, A.Ye., 152
Spencer, D.W., 311 Vdovyking, P., 319
Spiegler, K.S., 242 Veal, H.K., 425,427
Ssutu, L.,371 Vedder, J.G., 461
Stallman, R.W., 434 Veldink, R., 181
Stead, F.L., 425 Vermeulen, T., 399
Steelink, C., 186 Vetter, O.J.G., 368, 369, 372, 380, 461
Stiff, H.A., 131,132,371 Vilonov, V.A., 319
Stormont, D.H., 308 Vinogradov, A.P., 316
Stratton, G., 108 Vinogradov, V.L., 228,316
Stumm, W., 199 Viher, G.S., 200, 322
Subotta, M.I., 297 Von Engelhardt, W., 240, 294
Sudo, Y.,316
Sugawara, K., 165, 166 Wagner, H.C., 461
Sulin, V.A., 254, 257, 258, 347 Wallace, W.E., 8,343, 344
Swigart, T.E., 272 Walton, G., 183
Swinnerton, J.W., 181 Wangersky, P.J., 184
Waring, G.A., 135
Tabasaranskii, Z.A., 297 Warner, D.L., 425,426,461
Taggart, M.S.,12, 169,181,313 Washington, H.S.,197
Taguchi, K., 296 Water, C.J., 111
Takahashi, T., 202 Waters, Jr., O.B.,399
Tallmadge, J.A., 393, 395, 399 Watkins, J.W., 47, 108, 109, 110,461
Taylor, D.W., 206 Watson, J.A., 65
Taylor, S.S., 461 Weast, R.C., 34
Templeton, C.C., 369, 372 Weaver, C.E., 294
Terada, K., 165, 166 Weimer, J.C., 434
Terzaghi, K., 206, 294, 344 Weintritt, D.J., 369, 370,372
Tickell, F.G., 128 Welcher, F.J., 40
Thatcher, L.L., 31 Weller, J.M., 206
484 REFERENCE INDEX
Welte, D.H., 297, 298, 309 Witherspoon, P.A., 180,182

Weyl, P.K.,203,204 Woodcock, A.H., 91
White, D.E., 107,135,162,194,195, Wright, C.C., 437
206,232,240 Wright, J.L., 434
White, W.A., 3, 243 Wyllie, M.R.J., 32, 242
Whitehead, H.C., 108
Whiting, R.L., 429 Yakoylev, Yu.I., 316
Whitney, E.D., 379 Yarbrough, H.R., 313
Wilhelm, C.J., 461 Yasenev, B.P., 308
Willard, H.H., 92 Young, A., 244, 346
Williams, J.A., 302, 303 Yurovskii, Yu.M., 308
Williams, P.M.,184
Wilson, A.L., 186 Zarrella, W.M., 12,181, 314
Wilson, D.F., 182 Zinger, A.S., 180, 314
Wilson, J.E., 425 Zobell, C.E., 30, 167,302
Winter, J.A., 25 Zorkin, L.M., 313
Winters, J.C., 302,303
SUBJECT INDEX
Abiogenic, 205, 293 Ammonium pyrollidine dithiocarbamate,

Abnormal pressure, 344, 359 82
-, detection, 343 Anadarko Basin, 128,129,226, 297
Accumulation of petroleum, 2,298 Anaerobic bacteria, 169
Accuracy, in analyzing methods, 21 Analytical method, choosing of, 20
-, of measurements, 23 Anhydrite, 159,471
Acetic acid solutions, 120 -, solubility, 372
-, in oil-well acidizing, 120 Antigorite, 442
Acidified samples, 23 Apatite, 158
Acidity, 37, 156 Aquifer, 225, 244
Acid treatment, 464 -, contamination, 434
Additives, 463-466 Aragonite, 471
Aerobic bacteria, 169, 208, 213, 302, Arbuckle formation, 51, 330
304,314 -, chloride concentration, 332
Alabama, 333 -, potentiometric surface map, 331
Algae, 165 Arkansas, 333,335
Aliquot size, 41 Aromatic hydrocarbons, 314
Alkali metals, concentration during Arsenic, colorimetric methods, 108
evaporation, 134 -, constituent of oilfield waters, 158
-, properties, 134 -, diethyldithiocarbamate method, 108
Alkaline earth metals, concentration -, Gutzeit method, 108
during evaporation, 134 -, occurrence, 108
-, properties, 141 Artificial fracturing, 430
Alkalinity, 37, 155,254 Athabasca tar sands, 303
Alkanes, 298,309 Atomic absorption methods, 65-82
Allochthonous origin, 293 Authigenic deposition, 201, 206
Alteration of hydrocarbons, 299 Autochthonous origin, 293
Aluminum, abundance, 155
-, atomic absorption method, 65 Bacterial alteration of petroleum, 301
-, constituent of oilfield waters, 155 Bacterial reduction, 50, 235
-, emission spectroscopy method, 90 Barite, 471
-, properties, 148 Barium, 141,147
Amazonite, 140 -, atomic absorption method, 65,77
American Petroleum Institute, 19, 25, -, colorimetric methods, 114,115
27, 254, 347 -, emission spectroscopy method, 83
American Society for Testing and -, flame spectrophotometric method, 63
Materials, 19 -, gravimetric method, 115
Amino acids, 182 -, properties, 141, 171
-, chromatographic techniques, 182 -, qualitative test, 114
Ammonium nitrogen, 157 -, recovery, 415
-, concentration and economical profit, Barium sulfate, 367, 370,372
413 -, concentration and ionic strength, 374
-, determination by titrimetric methods, -, scale, 370
43 -,solubility, 372, 375-377, 379, 382
486 SUBJECT INDEX
Bathyal-abyssal deposits, 201 Brines (continued)

Bell Canyon formation, 322, 327-329 -,- high magnesium, 394
-, calcium concentration, 329 -,- high sodium chloride, 392
-, chloride concentration, 328 -, disposal, 411-417, 471
-, dissolved solids content, 327 -, evaluation, 427
-, potentiometric surface map, 323, 324 -, operations, 389,395
Bell Creek field, Wyoming, 303 -, ponds, 389,470
Benzene, 315 -, refinery, 390,395, 396,402,403
-, solubility, 179 -, stabilization, 380,381
Benzene method for prospecting, 181, -, state regulations, 434
314 Bromide, 45, 162,227, 360,408, 410,411
Beryl, 140 -, concentration, 361
Beryllium, atomic absorption method, 65 -,- in brines, 162,163
-, constituent of oilfield waters, 141,142 -,- and economic profit, 404,413
-, emission spectroscopy method, 89 -,- by evaporating, 163, 164
-, properties, 141 -, - and geologic age, 408
Bicarbonate, concentration and depth, 357 -, properties, 171
-,- and economic profit, 413 -, recovery, 41 5
Bioconcentration, 228 -, recovery from Catesville, 391, 392
Biogenic origin, 205, 293 -, seaweed and coral, 228
Biological degradation of hydrocarbons, -, sodium chloride relation, 162
301 -, titrimetric methods, 45
Biological weathering, 197 Bromine, 162,391,398
Biosphere, 193 -, abundance, 164
Bischofite, 162 -, locations with high concentration, 410
Bittern-type brines, 359 -, occurrence, 163
Boiling point, 172 -, recovery, 391
Bolivar Coastal field, 303 Buffers, 28
Borate boron, titrimetric method of boron, -, definition, 198
37 Bulk density, 363
Borehole temperature, 364 Burner height device, 67
Boron, 21, 37
-, abundance, 153 Cadmium, 152
-, concentration and economic profit, -, abundance, 152
404,413 -, atomic absorption method, 65
-,- by evaporating, 154 -, colorimetric method, 103
-, - and geologic age, 408 -, properties, 153
-, constituent of oilfield waters, 153 -, sphalerite as carrier, 153
-, emission spectroscopy method, 83 Calcite, 194,198, 201, 208,370, 371,
-, properties, 171 461,471
-, recovery, 415 -, solubility, 370
Bradford Sand, 1, 2 Calcium, 40, 73,140, 143, 283-289, 370,
Brine value, 414,416 395
Brine worth, 414,416 -, abundance, 144
Brines (see also Oilfield brines and Oilfield -, atomic absorption method, 65,72,75
waters), 13, 27, 117, 159,160, 181, -, complexometric method, 40
213, 269,319,348,380,395,427,471 -, concentration, 289,361
-, analyses, 272 -,- in Bell Canyon formation waters, 329
-, classification, 225, 254, 257, 260,267 -,- and depth, 358
-, commercial, 415,416 -,- and economical profit, 404,413
-, composition, 384 -,- by evaporating, 143,144
-, concentrations of elements, 395 -,- and geologic age, 408
-, containing high bromine, 391 -, constituent of oilfield waters, 141,143
SUBJECT INDEX 487
Calcium (continued) Chloride, concentration (continued)

-, locations with high concentrations, 409 -,- and relationships to other elements,
-, properties, 141,171 229-239
-, recovery, 415 -, properties, 171
-, solubility, 370 -, recovery, 415
Calcium carbonate, precipitate, 144, 171 -, titrimetric method (Mohr), 44
-, solubility, 202 Chlorine, 161,398
Calcium sulfate, solubility, 144, 381 -, abundance, 161
Calculated resistivity, 35 Chlorinity, 24
Calculating probable compounds, 125 Chocolate Bayou field, 245, 362
Caledonian Group, 151 Chromatographic techniques, 181-185
Calibration curve in flame spectrophoto- Chromium, atomic absorption method, 65
metric methods, 53 Chrysotile, 442
Caliche evaporite deposits in Chile, 161 Cinnabar, 151,152
Cambrian, 213 Classification of oilfield waters, 253, 276
-, concentrations of elements, 219 Classification systems, applications, 274-
Capillary cell method, 182 289
Capitan Limestone, 320 Clastics, 200
Carbonate, 146,152 -, depositional environments, 200
-, depositional environments, 201 Clay minerals, 140,209, 230, 240, 345,
-, recovery, 415 430,433,441,442
Carbonated waters, dissolved elements, 198 -, authigenic, 201
Carbon dioxide, 50, 142,155,170, 208, Cleaning pipelines, 421
299 Coal ashes, 133
-, determination, 50 Colorado, 323
Carboniferous age, 185 Colorimetric methods, interferences in, 93
Carnallite, 162, 228 Combination factor, 127
Case histories, geochemical, 308 Compaction, 206,294
Catesville field, 391 Compaction model, 344
Celestite, 471 Compatibility of oilfield waters, 367
Cell preparation, resistivity measurements, Completion of disposal wells, 432
34 Composition of minerals, 441
Central Basin platform, 320 Composition of oilfield waters, 213
Cesium, 141,393 Compressibility of rock and water, 428
-, abundance, 141 Concentrating by ion exchange, 95
-, constituent of oilfield waters, 134,141 Concentrating brines, 240
-, flame spectrophotometric method, 59, Concentration change during evaporation,
61 134, 204,227,229,231-234, 238
-, properties, 134 Concentration ratios in brines, 236, 237
Chebotarevk classification, 262, 267 Concentration versus proximity to an oil
Chelating agents, 40,82,96 accumulation, 315
Chemical analysis, 125 Confining beds, critical pressure of, 429
Chemical treatment of wells, 464 Connate water, 3, 169, 194, 270, 271
Cherokee Group, 311 Contamination of shallow aquifers, 434
Chile caliche evaporite deposits, 161 Continental slope drill hole, 223
Chloride, 44 Copper, 150
-, concentration, 238, 239 -, abundance, 150
-, - in Arbuckle formation waters, 332 -, atomic absorption method, 65,80
-,- in Bell Canyon formation waters, -, colorimetric method, 95,96, 150
328 -, properties, 148
-,- and depth, 357 Cordellera Isabella, Nicaragua, 323
-,- and economic profit, 413 Core samples analysis, 310
-,- and geologic age, 408 Corrosion inhibitors, 465,469
488 SUBJECT INDEX
Cretaceous age, 135, 154, 163,215, 227, Dolomite, 142, 194, 203, 234, 239
229 Dolomitization, 195,203, 204, 208, 234,
-, lithium concentration, 135 238
-, magnesium concentration, 143 Dolostone deposits, 234
-, potassium concentration, 139 Dow Chemical Company, 391,399
-, sodium concentration, 137 Dowex A-l,96
-, strontium concentration, 146 Drilling, 463
Cretaceous age rocks, brines from, 237 -, disposal wells, 432
Cretaceous system, concentration of Drilling fluids, 461,463
elements, 213 Drilling muds, 8,170,343, 461, 463,468
Critical pressure, 429 Drill-stem test, 8, 12, 181
Cymric fields, California, 152
Earthquakes, 434
Decarboxilization, 185 East Texas Basin, 224,232,272,283-
Deep well injection, 420 288, 321
-, acceptable geologic areas, 425 East Texas field, 420,424
Delaware Basin, 329 East Texas Salt-Water Disposal Company,
Delaware sand, 322, 323 423
Deltaic deposits, 201 Economic value of brine, 403
Density, 172 Eh, 14,19,29, 166, 170, 199
Denver Basin, 343 Eh, unfiltered and filtered petroleum
Deposition, organic matter, 205 producing wells, 320
-, silica, 206 Eh/pH plot, 159,168,170, 199
Depositional basin, 200 Electric log, 153, 341
Depositional environments, 200, 201,203 -, cross section of southwest Louisiana,
Depth, 226 349-352
-, versus concentration, 357, 358 Elements, minor, 220
Description for water sample, 16 Emission spectroscopy, 83
Deuterium, mass spectrophotometric -, calibration curve, 88
method, 91 -, emulsion calibration curve, 87
Devonian age, 234 -, gamma curve, 87
Devonian deposits, 317 Eocene age, 104,114,130, 152, 167,185,
Devonian system, 185, 213, 217, 261 212
Dexter formation, 335 Eolian deposits, 160
Diagenesis, 133, 207, 208, 232, 245, 267, Epm (equivalents per million), 24, 274
346 Equivalents per million (epm), 24, 126,
Diagenetic water, 194 269,275
Dilution technique, 41 Erosion, 198
Disodium 1,2-~yclohexanediaminetetra- Escape routes, 429
acetic acid, 40 Evaluation, economic, 402
Disposal, 471 Evaporites, 137, 223, 238, 239
Disposal brines, 411 -, basin, 203
Disposal costs of brines, 422 -, depositional environments, 203
Disposal systems, cost of, 437 Exchange reactions, 211
Disposal well, 412,424,425,432
-, cross section, 431 Fatty acids, 183,315
Disposal zone, 426,427 -, chromatographic techniques, 184
-, evaluation, 427 - in sea water, 184
Dissolved gases, 12 Feldspar, 138,140,242,433
Dissolved solids, 117,216, 284-286, 323, Ferromagnesian, 142
325,362,410,419 Fertilizers, 397
- in Bell Canyon formation waters, 3,327 Fertilizer production, flowsheet for, 397
Dnepr-Donets Basin, 313,318 Field sampling methods, 273
SUBJECT INDEX 489
Filtered petroleum-producing wells, pH Gravimetric methods, 24, 114

and eH values, 320 Green River formation waters, alkalinity,
Flame spectrophotometric methods, 53 156
Florida, 333-335 Groznyv oil district, 247, 302
Flow diagram of solubility equipment, Gulf Coast area, 128, 129, 181, 343, 346,
448 360
Flow line, 13 Gulf Coast Basin, 243
Flowsheet for descaled sea water and Gulf Coast shales, 179, 240, 360
fertilizers, 397 Gypsum, 194, 203,204,235,317,471
Flowsheet for fresh water and valuable
elements, 401 Hackberry field, Louisiana, 304
Fluid mechanics, 320 Halite, 162, 163, 194,203, 227-229
Fluid travel, 433 Heaving shales, 462
Fluoride, colorimetric method, 109 Holocene age, 165
Fluorine, abundance and properties, 161 Honaker Trail formation, 321
Fluorite, 161 Host rock, 133
Fluorspar, 124, 161 Hot brines, 227
Ford field, Texas, 323 Humic acids, 182,184,185,186
Formation damage, 368 -, chromatographic technique, 185
Formation interval tester, 10 Hydrocarbons, 12, 178, 194, 211,298,
Formation pressure, 12 310
Fossil water, 3 -, accumulation, 178,298
Freezing point, 172 -, alkanes, 309
Fresh-water conversion, 401 -, alteration, 299, 301
Fresh-water production, flowsheet for, -, aromatic, 298, 309
401 -, bacterial attack, 302
Fresh-water well, 471,472 -, biological degradation, 301
Fulvic acid, 185 -, compaction, 294
-, containing nitrogen, 182
Gas analysis, 12 -, diffusion, 180
Gas chromatographic methods, 147, 181, -, generation-migration, 295, 296, 362
182,184,185 -, hydrochemical indicators, 258
Gases in petroleum, reservoir waters, 271 -, maturation, 295
Gas/oil ratio, 11 -, migration, 299, 314, 316
Gasoline Association of America, 186 -, in natural gas, 180
Gaslwater relationships, 256 -, origin, 57
Generation and migration, 295 -, in petroleum, 181
Genetic indicators, 336 -, in recent sediments, 244
Genetic types of waters, 260 -, in sedimentary rocks, 245
Geochemical case histories, 308 -, solubility, 178, 179,296
Geochemical methods, exploration, 307 -, water washing, 300
Geochemistry versus geologic environment, Hydrochemical anomalies, 308, 316
266 Hydrochemical indicators, hydrocarbons,
Geologic maps, 426 258
Geopressure, 343 Hydrodynamic gradients, 295,325
Geopressured reservoirs, 343 Hydrodynamic potential, 382
Geopressured zones, 343, 362 Hydrodynamic zones, 266
Georgia, 335 Hydrogen chloride, free, 120
Geostatic ratio, 344 Hydrogen sulfide, 51,156,159,309,468
Gibbsite, 155 -, method for determination, 51
Glass electrode, 27 Hydrogeochemical exploration, 311
Graphic plots, 128 Hydrogeochemical research and exploration
Graphite, 295 313
490 SUBJECT INDEX
Hydrolysate, 139, 141, 171 Jurassic age, 135, 154, 163, 227, 229
-, rocks, 195,197 -, lithium concentration, 135
-, sediments, 141 -, magnesium concentration, 143
Hydrolysis reactions, 15 -, potassium concentration, 139
Hydrosphere, 193 -, sodium concentration, 137
Hydrothermal equipment, 447 -, strontium concentration, 146
Jurassic system, 213, 214
Igneous rocks, average composition, Juvenile water, 3, 195
196
Illinois Basin, 243 Kainite, 162
Illite, 138-140, 209,230, 239, 240, 269, Kansas brines, 226
441,443,444 Kaolinite, 155, 184, 209, 230, 269, 441,
Index base exchange (IBE), 267,270, 443,444
271,283 Kazhim stratigraphic well, 185
Inhibitors, 465 Kerogen, 205, 309, 311
-, chemical, 465
-, corrosion, 465,469 Lacustrine deposits, 200
Injection of subsurface brines, 420 Lanthanum, 83,84
Injection well, 424 Lea County, New Mexico, 320
Instrumental methods, 20 Lead, 152
Interior regions of the earth, 193 -, abundance, 152
Internal standard solution in emission -, atomic absorption method, 65,81, 82
spectroscopy, 84 -, colorimetric method, 95,99
Interstitial water, 3, 194, 206, 207, 209 -, dithizone method, 99
Iodate, 166 -, ion exchange, 99
Iodide, 45, 110,226 -, isotope ratio, 318
-, colorimetric method, 110 -, properties, 119,148, 152
-, concentration and economic profit, Lepidolite, 140
404,413 Limestone, 51, 197,201, 235, 320
-,- by evaporating, 166 -, dolomitization of, 204
-,- and geologic age, 408 Liquid exchange, chromatography, 182
-, properties, 171 Lithium, 133,392
-, recovery, 390,415 -, abundance, 133
-, seaweed and coral, 228 -, atomic absorption method, 68
-, titrimetric method, 45 -, concentration and economic profit,
Iodine, 164, 390, 398 404,413
-, concentration by algae, 165 -,- by evaporating, 135,136
-, recovery from brines, 390 -,- and geologic age, 408
Iodoargyrite, 164 -,- in Mississippian and Pennsylvanian
Iodoembolite, 164 age formation waters, 136
Ion association, 225 -,- in Tertiary, Cretaceous and Jurassic
Ion exchange, 230 age formation waters, 135
-, concentrating by, 95 -, constituent of oilfield waters, 133
Ionic potential, 133, 142,171 -, flame spectrophotometric method, 54
Ionic radii, 171 -, properties, 133,134,171
Ionization interferences, 66 -, recovery, 415
Iron, 149 -, toxicity, 133
-, abundance, 149 Lithology, 225
-, atomic absorption method, 65, 79 Lithophile, 147
-, colorimetric method, 94,95 Lithosphere, 193
-, emission spectroscopy method, 83 Locations of valuable brines, 406
-, properties, 148 Louisiana, 135, 142, 144, 146, 163,231,
Isotopic fractionation, 243 333,335,349-352, 359-361
SUBJECT INDEX 491
Magnesium, 40,142,283-289, 394 Mineral-acid acidity, 37

-, abundance, 142 Minerals, formation, 234
-, atomic absorption method, 65,71, 74 -, recovered from saline waters, 392
-, complexometric method, 40 Minor elements, 220
-, concentration, 289,238 Miocene age, 165
_ , _ and economic profit, 404,413 Mississippi, 333, 335
-,- and geologic age, 408 Mississippian age, 135, 145,154, 164,166,
-, -,in Mississippian and Pennsylvanian 213
age formation waters, 143 -, lithium concentration, 136
- ,_ ,in Tertiary, Cretaceous and Jurassic -, magnesium concentration, 143
age formation waters, 142 -, potassium concentration, 139
-, constituent of oilfield waters, 141,149 -, sodium concentration, 138
-, locations with high concentration, 410 -, strontium concentration, 146
-, properties, 141,171 Mississippian system, 216
-, recovery, 394,415 Mixed salts, 395
-, specific gravity versus concentration, Mixing of subsurface waters, 382
72 Molecular hydrogen, 180
Magnesium ammonium phosphate, 396 Montmorillonite, 140, 155,184, 209, 238,
Manganese, 40, 149 240, 294,360,441,445
-, abundance, 149 Mud filtrate, 11
-, atomic absorption method, 65,78
-, emission spectroscopy method, 83 Nalco Chemical Company, 186
--,flame spectrophotometric method, 61 Naphthenic acids, chromatographic
-, properties, 148 technique, 185
Maracaibo Basin, Venezuela, 303 National Bureau of Standards, 23
Mass spectrophotometric methods, 91 Natural gas, 178,307,309, 310
Maturation, 295 -, deposits, 178
Maximum worth, 414 Neogene age, 245
Membrane-concentrated brines, 240 Nernst equation, 167
Membrane effect, 240 New Mexico, 224
Metals, 65 Nickel, 98
Meteoric water, 194, 227, 253, 267, 270, -, atomic absorption method, 65
271,289 -, colorimetric method, 95,98
Methane, 178, 181, 295, 300, 316 -, ion exchange, 95
-, measurement of, 12,180,181 Nitrate nitrogen, colorimetric method, 107
-, solubility, 180 Nitrogen containing hydrocarbons, 182
Mercuric iodide in brackish water, 152 Nitrogen-free organic compounds, 178
Mercury, 151 Nodules of manganese oxide, 149
-, abundance, 151 Nontronite, 441, 442,445,446
-, atomic absorption method, 65 Normal pressure, 359
-, properties, 151 North Carolina, 335
Mexia-Talco Fault, 321
Mica, 140, 232,433 Ohm meter, 32
Michigan, 361 Oilfield brines (see also Brines and Oilfield
Michigan Basin, 205,226, 243, 359, 360 waters), 25, 193,219, 273, 389, 422
Microcline, 140 -, analysis, 25, 272
Microphotometer criteria, in emission -, disposal, 389, 420
spectroscopy, 85 -, economics, 422
Micropipet, 41 -, field sampling methods, 273
Microsyringe-evaporating flask, 186 -, origin, 193
Miersite, 164 -, pollution, 468
Migration, 2, 293, 295 Oilfield waters (see also Brines and Oilfield
Milligram per liter (mg/l), 25, 269, 275 brines), 215, 226, 367,372, 374, 389
492 SUBJECT INDEX
Oilfield waters (continued) Permian Basin, 473

-, altered, 242, 243 Petroleum, 195,210,211, 244, 295, 299,
-, analysis, 272 310,315
-, classification, 253 -, accumulation, 178,225, 298,309,310,
-, compared t o sea water, 227 313
-, concentration of elements, 217, 238 -, alteration, 293, 299, 304
-, incompatibility, 367 -, compaction, 294
-, physical properties, 133 -, degradation, 301
Oil in water, 186 -, exploration, 307,320
-, determination, 186 -, generation, 210, 212, 219, 295,297,
Oklahoma, 136, 139, 143, 145, 146, 157, 310
164,166,335 -, migration, 211,295, 297
Optical density, 94 Petroleum Abstracts, 307
Ordovician age, 213, 218, 345 pH, 10,14,15,27,37,168-170,198,
Organic acids, 186,216, 311,312,316 207
-, chromatographic technique, 186 -, temperature and, 28, 29
Organic compounds, 156,178,184-186, -, unfiltered and filtered petroleum-
188,205,317,433 producing wells, 320
-, deposits, 293 pH meter, performance characteristics, 29
-, in oilfield brines, 188 Phenol, chromatographic technique, 181
-, in petroleum-associated waters, 186 Phosphate, 105
-, in saline waters, 177 -, colorimetric method, 105,106
-, in subsurface water, 185 Phosphorus, 158
Organic matter, deposition, 205 Photosynthesis, 165,206
-, in sea water, 178 -, by algae, 165
Origin of brines, 219 Pinkerton Trail Limestone, 320
Origin of oilfield waters, 193 Plasma arc, 83,84
Osmotic pressure, 173 Plate development in emission spectroscopy,
Oxygen, 47,91,158 85
-, method for "0 Playa deposits, 223
-, solubility, 157 Pleistocene age, 150
-, titrimetric method, 47 Plugging of formations, 432
-, Winkler method, 47 Polarization, 171
Pollution, 466-468
Paleozoic age, 160 -, of uppersand by well blowout, 466
Palmer's classification, 254, 256 Porosity, 207,303,345
Paradox Basin, Utah, 224, 321 Porosity reduction, 345
Paraffins, 310 Potassium, 138, 393
Parts per million (ppm), 24, 25 -, abundance, 138
pE, 166 -, atomic absorption method, 70
Pedosphere, 193 -, concentration and economical profit,
Pennsylvanian age, 135,145, 154, 164, 404,413
166,238 -,- by evaporating, 139
-, lithium concentration, 136 -,- and geologic age, 408
-, magnesium concentration, 143 -,-, in Pennsylvanian and Mississippian
-, potassium concentration, 139 age formation waters, 139
-, sodium concentration, 138 -, -, in Tertiary, Cretaceous and Jurassic
-, strontium concentration, 146 age formation waters, 139
Pennsylvanian system, 213, 215 -, constituent of oilfield waters, 134, 138
Permeability, 207, 303,325, 345,430 -, depletion, 140,240
-, artificial fracturing, 430 -, flame spectrophotometric method, 58
-, calculations, 11 -, properties, 134,171
Permian age, 213, 214, 320, 322 -, recovery, 415
SUBJECT INDEX 493
Potentiometric surface map, 331 Saber field, 323, 326, 330

Powder River Basin, 303, 343 Sabkha sediments, 224
Precipitates, 432 Salinity, 24, 226, 254, 257
Precision, 20 -, concentration, 333-335
Precursors, 210, 212, 216, 293, 297, 311 Salt water disposal, 4 12
Preliminary evaluation, marketing research, Salting out effect, 179
405 Sample container, 16
Pressure equalizer, 1 2 Sample treatment, 22
Pressure-distance-time relationship, 430 Sampling, 8
Pressure head, 11 -, methods, 273
Pressure relationships, 433 Sand dike, 345
Pripvatsky Depression, 317 Sandstone, 152,158, 201, 345
Produced water, 4 San Juan Basin, 224, 343
Production, 467 San Juan Mountains, 321
Saturated hydrocarbons, 314
Quality control, 19, 20 Scale, 368, 370
Quartz, 194,201, 209,433 Scale inhibitors, 469
Quartzite, 155 Schoeller’s system,.267, 268
Quebracho, 464 Searles Lake, 135
Sea water, 137, 143, 149, 152, 158, 194,
Radioactive anomalies, 318 215,227,245,392, 394,402
Radioactive compounds, 317 -, average composition, 195
Radium, 318 -, composition, at dolomitization or
-, anomalies, 318 bacterial reduction, 235
-, concentration, 317 -, -, at gypsum precipitation, 234
Radium/uranium ratio, 317-320 -, fatty acids in, 184
Reacting values, 126 Sedimentary basins, 212,406
Reaction coefficients, 126 Sedimentary rocks, 11,51,135,140,147,
Reagent chemicals, 23 195,210,245
Reagent solutions, 23 -, average composition, 196
Recovery, 4, 390, 392,415,474 Sediment compaction, 206
Redox potential (Eh, pE), 29, 157, 166 Sediment diagenesis, 207
Reef carbonates, 202 Seleniferous vegetation in the U.S.,160
Regressive marine deposits, 200 Selenium, abundance, 160
Reistle diagram, 131 -, colorimetric method, 111
Replacing power of ions, 230 Sensitivity for metals in atomic absorption
Reporting analytical results, 25 methods, 65
Reservoir transmissibility, 428 Separation of gas, oil and brine, 400
Residual salt concentration, 470 Serpentine, 446,449,453,458
Resistivity, 32, 33, 35, 317 -, silicon solubility from, 453-458
Reverse exchange, 233 Shales, 141,152, 158,179, 206, 240, 245,
Reverse osmosis, 240, 241 360,430
Rio Bravo fields, California, 152 Shallow aquifers, contamination of, 434
Rocky Mountain area, 156, 343 Siderite, 471
Rodessa formation, 284, 286-288 Significant figures, 25
Rounding-off numbers, 26 Silica, 107
Rubidium, 59, 140, 393 -, abundance, 156
-, abundance, 140 -, analytical procedure when dissolved,
-, constituent of oilfield waters, 139, 140 442
-, flame spectrophotometric method, 59, -, colorimetric method, 107
61 -, deposition, 206
-, properties, 134 -, spectrophotometric method, 107
-, standard-addition technique, 6 1 Silicate, 149, 240,443
494 SUBJECT INDEX
Silicate (continued) Stable-isotopes analysis, 15

-, chemical composition, 441 Standard-addition technique, 50, 55, 57,
-, solubility, 441 60,61
Silicon, concentration, 443-446 Standard solutions, 23
-, solubility, 458 State regulations, 434
-,-, from serpentine, 453-458 Stiff diagram, 131
Silurian age, 213, 218 Stinkfluss, 161
Silver, atomic absorption method, 65 Stratigraphic interval, 11
Simpson Sand, 345 Stratigraphic problem, 325
Sloughing, 462 Stratigraphic traps, 312, 323,325, 326
Smackover formation, 12, 135, 230, 232, Strontium, 62,76, 83, 145,385
234,238,239,391 -, abundance, 145
-, brines, 233, 235,236, P38,240, 392 -, atomic absorption method, 76
-, -, concentration ratios, 236 -, concentration, 239
Smackover Limestone water, drill-stem -,- and economic profit, 404
test, 12 -,- and geologic age, 408
Sodium, 136, 283, 284, 286, 287, 289, -, -, in Pennsylvanian and Mississippian
360 age formation waters, 146
-, abundance, 137 -, -, in Tertiary, Cretaceous and Jurassic
-, atomic absorption method, 65,68 age formation waters, 146
-, concentration, 289, 361 -, constituent of oilfield waters, 141,145
-,- and economic profit, 404,413 -, emission spectroscopy method, 83
-,- and geologic age, 408 -, flame spectrophotometric method, 62
-,-,in Pennsylvanian and Mississippian -, properties, 141, 171
age formation waters, 138 -, recovery, 415
-,-,in Tertiary, Cretaceous and Jurassic Strontium sulfate, 385
age formation waters, 137 -, concentration, 145
-, constituent of oilfield waters, 134, 136 -, concentration and ionic strength, 378
-, determination by calculation, 116 -, saturation in waterflood makeup brines,
-, flame spectrophotometric method, 57 383
-, locations with high concentrations, 409 -, solubility, 145,370, 372,375-377,
-, properties, 134,171 379,381,383,385
-, recovery, 415 Structural trap, 312
Sodium chloride, 33,392 Structure of minerals, 441
-, bromide relation, 163 Subsurface brines, 143, 159, 225, 315
-, resistivity, 33 420
Solids, dissolved (see also Dissolved solids), -, analyses, 321
specific gravity and concentration, 36 -, classification, 216, 234
-, -, in brines and sea water, 413 -, contents, 159,313-315
Solubility, 144, 178, 296, 370, 372, 375- -, hydrochemical anomalies, 316
377,382,441 -, maps, 320
Solubility equipment, flow diagram, 448 -, mixing, 382
Source rocks, 309 -, properties, 214
South Carolina, 335 -, sampler, 9,224
South Caspian Basin, 316 Subsurface disposal, 419,421, 426, 471
South Pyote field, Texas, 323 Subsurface waters, 7,9,135, 253, 289,
Specific gravity, 35,410 382
-, versus concentration for magnesium Sugars, chromatographic technique, 181
solution, 72 Sulfate, 114,385
-, chloride solution, 45 -, concentration and economic profit,
Specific heat, 172 413
Spent acid, 118 -, determination, 53
Sphalerite, 153 -, gravimetric method, 114
SUBJECT INDEX 495
Sulfate (continued) Uinta Basin, 157, 210, 212

-, properties, 171 Uncompahgre Uplift, 321
-, recovery, 415 Underground waste disposal, 434
-, reduction, 208,317 Unfiltered petroleum producing wells,
-, solubilities in synthetic brines, 379 pH and eH values, 320
Sulfide, 51 Units for water analysis, 24
-, determination, 52,53 Unsaturated hydrocarbons, 314
Sulfur, 52, 159 Unstable properties, 15
-, abundance, 159 Upper Cretaceous, 321-
-, concentration and economic profit, Uranium, 318
401 Utah, 157
-,- and geologic age, 408
-, determination, 52 Valuable brines, 406,407
Sulin’s classification, 257, 259 Valuable elements, 474
Suiyosei-ten’nengasu, 178 -, production, flowsheet for, 401
Surat Basin, 245 Value of brine, constituents, 415
Surface tension, 173 Value of dissolved chemicals versus depth,
Suspended solids, 31 403
Sylvania formation, 226 Value estimate, 414
Sylvite, 139, 163, 228 Vapor pressure, 172
Synthetic brines, 27,383 Viscosity, 173
Valence, 171
Tar mats, 299 Volcanic waters, 58
Tax incentives, 436 Volga region, 180
Temperature, 15, 29, 212 Vorobyevite, 140
Temperature gradient, 212
Ternary diagrams, 288 Waste disposal, 471
Tertiary age, 154, 163,212, 213, 227,229 Waste disposal well, 472
-, lithium concentration, 135 Water, compatibility, 12, 290
-, magnesium concentration, 142 -, compressibility, 428
-, potassium concentration, 139 -, properties, primary, 214, 261
-, sodium concentration, 137 -, washing, 300
-, strontium concentration, 146 Water analysis, 24, 130, 321
Tertiary age rocks, brines from, 237 -, interpretation, 130, 131
Texas, 333-335 Water and hydrocarbons, 311
Thermal conductivity, 172 Water exchange coefficient, 265
Thermodynamic equations, 449 Waterflood makeup brines, 383
Thiosulfate, determination, 53 Waterflooding, 4, 369, 420,473
Tickell diagrams, 128 Waters, composition, 280
Titrimetric analysis, 24, 37 -, connate, 3, 194
Toluene, 181,314 -, diagenetic, 194
Transgressive marine deposits, 200 -, formation, 195
Transmissibility, 428 -, fossil, 3
Traps, 299, 312, 326 -, interstitial, 3, 194
-, stratigraphic, 299, 312,323 -,juvenile, 3, 195
-, structural, 312 -, meteoric, 194
-, types, 247 -, subsurface (see also Subsurface waters),
Treatment facilities, 420,421 297
Troilite, 471 Weathering, 197,198, 200,262
Turbidity, 31 -, cycles and types of products, 263
Tuscaloosa formation, 335 Well blowout, pollution by, 466
Wellbore, 368
Wellhead, 14,15
496 SUBJECT INDEX
Wellhead sampling, 13 Zinc, 80, 101, 151

Well treatment, 13 -, abundance, 151
-, pH for determining of, 170 -, atomic absorption method, 65, 80
Wilcox formation, 331-334 -, colorimetric method, 101
Williston Basin, 129 -, ion exchange, 151
Woodbine (Dexter) formation, 335 -, properties, 148
Worth estimate, 414
X-ray diffraction, 201, 471

Geochemistry of Oilfield Waters - Collins AG - 1975 PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Geochemistry of Oilfield Waters - Collins AG - 1975 PDF

Uploaded by

Copyright:

Available Formats

Developments in Petroleum Science, 1

ELSEVIER SCIENTIFIC PUBLISHING COMPANY

AMERICAN ELSEVIER PUBLISHING COMPANY, INC.

Library of Congress Card Number: 73-89149

With 132 illustrations and 87 tables

Copyright 0 1975 by Elsevier Scientific Publishing Company, Amsterdam

Printed in The Netherlands

The purpose of this book is to provide information relevant to the analyt-

Chapter 2 . Sampling subsurface oilfield waters . . . . . . . . . . . . . 7

ATOMIC ABSORPTION METHODS . . . . . . . . . . . . . . . . 65

Chapter 4 . Interpretation of chemical analysis of oilfield waters . . . . . . . 125

Determining a sought compound . . . . . . . . . . . . . . . . . 127

Chapter 5 . Significance of some inorganic constituents and physical properties of oil-

Chapter 6. Organic constituents in saline waters .

Chapter 7. Origin of oilfield waters . . . . . . . . . . . . . . . . . 193

Chapter 8 . Classification of oilfield waters . . . . . . . . . . . . . . 253

Chapter 11. Geopressured reservoirs . . . . . . . . . . . . . . . . 343

Chapter 12. Compatibility of oilfield waters . . . . . . . .

Minerals recovered from saline waters . . . . . . . . . . . . . . . . . 392

Chapter 1 4.Subsurface disposal . . . . . . . . . . . . . . . . . . 419

Reference Index . . . . . . . . . . . . . . . . . . . . . . . 477

these waters were considered waste and had t o be disposed of in some

Subsurface waters associated with petroleum are subjected to forces that

Changes in the composition of petroleum-associated waters (mg/l)

Constituent Well 1 Well 2 Well 3

1947 1957 1947 1957 1956 1959

There is a tendency for some petroleum-associated waters t o become more

The drill-stem test, if properly made, can provide a reliable formation

Fig. 2.1. Multiple closed-in-pressure subsurface sampler. (Courtesy of the Halliburton

Tester positioning depth

the oil. At some point down the column a representative formation-water

Drill-stem test recovery of Smackover Limestone water

Constituent Concentration (mg/l)

top middle bottom

Sample containing dissolved gases

Knowledge of certain dissolved hydrocarbon gases is used in exploration.

sample should be analyzed immediately. If this is not possible, cool or freeze

Sampling at the flow line

Another method of obtaining a sample for analyses for dissolved gases is

Fig. 2.3. Flow-line sampler.

Sampling at the wellhead

It is common practice in the oil industry to obtain a sample of formation

Fig. 2.4. Schematic of method of obtaining a sample at the wellhead.

Sample for determining unstable properties or species

Ferrous iron will oxidize to ferric and should be determined immediately

Sample for stable-isotope analysis

A sample that is t o be analyzed for stable isotopes should be collected

it may be feasible t o collect the sample from a nonheated separator or

Various factors influence the type of sample container that is selected.

Description needed for each petroleum-associated water sample

Sample obtained from (lead line, separatory flow tank, etc.)

Tabulation of sample description

The sample is of little value if detailed information concerning it is not

Water analyses are used by the petroleum industry in studies related to

Data provided by the analytical laboratory are used in decision-making,

documented. The value of research investigations which use oilfield brine

Widespread use of an analytical method indicates that it probably is

Precision is the reproducibility among replicate observations, and in

Precision data on oilfield brine samples for boron

Sample Concentrations of boron found (mg/l)

Taylor Eagle Ford Paluxy Douglas

10.1 15.2 20.1 30.3