Developments in Petroleum Science, 13

enhanced oil recovery

FURTHER TITLES IN THIS SERIES

1A, GENE COLLINS GEOCHEMISTRY O F OILFIELD WATERS
2 W.H. FERTL ABNORMAL FORMATION PRESSURES 3 A.P. SZILAS PRODUCTION AND TRANSPORT O F OIL AND GAS
4 C.E.B. CONYBEARE GEOMORPHOLOGY O F OIL AND GAS FIELDS IN SANDSTONE BODIES

5 T.F. YEN and G.V. CHILINGARIAN (Editors) OIL SHALE
6 D.W. PEACEMAN FUNDAMENTALS O F NUMERICAL RESERVOIR SIMULATION

7 G.V. CHILINGARIAN and T.F. YEN (Editors) BITUMENS, ASPHALTS AND TAR SANDS
8 L.P. DAKE FUNDAMENTALS OF RESERVOIR ENGINEERING

9 K. MAGARA COMPACTION AND FLUID MIGRATION 10 M.T. SILVIA and E.A. ROBINSON DECONVOLUTION O F GEOPHYSICAL TIME SERIES IN THE EXPLORATION FOR OIL AND NATURAL GAS
1 G.V. CHILINGARIAN and P. VORABUTR 1 DRILLING AND DRILLING FLUIDS
1 2 T. VAN GOLF-RACHT FRACTURED HYDROCARBON-RESERVOIR ENGINEERING

Developments in Petroleum Science, 1 3

Proceedings of the third European Symposium on Enhanced Oil Recovery, held in Bournemouth, U.K., September 21-23,1981

Edited by

EJOHN FAYERS
Atomic Energy Establishment, Winfrith, Dorchester, England

ELSEVIER SCIENTIFIC PUBLISHING COMPANY AMSTERDAM -OXFORD -NEW YORK 1981

ELSEVIER SCIENTIFIC PUBLISHING COMPANY Molenwerf 1 P.O. Box 211, 1000 AE Amsterdam, The Netherlands

Distributors for the United States and Canada:
ELSEVIER/NORTH-HOLLAND INC. 52, Vanderbilt Avenue New York, N.Y. 10017

ISBN 0-444-42033-9 (Vol. 13) ISBN 0-444-41625-0 (Series)
0 Elsevier Scientific Publishing Company, 1981

All rights reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without the prior written permission of the publisher, Elsevier Scientific Publishing Company, P.O. Box 330, 1000 AH Amsterdam, The Netherlands Printed in The Netherlands

V

TABLE OF CONTENTS

Foreword . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

ix

CHEMICAL FLOODING
1.

Keynote Paper: “Fundamental Aspects of Surfactant-Polymer Flooding Process” . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D. 0. SHAH, University of Florida, USA Surfactants for EOR Processes in High-Salinity Systems; Product Selection and Evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . M. H. AKSTINAT, Institute of Petroleum Engineering, Clausthal, West Germany

1

2.

43

3 Preliminary Studies of the Behaviour or Some Commercially Available . Surfactants in Hydrocarbon-Brine-Mineral Systems . . . . . . . . . . . . . . . . . 63 C. ANDREWS, N. COLLEY and R. THAVER, British Gas Corporation, London Research Station, UK 4. The Provision of Laboratory Data for EOR Simulation. . . . . . . . . . . . . . . . 8 1 C. E. BROWN and G. 0. LANGLEY, BP Research Centre, Sunbury, UK

5. Experimental Study and Interpretation of Surfactant Retention in Porous 101 Media . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . J. NOVOSAD, Petroleum Recovery Institute, Calgary, Canada
6. The EACN of a Crude Oil: Variations with Cosurfactant and Water Ol Ratio 123 i MIN KWAN THAM and P. B. LORENZ, US Department of Energy, Bartlesville, OK, USA
7.

Dynamic Interfacial Phenomena Related to EOR. . . . . . . . . . . . . . . . . . . . 135 J. H. CLINT, E. L. NEUSTADTER and T. J. JONES, BP Research Centre, Sunbury, UK

8. Behaviour of Surfactants in EOR Applications at High Temperatures . . . . . . 149 L. L. HANDY, University of Southern California, Los Angeles, CA, USA 9. Surfactant Slug Displacement Efficiency in Reservoirs . . . . . . . . . . . . . . . . 161 R. J. WRIGHT and R. A. DAWE, Imperial College, University of London, UK 10. Some Aspects of the Injectivity of Non-Newtonian Fluids in Porous Media. . . 179 P. VOGEL and G. PUSCH, Institute. of Petroleum Engineering, Clausthal, West Germany

vi 11. Basic Rheological Behaviour of Xanthan Polysaccharide Solutions in Porous Media: Effects of Pore Size and Polymer Concentration . . . . . . . . . . . . . . . 197 G. CHAUVETEAU and A. ZAITOUN, Institut Franqais du PCtrole, RueilMalmaison, France The Chateaurenard (France) Polymer Flood Field Test. . . . . . . . . . . . . . . . 213 A. LABASTIE, Elf Aquitaine, Boussens, Saint-Martory, France L. VIO, Elf Aquitaine, Centre de Recherche de Lacq, Artix, France Caustic Flooding in the Wilmington Field, California, Laboratory Modelling and Field Results. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 223 V. S. BREIT, Scientific Software Corp., Denver, TX, USA E. H. MAYER, THUMS Long Beach Co, CA, USA J. D. CARMICHAEL, City of Long Beach Department of Oil Properties

12.

13.

MISCIBLE GAS DISPLACEMENT 14. Keynote Paper: “Miscible Displacement: Its Potential for Enhanced Oil 237 Recovery” . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . R. J. BLACKWELL, Exxon Production Research, Houston, TX, USA Theoretical Aspects of Calculating the Performance of COz as an EOR 247 Process in North Sea Reservoirs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D. S. HUGHES, J. D. MATTHEWS and R. E. MOTT, AEE Winfrith, DorChester, Dorset, UK A New Linear Displacement Model with Mass Transfer Between Phases, n.a. Adapted to C 0 2 Injection. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. JAIN, Institut Franqais du PCtrole, Rued-Malmaison, France (This paper will be distributed at the Conference) Oil Recovery by Carbon Dioxide, the results of Scaled Physical Models and 267 Field Pilots. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . T. M. DOSCHER, M. EL ARABI, S. GHARIB and R. OYEKAN, University of Southern California, Los Angeles, CA, USA Laboratory Testing Procedures for Miscible Floods . . . . . . . . . . . . . . . . . . 285 S. G. SAYEGH and F. G. MCCAFFERY, Petroleum Recovery Institute, Calgary, Canada 299 Complex Study of C 0 2 Flooding in Hungary . . . . . . . . . . . . . . . . . . . . . . s. DOLESCHALL, G. ACS, v. BALINT, z. BIRO, E. FARKAS, T. PAAL, J. TOROK, Hungarian Hydrocarbon Inst., Szazhalombatta, Hungary An Iterative Method for Phase Equilibria Calculations with Particular Appli313 cation t o Multicomponent Miscible Systems . . . . . . . . . . . . . . . . . . . . . . . N. VAROTSIS, A. C. TODD and G. STEWART, Heriot-Watt University, Edinburgh, UK

15.

16.

17.

18.

19.

20.

vii 2 1. Phase Equilibrium Calculations in the Near-Critical Region . . . . . . . . . . . . . 329 R. RISNES, Norsk Agip, Norway V. DALEN, J. I. JENSEN, Continental Shelf Institute, Trondheim, Norway The Effect of Simulated COz Flooding on the Permeability of Reservoir Rocks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 351 G. D. ROSS, A. C. TODD and J . A. TWEEDIE, Heriot-Watt University, Edinburgh, UK

22.

NUMERICAL METHODS 23. 24. 25. Keynote Paper: “Computer Modelling of EOR Processes”. K. AZIZ, University of Calgary, Canada

.............

367

Three-Dimensional Numerical Simulation of Steam Injection. . . . . . . . . . . . 379 P. LEMONNIER, Institut Frangais du PCtrole, Rueil-Malmaison, France Special Techniques for Fully Implicit Simulators. . . . . . . . . . . . . . . . . . . . 395 J. R. APPLEYARD, I. M. CHESHIRE and R. K. POLLARD, Operatings Research Group, AERE Harwell, UK Some Considerations Concerning the Efficiency of Chemical Flood Simulators 409 R. W. S. FOULSER, AEE Winfrith, Dorchester, Dorset, UK Control of Numerical Dispersion in Compositional Simulation. . . . . . . . . . . 425 D. C. WILSON, T. C. TAN and P. C. CASINADER, Imperial College, University of London, UK Interphase Mass Transfer Effects in Implicit Black Oil Simulators. D. BANKS and D.K. PONTING, AERE Harwell, Oxfordshire, UK

26. 27.

28.

........

441

EXPERIMENTAL TECHNIQUES 29. 30. 45 1 A Novel Device for COz Flooding. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . V. MEYN, Institute of Petroleum Engineering, Clausthal, West Germany The Use of Slim Tube Displacement Experiments in the Assessment of Miscible Gas Projects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 467 B. J. SKILLERNE DE BRISTOWE, BP Research Centre, Sunbury, UK Nuclear Measurements of Fluid Saturations in EOR Flood Experiments 483 N. A. BAILEY, P. R. ROWLAND and D. P. ROBINSON, AEE Winfrith, Dorchester, Dorset, UK Characterisation of EOR Polymers as to Size in Solution . . . . . . . . . . . . . . 499 R. DIETZ, National Physical Laboratory, Teddington, UK

31.

32.

viii 33. Visualization of the Behaviour of EOR Reagents in Displacements in Porous Media . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 511 E. G. MAHERS, R. J. WRIGHT and R. A. DAWE, Imperial College, University of London, UK

THERMAL RECOVERY METHODS 34. Keynote Paper: “The Interplay Between Research and Field Operations in the Development of Thermal Recovery Methods” . . . . . . . . . . . . . . . . . . . 527 J. OFFERINGA, R. BARTHEL and J. WEIJDEMA, Shell Exploration and Production Laboratories, Rijswijk, Holland U.S. Department of Energy R & D on Downhole Steam Generator for the Recovery of Heavy Oil . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 543 R. L. FOX, Sandia Laboratories, NM, USA J. J. STOSUR, U.S. Department of Energy, Washington, DC, USA Steam Drive - The Successful Enhanced Oil Recovery Technology. . . . . . . . 549 T. M. DOSCHER and F. GHASSEMI, University of California, Los Angeles, CA, USA Down Hole Steam Generation using a Pulsed Burner . . . . . . . . . . . . . . . . . 563 D. A. CHESTERS, C. J. CLARK, F. A. RIDDIFORD, BP Research Centre, Sunbury, UK Hot Caustic Flooding. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 573 R. JANSSEN-VAN ROSMALEN and F. Th. HESSELINK, Shell Exploration and Production Laboratories, Rijswijk, Holland

35.

36.

37.

38.

UNITED STATES RESEARCH PROGRAMME 39. Enhanced Oil Recovery Research and Development in the United States and in the U.S. Department of Energy. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 587 J. J. G. STOSUR, U.S. Department of Energy, Washington, DC, USA

AUTHORINDEX

............................................

595

ix

FOREWORD

T h i s r e s i d e n t i a l symposium i s t h e t h i r d i n t h e series of symposia which have been h e l d on t h e s u b j e c t of enhanced o i l r e c o v e r y i r , t h e United Kingdom; t h e o t h e r two b e i n g h e l d a t B r i t a n n i c House of BP i n London i n May 1977, and a t Heriot-Watt U n i v e r s i t y i n Edinburgh i n J u l y 1978. S i n c e 1977, when t h e f i r s t symposium was h e l d i n London, t h e a n n u a l p r o d u c t i o n and t h e number o f f i e l d s i n o p e r a t i o n i n t h e UK s e c t o r o f t h e North Sea h a s roughly doubled and i t i s perhaps r i g h t t o r e - i t e r a t e t h e remarks of t h e Chairman of t h e o r g a n i s i n g committee of t h e f i r s t meeting. H e s a i d t h a t , "There i s an u r g e n t need t o d e c i d e which enhanced o i l r e c o v e r y t e c h n i q u e s a r e s u i t a b l e f o r u s e i n t h e North Sea. Once t h i s d e c i s i o n i s made, t h e s e l e c t e d R&D g o a l s s h o u l d b e v i g o r o u s l y p u r s u e d , l e a d i n g , h o p e f u l l y , t o t h e development of s p e c i f i c t a i l o r - m a d e t e c h n i q u e s e f f e c t i v e i n t h e i n d i v i d u a l f i e l d s i n t h e North Sea a r e a " . Although t h e s e remarks a r e s t i l l v a l i d t o d a y , i n t h e i n t e r vening p e r i o d t h r o u g h o u t Europe s i g n i f i c a n t p r o g r e s s has been made. W e have s e e n an i n c r e a s e i n t h e number o f p i l o t f i e l d experiments u n d e r t a k e n by t h e o i l i n d u s t r y , an i n c r e a s e i n t h e r e s e a r c h work c a r r i e d o u t a t u n i v e r s i t i e s , r e s e a r c h i n s t i t u t e s and o i l company laboratories. A number of Government programmes have been i n i t i a t e d o r expanded. A g a i n s t t h i s background of an i n c r e a s e d R&D a c t i v i t y , some s i g n i f i c a n t , a l b e i t t e n t a t i v e , s t e p s i n t h e a p p l i c a t i o n of enhanced o i l r e c o v e r y o f f s h o r e have been t a k e n . The c o n t i n u i n g i n c r e a s e i n t h e p r i c e of o i l o v e r t h e p a s t few y e a r s r e n d e r s t h e t i m i n g of t h e p r e s e n t symposium t o be p a r t i c u l a r l y r e l e v a n t t o t h e q u e s t i o n of improvement i n o i l recovery i n a l l t h e s e c t o r s of t h e North Sea and f o r t h e p r o v i s i o n of f u t u r e s u p p l i e s of e n e r g y t o Europe. The o c c a s i o n of t h e p r e s e n t c o n f e r e n c e p r o v i d e s an i n t e r n a t i o n a l forum f o r r e s e a r c h workers ins enhanced o i l r e c o v e r y t o exchange i n f o r m a t i o n and t o develop an i n c r e a s e d awareness of t h e r e s e a r c h s t u d i e s c u r r e n t l y b e i n g pursued e l s e w h e r e . I t i s hoped t h a t new d i r e c t i o n s f o r r e s e a r c h , a p p l i c a b l e t o t h e European C o n t i n e n t a l S h e l f , may become a p p a r e n t and t h e f u t u r e a d o p t i o n of enhanced o i l r e c o v e r y t e c h n i q u e s i n t h i s a r e a advanced. T h i s volume i s a c o l l e c t i o n of t h e p a p e r s t o b e p r e s e n t e d and d i s c u s s e d a t t h e Symposium.

F J FAYERS

Chairman, O r g a n i s i n g Committee September 1981

This Page Intentionally Left Blank

CHEMICAL FLOODING

1

FUNDAMENTAL ASPECTS OF SURFACTANT-POLYMER FLOODING PROCESS

D.0. SHAH
Department of Chemical Engineering and Anesthesiology, University of Florida, Gainesville, Florida 3261 I
ABSTRACT Surfactant-polymer f l o o d i n g process o f f e r s a promising approach t o recover a d d i t i o n a l o i l from the water flooded r e s e r v o i r s which may conThe c a p i l l a r y number, t a i n as much as 70% o f o r i g i n a l o i l - i n - p l a c e . which determines the microscopic displacement e f f i c i e n c y o f o i l , can be increased b y 3 t o 4 orders o f magnitude by reducing the i n t e r f a c i a l tension (IFT) of o i l ganglia below 10-3 dynes/cm. Conceptual events involved i n the m o b i l i z a t i o n and displacement o f o i l ganglia are described i n c l u d i n g the r o l e o f u l t r a l o w i n t e r f a c i a l tension, t h e r o l e o f i n t e r f a c i a l v i s c o s i t y i n coalescence o f o i l ganglia and formation o f t h e o i l bank, the propagation o f the o i l bank, the surfactant-polymer incomp a t i b i l i t y , the formation and f l o w o f emulsions i n porous media, the r o l e o f w e t t a b i l i t y as well as the i n f l u e n c e o f surface charge d e n s i t y o f the r o c k / f l u i d i n t e r f a c e and o i l - b r i n e i n t e r f a c e i n o i l displacement e f f i ciency. It i s shown t h a t t h e r e are two regions o f u l t r a - l o w IFT; 1) i n the low surfactant concentration (0.1-0.2%) and the other i n t h e high s u r f a c t a n t concentration region (2.0%-10.0%). I n t h e low concentration systems, the u l t r a - l o w i n t e r f a c i a l tension occurs when t h e aqueous phase of the surfactant s o l u t i o n i s about the apparent c r i t i c a l m i c e l l e concent r a t i o n . And, the s a l i n i t y i s a t the c r i t i c a l e l e c t r o l y t e concentration for the coacervation process. The m i g r a t i o n o f surfactant from the aqueous phase t o the o i l phase v i a coacervation process appears t o be r e sponsible f o r the u l t r a l o w i n t e r f a c i a l tension. I n high surfactant concentration systems, a middle phase microemuls i o n i n e q u i l i b r i u m w i t h excess o i l and b r i n e forms i n a narrow s a l i n i t y range. T h e ' s a l i n i t y a t which equal volumes o f o i l and b r i n e are s o l u b i l i z e d i n the middle phase microemulsion i s r e f e r r e d t o as the optimal s a l i n i t y o f the system. A t t h e optimal s a l i n i t y , t h e i n t e r f a c i a l tension a t both i n t e r f a c e s i s equal. Evidence i s presented t h a t the middle phase microemulsion a t the optimal s a l i n i t y i s a water external microemulsion formed due t o coacervation and subsequent phase separation o f m i c e l l e s f r a n t h e aqueous phase. The optimal s a l i n i t y can be s h i f t e d t o a desired value by varying t h e s t r u c t u r e and 'concentration o f alcohol. The s h i f t i n optimal s a l i n i t y can be c o r r e l a t e d w i t h t h e b r i n e s o l u b i l i t y o f the alcohol used i n a given s u r f a c t a n t formulation. It was f u r t h e r observed t h a t the optimal s a l i n i t y increases w i t h the o i l chain length. I n order t o form middle phase microemulsions at very high s a l i n i t y , ethoxylated surfactants o r alcohols can be incorporated i n t o a s u r f a c t a n t formulation which can s h i f t the optimal s a l i n i t y t o as high as 32% NaCl concentration. Such high s a l i n i t y formulations c o n s i s t i n g o f petroleum sulfonates and ethoxylated sulfonates are r e l a t i v e l y i n s e n s i t i v e t o diva1ent c a t ions.

The surfactant-polymer i n c o m p a t i b i l i t y can lead t o a phase separat i o n o f a surfactant and polymer even i n the absence o f o i l . The r a t e o f f l a t t e n i n g o f an o i l drop i n a surfactant f o r mulation increases s t r i k i n g l y i n the presence o f alcohol. The a d d i t i o n o f alcohol also promotes the coalescence o f o i l drops. I t appears t h a t the presence o f alcohol promotes the mass t r a n s f e r o f s u r f a c t a n t from the aqueous phase t o the i n t e r f a c e . Some novel concepts f o r surfactant-polymer f l o o d i n g process have been discussed i n c l u d i n g the use o f tm. 1 Schematic diagram o f an o i l r e s e r v o i r and the displacement of o i l by water o r chemical flooding. the formation o f middle phase microemulsion i s promoted by the presence o f polymer i n the aqueous phase. The surfactant-polymer i n c o m p a t i b i l i t y i s explained i n terms o f excluded volume e f f e c t s and the maximization o f solvent f o r polymer molecules. .2 The coalescence r a t e o r the phase separation time was minimum at optimal s a l i n i t y . s a l i n i t y gradient design and the i n j e c t i o n o f an o i l bank t o promote o i l recovery. PRODUCTION BANK PRODUCTION WATER WELLS A B C D Fig. d i f f e r e n t s u r f a c t a n t slugs. I n the presence o f o i l . presumably due t o a decrease i n the i n t e r f a c i a l v i s c o s i t y . two d i f f e r e n t polymer slugs. I t was also observed t h a t the apparent v i s c o s i t y was minimal a t the optimal s a l i n i t y f o r the f l o w o f microemulsions through porous media.

chemical processes such as polymer flooding. INJECTION S URFACTANT SLUG PRODUCTION ---. the importance o f enhanced o i l recovery t o increase t h e supply o f crude o i l i s obvious and various enhanced o i l recovery processes have been proposed and tested both on a l a b o r a t o r y scale and i n the f i e l d . For heavy o i l s . i f a s u r f a c t a n t s o l u t i o n i s i n j e c t e d t o lower the i n t e r f a c i a l tension o f the o i l ganglia fo i t s value o f 20-30 rm dynes/cm t o 10-3 dyneslcm. caustic f l o o d ing . It also serves as a raw m a t e r i a l f o r feed stocks i n chemical industry.--- - THICKENED FRESH WATER Fig. f o s s i l f u e l s o r crude o i l s p l a y an important r o l e i n providing the energy supply o f t h e world. t h e o i l t h a t remains i n the r e s e r v o i r i s believed t o be i n the form o f o i l ganglia trapped i n t h e pore s t r u c t u r e o f the rock as shown i n Figure 1A. Figures 1C and 1D schematically show the o i l bank approaching the production well and the subsequent breakthrough o f t h e d r i v e water. Figure 1 schematically shows a three-dimensional view o f a petroleum reservoir. 2 Schematic diagram o f the surfactant-polymer f l o o d i n g process. l i v i n g o f a s o c i e t y are i n t e r r e l a t e d .. Figure 2 schematically i l l u s t r a t e s a twodimensional view o f the surfactant-polymer f l o o d i n g process. thermal processes have been used e x t e n s i v e l y whereas f o r l i g h t o i l s . INTRODUCTION I t i s well recognized t h a t the energy consumption per c a p i t a and the standard of. m i s c i b l e f 1ood ing and s u r f act ant-pol ymer f 1oodi ng have a t t r a c t e d great i n t e r e s t . The s u r f a c t a n t slug i s followed by a polymer slug f o r a proper m o b i l i t y c o n t r o l o f the process. the o i l ganglia can be mobilized and can move through narrow necks o f t h e pores. The present paper focuses on the fundamental aspects o f the surfactant-polymer f l o o d i n g process and r e 1ated phenomena. Such mobilized o i l ganglia form an o i l bank as shown i n Figure 16. . However.. A t the end o f water-flooding. I n view o f the worldwide energy c r i sis.3 A. Among various sources o f energy. These o i l ganglia are entrapped due t o c a p i l l a r y forces. The major research f i n d i n g s i n the enhanced o i l recovery area have been reported i n recent l i t e r a t u r e and the symposia proceedings of various conferences during the l a s t decade (1-11).

. i n an excellent review a r t i c l e . the c a p i l l a r y number i s around 10-6 and t h i s number has t o be increased by 3 t o 4 orders o f magnitude f o r t e r t i a r y o i l recovery processes (14). 1973). 206. F i g u r e 3 shows such a c o r r e l a t i o n reported by Foster (13) between t h e c a p i l l a r y number and r e s i d u a l o i l i n porous media. . The c a p i l l a r y number represents the r a t i o o f viscous t o c a p i l l a r y forces uv/u+ where 11 and v are the v i s c o s i t y and v e l o c i t y of ! k e i i $ ? l i n g f l u i d . W.l o w i n t e r f a c i a l tension a t the o i l ganglia/surfactant formulation i n t e r f a c e . Under p r a c t i c a l r e s e r v o i r rm conditions. Feb. Taber (12) has summarized various emperical dimensionless numbers proposed by several i n v e s t i g a t o r s t o c o r r e l a t e t h e o i l displacement e f f i c i e n c y i n porous media. PERCENT PORE VOLUME Fig. the reduc i o n i n ' n t e r f a c i a l tension fo a high value of 20 o r 30 dyneslcm t o 1 0 . I t i s known t h a t h i g h i n t e r f a c i a l v i s c o s i t y r e s u l t s i n the formatin o f s t a b l e emulsion (15). Tech. J. For t h i s a very low i n t e r f a c i a l v i s c o s i t y i s d e s i r a b l e (Figure 5). CAPILLARY NUMBER AND CONCEPTUAL ASPECTS O THE PROCESS F Recently.i o r 10-4 dynes/cm o f f e r s such a p o s s i b i l i ty. Subsequently.. u i s the i n t e r f a c i a l tension and 4 i s t h e pore volume). p. 3 l Dependence o f r e s i d u a l o i l s a t u r a t i o n on C a p i l l a r y Number (Foster. I1 ~ 1 0 2 0 3 0 4 0 5 0 6 0 0 0 RESIDUAL OIL. Figure 4 schematically shows the r o l e o f u l t r a l o w i n t e r f a c i a l tension i n promoting the m o b i l i z a t i o n o f o i l ganglia i n porous media. Therefore.R. t h e displaced o i l ganglia must coalesce t o form an o i l bank. A t the end o f water flooding. the main f u n c t i o n o f t h e s u r f a c t a n t i s t o produce such an u l t r a .4 B. Pet.

5 Schematic diagran of the role of low interfacial viscosity in the surfactant-polymer flooding process.OOl DYNES/CM Fig. SURFACTANT Y " d SLUG COALESCENCE OF OIL GANGLIA WITH OIL BANK CAUSES FURTHER DISPLACEMENT OF OIL Fig.5 FOR THE MOVEMENT OF OIL THROUGH NARROW NECK OF PORES. i t has to be propagated through the porous medium without increasing the entrapment of o i l at the t r a i l i n g edge of the oil bank. A VERY LOW OIL / WATER INTERFACIAL TENSION IS DESIRABLE z . . 4 Schematic diagran of the role of low interfacial tension in the surf actant-pol ymer flooding process. the maintenance of ultralow interfacial tension at the o i l bank/surfactant/ slug interface i s essential for minimizing the entrapment of the oil i n the porous medium whereas the leading edge will coalesce with the o i l gang1 i a. As shown in Figure 6 . Once an o i l bank is formed i n the porous medium. SURFACTANT SLUG 4 DISPLACED OIL GANGLIA MUST COALESCE TO FORM A CONTINUOUS OIL BANK : FOR THIS A VERY LOW INTERFACIAL VISCOSITY IS DESIRABLE Fig. 6 Schematic diagran of the role of coalescence of o i l ganglia i n the surf act ant-pol ymer flooding process.

6 Figure 7 schematically i l l u s t r a t e s the movement o f the o i l bank. emulsions w i l l be formed a t the o i l bank/surfactant slug i n t e r face and a mixed surfactant-polymer zone w i l l occur a t t h e s u r f a c t a n t polymer s o l u t i o n i n t e r f a c e . 8 The r o l e o f w e t t a b i l i t y and contact angle on o i l displacement. INTERFACES t Since the f l o w i s through porous m e d i a .surfactant dispersions i n SURFACTANT SLUG OIL Fig. A l s o a v o i d t h e formation of high v i s c o s i t y structures i n the o i l .water .s t r u c t u r e s a t both these i n t e r f a c e s should be avoided i n order t o improve the e f f i c i e n c y o f t h e process. 7 Schematic i l l u s t r a t i o n o f t h e e f f e c t s o f dispersion and emulsif i c a t i o n between the various slugs during the surfactant-polymer f l o a d i n g process. The mass t r a n s f e r o f s u r f a c t a n t t o t h e o i l bank can i n f l u e n c e the magnitude o f i n t e r f a c i a l tension (16). surfactant s l u g and the m o b i l i t y c o n t r o l polymer slug i n the porous med iu m . Trushenski (17) has shown t h a t surfactant-polymer i n c o m p a t i b i l i t y leading t o a phase separation o f surfactant and polymer s t r i k i n g l y reduces t h e e f f i c i e n c y o f t h e process. Since the f l o w through the porous mediun causes dispersion o f these f l u i d s . High v i s c o s i t y . PROPER CHOICE OF SURFACTANT CAN CHANGE@TO@ Fig. t h e e f f e c t o f dispersion for emulsification should b e minimized a t a l l 3 interfaces. .

* .. L W SURFACTANT CONCENTRATION SYSTEMS O Figure 10 shows the i n t e r f a c i a l tension as a f u n c t i o n o f s u r f a c t a n t concentration i n a dodecane/brine system.*. Another parameter t h a t we have found (18.High Interfacial V k 4 i t y -Low Ebctrical Repulsion Between Oil Droplei a sand High Interfacial Tension Fig. . W e found t h a t a high surface charge d e n s i t y leads t o a lower i n t e r f a c i a l tension.Sand ...l o w i n t e r f a c i a l tension (IFT)... . The conceptual processes described i n Figures 3 t o 9 are supported by t h e r e s u l t s o f our studies described i n the following sections.7 Figure 8 schematically i l l u s t r a t e s the r o l e o f w e t t a b i l i t y o f s o l i d surface on the o i l ganglia. . The choice o f s u r f a c t a n t can i n f l u e n c e the w e t t a b i l i t y o f t h e rock surface t o o i l and b r i n e (12)..' . whereas a t high s u r f a c t a n t concentration systems (around 4 t o 8% s u r f a c t a n t concentration). . a middle phase microemulsion e x i s t s i n e q u i l i b r i u n w i t h excess o i l and brine. *.. A t low surfactant concentrations. .. lower i n t e r f a c i a l v i s c o s i t y and higher o i l recovery (Figure 9). . . High Surface Charge Densky -Low Interfacial Tenslon Low Interfacial Vicorlty High Electrkal Repulsion Between Oil Droplets (i Sand - '. . . 19) t h a t influences the i n t e r f a c i a l tension and i n t e r f a c i a l v i s c o s i t y and o i l recovery i s t h e surface charge a t the o i l ... Low Surface Charge Density .. * ' a .b r i n e as well as rock-brine i n t e r f a c e s . namely. . . ' I t i s evident t h a t there are two regions o f u l t r a . .. . *. o i l and b r i n e i n e q u i l i b r i u m w i t h each other. 9 Schematic diagran o f the r o l e o f surface charge i n the o i l d i s placement process.:. .. C.. The formation o f middle phase microemulsion and r e l a t e d phenomena w i l l be discussed i n section D. . . the system appears t o be a two-phase system.

1 Fig.01 I I I 1 I I I I 0. %I Effect o f surfactant concentration on the i n t e r f a c i a l tension of TRS 10-410 + IBA i n 1.005 0 0 .005 001 .0.0 0.5 00 ' .1 0.0005 000 .5% NaCl with dodecane.11 00 . . 10 TRS 10-410 CONCENTRATION ( wt.

we have show t h a t t h e u l t r a l o w IFT occurs when s u r f a c t a n t molecules migrate from t h e aqueous phase t o the o i l phase (19-21). 0 15 2D 25 NoCl CONCENTRATION (wt. The u l t r a low IFT occurred around a p a r t i t i o n c o e f f i c i e n t o f u n i t y i n t h i s system (19.TXTRS 10-80 + b r i n e + octane. 11 E f f e c t o f added e l e c t r o l y t e on i n t e r f a c i a l tension and surfact a n t p a r t i t i o n c o e f f i c i e n t o f t h e system O. W bee l i e v e t h a t a reasonable conclusion i s t h a t lowering o f i n t e r f a c i a l tension i s observed when m i c e l l e s leave the aqueous phase due t o coacervat i o n process (19-23). Since c m e r c i a l petroleum sulfonates i n v o l v e a d i s t r i b u t i o n o f molecular weights and isomeric s t r u c t u r e s we also Investigated t h e i n t e r f a c i a l tension using i s o m e r i c a l l y pure sulfonates.2% \ 5. 0. 0 --. I( SURFACTANT CONCENTRATION. Figure 11 shows t h e i n t e r f a c i a l t e n s i o n and t h e p a r t i t i o n c o e f f i c i e n t o f a s u r f a c t a n t i n an octane/brine system. However.56) Fig.W 0. and o i l chain length e f f e c t s were s i m i l a r f o r both these classes o f sur- .ZO).W 05 1 .9 For low s u r f a c t a n t concentration systems. Figure 12 shows the IFT behavior as a f u n c t i o n o f s a l i n i t y . o i l 'chain length and s u r f a c t a n t concentration using petroleum sulfonates (TRS 10-80 o r TRS 10-410 and an It i s evident t h a t both t h e s a l i n l t y i s o m e r i c a l l y pure s u r f a c t a n t UT-1). e a c $ 2 0 v) z W I- 2 d a W I - 1 ox I 0. i t should be emphasized t h a t since t h e p a r t i t i o n c o e f f i c i e n t changes a b r u p t l y i n t h i s region the exact value o f p a r t i t i o n c o e f f i c i e n t can vary s i g n i f i c a n t l y around u l t r a l o w IFT. \ \ .

However. 12 Schematic diagram o f the e f f e c t o f s a l t concentration. Figure 13 i s a generalized diagran showing t h e IFT. B d i r e c t analysis o f surfactant concentrations i n each phase. we found y (21) t h a t t h e s a l i n i t y a t which s u r f a c t a n t molecules leave t h e aqueous phase i s lower than the s a l i n i t y at which they enter t h e o i l phase. m a y p r e c i p i t a t e o r may form a coacervate phase below t h e aqueous phase o r i n between the aqueous and the o i l phase depending upon i t s d e n s i t y (21). t o represent the e l e c t r o l y t e concentrations a t which t h e surf a c t a n t concentration begins t o decrease i n t h e aqueous phase and begin e t o increase i n t h e o i l phase respectively. These conclusions were i n aggreement w i t h osmotic pressure. t h e s u r f a c t a n t e l e c t r o l y t e concentration. and CEC2. namely. Thus. For low surfactant concentration systems. CEC1. phase behavior and the two c r i t i c a l e l e c t r o l y t e concentrations f o r both pure and ctmwnerc i a 1 surfactants a t low as w e l l as high s u r f a c t a n t concentrations. factants. we also observed t h a t the u l t r a low IFT appears when the aqueous phase i s a t t h e apparent anc f o r the surfactant remaining i n t h e aqueous phase. o i l chain length and s u r f a c t a n t concentration on t h e i n t e r f a c i a l tension o f pure and impure a1 k y l benzene sulfonates. l i g h t s c a t t e r i n g and spectroscopic measurements on t h e e q u i l i b r a t e d aqueous phase (22). the e f f e c t o f surfact a n t concentration on IFT was d i f f e r e n t f o r commercial and i s o m e r i c a l l y pure surfactants.10 PETROLEUM SULFONATES: ISOMERICALLY PURE ALKYL BENZENE SULFONATES: Fig. . namely. t h e r e i s a s p e c i f i c s a l i n i t y and s p e c i f i c o i l chain length where we o b t a i n an u l t r a l o w IFT. we d e f i n e two c r i t i c a l e l e c t r o l y t e concentrations. W f u r t h e r observed t h a t t h e minimun i n t e r f a c i a l tension occurs i n the v i c i n i t y o f t h e f i r s t c r i t i c a l I n between CECl and CEC2.

11 I n low concentration systems. we observed t h a t the u l t r a l o w IFT o n l y occurs f o r o i l chain lengths below the c r i t i c a l o i l chain length (< C11). From our extensive studies on i n t e r f a c i a l tension and p a r t i t i o n i n g o f surfactant i n r e l a t i o n t o many parameters. phase behavior and i n t e r f a c i a l tension. Figure 14 shows the e f f e c t o f o i l chain length on CEC and CEC2 i n Aerosol OT/brine/oil systems. W propose t h a t a l l the o i l s below the c r i t i c a l o i l chain length are e able t o s o l u b i l i z e i n the m i c e l l e s whereas the o i l s having chain length above the c r i t i c a l o i l chain length are unable t o s o l u b i l i z e i n the m i c e l l a r s o l u t i o n . I t i s evident t h a t the EEC! i n creases w i t h o i l chain length u n t i l i t reaches the c r i t i c a l o i l chain length (C11) above which the value o f CECl remains constant. This suggestion i s i n agreement w i t h observation t h a t the volume o f the middle phase microemulsion increases l i n e a r l y w i t h the surfactant concentration and the s t r a i g h t l i n e passes through the o r i g i n (24). we have proposed the f o l l o w i n g 5 necessar y conditions t o achieve u l t r a l o w IFT's. On t h e other hand. NaCl Fig.r e l a t i o n s h i p shown i n Figure 13 i s v a l i d f o r both commercial and i s o m e r i c a l l y pure surfactants (21). I n t e r e s t i n g l y . whereas the i n t e r f a c i a l tension remains high for o i l s above t h e c r i t i c a l o i l chain length (21). CONCENTRATION Generalized d i a g r m o f the e f f e c t o f s a l t concentration on surf a c t a n t p a r t i t i o n i n g . Thus. It should be emphasized t h a t the general behavior and i n t e r . i t i s possible t h a t an extremely small volume o f middle phase may e x i s t between the o i l and b r i n e phases even though i t may not be v i s i b l e . . i t appears t h a t s o l u b i l i z a t j o n o f o i l w i t h i n the m i c e l l e s i s an important requirement f o r producing u l t r a l o w IFT. CEC2 continues t o increase w i t h the o i l chain length.

12 Fig.0% AOT/brine/oil . O/o AOT I BRINEI OIL DODECANE TETRADECANE A HEXADECANE 0 0 I I 1 2 NaCl CONCENTRATION 3 (Wt. . 14 E f f e c t of o i l chain length on the f i r s t and second c r i t i c a l e l e c t r o l y t e concentrations of Aerosol OT. 15 E f f e c t of o i l chain length on the i n t e r f a c i a l tension o f t h e systems 1. '/o) Fig.

E < 8 - $ t 0 i m 0) =! 7. Micelles i n the aqueous phase should be able t o s o l u b i l i z e o i l fo rm the hydrocarbon phase. o 6I W 0 p 0 W 5- n I - a E 1 4- 3 .4 SODIUM HYDROXIDE (NaOH) CONCENTRATION ( w t ./ . % ) . .13 The t o t a l surfactant concentration should be appreciably above the apparent anc i n the aqueous phase.2 0. The surfactant should be soluble i n both the aqueous and the hydrocarbon phase. a steep p a r t i t i o n coe f f i c i e n t curve f o r surfactant). There should be a large slope i n the surfactant p a r t i t i o n c o e f f i c i e n t curve i n the region o f the ultralow IFT.e.004 0. 16 A correlation between i n t e r f a c i a l tension and electrophoretic m o b i l i t y f o r crude oil-NaOH solutions.02 0. 3 0. The aqueous phase s a l i n i t y should be near the f i r s t c r i t i c a l electrol y t e concentration (CECI).005 2 4 Fig. - : I - =.002 0.0004 0. / SPONTANEOUS EMULSIFICATION A 2- I - 0. (i.04 0.

dyneskm Fig. Figure 17 schematically i l l u s t r a t e s 3 components of the interfacial tension.19. and 3) mutual solubilization of o i l and brine. . T h i s correlation was also observed for the effect of s a l i n i t y and surfactant concentration (19). The Klearfac AA-270 containing a phosphate group possesses two ionic oxygen atoms and hence can generate a high surface charge density a t the interface. W propose that by a d j u s t i n g any of e these variables one can influence the magnitude of interfacial tension. namely. 1 ) surface concentration of the surfactant. This presumably i s responsible for lowering the magnitude of IFT and broadening the s a l i n i t y range over which the ultralow IFT occurs for the mixed surfactant systems (27). 17 A schematic i l l u s t r a t i o n of the factors affecting the magnitude of the interfacial tension. e Using the conceptual approach shown i n Figure 17.14 Figure 16 shows the correlation of interfacial tension with electroe phoretic mobility in crude oil/caustic systems (18.25. 2) surface charge density. This suggests that the maximun in surface charge density coincides w i t h the m i n i m u m i n i n terfacial tension. w were able to broaden and lower the magnitude of interfacial tension as well as increase the s a l t tolerance limit of the surfactant formulation.26). Figure 18 shows the interfacial tension of a petroleum sulfonate TRS 10-410/n-octane/brine system when gradually the petroleum sulfonate i s replaced w i t h an ethoxylated phosphate e s t e r (Klearfac AA-270). W have observed for several crude o i l s that the u l t r a low IFT occurs in the r e gion where the electrophoretic mobility is maximum.

-- W 15 . 18 An i l l u s t r a t i o n o f the reduction and broadening o f the minimum i n i n t e r f a c i a l tension by the addition o f Klearfac AA-270. D ! 2 7 * . .. .v F i oww e ' N. N F ' 0 0 \ L ."' 5 N Y 2 ae a al k y I 7 0 0 0 0 O O Q O 9999 9 O N * W ++++ + II .- 0 c co I 0 * I * c I I N 0 1 I F i g .

The m i c r o s t r u c t u r e o f such b i r e f r i n g e n t formulations should be o f i n t e r e s t i n understanding the changes i n molecular associations o c c u r r i n g i n these systems. e . I t appears t h a t when t h e s a l i n i t y i s increased i n the s u r f a c t a n t formulation. Figures 19-21 i l l u s t r a t e the m i c r o s t r u c t u r e o f a b i r e f r i n g e n t surfactant formulation c o n s i s t i n g o f 5% TRS 10-410 + 3% isobutanol and 2% NaCl brine. t h e s u r f a c t a n t molecules form the m u l t i l a y e r s t r u c t u r e w h i l e keeping t h e i r p o l a r groups i n contact with b r i n e and form such c e l l s o r f o m l i k e s t a b l e s t r u c t u r e . 19 Freeze-fracture e l e c t r o n micrograph o f t h e a n i s o t r o p i c system 5% TRS 10-410 + 3% Isobutanol + 2% NaCl ( 8 5 5 0 1x). These e l e c t r o n micrographs c l e a r l y i n d i c a t e t h a t the b i r e f r i n g e n t formulations c o n s i s t o f bubbles f i l l e d w i t h ' b r i n e and separated f r a n each other by a t h i n s u r f a c t a n t membrane. HIGH SURFACTANT CONCENTRATION SYSTEMS AND THE OPTIMAL SALINITY Many s u r f a c t a n t formulations e x h i b i t extreme flow o r s t a t i c b i r e f r i n g e n c e i n a given s a l i n i t y range o r i n a given temperature range. The dimension o f each l a y e r i s close t o a surfact a n t b i l a y e r (approximately 70A). The freeze-fracture e l e c t r o n microscopic technique used t o o b t a i n these p i c t u r e s i s believed t o preserve the m i c r o s t r u c t u r e o f the sanples due t o the very r a p i d c o o l i n g r a t e (24).16 D. Therefore. Figure 21 c l e a r l y shows the s t r u c t u r e o f t h i s membrane c o n s i s t i n g o f several t h i n layers. Fig. Often these o p t i c a l l y a n i s o t r o p i c formulations e x h i b i t u l t r a l o w IFT w i t h o i l . W have c a l l e d these s t r u c t u r e s b i r e f r i n g e n t c e l l u l a r f l u i d s (24).

20 Freeze-fracture electron micrograph o f the above system at 18. Fig.OOOX. 21 Freeze-fracture electron micrograph of the above system at 30. OOOX .17 Fig. .

Fig. . middle phase and upper phase microemulsions. 23. 24).18 Figure 22 shows the s i m i l a r i t y between coacervation o f a m i c e l l a r s o l u t i o n i n the absence o f o i l and the formation o f a middle phase microemulsion i n t h e presence o f o i l . t h e same formulation i n the presence o f an equal volume o f dodecane shows the formation o f lower phase. The lower p a r t o f F i g u r e 22 shows t h e t r a n s i t i o n o f a b i r e f r i n g e n t s u r f a c t a n t formulation t o an i s o t r o p i c coacervate phase upon a d d i t i o n o f s a l t . Additional studies i n support o f these models have been reported elsewhere (21. the other hand. 22 A comparison o f coacervation i n aqueous s o l u t i o n with middle phase format i o n i n s u r f act a n t / o i 1/ b r i ne/al coho1 systems. W propose t h a t the middle phase microemulsion i s s i m i l a r t o t h e coacere vated phase containing some s o l u b i l i z e d o i l . (x.

(15-20%) the existence o f anamalous s t r u c t u r e which are n e i t h e r conventional water external o r o i l ext e r n a l microemulsions have been proposed t o a. I n very high s u r f a c t a n t concentration systems. t h e higher t h e s o l u b i l i z a t i o n o f oi. The compression o f the e l e c t r i c a l double l a y e r around m i c e l l e s w i l l occur. @ 0 oil swollen micelles (microdraplets of oil) reverse micelles (microdroplets of water 1 I 2 3 4 5 6 7 Fig. It c l e a r l y shows t h e d i s c r e t e spherical s t r u c t u r e s embedded i n a continuous aqueous phase consistent w i t h the mechanism proposed i n Figure 23. t h e lower i s t h e i n t e r f a c i a l tension w i t h respect t o these excess phases (28).o i l .c u i s not o n l y achieved by increasing the s a l i n i t y but i s also possible by changing any o f t h e other 8 variables. . 30). The s a l i n i t y a t which equal volumes o f o i l and b r i n e are s o l u b i l i z e d i n the middle phase microemulsion i s r e f e r r e d t o as optimal s a l i n i t y f o r the s u r f a c t a n t . the aggregation number o f the m i c e l l e s increases and the s o l u b i l i z a t i o n o f o i l w i t h i n m i c e l l e s increases. the solub i l i z a t i o n o f o i l i n t h i s middle phase increases whereas t h a t o f b r i n e decreases. As one increases the s a l i n i t y .f r a c t u r e e l e c t r o n micrograph o f a middle phase microemulsion formed i n the system e x t e n s i v e l y studied by Reed and Healy (28-30). It should be pointed out t h a t other i q v e s t i g a t o r s (40-47) have proposed the p o s s i b i l i t y o f bicontinuous s t r u c t u r e o r t h e coexistance o f o i l external and water external microemulsions i n the middle phase. Figure 24 shows the f r e e z e . t h e anc decreases. hence reducing the r e p u l s i v e forces between the m i c e l l e s .1 o r b r i n e i n the middle phase microemulsion. Figure 25 shows t h a t the t r a n s i t i o n l + m . I n general.b r i n e systems under given physical chemical conditions (29.ccount f o r some unusual prop e r t i e s o f these systems (43-46).19 Figure 23 schematically shows the mechanism o f formation o f middle phase microemulsions as s a l i n i t y i s increased. The magnitude o f i n t e r f a c i a l tension o f t h e middle phase w i t h o i l o r b r i n e depends upon the extent o f s o l u b i l i z a t i o n o f o i l and b r i n e i n the middle phase. 23 A schematic i l l u s t r a t i o n o f t h e l + m + u t r a n s i t i o n f o r t h e TRS 10-410/Isobutano1/0il /Brine System. Upon f u r t h e r increase i n s a l i n i t y . Thus t h e reduction i n the r e p u l s i v e forces and increase i n the a t t r a c t i v e forces between t h e m i c e l l e s w i l l b r i n g t h e m i c e l l e s c l o s e r and u l t i m a t e l y lead t o a separation o f a m i c e l l e r i c h phase forming the middle phase microemulsion.

20 Fig. Decreasing oil chain length Increasing alcohol concentration (C. 7.. 4. C. . 5.. U m m m Oil Brine The transition I - D m --c u occurs by: Parameter Increasing 1. 6. C. 8. 24 Freeze-fracture electron micrograph of the middle phase o f the Exxon system at the optimal salinity. Increasing Salinity 2. ) Decreasing temperature Increasing total surfactant concentration Increasing brine/oil ratio Increasing surfactant solution/oil ratio Increasing molecular weight of surfactanf Fig. 25 Schematic illustration of the factors influencing the l + m + u transit ion in surf act ant /oi 1 /bri ne/al coho1 systems. 3.

26 Effect of s a l i n i t y on the capillary number and t e r t i a r y oil recovery i n sand packs. SURFACTANT SLUG =O. Figure 27 shows the effect of o i l chain length on optimal s a l i n i t y of the TRS 10-410 + isobutanol systems (49) and the corresponding interfacial tension at the optimal s a l i n i t y for dach o i l chain length. NaCl w t 7 0 Fig. The importance of the optimal s a l i n i t y concept for enhanced oil recovery i s shown in the data i l l u s t r a t e d in Figure 26. I t i s evident that the o i l recovery is maximum at optimal s a l i n i t y f o r the systems reported here. I t was observed that as the o i l chain length increases.21 Thus. by choice of a suitable parameter.3 ft/day DOW PUSHER 700: SALINITY. The range over which the middle phase microemulsion exists also increases as the o i l . the optimal s a l i n i t y i n creases and the volume of the middle phase decreases. one can obtain the transition i n the structure of these microemulsions. In view of this observation. the surfactant formulation for a practical application should be designed such that the reservoir s a l i n i t y becomes the optimal s a l i n i t y under the given reservoir conditions. FLOODING RATEm2.V. An excellent correlation between the capillary number and oil recovery i s also evident from Figure 26 (48). A t optimal s a l i n i t y . the partition coefficient of surfactant i n the excess o i l and brine phases i s found to be near unity and the interfacial tension between the excess oil and excess brine is also m i n i m u n (19).OS P.

27 E f f e c t o f o i l chain length on t h e optimal s a l i n i t y and i n t e r f a c i a l tension a t t h e optimal s a l i n i t y . t h a t a r e l i k e l y t o occur i n the surfactant-polymer f l o o d i n g process. This i s a very useful c o r r e l a t i o n and eliminates t h e t i m e consuming and laborious procedure o f o b t a i n i n g t h e optimal s a l i n i t y i n t h e presence o f each alcohol. . This c o r r e l a t i o n suggests t h a t i f one determines t h e optimal s a l i n i t y i n the presence o f 2 o r 3 alcohols. F i g u r e 28 sumnarizes the data obtained by three research groups (49.22 F i g . 27 Fig. 53). such as the formation and coalescence o f drops as w e l l as t h e i r f l o w through porous media. Figure 28 shows t h e c o r r e l a t i o n o f optimal s a l i n i t y i n t h e presence o f various alcohols w i t h t h e i r s o l u b i l i t y i n brine. one can p r e d i c t the optimal s a l i n i t y i n t h e presence o f other alcohols from t h e i r b r i n e s o l u b i l i t y data. Most l i g h t o i l s o r t h e EACN f o r most l i g h t crude o i l s were found t o be between C7 and c11. It i s i n t e r e s t i n g t h a t the optimal s a l i n i t y o f a given o i l and surf a c t a n t formulation l i e s near t h e i n t e r s e c t i o n o f the b r i n e s o l u b i l i t y . the concept o f -Equivalent A1 kane Carbon Number (EACN) has been proposed t o c o r r e l a t e the i n t e r f a c i a l tension o f pure alkanes w i t h those o f the mixtures (50). TRANSIENT PROCESSES There are several t r a n s i e n t processes. chain length increases. Figure 29 shows the coalescence o r phase separation t i m e o f handshaken and sonicated macroemulsions as a f u n c t i o n o f s a l i n i t y .52. Many l i g h t crude o i l s have been simulated by the mixtures o f pure hydrocarbons (51). E. It should be pointed out t h a t from extensive studies on mixed a1 kanes.

Surfoclont Alcohol Brine 5 0 % TRS 10-410 30% Variable Dodecone NoCl 40% 0 7% Amoco A A .5 % Xylene Sulfonate 0. 5 % NoCl Crude Variable NaCl M/HAN Vorioble Wyomlng 0 I 1 Oil 90/10 lsopor I . 28 A c o r r e l a t i o n o f optimal s a l i n i t y i n t h e presence of various alcohols w i t h t h e i r s o l u b i l i t y i n b r i n e . SPE 6843 Puerto and Gale. SPE 6594 I0 Salter.o WOR Ref I0 Shah ond Hsleh.Sulfonate 1. SPE 5 8 1 4 BRINE SALINITY (wt % NoCl 1 BRINE SALINITY (wt% doc1 I BRINE SALINITY ( u t % No CI I F i g . N W .

(b1T. .X) F i g . 29 E f f e c t o f s a l i n i t y on t h e phase separation o r coalescence r a t e o f sonicated and hand-shaken emulsions.24 16C 12c 8 C E w l - E 4 c SONICATEO C 0 2 4 6 8 SODIUM CHLORIDE CONCENTRATION.

Figure 30 shows the pressure drop across a porous medium hhen emulsions prepared at various s a l i n i t i e s flow through i t . t L SONICATED EMULSION CONTAINING EQUAL VOLUME OF DODECANE AND 0 AQUEOUS PHASE : TRS 1 -410 + IBA (5:3W/W) + NaCl +WATER / EMULSION FLOW RATE QE (rnl/rnin) Fig. This also suggest that at the optimal s a l i n i t y the interfacial viscosity must be very low to promote the rapid coalescence. T h i s also suggests that the interfacial tension is an important factor influencing the pressure drop across porous media ( 5 4 ) . I t is evident that the minimum pressure drop occurs at and around the optimal s a l i n i t y of the surfactant formulation. 30 Effect of s a l i n i t y on the pressure drop-flow r a t e curves of soni cated emu1 sions. .25 I t i s obvious that minimal phase separation time or the f a s t e s t coalescence r a t e occurs at the optimal s a l i n i t y (54). The rapid coalescence could contribute significantly to the formation of an oil bank from the mobilized oil ganglia.

SYSTEM: SONICATED EMULSION CONTAINING 30 1 TRS 10-410 t IBA(5:3 W/W) t WATER + NaCl (x%) AND EQUAL VOLUME OF DODECANE 0 1 0 F i g . 31 I 2 I I 4 I I 6 I I NaCl CONCENTRATiON (WT.26 Figure 31 shows a very interesting and important correlation between the coalescence r a t e in enulsions and the apparent viscosity i n the flow through porous media. . %) A correlation between the apparent viscosity and coalescence r a t e of sonicated emulsions. The minimun apparent viscosity for the flow of emulsions in porous media coincides with minimum phase separation time a t the optimal s a l i n i t y . I t i s l i k e l y that a rapidly coalescing m u l sion will give a lower apparent viscosity for the flow i n porous media (54). This correlation between the phenomena occurring in porous media and outside the porous medium allows us to use coalescence measurements as a screening criterion for many surfactant formulations for their possible behavior i n porous media.

we have found t h a t t h e s u r f a c t a n t l o s s i n porous media i s also minimum a t t h e optimal s a l i h . yp I Oi I Recovery Efficiency Apporent viscosity (or AP) of emulsions in porous media Coolescence or phase-seporotion time of emulsions * VW vo v x v Surfoctont loss in Porous Media w Solubilirotion of Oil and Brine in m+ microemulsions lnterfociol tension Ymo OPTIMAL SALINITY I/ Fig.o i l . I n our p r e l i m i n a r y studies. the maximum i n o i l recovery at optimal s a l i n i t y might be a combined e f f e c t o f a l l these processes t a k i n g place a t the optimal s a l i n i t y . optimal s a l i n i t y of a mixed s u r f a c t a n t formulation c o n s i s t i n g of a petroleum sutfonate and ethoxylated s u l f o n a t e (EOR-200). i t y presumably due t o r e d u c t i o n i n the entrapment process f o r the surfactant phase. .b r i n e systems. 32 SALl NlTY A sumnary o f various phenomena occurring a t t h e optimal s a l i n i t y i n r e l a t i o n t o enhanced o i l recovery by surfactant-polymer flooding - . one would l i k e t o design approaches t o a l t e r the optimal s a l i Figure 33 shows the n i t y o f a given s u r f a c t a n t formulation (55-57). It i s evident t h a t the maximum i n o i l recovery e f f i c i e n c y c o r r e l a t e s w e l l with t r a n s i e n t and equilibrium p r o p e r t i e s o f s u r f a c t a n t . Therefore. Since optimal s a l i n i t y leads t o favorable conditions f o r optimal o i l recovery.27 Figure 32 sumnarizes a l l the phenomena occurring a t t h e optimal s a l i n i t y i n r e l a t i o n t o enhanced o i l recovery by surfactant-polymer flooding.

I n t e r e s t i n g l y .00/0 3. Increase i n the optimal s a l i n i t y o f surfactant formulation by a d d i t i o n o f EOR-200. As one replaces petroleum sulfonates w i t h the ethoxylated s u l f o n a t e t h e optimal s a l i n i t y increases and can reach as h i g h as 32% NaCl brine. Figure 34 shows t h e shape o f an o i l drop upon contacting a surfact a n t formulation c o n s i s t i n g o f 0. It i s e v i rm dent t h a t as s u r f a c t a n t molecules migrate fo t h e aqueous phase t o the i n t e r f a c e and subsequently t o the o i l phase the i n t e r f a c i a l tension decreases and the spherical drop g r a d u a l l y f l a t t e n s out. 5 6 ) . This f l a t t e n i n g . 33.0% 4 2 I 1 1 I I TRSIO-410 I €OR-200 5 '0 C 4 2 3 4 3 2 I SURFACTANT CONCENTRATION wt. the mixed s u r f a c t a n t formulations containing p e t r o l e m sulfonates and ethoxylated sulfonates o r alcohol are promising candidates f o r h i g h s a l i n i t y formulations ( 5 5 .% Fig. these formulations when e q u i l i b r a t e d with o i l produced middle phase microemulsions having very low i n t e r f a c i a l tension.05% TRS 10-80 i n 1% NaCl. Thus.28 I 1 I SURFACTANT FORMULATION: TRS 10-410 t EOR-200 t ISOBUTANOL 5.

An i l l u s t r a t i o n o f t h e drop f l a t t e n i n g phenomenon f o r a drop of octane i n an e q u i l i b r a t e d s o l u t r i o n o f 0. . IFT and the o i l recovery.48). there i s a good c o r r e l a t i o n between the f l a t t e n i n g time. time r e f l e c t s the r a t e a t which molecules accumulate a t the o i l .05% TRS 10-80 I n 1% NaCl .b r i n e i n t e r f a c e . The reduction i n t h e f l a t t e n i n g time leads t o favorable o i l recovery efficiency (16. As shown i n Table 1.29 Fig. 34.

8 mNlm. f l a t t e n i n g t i m e and o i l recovery e f f i c i e n c y i n the presence and absence o f alcohol (16). o r changes i n the aggregation process o f m i c e l l e s (60). 2 0 ' 2. W'dia x 7" long: Permeability= 3 darcy: flow rate: I n general. n-octane at 25OC SYSTEM 1. . 1 Table I. This include the chemical degradation (58. I n t e r e s t i n g l y .30 TABLE 1 IFT. (m IFT i m) FLATTENING TIME* OILRECOVERY+ (%OIP) (seconds) 00 Ii1.59).0267 w VI. a s u r f a c t a n t formulation f o r enhanced o i l recovery i n cludes a short chain alcohol. It i s evident t h a t the f l a t t e n i n g time decreases s t r i k i n g l y i n t h e presence o f alcohol suggesting t h a t t h e alcohol promotes the mass t r a n s f e r t o the i n t e r f a c e and a r a p i d r e d u c t i o n i n t h e magnitude o f the i n t e r f a c i a l tension. I T = 50. As shown i n Table 2 i t compares t h e many i n t e r f a c i a l properties. II. Several i n v e s t i gators have show t h a t the i n t e r f a c i a l tension changes w i t h t i m e (61). p. There are also time dependent changes i n the surface p r o p e r t i e s of a surfactant formulation. a O0 O 29 1 5 *Flattening time i s defined as the time required for the n-octane drop to gradually flatten out * *OctanelH@. lowering o f the i n t e r f a c i a l tension. r e d u c t i o n i n i n t e r f a c i a l v i s c o s i t y o r change i n s u r f a c t a n t p a r t i t i o n i n g and modifying the s o l u b i l i t y o f s u r f a c t a n t i n o i l o r b r i n e phase.3 f t /day.Y. EquilibratedOil/ Equilibrated Surfactant Solution EquilibratedOill Fresh Surfactant Solution 0. 7 + San@ckdimension9 1. C. 1963.and O i l Recovery Efficiency of 0. we have observed t h a t the presence o f alcohol has a much more s t r i k i n g effect on the f l a t t e n i n g t i m e o f an o i l drop i n the presence o f a surfact a n t formulation. Flattening Time. Chapter 1. N. The p o s s i b l e e f f e c t o f alcohol can be t h e change i n v i s c o s i t y .627 Surfactant Solution EquilibratedOill% a 121 NaCl 61-63 44-52 15-11 6600 480 900 240 83 V. W have also shown t h a t using several physical techniques t h a t molecular e association also changes w i t h t i m e leading t o t h e aging e f f e c t s o f the surfactant formulation (58). The aging processes may occur over a period o f months o r years. 052TRS 10-80 in I%NaCl vs. Fresh Oilll% NaCl ~50. Academic Press.731 Solution Fresh OillFresh 0.8* * FreshOillEquilibrated Surfactant 0. IV. Davies and F Rideal. "Interfacial Phenomenb'.

This Page Intentionally Left Blank .

Figure 35 shows the e f f e c t o f mixing surfactant and a polymer solut i o n i n the absence o f o i l . . SURFACTANT-POLYMER INCOMPATIBILITY Trushenski (17) has shown t h a t surfactant-polymer i n c o m p a t i b i l i t y can lead t o a considerable r e d u c t i o n i n the e f f i c i e n c y o f the process. 35. one a t low s a l i n i t y and the other a t h i g h s a l i n i t y separated by a metastable c o l l o i d a l dispersion. h e r e a s t h e separation o f m i c e l l e s i n region 1 i s induced by the presence o f p o lymer molecules. The surfactant-polymer i n c o m p a t i b i l i t y manifests i t s e l f as a phase separ a t i o n and a l t e r a t i o n o f the v i s c o s i t y o f t h e separated phases. the t r a n s i t i o n middle phase t o upper phase microemulsion i s not influenced a t a l l by t h e presence o f polymer. The entrapment o f the h i g h v i s c o s i t y phase w i l l e f f e c t i v e l y remove t h a t component fo the f l o o d i n g process. I t i s evident t h a t there are tho regions o f phase separation. The s u r f act ant -polymer i ncompat i i 1i y b t shows up s t r i k i n g l y i n t h e formation o f region 1 (62). W r e f e r t o the separation a t the lower s a l i n i t y as r e e gion 1 and those a t h i g h s a l i n i t y as region 2. However. W have f u r t h e r s h o w (62. E f f e c t o f a d d i t i o n o f polymer on t h e phase behavior o f aqueous s u r f actant solutions.r i c h phase i n r e g i o n 2 i s s i m i l a r t o t h a t i n coacervation process. The mixing o f the surfactant and polymer rm i n the porous medium occurs due t o both dispersion e f f e c t s as w e l l as excluded volume e f f e c t s f o r the f l o w o f polymer molecules i n porous media. Fig.63) t h a t t h e optimal s a l i n i t y i s n o t e s i g n i f i c a n t l y influenced by the presence o f polymer i n t h e o i l l b r i n e l s u r f a c t a n t l a l c o h o l system. The separation o f a surf a c t a n t .32 F. The a d d i t i o n o f polymer t o an oillbrinelsurfactantlalcohol system shows t h a t the formation o f middle phase microemulsion i s promoted by the presence o f polymer (Figure 36).

8 - 0.0 - 5% W/V TRS 10-410 Polymer Concentration 3% w/v IBA O i l : n-dodecane (MOR=l .0 I I I I 1 I I Fig. 36.4 0.0 0. Effect of polymer concentration on the s a l i n i t y range f o r formation of middle-phase microemulsion.4 0.6 0. .2 - 2500 ppm 0. Figure 37 shows the schematic explanation of the surfactant polymer incompati b i 1 it y and concomittant phase separation.0 1500 ppm 1.2 0.6 0.O) 0.33 1.8 0.

This Page Intentionally Left Blank .

This Page Intentionally Left Blank .

1% NaCl and t h e r e s e r v o i r b r i n e was 3% NaCl p l u s 1% calcium c h l o r i d e . c t OPTIMAL SALINITY SALINITY. the s a l i n i t y o f the p r e f l u s h water i s made higher while t h e s a l i n i t y o f the polyner s o l u t i o n i s made lower than the optimal s a l i n i t y o f the s u r f a c t a n t formulation. With t h i s unique design h i g h o i l recovery and h i g h s u r f a c t a n t recovery can be obtained f o r soluble o i l f l o o d i n g i n sandpacks.36 surfactant slug i n porous media i s l a r g e l y determined by the s a l i n i t y o f t h e polymer s o l u t i o n (65). . The proposed s a l i n i t y shock design o f m o b i l i t y polymer s o l u t i o n s employs two slugs o f polymer s o l u t i o n s i n which t h e f i r s t polymer slug i s a t the optimal s a l i n i t y o f the preceeding s u r f a c t a n t f o r mulation and the second polymer slug i s at a much lower s a l i n i t y . With t h i s design. The optimal s a l i n i t y f o r t h e s u r f a c t a n t f o r m u l a t i o n used was 2. i t has been shown (66-69) t h a t a c o n t r a s t s a l i n i t y design o f t h e p r e f l u s h micellar-polymer f l o o d i n g process may produce a b e t t e r o i l recovery than t h a t obtained from a constant s a l i n i t y process. For b e t t e r m o b i l i t y c o n t r o l and minimal surfactant loss a two-slug design o f a s u r f a c t a n t formulation was employed (23). In t h i s design. the f i r s t surfactant s l u q has an optimal s a l i n i t y close t o the connate water s a l i n i t y and the second s u r f a c t a n t slug has a much lower optimal s a l i n i t y . while t h e polymer consumption can be g r e a t l y reduced. In t h e c o n t r a s t s a l i n i t y design. %NaCI Schematic representation o f the g r a d e d . 39. The polymer s o l u t i o n s a l i n i t y i s made equal t o the optimal s a l i n i t y o f the second s u r f a c t a n t slug. The optimal s a l i n i t y concept i s f u r t h e r extended t o include t h e e f f e c t o f m o b i l i t y c o n t r o l and s u r f a c t a n t dispersion and entrappment i n porous media (65). high o i l recovery i n berea cores can be obtained even i n the presence of h i g h s a l i n i t y (6% NaCl + 1% calcium c h l o r i d e ) connate water. For aqueous micellar-polymer f l o o d i n g w i t h crude o i l i n Berea cores. INJECTION PRODUCTION FLOW c n A 21 0 Fig. The r a t i o n a l e o f t h i s design i s t h a t an opt i m a l s a l i n i t y p r o f i l e can be established i n t h e v i c i n i t y o f the surfact a n t slug upon mixing o f t h e i n j e c t e d f l u i d s i n t h e porous medium. By the use o f two polymer slugs w were able t o o b t a i n i n e berea cores 88% t e r t i a r y o i l recovery and 48% s u r f a c t a n t recovery.s a l i n i t y design o f polymer b u f f e r s o l u t i o n f o r enhanced o i l recovery. Figure 40 schematically shows our r e s u l t s obtained using t h e s a l i n i ty shock design.

37

1

CHASE WATER

~

~

~

a

~

L

u

G

'IiASE

.prrRI

POLYMER SLUG 0.05% NaCl

42 %

25 %

88 x

48 x

Fig. 40.

The e f f e c t o f s a l i n i t y shock o f polymer buffer s o l u t i o n an o i l displacement e f f i c i e n c y and surfactant loss.

I t i s hoped t h a t the experimental r e s u l t s presented i n t h i s paper c o n t r i b u t e i n a small way t o increasing our understanding of phenomena occurring i n porous media. It should be enphasized t h a t r e s u l t s we have obtained using laboratory scale experiments are n e i t h e r conducted nor intended t o be extrapolated t o the o i l f i e l d processes. It i s recognized t h a t the processes occurlng i n petroleun r e s e r v o i r s are f a r more complex than those t h a t we can design and c o n t r o l using a laboratory setup.

ACKNOWLEDGEMENTS The author wishes t o express h i s sincere thanks and appreciation t o the National Science Foundation RANN, ERDA and t h e Department of Energy (Grant N DE-AC1979BC10075) and the consortium o f the f o l l o w i n g Induso : t r i a l Associates f o r t h e i r generous support o f t h e U n i v e r s i t y o f F l o r i d a Enhanced O i l Recovery Research Program during the past seven years: 1) A l b e r t a Research Council, Canada, 2) Pmerican Cyananid Co., 3 ) Ammo Production Co., 4) A t l a n t i c - R i c h f i e l d Co., 5 ) BASF-Wyandotte Co., 6 ) B r i t i s h Petroleum Co., England, 7) Calgon Corp., 8) C i t i e s Service O i l Co., 9) Continental O i l Co., 10) E t h y l Corp., 11) Exxon Production Research Co., 12) Getty O i l Co., 13) Gulf Research and Development Co., 14) Marathon O i l Co., 15) Mobil Research and Development Co., 16) Nalco Chemical Co., 17) P h i l l i p s Petroleum Co., 18) Shell Development Co., 19) Standard O i l o f Ohio Co., 20) Stepan Chemical Co., 21) Sun O i l Chemical Co. 22) Texaco, Inc., 23) Union Carbide Corp., 24) Union O i l Co., 25) Westvaco, Inc., 26) Witco Chemical Co., and the U n i v e r s i t y o f Florida. He also wishes t o convey h i s sincere thanks t o h i s Colleagues i n Chemical Engineering, Petroleum Engineering and I n s t i t u t e f o r Energy Studies o f Stanf o r d U n i v e r s i t y f o r t h e i r c o l l a b o r a t i o n during h i s s t a y a t Stanford University.

-

This Page Intentionally Left Blank

This Page Intentionally Left Blank

40

38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49.

Chou, S. I.and Shah, D.O., ( 1981 )

.

J. C o l l o i d I n t e r f a c e Sci.,
J. C o l l o i d I n t e r f a c e Sci

80(1), 49 311

Chou, S. I. and Shah, D.O., (1981). Scriven, L.E., Nature,

., 80(2),

263,

123 (1976).

Scriven, L.E., i n "Mice1 1i z a t i o n , Sol u b i l i z a t i o n and Microemulsions" ed. K.L. M i t t a l , Vol. 11, pp. 877, Plenum Press, N.Y. (1977). Ranachandran, C., 1561 (1980). Vijayan, S. and Shah, D.O., and Fort, T.

,

J. Phys. Chem.,

84,

Hwan, R., M i l l e r , C.A. 68, 221 (1979).

-

Jr., J. C o l l o i d I n t e r f a c e Sci.,

Shinoda, K. J. C o l l o i d I n t e r f a c e Sci., Shinoda, K. and Saito, H.,

24, 4

(1967).

J. C o l l o i d I n t e r f a c e Sci.,

g,70

(1968).

. Friberg, S., Lapczynska, I and Gilberg, G., Sci., 56, 19 (1976).

J. C o l l o i d I n t e r f a c e

M i l l e r , C.A., Hwan,.R., Benton, WJ and Fort, T. Jr., J. C o l l o i d .. I n t e r f a c e Sci , 61 (3), 554 ( 1977).

.

Chiang, M.Y.,

Ph.D.

Dissertation, l k r i v e r s i t y o f F i o r i d a (1978).

"The E f f e c t o f Chain Length o f O i l and Hsieh, W.C. and Shah, D.O., Alcohol as w e l l as Surfactant t o Alcohol R a t i o on t h e S o l u b i l i z a tion, Phase Behavior and I n t e r f a c i a l Tension o f Oil/Brine/Surfactant/Alcohol Systems",SPE 6594, presented a t t h e 1977 SPE-AIME I n t l . Symposiun on O i l f i e l d and Geothermal Chemistry, LaJolla, CA, June 27-28, 1977. Morgan, J.C., Schechter, R.S. and Wade, W.H., i n "Improved O i l Recovery by Surfactant and Polymer Flooding", D.O. Shah and R.S. Schechter, eds., Acad. Press, Inc., N.Y. (1977). Cash, R.L., Cayias, J.L., Fournier, G., Jacobson, J.K., Schares, T., Schechter, R.S. and Wade, W.H., "Modeling Crude O i l s f o r Low I n t g r f a c i a l Tension", SPE 5813, presented a t t h e SPE Symposium on I m proved O i l Recovery, Tulsa, OK, March 22-24, 1979. Satter, S.J., "The I n f l u e n c e o f Type and Amount o f Alcohol on Surfact a n t - O i l - B r i n e Phase Behavior and Properties," SPE 6843, presented a t the 52nd Annual F a l l Conference and E x h i b i t i o n o f SPE-AIME, Denver Co., Oct. 9-12, 1977. "Estimation o f O p t i m a l S a l i n i t y and SolPuerto, M.C. and Gale, W.W., u b i l i z a t i o n Parameters f o r A l k y l Orthoxylene Sulfonate Mixtures", SPE 5814, presented a t the SPE Improved O i l Recovery Symposium, Tulsa, OK March 27-24, 1976. Vijayan, S., Ramachandran, C., Doshi, H. and D.O. Shah, i n "Surface Phenomena i n Enhanced O i l Recovery", D.O. Shah ed., pp. 327-376, Plenum Publishing Co., N.Y. ( i n press).

50.

51.

52.

53.

54.

41
55. 56. 57. 58. 59. 60. Bansal, V.K. (1978). Bansal, V.K. Bansal, V.K. (1978). Vijayan, S., SOC., Vijayan, S., SOC., and Shah, D.O., and Shah, D.O., and Shah, D.O., Ramachandran, C., (1981).
J. C o l l o i d I n t e r f a c e Sci.,

650, 451
55 (3), 367

SPE J.,

167 (June, 1978).

J. h. O i l Chemists SOC.,

580,566

and Shah, D.O.,

J. h . O i l Chemists J. An. O i l Chemists

580,746

Ramachandran, C. and Shah, D.O., (1981).

Vijayan, S., Ramachandran, C. and Doshi, H., " U n i v e r s i t y of F l o r i d a Research on Chemical O i l Recovery Systems Semi-Annual Report", pp. Bl l -857, June, 1978. Cash, R.L., Cayias, J.L., Hayes, M., M c A l l i s t e r , D.J. %hares, J. Pet. Tech., 985 (Sept. 1976). Wade, W.H., T. and

61. 62. 63. 64.

i n " U n i v e r s i t y o f F l o r i d a Research on Surfactant-Polymer Desai, N.N.; O i l Recovery Systems-Annual Report", pp. 127-149, Dec. 1979. Desai, N.N., i n " U n i v e r s i t y o f F l o r i d a Research on Surfactant-Polymer O i l Recovery Systems-Annual Report", pp. 135-1-48, Dec. 1980. Hesselink, F. Th. and Faber, M.J., i n "Surface Phenomena i n Enhanced O i l Recovery," D.O. Shah, ed., pp. 861-869, Plenum Publishing Co., N.Y. ( i n press). Chou, S . I . and Shah, D.O., covery", D.O. Shah, ed., ( i n press) Paul, G.W. Gupta, S.P. and Froning, H.R., i n "Surface Phenomena i n Enhanced O i l Repp. 843-860, Plenum Publishing Co., N.Y.
J. Pet. Tech.,

65.

66. 67. 68.

25, 957

(1973).

and Trushenski, S.P..

SPE J.,

2, 116

(1979)i

Nelson, R.C., "The S a l i n i t y Requirement Diagram-A Useful Tool i n Chemical Flooding Research Development1', SPE 8824, presented a t t h e SPE Improved O i l Recovery Symposium, Tulsa, OK, A p r i l 20-23, 1980. Hirasaki, G.J., Van Dmselaar, H.R. and Nelson, R.C., "Evaluation o f . t h e S a l i n i t y Gradient Concept i n Surfactant Flooding", SPE 8825, presented a t t h e SPE Improved O i l Recovery Symposium, Tulsa, OK, A p r i l 20-23, 1980.

69.

This Page Intentionally Left Blank

This Page Intentionally Left Blank

44

The c h a r a c t e r i s t i c d a t a p e r t a i n i n g t o a s p e c i f i c g r o u p of res e r v o i r s must be e v a l u a t e d i n o r d e r t o p r o v i d e a r e p r e s e n t a t i v e s u r v e y o f t h e boundary c o n d i t i o n s r e q u i r e d f o r b o t h t h e r a p i d t e s t methods and t h e p r e l i m i n a r y f l o o d i n g e x p e r i m e n t s . C o n c l u s i o n s w i t h t h e most g e n e r a l a p p l i c a b i l i t y : S i n c e t h e r e s h o u l d be n o r e s t r i c t i o n t o one o i l r e s e r v o i r o n l y , t h e r e s u l t s s h o u l d be of t h e most g e n e r a l p o s s i b l e v a l i d i t y , and s u r f a c t a n t s o l u t i o n s w i t h a b r o a d r a n g e of a p p l i c a t i o n s h o u l d be s o u g h t . Hence, maximal demands s h o u l d be imposed on t h e f l o o d i n g media, i n o r d e r t o e n s u r e t e s t r e s u l t s which a r e a p p l i c a b l e t o a s many o i l r e s e r v o i r s a s p o s s i b l e . TEST PROGRAM/STANDARDIZATION Reproducible t e s t c o n d i t i o n s a r e always r e q u i r e d f o r i n v e s t i g a t i n g s u r f a c t a n t s , i n o r d e r t h a t v a r i o u s p r o d u c t s be a p p r a i s e d and compared. Such c o n d i t i o n s c a n be f u l f i l l e d o n l y by model s y s t e m s , s i n c e t h e p r o p e r t i e s of r e a l s y s t e m s ( r e s e r v o i r w a t e r , c r u d e o i l , reservoir rock) are usually subject t o variations. Hence a t e s t program r e q u i r e s t h e d e s i g n i n g of model s y s t e m s , i n which a s many p a r a m e t e r s of r e a l s y s t e m s a s p o s s i b l e a r e c o n s i d e red. S i n c e s u r f a c t a n t s a r e generally d i s s o l v e d i n w a t e r , t h e t o t a l s a l i n i t y and t h e c o m p o s i t i o n of r e s e r v o i r b r i n e a r e of u t m o s t i m p o r t a n c e f o r t h e s e l e c t i o n of s u i t a b l e s u r f a c t a n t s . Based on s e v e r a l hundred c h e m i c a l a n a l y s e s o f . w a t e r s a m p l e s from German o i l r e s e r v o i r s , a c l a s s i f i c a t i o n i n t o t h r e e b r i n e c a t e g o ries was p o s s i b l e : 1 t y p e AM ( l o w s a l i n i t y ) TDS < l o g.11 t y p e B ( i n t e r m e d i a t e s a l i n i t y ) 1 0 .g-l-’C TDS (165 g.1M t y p e CM ( h i g h l y s a l i n e ) TDS > 1 6 5 g.1-1

-

Observed s i g n i f i c a n t c h a r a c t e r i s t i c s of h i g h l y s a l i n e r e s e r v o i r brines are that many s a l t s o c c u r i n t h e d i s s o l v e d s t a t e a t c o n c e n t r a t i o n s exceeding t h e i r usual s o l u b i l i t y products 2+ a l l w a t e r s a m p l e s c o n t a i n heavy m e t a l i o n s , s u c h a s Fe t h e pH-values of b r i n e s BM and CM l i e i n a r e l a t i v e l y a c i d i c range (3,0-6,5) a l l b r i n e s show c o m p a r a t i v e l y h i g h s u l f a t e c o n t e n t s a l l b r i n e s c o n t a i n large q u a n t i t i e s of Ca2+ and Mg2+ i o n s and lower c o n c e n t r a t i o n s of Sr2+ and Ba2+.

-

T r a c e e l e m e n t s 0 1 0 mg.1-l) were n o t c o n s i d e r e d . A t r e s e r v o i r p r e s s u r e bgtween 50 and 100 b a r and r e s e r v o i r t e m p e r a t u r e s between 4 0 and 80 C a b o u t 4 g of C 0 2 d i s s o l v e s i n 100 g of w a t e r . However, above 6 b a r t h e pH-value of w a t e r c o n t a i n i n g CO a l r e a d y t e n d s t o ward a c o n s t a n t v a l u e of 3 , 3 / 1 5 / , which p r o b a b l y a l s o d o m i n a t e s i n most r e s e r v o i r b r i n e s of t y p e CM. For t h e s u r f a c t a n t i n v e s t i g a t i o n s a s t a t i s t i c a l c o m p o s i t i o n was a s c e r t a i n e d f o r a h i g h l y s a l i n e model r e s e r v o i r w a t e r CM ( t a b l e 1 ) . T h i s s t a n d a r d i z e d b r i n e CM was employed f e r a l l s u b s e q u e n t t e s t s , unless otherwise indicated. The p r i m a r y s c r e e n i n g c r i t e r i a of s u r f a c t a n t s f o r EOR p r o c e s s e s i n h i g h l y s a l i n e s y s t e m s may be l i s t e d a s f o l l o w s / l o / : s o l u b i l i t y i n r e s e r v o i r b r i n e (TDS >165 g . 1 - l ) long-term s t a b i l i t y i n t h e t e m p e r a t u r e r a n g e of 30-80°C low i n t e r f a c i a l t e n s i o n s i n t h e s y s t e m b r i n e / c ? u d e o i l (v< 1 mN.m-l)

-

45 Table 1: Composition of synthetic reservoir brine CM Salt NaCl CaC12 MgC12 KC1 KBr KJ LiCl NH4Cl SrC12 BaC12 NaHC03 Na2S04 Concentrations in mg.1-l

. 6H20

. 6H20

165 000 49 349 ( 2 5 O O O ) * 12 810 ( 6 O O O ) * 750 400 20 100 350 1 681 ( 1 O O O ) * 58 ( SO)* 650 680 ( 523 ( 366 ( 300)*
250)*
200)*

. 6H20 . 2H20
. 10H20 .

C02 injected for at least 1 h NaCQ3 4H20 FeS04 9H20

.

TDS: 200 070 mg.1-'
*The concentration in parentheses refers to the quantity without water of crystallization Based on these fundamental requirements a standard test program for surfactants was developed (see fig. 1).

.solubility stobility wrfoctont in brine CM

interfociol tension Y e 1 mN.rn-1 ogoinst oil

'T

,- nwhthenic
L poroninic
'1:

30 C ' 60 'C

80 ' C Figure 1: Test program for water-soluble surfactants

swfoctont water soluble

1: 3a5 Y. 1.0

Besides the aforementioned criteria suitable surfactants should show further, - low adsorption on reservoir rock - favourable partition coefficients and. broad range of application.

-

'

This Page Intentionally Left Blank

47

T a b l e 2 : Commercially a v a i l a b l e s u r f a c t a n t g r o u p s
Structure Des isnation Na-salts of fatty acids Alkylbenzenesulfonate
+

Ionic surfactants
R-CH,-COONa R-C,H,-S0,Na
:)CH-SO,N~
R -CH,--CH=CH-CH,-SO,Na

R'-CH,-CH-(CH,)n-CH,-SO,Na
OH
R-CH-C~' $O,Na'OCH' R-CH,-0-S0,Na
R; R,CH -O-(C,H,O),-SO,Nn

Alkane sulfonate a-olefinsulfonate Hydroxyalkanesulfonate as byproduct

a-sulfa fatty acid ester
Fatty alcohol sulfate Fatty alcohol ether sulfate* Fatty alcohol ethoxylate acetate** (luaternary ammonium salts

R -CH,-O-(C,H,O),

-CH,COONa

R',

+ /R'

CP

R'/ N\R,

N o n I o n i c Surfadants
R'

H

1. R=C,,, bei R'=H

R y - ~ - t ~ , ~ , ~ ) , - ~

n=3-15 2. R+R=C,,,, n=3-12
R=C,,,

Primary or secondary alcohol ethoxylates Alkylphenol ethoxylates Amine oxides

R-C,H.-O-(C,H.O)n-H YH, CA-Y+C' CH,

n=7-10

h p l m l y t i c Surfactants
R'

Sulfobetains
R'

R' - m N -CH,-COO~ R'

Betains

* Fatty alcohol polyethylene glycol ether sulfate, Na-salt ** Fatty alcohol polyethylene glycol ether carboxymethylate,Na-salt

-

-

-

t y p e of c r u d e ( p a r a f f i n i c , n a p h t h e n i c , a r o m a t i c o r mixed t y p e ) c o l l o i d a l c h e m i s t r y of c r u d e ( c o n t e n t of a s p h a l t e n e s , r e s i n s , etc.) a d s o r p t i o n phenomena ( c o m p o s i t i o n of r e s e r v o i r r o c k ) c h a r a c t e r i s t i c s of r e s e r v o i r e n v i r o n m e n t (pH, t e m p e r a t u r e , wetting conditions, s a l i n i t y ) d i f f u s i o n phenomena ( r a p i d d i f f u s i o n t o t h e O / W - i n t e r f a c e ) .

The i m p o r t a n c e of t h e i n d i v i d u a l p a r a m e t e r s c a n v a r y g r e a t l y dep e n d i n g on t h e c o n d i t i o n s of a p p l i c a t i o n , and c a n n o t be g e n e r a l i z e d . For t h i s r e a s o n , two complex p a r a m e t e r s w i l l be d i s c u s s e d i n detai1. DIFFUSION PHENOMENA
I t i s known from s t u d i e s on O/W s y s t e m s t h a t t h e d i f f u s i o n i s dep e n d e n t , among o t h e r s , on t h e s t r u c t u r e of s u r f a c t a n t s /l/. Y e t , t h e d i f f u s i o n c o e f f i c i e n t of s u r f a c t a n t s i t s e l f i s of d e c i s i v e imvortance. I n g e n e r a l , t h e following r e l a t i o n s h i p s apply:

48

-

t h e d i f f u s i o n c o e f f i c i e n t decreases w i t h i n c r e a s i n g d e g r e e of alkoxylation the d i f f u s i o n c o e f f i c i e n t increases with increasing concentrat i o n of s u r f a c t a n t ( u p t o c.m.c.) t h e d i f f u s i o n c o e f f i c i e n t i s d i r e c t l y p r o p o r t i o n a l t o temperature t h e d i f f u s i o n c o e f f i c i e n t f o r d i s s o l v e d s u r f a c t a n t s is i n v e r s e l v p r o p o r t i o n a l t o t h e v i s c o s i t y of t h e s o l v e n t branched b l o c k copolymers d i f f u s e more r e a d i l y t h a n l o n g - c h a i n l i n e a r types.

The s t r u c t u r e of t h e p o l y e t h e r c h a i n s of s y n t h e t i c s u r f a c t a n t s c a n a l s o be of i m p o r t a n c e f o r d i f f u s i o n p r o c e s s e s . I t is w e l l known t h a t p o l y e t h e r c h a i n s , depending on t h e d e g r e e of a l k o x y l a t i o n , c a n e x i s t i n t h e s o - c a l l e d z i g - z a g form o r i n t h e meander form ( s e e fig. 3 /2/).

Zig-zag form

Meander form

- F i g u r e 3 : Shapes of p o l y e t h e r c h a i n s / 2 / With i n c r e a s i n g EO number, t h e w i d t h / l e n g t h c o e f f i c i e n t of t h e noni o n i c s i n c r e a s e s , and d i f f u s i o n c o e f f i c i e n t t h u s d e c r e a s e s . By b l o c k i n g t h e p o l y e t h e r oxygen f o r h y d r a t i o n a s a r e s u l t of 0 4 H 2 d i p o l e f o r c e s , a change c a n a l s o o c c u r i n t h e c l o u d p o i n t s , t h e c r i t i c a l micelle f o r m a t i o n c o n c e n t r a t i o n (c.m.c.), and t h u s t h e i n t e r f a c i a l a c t i v i t y o r s o l u b i l i t y behavior / 3 / . ADSORPTION For t h e q u e s t i o n of a d s o r p t i o n phenomena a s a f u n c t i o n of s u r f a c t a n t s t r u c t u r e o r r e s e r v o i r r o c k , numerous f i n d i n g s a r e of importance /l, 9,12, l 4 / .
GENERAL C R I T E R I A Swf aetants

-

Amphiphatic s u r f a c t a n t s a r e r e a d i l y a d s o r b e d on h y d r o p h o b i c r o c k s u r f a c e s , d e p e n d i n g on t h e i r s t r u c t u r e The g r e a t e r t h e s o l u b i l i t y of a s u r f a c t a n t , t h e s m a l l e r i s i t s a d s o r p t i o n ( g r e a t e s t a d s o r p t i o n of s u r f a c t a n t o c c u r s i n h i g h s a l i n i t y w a t e r b e c a u s e of diminished s o l u b i l i t y ) With i n c r e a s i n g t e m p e r a t u r e and v i s c o s i t y of t h e s o l v e n t a d s o r p tion decreases With i n c r e a s i n g s u r f a c t a n t c o n c e n t r a t i o n a d s o r p t i o n i n c r e a s e s Low t o t a l a d s o r p t i o n b u t h i g h r a t e of a d s o r p t i o n u p t o t h e s a turation concentration

AIM:

R e s e r v o i r system

-

-

Hydrophilic e a s i l y water-wettable rocks: quartz, c l a y Hydrophobic, p o o r l y w a t e r - w e t t a b l e r o c k s : c a r b o n a t e s

49 SPECIAL CRITERIA Surf a c t a n t s

-

-

T o t a l a d s o r p t i o n d e c r e a s e s w i t h i n c r e a s i n g m o l e c u l a r mass of s u r f a c t a n t ( t h e t o t a l a r e a a c c e s s i b l e t o a d s o r p t i o n becomes smaller) Nonionic s u r f a c t a n t s / 9 / a r e adsorbed mostly i n unimolecular l a y e r s , a d s o r p t i o n d e c r e a s e s w i t h increasing EO d e g r e e , b u t a d s o r p t i o n i n c r e a s e s w i t h i n c r e a s i n g l e n g t h of t h e h y d r o c a r b o n c h a i n ; d e r i v a t i v e s w i t h an a l i p h a t i c h y d r o c a r b o n c h a i n a r e more s t r o n g l y a d s o r b e d t h a n d e r i v a t i v e s w i t h an a r o m a t i c h y d r o c a r b o n c h a i n I o n i c s u r f a c t a n t s a r e a d s o r b e d f o r t h e most p a r t i n polymolecul a r l a y e r s ( c a t i o n i c s : a b o u t 250 l a y e r s ) . Limiting concentration: s y n t h e t i c s u r f a c t a n t s (0.05-0.07 %) < < n a t u r a l s u r f a c t a n t s (0.25 %).

Reservoir rock

-

C l a y : a d s o r p t i o n of u n s a t u r a t e d h y d r o c a r b o n s ( i n p a r t p o l y m e r i z e d ) > a r o m a t i c s > n a p h t h e n a t e s > a l k a n e s . C a t i o n i c s >>nonionics > anionics. S i l i c a t e s : s l i g h t a d s o r p t i o n of n o n i o n i c s ( o i l - w e t t e d > w a t e r w e t t e d ) , a d s o r p t i o n i n c r e a s e s with temperature; s t r o n g adsorpt i o n of c a t i o n i c s on q u a r t z ( l o w e r e d by a d d i t i o n of n o n i o n i c s ) .

The i n c r e a s e d a d s o r p t i o n of c a t i o n i c s , i n d e p e n d e n t of t h e reserv o i r r o c k i s t h s c l e a r l y e v i d e n t . A s a g u i d e , v a l u e s of a b o u t l o d 4 mg/cm c a n be g i v e n f o r t h e a d m i s s i b l e a d s o r p t i o n on 0,s quartz surfaces.

.

Y

PHYSICOCHEMICAL P R O P E R T I E S OF SURFACTANTS

P r i o r t o t e s t i n g , a f e w g e n e r a l l y - k n o w n r u l e s and some e m p i r i c a l d a t a from t h e c h e m i s t r y of s u r f a c t a n t s c a n be u s e d : range ( < C s m a l l m i c e l l e s , low Good w e t t i n g a c t i o n : C -C s u r f ace a c f i v $ $ y ) B r a n c h e a and s o l v a t a b l e g r o u p s s h o u l d l i e close t o t h e c e n t r e of t h e m o l e c u l e .

.

,:

H y d r o p h i l i c c h a r a c t e r : 3 CH2 g r o u p s P 1 OH-group -$-NH-group P -0-group ( h y d r o g e n b r i d g i n g )
3 CHp-groups

beginning water s o l u b i l i t y S o l u b i l i t y : n / 3 EO n / 2 EO -medium w a t e r s o l u b i l i t y 1 - 1 , 5 n EO-good water s o l u b i l i t y ( n = number of c a r b o n atoms i n h y d r o p h o b i c c h a i n ) Solubility decreases with r i s i n g temperature ( - c l o u d point/through dehydration and i n c r e a s i n g e l e c t r o l y t e c o n t e n t (see f i g . 4 / l / ) .
HLB v a l u e : W/O e m u l s i f i e r s

-

3 -6

Wetting a g e n t s 7-9 O/W e m u l s i f i e r s 8-12 O/W d i s p e r s i n g a g e n t s , W/O demuls i f i e r s , solubilizing agents

/13

-18/

50

rurtactont nonylphsnol/l5 EO
in water

0

1.o 2.0 3.0 electrolyte concentration /ma1 1-1

F i g u r e 4 : E f f e c t o f e l e c t r o l y t e c o n c e n t r a t i o n and t y p e o n c l o u d p o i n t TP / 7 / On t h e b a s i s o f t h i s p r e l i m i n a r y i n f o r m a t i o n , i t i s now a l r e a d y p o s s i b l e t o g e t the most i m p o r t a n t r e q u i r e m e n t s o n s u r f a c t a n t s f o r EOR p r o c e s s e s / 4 , 5 / : - Enrichment a t t h e i n t e r f a c e - Formation of o r i e n t e d monolayers - Permanent l o w e r i n g of i n t e r f a c i a l t e n s i o n i n t h e s y s t e m o i l / w a t e r t o (1 m N . m - 1 a t low s u r f a c t a n t c o n c e n t r a t i o n / 1 3 / - Tendency t o m i c e l l e f o r m a t i o n - Partial o i l solubility - S t a b i l i z a t i o n of O/W e m u l s i o n s Solubility or d i s p e r s a b i l i t y i n highly s a l i n e formation water - Long-term s t a b i l i t y ( 1 - 2 y e a r s ) u n d e r r e s e r v o i r c o n d i t i o n s - Low a d s o r p t i o n on r e s e r v o i r r o c k - Low c o s t c o u p l e d w i t h h i g h e f f e c t i v e n e s s

-

A l i s t of p o s s i b l e b u i l d i n g s t o n e s a v a i l a b l e c o m m e r c i a l l y f o r t h e s y n t h e s i s of s u r f a c t a n t s i s g i v e n i n t a b l e 3 . These c o n s i d e r a t i o n s t h e n l e a d t o classes of p r o m i s i n g p r o d u c t s , which i n p a r t s h o u l d e x h i b i t v e r y s t r o n g i n t e r f a c i a l a c t i v i t y and a r e described i n t h e U S - l i t e r a t u r e a s e f f e c t i v e f o r EOR p r o c e s s e s ( s e e t a b l e 4 and 5 ) .

T h e s e known s u r f a c t a n t s a r e s u i t a b l e p r i m a r i l y f o r low s a l i n i t i e s (1 % N a C l w i t h a b o u t 100-200 ppm Ca2+ and Mg2+ o n l y ) . W i t h o u t a p o l y e t h e r c h a i n w i t h s u f f i c i e n t d i s p e r s i n g power, however, t h e sol u b i l i t y i n h i g h s a l i n i t y s y s t e m s ( 1 5 - 2 5 % NaC1, 20 000-40 000 ppm C a 2 + and Mg2+) i s f o r t h e m o s t p a r t t o o low o r t h e e l e c t r o l y t e s e n s i t i v i t y t o a l k a l i n e e a r t h i o n s t o o h i g h . Even i n t h e case o f pol y e t h o x y l a t e s t h e e l e c t r o l y t e c o n t e n t of t h e r e s e r v o i r w a t e r c a n l o w e r t h e c l o u d p o i n t s t r o n g l y ( f i g . 4 ) and t h u s c a n b r i n g a b o u t a d e c r e a s e d s o l u b i l i t y as w e l l a s i n c r e a s e d a d s o r p t i o n and a p a r t i a l p a s s a g e o f t h e s u r f a c t a n t i n t o t h e o i l p h a s e /6/. I n g e n e r a l t h e i n t e r f a c i a l a c t i v i t y o f t h e anionics i s l i k e w i s e r e d u c e d s t r o n g l y i n water w i t h a h i g h e l e c t r o l y t e c o n t e n t / 5 / , F r e q u e n t l y a l s o s u r f a c t a n t m i x t u r e s f o r EOR p r o c e s s e s h a v e b e e n d e s c r i b e d and a p p l i e d . T h e r e r e m a i n s u n c l e a r t h e q u e s t i o n o f c h r o m a t o g r a p h i c phenomena i n t h e u s e of c o m p l i c a t e d s u r f a c t a n t m i x t u r e s i n r e s e r v o i r , i n which t h e q u i t e d i f f e r e n t components of t h e m i x t u r e c a n e x h i b i t c o m p l e t e l y d i f f e r e n t r a t e s of m i g r a t i o n .

51 T a b l e 3 : Possible b u i l d i n g s t o n e s f o r s u r f a c t a n t s
Surfactant building stones a-olefins oligomeric alkenes fatty acids (saturated and unsaturated) and derivatives, natural oils alkanols (alfols, 0x0-alcohols, fatty alcohols) alkylaromatics isoalkylphenols alkylamines (fatty amines) polyalkylene glycol ethers polybutylene oxide (polypropylene oxide)

SO,, ( ~ O , ) , C I S O , H , , H,NSO,H,Na,SO,,NaHSO,

8 8-

),

HOC,H,SO,Na, ( C H , ) , < P ,

(CH2)(_So2\0

c'

H,O, (N), CICHFO,H, (HNO,)
1

0

(Formaldehyde, epichlorohydrin, RO-CHSH-CH,, aliphatic oligoamines, polyols, etc.)
Table 4 :

/o\

A n i o n i c s u r f a c t a n t s f o r EOR p r o c e s s e s ( i n t e r n a t i o n a l literature)
Anionic surfactants Designation a-Sulfo fatty acid esters OH fatty acid sulfates Sulfated amide oils Didecyldiphenyl ether disulfonates structural type

Chemical constitution
R'-FH-CO,R' SO,H (Na) R-CH,-CH-R-COONa OS0,Na
R ~ C O N ~ ~ '

--T

7

S0,Na R 2

7x
-77
).wv,

'P c
% R-CH,

H H R-C -C,H,-N-CH,SO,Na bH N + S O , N ~ R

R-N -R'-(OC,H,)xOSO,Na R PH

Hydroxyalkylaminosulfonic acids Alkenylsuccin-N-(alkyl)phenylimidesulfonates Dialkylamino polyether sulfates Alkenyl-, OH-Alkane sulfonates Sulfates of iosalkylphenyl polyether sulfonates Bisfatty acid esters of triethanolamine polyglycol ether sulfocarboxy lates Alkanol polyether phos-

RG@-o(c.H,,o)~R S0,Na (rnFXAC0) ,C,H,O-C-R HN' -C,H,O-C-R \ d H C,H,(OC,H,),OC-~-CH,-CO.Na

p

b sop

R-0-(CH,-CH,-0)"-P-ONa ONa

P

-

52
T a b l e 5 : Amphoteric s u r f a c t a n t s f o r EOR p r o c e s s e s ( i n t e r n a t i o n a l literature)
Amphoteric Surfactants Chemical constitution Sulfobetains Structural type

R’ H R - ~ C H , - C -cH,-so,~

Sulfobetains

4
--f.)

R’

OH

YH’
H,,.,,c,,,:-I]I-cH,-coo~

CH,

Betain Alkylimidazoliniumbetains
e

R-&

HO-CH,~CH,/N\CH,~COO~
CH;

-79

H,,,,-c,-co-NH-(cH,):-$-cH,-coo
CH,

Amidoalkylbetains

-a . -

O t h e r w i s e i t is p r o b a b l y p o s s i b l e t o i n c r e a s e , by t h e u s e of s u c h s u r f a c t a n t m i x t u r e s , t h e p a c k i n g d e n s i t y a t t h e i n t e r f a c e and t h u s t h e d e g r e e of w e t t i n g ( s e e f i g . 5 / l o / ) ; f u r t h e r , a l s o t h e format i o n of mixed micelles i s p o s s i b l e ( f i g . 6 /4/).

lipophilic port of molecule surfactant molecule
0 hydrophtlic part of mokcuie

F i g u r e 5 : I n c r e a s e d p a c k i n g d e n s i t y by s u r f a c t a n t m i x t u r e s a t O/W i n t e r f a c e s

When p u r e n o n i o n i c s a r e u s e d , s u c h a s i s o a l k y l p h e n o l e t h o x y l a t e s , a t t a i n m e n t of s a t i s f a c t o r y i n t e r f a c i a l a c t i v i t i e s demands a h i g h e r d e g r e e of a l k o x y l a t i o n . These p r o d u c t s a r e n o t e l e c t r o l y t e - s e n s i t i v e and h a v e a good s o l u b i l i t y i n b r i n e ( a r u l e of thumb i s t h a t a t 5OoC, o n l y n o n i o n i c s w i t h a d e g r e e of e t h o x y l a t i o n n of 1 0 o r more a r e s o l u b l e i n 1 0 % NaC1).

53
oure onionic surfoctont
011

oil

transitional interface

L1

transitional interface

water

water

1 : 12 1

- 1.1

surfoctont mixture (anionics/nonionics I

F i g u r e 6 : a ) p a c k i n g d e n s i t y of p u r e a n i o n i c s u r f a c t a n t a t i n t e r face b ) p a c k i n g d e n s i t y of a n i o n i c - n o n i o n i c s u r f a c t a n t a t interface

EXPERIMENTAL

Under c o n s i d e r a t i o n of a s many s e l e c t i o n c r i t e r i a , physico-chemic a l p r o p e r t i e s and p o s s i b i l i t i e s of s u r f a c t a n t s y n t h e s i s a s poss i b l e more t h a n 1 2 0 0 s u r f a c t a n t s were t e s t e d f o r t h e i r a p p l i c a b i l i t y t o EOR p r o c e s s e s . The s c r e e n i n g of t h e s u r f a c t a n t s was c a r r i e d o u t a c c o r d i n g t o t h e r a p i d s c r e e n i n g program a l r e a d y i n troduced. F u r t h e r t e s t s on a s u r f a c t a n t w e r e proposed o n l y , i f i t had pass e d t h e s c r e e n i n g program. SOLUBILITY I N BRINE CM
A l l e x p e c t a t i o n s on t h e s o l u b i l i t y of s u r f a c t a n t s i n h i g h - s a l i n i t y b r i n e s were c o n f i r m e d i n a l l r e s p e c t s . E s p e c i a l l y t h e s u r f a c t a n t s w i t h p o l y e t h e r c h a i n s and a n i o n i c g r o u p s have shown good s o l u b i l i t i e s up t o t h e mark. T a b l e 6 p r e s e n t s some t y p i c a l p r o d u c t s , which were s e l e c t e d on t h e b a s i s of t h e a b o v e - d e s c r i b e d s o l u b i l i ty criteriaa.

INTERFACIAL ACTIVITY

During t h e measurements of t h e i n t e r f a c i a l a c t i v i t y (Lecomte du Nouy) a s t r o n g dependence of t h e i n t e r f a c i a l t e n s i o n on t h e temper a t u r e and s a l i n i t y was e s t a b l i s h e d i n t h e system o i l / w a t e r w i t h o u t a d d i t i o n of any s u r f a c t a n t s ( s e e f i g . 7 ) . The e x p e r i m e n t s have shown t h a t a l l s t a n d a r d o i l s a r e c h a r a c t e r i z e d by a t y p i c a l i n t e r f a c i a l t e n sion - relationship - i n t e r f a c i a l t e n s i o n depends s t r o n g l y on t h e o i l c o m p o s i t i o n - n a p h t h e n i c o i l shows t h e h i g h e s t v a l u e s of i n t e r f a c i a l t e n s i o n against high-salinity brines - i n g e n e r a l an i n c r e a s e i n s a l i n i t y is accompanied by a d e c r e a s e i n i n t e r f a c i a l t e n s i o n / A minimum i n i n t e r f a c i a l t e n s i o n w i l l be passed. T h e r e s u l t s a r e summarized i n f i g u r e 7 .

-

54

T a b l e 6 : Some s u r f a c t a n t s w i t h good s o l u b i l i t y , t e s t e d i n h i g h salinity brine
Chem. constitution
I-C,H,,~(C,H.OJ.,CH,CD,N~
IC,H,.~O(C,H.OI,. l-C,Ha,

Designation Isoalkylphenylpolyether acetates Diisoalkylphenylpolyether sulfates

Structural type-

SO, Na

Rj-N/(C,H,O)x '(C,H,O), (2 x = 5 )

SO,Na S0,Na

Acylamidopolyether sulfates

(R-C-~,H,-N/(C1H40)y
( 2 y t 1= 2 X I

fl

'(C,H,Oly

S0,Na)

(Esteramine polyether sulfates)
OR'

N

i-C,,H,,~-CH,-~-CH,~(OC.H,.Jx
SO,N~
x-0-20 R'= H, SO,Na,

( W n H x n)x O R

(Sulfone/sulfate-isoalkylphenylpolyethoxyglycerol ether)

-CH,COONa n=2,3

Ditert.-alkylphenyl polyethers

Isoalkylphenylpolyethers

I f s u r f a c t a n t s were added t o t h e s t a n d a r d o i l s , c h a r a c t e r i s t i c c u r ves r e s u l t e d f o r t h e f u n c t i o n i n t e r f a c i a l t e n s i o n y = f ( s u r f a c t a n t concentration CT). T h e s h a p e of t h e c u r v e s d e p e n d on - t y p e of o i l ( c o m p o s i t i o n ) - temper a t u r e - salinity - t y p e of s u r f a c t a n t and c o n c e n t r a t i o n of s u r f a c t a n t . T y p i c a l d i a g r a m s of s u r f a c t a n t m i x t u r e s ( a n i o n i c s - n o n i o n i c s ) a r e presented i n f i g u r e 8 a/b. -1 I f t h e i n t e r f a c i a l t e n s i o n r e a c h v a l u e s of < 1 mN.m , t h e accur a c y of t h e method of Lecomte d u Nouy i s no l o n g e r s u f f i c i e n t . Furt h e r t e s t s on " s u c c e s s f u l " s u r f a c t a n t s makes t h e a p p l i c a t i o n of a spinning-drop-tensiometer (SITE) n e c e s s a r y . A " s u c c e s s f u l " s u r f a c t a n t m u s t comely w i t h t h e f o l l o w i n q c r i t e r i o n : The i n t r f a c i a l t e n s i o n of a s u c c e s s f u l s u r f a c t a n t must be < 1 mN.m-f against a l l three standard o i l s i n the temperature range of 30-80°C. On t h e b a s i s of t h i s c r i t e r i o n a g e n e r a l t e m p e r a t u r e i n t e r f a c i a l t e n s i o n - r e l a t i o n s h i p was d e r i v e d f o r t h e s u r f a c t a n t s t e s t e d ( s e e fig. 9). By t h i s i t was e v i d e n t , t h a t a n i o n i c s ( w i t h an o p t i m a l c o n t e n t of nonionics) w i l l e x h i b i t the lowest i n t e r f a c i a l tension.
TEMPERATURE STABILITY

For t h e i n v e s t i g a t i q n of t h e t e m p e r a t u r e s t a b i l i t y , s u r f a c t a n t s o l u t i o n s of v a r i o u s c o n c e n t r a t i o n s i n b r i n e CM were k e p t f o r

55

"I
20

N

Figure 7 : I n t e r f a c i a l tens i o n a s a f u n c t i o n o f temp e r a t u r e , s a l i n i t y and o i l composition (naphthenic, aromatic, p a r a f f i n i c ) 6 month a t a t e m p e r a t u r e o f 8OoC. After t h i s t i m e the interf a c i a l a c t i v i t y was comp a r e d t o t h a t of a s t a n d a r d solution On t h e b a s i s of t h e s e e x periments, the following s t a t e m e n t was p o s s i b l e : ether Ether phosphates s u l f a t e s - e t h e r carboximethylates -ether sulfonates a r r o w i n d i r e c t i o n of increasing s t a b i l i t y

Y ogoinst dist woter

-

0

30

60

80

temperature 9/"C

5 8
.-

s

\:I y
12 10
0 : .

20t

= f(8.011 comp

1

-

Y against brine EM

N

SPECIAL FINDINGS P o l y e t h e r s u l f a t e s , -carbo x i m e t h y l a t e s and - s u l f o n a t e s of t h e f o l l o w i n g s t r u c ture are particularly suita b l e f o r EOR g r o c e s s e s : R(OC2H4)xY-Me (Fig.10).

.-I

1

v

.

I

.

I

.

,

, -

0

30

20

t

80 temperot ure 8 / "C
y = t(;t.oil comp I

60

Numerous s u r f a c t a n t s w i t h e s p e c i a l l y low v a l u e s of t h e i n t e r f a c i a l t e n s i o n may be c l a s s i f i e d a s m i x e d s u r f a c t a n t s (Mischtenside) (ani o n i c / n o n i o n i c ) . The compos i t i o n of t h e m i x e d s u r f a c t a n t i s u s u a l l y g o v e r n e d by t h e manufacturing process o r t h e d e g r e e of c o n v e r s i o n . I n t h i s r e s p e c t i t was ob30 60 eo s e r v e d t h a t a d e g r e e of contemperature ;t/' C v e r s i o n o f 50 t o 8 0 p e r c e n t N naphthenic oil nonionic t o anionic surfacA aromottc oil t a n t g i v e s rise t o p a r t i c u P poroffinic oil l a r l y favourable surfactant p r o p e r t i e s . A t y p i c a l homoloaue d i s t r i b u t i o n f o r f o r s u c h a s u r f a c t a n t i s shown i n f i g u r e 1 1

Y agoinst brine CM

.

The f o l l o w i n g i m p o r t a n t d a t a and r e s e a r c h r e s u l t s a r e w o r t h mentioning: - The d i s t r i b u t i o n c u r v e f o r t h e s u r f a c t a n t homologs ( a l k o x i l a t e s ) s h o u l d be a s b r o a d a s p o s s i b l e (more p d l d i s p e r s e ) i . e . , a l k a l i catalyzed alkoxilation (not Lewis-acid catalyzed). - The d e g r e e of a l k o x i l a t i o n n m u s t be a d j u s t e d a c c o r d i n g t o t h e c r u d e type (and Y - ) . General r u l e s a r e HLB: 8-10 p a r a f f i n i c c r u d e s : n = 4 5 2 EO (partial o i l solubility) n a p h t h e n i c c r u d e s : n = 6 5 2 EO

56
rn namlhanic oil
y = f (c,,

oil

type)

A poraflinic oil

x ommolic oil
suiodoni 63 tenpirniure ’ 3. 0C surfuianl in brine CM

Figure 8 a : I n t e r f a c i a l tension a s a funct i o n of o i l composition, surf a c t a n t concent r a t i o n and temp e r a t u r e (30°C) surfactant: C12/14-fatty alkohol-polyglycolether-(4,5 EO)-carboxmet h y l a t e , Na-salt

naphthanic oil

y = f (cT, oil type 1

A poraflinic oil

x oromalic oil
surfactant 63 impKa1ua:I0.C surfoclani in brina CM

10

I

Figure 8 b: same a s f i g . 7a t e m p e r a t u r e 80 C

6

1n

1
A h htypical

high content 1 of
nonionics

onionics and mixed surfactants

/

temperoture

a/

0;

area of spontaneous

emulsification

Figure 9 : Temperature/ i n t e r f a c i a l tension-rel a t i o n s h i p f o r some i m portant s u r f a c t a n t groups

57

hydrophobic chain

Polyether group

polar hydrophilic group

counter ion

R

-+X+

YQ

Z@
OlkOll earth alkali ornines etc

fatty OlCOhd tatty ocid nonylphenol naphthenic ocid tatty ornines

ethyleneoxide propyleneoxide etc

car boxylote sulfate sulphonate phosphate propionate etc

F i g u r e l o : S u r f a c t a n t s s u i t a b l e f o r EOR p r o c e s s e s

nonionic port

anionic part

Hol EO

Hal EO
= 8 & 2 EO

F i g u r e 11: Q u a n t i t a t i v e a n a l y s i s of i-nonylphenol-polyglycole t h e r (6 E O ) - c a r b o x i m e t h y l a t e , Na s a l t , by HPLC aromatic crudes:

n

-

( P r o p o x i l a t e s a r e i n g e n e r a l less e f f e c t i v e , a s a r e EO-PO ducts) The h y d r o p h o b i c c h a i n R must be t a i l o r e d w i t h p r e c i s i o n . p a r a f f i n i c c r u d e s : C14 5 4 ( s a t u r a t e d , u n b r a n c h e d ) naphthenic crudes: C,, + 2 arbmatic rudes: a k y l r o m a t i c s iiso-C8-12-alkyl). C a t i o n ( 2 ) : Na', ,'K I i'R4 o r NH4

ad-

The l e n g t h of t h e h y d r o p h o b i c and h y d r o p h i l i c m o l e c u l a r p a r t s s h o u l d be r o u g h l y i n t h e 1 : 1 r a t i o ( p a r t i a l o i l s o l u b i l i t y ) .

58
Example: C 1 2 , 1 4 - a l k y l p o l y g l y c o l e t h e r s u l f a t e - ( 4 , 5 EO), N a - s a l t

( h y d r o p h o b i c : h y d r o p h i l i c c h a i n l e n g t h = 2 , 2 nm: 2 , 3 n m ) o r f o r n o n y l p h e n o l p o l y g l y c o l e t h e r s u l f a t e - ( 4 EO), Na-salt (hydrophobic: hydrophilic chain l e n g t h = 2,O: 2 , l n m (see f i g . 1 2 ) .

oil

transitional interface

hydrophobic part A

water

~~~

hydrophilic port B

ratio at optimum A B - 1 1

F i g u r e 1 2 : Optimal c h a i n r a t i o of s u i t a b l e s u r f a c t a n t s f o r EOR p r o c e s s e s

-

The d e g r e e of c o n v e r s i o n of n o n i o n i c s i n t o s u l f a t e s , c a r b o x i m e t h y l a t e s , s u l f o n a t e s , e t c . , s h o u l d be 50-80 % ( m i x e d s u r f a c t a n t f o r m a t i o n f r o m n o n i o n i c s and a n i o n i c s ) .

The c a r a c t e r i s t i c b e h a v i o u r of a n i o n i c - n o n i o n i c m i x e d s u r f a c t a n t s w i t h t e m p e r a t u r e (minimum of i n t e r f a c i a l t e n s i o n ) c a n be e x p l a i n e d w i t h t h e h e l p of t h e p h a s e d i a g r a m f o r s u c h s y s t e m s ( s e e f i g u r e 13), w h e r e b y t h e o c c u r r e n c e o f a m i s c i b i l i t y g a p is d e c i s i v e .

A
L a I

aI

1 1 MST=T?

0% 100%
MSTK
CK
CX
=

-nonionic surfnctant + c a n i o n i c surfactant -

100% 0%

F i g u r e 1 3 : Schematic p h a s e d i a g r a m f o r mixed surf actant (anionicn o n i o n i c ) w i t h miscib i l i t y gap

lower critical micell- splitting - temperature

= critical splitting -concentration of mixed micelles at MSTk = composition of mixed micelles

MST=T2= splitting - temperature of mixed micelles

AT
a,p

=

1 -12 1

= coexistent phases with concentrations c, and c ,

. As a r e s u l t of t h e manufact u r i n g p r o c e s s .s a l i n i t y s y s t e m s from a l a r g e number of p r o d u c t s . When t h e MST i s e x c e e d e d . tailor-made p r o d u c t s .59 A s e p a r a t i o n i n t o w a t e r . I n t h e long term. S i n c e m i x e d micelles a r e formed. t h e s e p r o d u c t s p o s s e s s s p e c i a l t e m p e r a t u r e . t h e s e a r e accompanied by t r a n s p o r t p r o c e s s e s a t t h e i n t e r f a c e s .d e p e n d e n t p r o p e r t i e s which a r e i n t e r e s t i n g f o r EOR p r o c e s s e s . e s p e c i a l l y s u r f a c t a n t mixt u r e s o r mixed s u r f a c t a n t s .a n i o n i c ) . t F i g u r e 1 4 : Phenomena a t the micelle-splitting t e m p e r a t u r e (MST) f o r m i x e d s u r f a c t a n t (nonionic-anionic) CONCLUSIONS With t h e t e c h n o l o g i c a l p o s s i b i l i t i e s t a k e n i n t o c o n s i d e r a t i o n . o f f e r s p e c i a l p r o m i s e from t h e economic p o i n t of view. i t was p o s s i b l e t o s e l e c t s u r f a c t a n t s s u i t e d f o r EOR p r o c e s s e s i n h i g h . The s e l e c t e d s u r f a c t a n t s a r e ani o n i c s and b e l o n g t o t h e c l a s s e s of polyglycolethercarboximethyl a t e s and polyglycolethersulfonates. and w i t h t h e h e l p of a r a p i d t e s t p r o c e d u r e ./ o i l .s o l u b l e s u r f a c t a n t s o c c u r s when t h e m i x e d m i c e l l e s formed from anionics & nonionics reach the micelle s p l i t t i n g t e m p e r a t u r e ( M S T ) . t h e s e p r o d u c t s may be c l a s s i f i e d a s mixed s u r f a c t a n t s ( n o n i o n i c . v a r i o u s i n t e r e s t i n g phenomena may be o b s e r v e d ( s e e f i g u r e 1 4 ) .

nm length of hydrophobic chain.c CS . OC Abbreviations for fiqures CT A1 # A 2 Au L1 L2 Y surfactant concentration in ppm or % distance between surfactant molecules at interface. nm thickness of transitional interface length of hydrophilic chain. % water-in-oil temperature.m-1 . 1-1 ethylene oxide enhanced oil recovery tertiary oil recovery phase) hydrophilic/lipophilic balance hydrocarbons original oil in place. g. mN. % oil/water propylene oxide parts per million iota1 dissolved solids. EO EOR - HLB HC OOIP O/W PO ppm TDS W/O 0 - critical micelle formation concentration salinity/concentration of salts dissolved.m . nm interfacial tension.60 Nomenclature c .

Park Ridge. D. 302313 T e n s i d e D e t e r g e n t s 9 ( 1 9 7 2 ) . J . 256 6 261 S u r f a c e f i l m s of n o n i o n i c d e t e r g e n t s I.J. t e s t i n g methods and e x p e r i m e n t a l r e s u l t s ( i n German) Ph.: t e " ( I n t e r f a c e . C o l l . 1974 ( i n R u s s i a n ) The c o n f i g u r a t i o n of t h e p o l y e t h y l e n e o x i d e c h a i n of n o n i o n i c s u r f a c t a n t s ( p a r t 1 ti 2 ) ( i n German) Tenside Detergents ( 1 9 7 1 ) .H. Moscow. 191-204 E f f e c t of t e m p e r a t u r e on t h e a d s o r p t i o n of n o n i o n i c s u r f a c e .. S. TU C l a u s t h a l 1 9 7 8 " S u r f a c t a n t s and S e q u e s t r a n t s " . . t h e s i s . : 10 A k s t i n a t . 1 9 6 7 / C h a p t e r 22 V i s c o u s f l o o d i n g media f o r t e r t i a r y o i l recovery i n h i g h l y s a l i n e systems s e l e c t i o n c r i t e r i a . N .H.i n v e r s i o n . S c i .: - 5 Gutscho. .: 1 7 8 Crook. M.. 5763 . p p . 23-28 " G r e n z f l a c h e n a k t i v e E t h y l e n o x i d -AddukAdducts) . J . F o r d y c e . Wiss.J. Kosswig .: Surface-active agents f o r t e r t i a r y o i l r e c o v e r y : s e l e c t i o n c r i t e r i a and s e l e c t i o n m e t h o d s ( i n German) 4 T e n s i d e D e t e r g e n t s 1 ( 1 9 7 7 ) . pp. p p . Sci. Surface tension study J. D. USRR 33. ( 1 9 6 2 ) . N. t . J.: 6 B a l z e r . : - 7 Schick. ( 1 9 7 1 ) . I n c . M. 801-813 4 A k s t i n a t . C o l l . N e w York. J . : 3 Schonfeldt.A. G. B.: Physicochemical p r o c e s s e s i n o i l prod u c t i o n . M a r c e l Dekker.Octylphenoxyethoxie t h a n o l s (OPES) J .A c t i v e E t h y l e n e Oxide Schick. 20 ( 1 9 6 5 ) . pp. 379-381 9 Kravchenko. E. . K . pp.F. S t u t t g a r t 2 Rosch.t e m p e r a t u r e a s a c r i t e r i a f o r s e l e c t i o n of s u r f a c t a n t s f o r EOR ( i n German) T e n s i d e D e t e r g e n t s 1 ( 1 9 7 9 ) . Noyes Data Corp. V e r l a g s GmbH.: M o l e c u l a r w e i g h t d i s t r i b u t i o n of noni o n i c s u r f a c t a n t s / I I . .H.. J. The p h a s e . M .61 LITERATURE 1 Babalyan. M. " P u b l i s h i n g House " N e d r a " . : "Nonionics Surf a c t a n t s " .a c t i v e subs t a n c e s on s o l i d a d s o r b e n t s C o l l . T r e b b i . pp. 1977. M. P a r t i t i o n coeffic i e n t s o f n o r m a l d i s t r i b u t i o n and homogeneous p . p p . G. D.

et al. 1 9 6 3 . H. Conf. . : Cash. F a l l Mtg.J. 577 67 14 Trogus. SPE of AIME Ann..1975. M.c r i t e r i a f o r t h e s u r f a c t a n t s e l e c t i o n and a p p l i cation T e n s i d e D e t e r g e n t s 1 ( 1 9 8 0 ) . 5 .C. & Exh..: 16 Oppenlander .. Murtada. Mtg. 15 W r i g h t . SPE-Paper 5564 A d s o r p t i o n of m i x e d s u r f a c t a n t s y s t e m s 5 2 . A k s t i n a t .: Surfactant aging: a possible detriment t o t e r t i a r y o i l recovery 5 0 . 12 R u d i . SPE-Paper 6845 The u s e o f C a r b o n D i o x i d e i n w a t e r floods A P I P r o d . : T e r t i a r y o i l r e c o v e r y methods ( i n German) Erdo14rdgas-Z.1977. 2 1 .62 1 1 Rieckmann. E. M . R. SPE of AIME A n n . pp.R.2 3 . C.: .. 119-122 K o l l o i d Zh. 9. S o b k i v . V. Los A n g e l e s P r e p r i n t 801-39 k S u r f a c t a n t s f o r enhanced o i l recovery i n hiqh-salinity systems . F. Dallas/Tx. F a l 1 Techn.P.3.10. D i v .1.H. .10. P a c i f i c Coast D i s t r . 28.. pp. K. Denver/Col.-12. 91 ( 1 9 7 5 1 . p p .L. S 13 et al. 348359 I n f l u e n c e o f s u r f a c t a n t s on t h e p r o p e r t i e s of c l a y s ( i n R u s s i a n ) ( 1 9 6 6 ) .

The phase inversion temperature decreases with increasing lipophilic alcohol content of the systems. Their adsorption onto rock surfaces 3. COLLEY and R. M. 1 . 3. the interfacial tension of a surfactant-brine. . For constant salinity and surfactant concentration phase inversion temperature increases with increasing equivalent alkane carbon number. Their interfacial tension behaviour with crude oil and pure alkanes. We have found: 1. N. London Research Station ABSTRACT Some commercial surfactants have been studied with a view to their usefulness for enhanced oil recovery applications. 5. Winfrith. The phase inversion temperature increases with ethylene oxide content of the surfactant. Some surfactants have given interfacial tensions approaching 10-3 dynes cm-1. The variation of phase inversion temperature with different variables. 2. After a review of information available to us on the reservoirs operated by British Gas Corporation .hydrocarbon mixture passes through a minimum. As the temperature varies. Static adsorption tests on reservoir rock show Langmuir adsorption isotherms. . Introduction London Research Station. THAVER British Gas Corporation. the corporate laboratory of British Gas became involved in enhanced oil recovery after an invitation by the Department of Energy to take part in its research programme coordinated by A E E . salinity and hydrocarbon remaining constant.CHEMICAL FLOODING 63 PRELIMINARY STUDIES OF THE BEHAVIOUR OF SOME COMMERCIALLY AVAILABLE SURFACTANTS IN HYDROCARBON-BRINE-MINERAL SYSTEMS C. the hydrocarbon and the surfactant concentration and composition remaining constant. 4. The interfacial tensions were measured by the spinning drop technique. 2. The following aspects of their behaviour have been assessed. The phase inversion temperature decreases with increasing salinity. ANDREWS.

000 mgNaCl/litre 400 mgCa/litre 1.(11 '* ethylene ethylene oxide units.. . Surfactant B .. lack of sensitivity being a desirable (but attainable?) ideal. Sapogenate and Arkopal are trade marks of Hoechst AG. and to assess their sensitivity to changes in reservoir variables.) T130 oxide units.) Hydrocarbons used in this work were specified to be greater than 99% pure. The Reservoir Conditions in the target reservoir are similar to those listed below: Oil type E A C N . AG . Surfactant A straight chain alkyl phenol ether acetate. This paper describes the work we have performed so far to identify commercially available surfactants with interfacial tensions-lowering properties to suit the conditions prevailing in our reservoirs.Na salt. Non ionics: units ) TlOO" " . (6 ethylene oxide units. 4 ethylene oxide units.300 mgCa/litre 500 mgMg/litre 30. 6 ethylene oxide units. Sapogenate* T80 tri-butyl phenylpolyglycol ether(8 ethyleae oxide " 1 ethylene oxide units.64 we decided that our resources would be most usefully employed studying micellar/polymer and miscible flooding. an alkylaryl polyglycol ether sulphate ( 7 ethylene oxide (e.) (0 T110 '* '* " . interfacial tensions and adsorption onto mineral surfaces have been made.oJunits.200 mgMg/litre Chemicals Surfactants. 50% w/w active). Samples of the surfactants listed below were obtained from Hoeschst Anionics: Hostapal" BV.) (13 Arkopal* NO60 Nonyl phenylpolyglycol ether (All 100% active).. * Hostapal. .000 mgNaCl/litre 1. Temperature Formation water Flood water (sea water) (Reservoir) (Stock tank) 7-a 10 43oc 90. Measurements of phase inversion temperature (PIT).. 1 ..

PIT increases with increasing number of ethylene oxide groups in ethoxylated surfactants We have extended this work to nonionic surfactants in studying the following variables on PIT.(Shinoda 1965). it partitions mainly into the oleic phase and forms a 'water-in-oil' emulsion (Balzer and Kosswig. 2.. PIT increases with increasing equivalent alkane carbon number (EACN) of the oil and that aromatic hydrocarbons show very low values of EACN. 1 . 3. The effects of these variables must be considered if a surfactant flood is to maintain its oil mobilising properties as it passes through the reservoir. reservoir rock will change PIT. PIT. PIT increases with decreasing salinity..l979). 3. Mixtures of aromatic compounds and alkanes give intermediate PITS. Since the lowering of interfacial tension is a requirement for the mobilisation of oil trapped in constricted capillaries and all oil reservoirs are essentially isothermal. The hydrophilic/lipophylic balance of a surfactant will depend upon its ethylene oxide content. 5 . Oil type expressed as EACN. They found: 1. 6. Adsorption measurements are therefore important. Above the PIT. The salinity of the brine in contact with residual crude in a waterflooded reservoir may vary from pure injection water to pure formation water. If PIT is dependent upon the number of e 0 units.65 Phase Inversion Temperature The phase inversion temperature. The EACN of our reservoir crude has been assessed from measured EACN 6f stock tank crude and calculated from a well stream analysis. 4. . 3. The EACN for a given reservoir oil should be constant. must be taken into account when performing laboratory tests. 6 . PIT represents a useful parameter for the selection of surfactant for a given reservoir. . Balzer and Kosswig (1979) have carried out some parametric studies of PIT with a range of anionic carboxy methylated nonyl phenol ethyoxylate surfactants. . then selective adsorbtion by . This 5 Variable oil/brine ratios will occur in a reservoir as a flood proceeds. 2. 4. but actually contain a distribution of e 0 chain lengths. The parameters studied in this work are: 1. The concentration of surfactant at some point away from the injection well is likely to change because of adsorption onto the reservoir rock surfaces. 2. For nonionic surfactants below the PIT the surfactant partitions preferentially into the aqueous phase and the emulsion formed between the two phases is predominately 'oil-in-water'. of a hydrocarbon/brine/ surfactant system indicates the existence of a minimum interfacial tension at that temperature. Commercial surfactants are usually assigned a nominal ethylene oxide content. Surfactant type and concentration Salinity Co-surfactant type and concentration Phase ratio Number of ethylene oxide units in surfactant molecule. Co-surfactant effectiveness may change with concentration and type.

water emulsion with a non-ionic surfactant initially below the PIT the conductivity slowly increased with temperature but fell rapidly as the emulsions inverts and the aqueous phase became discontinuous and therefore non-conducting. :tivity T°C Temperature 4' $ 4 (I *I Figure 2 Adsorption The investigation consisted of a series of experiments to measure A. More than one 'minimum' may occur for impure surfactants. The method of analysis for surface-active material was the titration procedure of Reid et al.66 Interfacial Tension It is necessary to augment the data obtained from PIT measurements. . against temperature is sketched in Fig. Interfacial Tension Measurements were made at ambient pressure with the University of Texas spinning drop tensiometer. The method used was based on that of Somasundaran & Hannah (1979). For an oil-in. IFT.1.. Methods Phase Inversion Temperature was determined by means of electrical conductivity measurements (Baker and Kosswig 1979).F. A typical plot of I. 10-L dynes/cm t0-L lo-& J V t I I I PIT T°C Measurements have been made to determine the way IFT changes with temperature. The inversion of emulsions occurs over a small temperature range.T. the adsorptive capacity of the reservoir rock for surfactant material. For this to occur with minimum energy an interfacial tension minimum is implied. (1967). Figure 2 shows a typical curve.

50 . RESULTS AND DISCUSSION 1 . I. butyl cyclohexane phenyl heptane phenyl octane 4:l Table 1 Table 1 Table 1 *. Results and Discussion The results presented in Figs 3 to 5 show a linear relationship between EACN and PIT over the EACN range studied. . . . PITS were determined on the following mixtures at the phase ratios stated (brine/oil). . I. Surfactant A A+B Concentration /litre ~ Brine Seawater Seawater Hydrocarbons n-alkanes C7-C10 n-heptane-toluene mixtures EACN 4 to 7 N-alkanes c6 to cll methylcyclohexane Phase Ratio brine/oil 5: 1 Results Fig 3 line A Fig 3 line B 10 5 of each 5:l T80 10 Seawater 5:l Fig 4 line A Fig 4 line B T80 T8 0 T80 TlOO TlOO TlOO As above after storage for 3 months at room temperature 10 50 30g /NaClI n-alkanes litre cg-cl1 11 4: 1 Fig 5 line A Fig 5 line B Fig 5 line C Fig 5 line D ** 1.67 TESTS. Variation of PIT with EACN and determination of EACN value of the p h e n y u cyclohexyl groups.

68 3 4 5 6 7 8 9 10 EACN Figure 3: Variation of PIT with EACN for two different surfactant solutions. 30 g /litre NaCl . PIT "C 70 60 50 40 Line D Line A 30 20 6 7 8 9 10 11 EACN 12 Figure 5: Variation of PIT with EACN. PIT "C 70 60 50 40 30 20 I I 4 5 I 6 I 7 I 8 I 9 I 1 0 1 1 EACN I I 12 I Figure 4 : Variation of PIT with EACN for log /1 Sapogenate T80 in seawater. pure n-alkanes.

nor on a freshly prepared solution stored at 40OC for 3 weeks. TABLE I EACN found butylcyclohexane phenyl heptane phenyl octane 6. Tests Surfactant T80 T80 TlOO TlOO Brine g Hydrocarbon heptane octane heptane octane NaC111 30 Phase ratio brineloil 4:1 4:1 Results Fig 6 Curve 1A Fig 6 Curve 1B Fig 6 Curve 2A Fig 6 Curve 2B *I . Variation of PIT with surfactant concentration h e effect of increasing surfactant concentration was studied on the following mixtures. At the higher concentration of T80 the dependance of PIT on EACN is reduced.75 fig 5 line B 5.5 (Fig 5 Line C) 6. Rates of change of PIT of up to 140C/EACN unit (line B Fig 3) are possible and the slope can change with surfactant concentration. A shift of PIT of 2 to 30C was observed on storage at room temperature for 3 months of the stock 50 gms TlO/litre solution from which line B fig 4 was obtained. The The difference in slopes shown between lines A and B.0(" Assigned EACN of ring 26 . This may enable us to dilute stocktank crude with hydrocarbons with rings to obtain mixtures of hydrocarbon at ambient pressures having EACNs more representative of reservoir crudes. 4:l 4:l I. The values found are listed in Table 1 . . ) 2 . Thus we are able to assign an EACN of 3 6 to methyl cyclohexane. PIT can change rapidly with EACN. -2. Rates of change of PIT as low as 30C/EACN unit are possible (line A Fig 3.* I . phenyl heptane and phenyl octane were determined relative to Fig 5 line C .0 '. Fig 3 suggest the possibility of modifying the sensitivity of a system to change in temperature by the addition of another surfactant. I . affect is not as marked in the case of T100. No further change was observed on storage for a further 5 months.69 The EACN of butyl cyclohexane was determined relative to the Fig 5 line B and line C. Differential adsorption within a reservoir could cause problems in practice.0(" 6. Concentration 8 hitre Various 10 to 70 I.0 -2. " ") ") - .

Both surfactants exhibit a non-linear relationship. Qo 7 b ad Surfactant concentration. Variation of PIT with Salinity Tests PIT'S were determined on the following mixtures.I0 Results and Discussion The results are shown in Fig 6 . 3. PIT *C 60 50 40 30 20 I ' 16 o i 40 4b o' . This The rate of change of is in agreement with the work of Shinoda and Arai (1964). with PIT increasing with decreasing surfactant concentration. g /litre Figure 6: Variation of PIT with concentration of Sapogenate T80 and T100. Surfactant Ta0 Concentration g /litre 10 Brine Hydrocarbon Phase ratio brine/oil 5:l Results Fig 7 line A Fig 7 line B Fig 7 line C Fig 7 line D NaCl only Stock Tank (various Crude EACN 10 concentrations) hexane heptane octane T80 Ta0 T80 10 4:l 4:l 10 10 4:l . PIT is lower at high surfactant concentrations which indicates that a high concentration flood could be less susceptible to concentration changes.

60 PIT OC 50 . NaCl/litre. Variation of PIT with Alcohol (Cosurfactant) type and concentration Tests - Measurements were made on 50 g T100flitre brine. Alcohols studied were: () a' iso-butanol (lipophilic) (b) is0 pentanol( " 1 (c) isopropanol (hydrophilic) Results and discussion. . 10 g /litre Sapogenate T80 solution. the rate of change of The results are shown in Fig 7. - 40 \ Line A 30 2o 1 10 20 30 40 50 60 70 80 90 Brine concentration. Brine concentration was 30 g /NaCl/litre and oil EACN 7. The hydrophilic alcohol (line C) has the opposite affect but less pronounced. g 4. in that increasing the concentration of a lipophilic alcohol (lines A and B) will increase the partitioning of the surfactant into the oleic phase and tend to lower the PIT. (Knickerbocker et al. The results shown in Fig 8 agree with trends predicted in the literature. 1979). Increasing the concentration of hydrophilic alcohols has the opposite effect on PIT as increasing surfactant concentration. 1979). The decrease in PIT with increasing salinity is to be expected as the surfactant partitions more readily into the oleic phase as salinity increases (Knickerbocker et al. If an alcohol has to be used as a viscosity modifier then hydrophilic alcohols may be more manageable with respect to their effect on PIT than lipophilic alcohols.71 Results and discussion For all hydrocarbons tested. PIT with salinity is independent of the hydrocarbon used. Figure 7:.Variation of PIT with brine salinity..5 at a phase ratio of 4 : l .

Arai (1965) reports the same effect as Balzer and Kosswig (op. *. The results show that PIT increases as the proportion of oleic phase increases. .5 50 40 30 10 20 30 40 50 Alcohol concentration. Surfactant T80 Concentration 50 g 10 g /litre /litre I. 9 line D T80 T80 T80 .) . Variation of PIT with brine/oil phase ratio Tests PITs were measured on the following mixtures to find out if PIT varied with phase ratio. *. I. Brine 30 g Nacl /litre " " " Hydrocarbon n heptane n hexane n heptane Results Fig 9 line A Fig 9 line B Fig 9 line C Fig. . . .72 PIT OC 60 50 g 30 g EACN /litre Sapogenate TlOO /litre NaCl 7. " .cit. : Figure 8 Variation of PIT with alcohol concentration 5. This is contrary to the findings of Balzer and Kosswig (1979) who reported smaller changes with the opposite slope. ** n octane Results and Discussions The possibility that PIT would depend upon phase ratio was indicated when PITs obtained from the measurements in the spinning drop tensiometer did not correspond exactly to those made by the conductivity measurements. g /litre.

EACN 7 5 . in PIT is explained by the increased hydrophilic properties with increased eo content (Shinoda. The Arkopal series of surfactants probably exhibits a similar trend but only These gave PITS of two have been tried i e NO60 ( 6 eo's) and N080(8 e ' ) . - . 11 and 13 ethylene oxide units respectively.73 Line D PIT "C ' 60 Line C Line B 50 40 Line A 30 20 10 20 30 40 50 Volume I hydrocarbon Figure 9: Variation of PIT with phase ratio. about 30C and 70-75OC respectively at a concentration of 50 g /litre in the same brine/hydrocarbon system.(1980).alkanes. Where a surfactant contains a spectrum of eo contents. Results and Discussion The results presented in Fig 10 show a linear relationship between PIT and the number of eo units per molecule. 1965). with 30 g NaCl/litre brine containing 50 g of surfactant/litre. 6. os. The change The findings are in agreement with those of Bourell et a1. selective adsorption by the reservoir rock may change its effective eo value and thus affect the PIT of the system..T110 and T130 which contain (nominally) 8. The effect of the number of eo units is greater with the Arkopal series than the Sapogenates. The surfactants used were Sapogenate T80. The deviation from linearity above eo 11 is probably due to evaporation of the hydrocarbon during the test. Intermediate eo contents were obtained from mixtures of the adjacent surfactants as supplied and were calculated on a molar basis. 10. Variation of PIT with Ethylene Oxide (eo) content of surfactant - Tests PITS were measured using a mixture of pure normal .TlOO.

** *' *I Crude EACN =10 ' Results and Discussion The results obtained are shown in Figs 11 to 13. Measurements were made at various temperatures after heating . . Figs 11 and 12 indicate that a minimum does occur at the PIT but that more than one 'minimum' can occur.5 Results Fig 11 Fig 12 Fig 13 4: 1 4: 1 5: 1 . were usually found to agree within 2 5%. Repeat determinations of 1FT . Variation of IFT with temperature Tests Interfacial tension measurements were made between the upper and lower phases obtained from mixtures whose PIT'S had been determined in an attempt to confirm the presence of an IFT minimum at fhe PIT. . 1 . The equipment available only allowed for the transfer of phases into the tensiometer at room temperature.. Tests were performed with the following mixtures. Figure 10: Variation of PIT with surfactant ethylene oxide number 7. . I. This is supported by conductivity traces made during PIT measurements and is probably due to a proportion of surfactant having a different number of eo units than the stated nominal value. I. Hydrocarbons Phase ratio n hexane n octane PIT 39 49. ~~~ Surfactants T80 T80 NO60 Concentration 10 g /litre Brine 30 gm NAClf litre *..5 'Surfactant ethylene oxide number.5 23. The PITS obtained from conductivity measurements are included in the above table.74 PIT "C. 1. tion = 50 g - 7.

Oil : n-octane Brine : 30 g /litre NaCl Surfactant : Sapogenate T80. sampling and measurements should be carried out at the same temperature. . 38 42 46 50 54 Temperature OC Figure 12: Variation of IFT with temperature.75 dynes/cm Oil : n-hexane Brine : 30g /litre NaCl Surfactant : Sapogenate T80. 1Og /litre 30 34 38 42 46 Temperature. Ideally. from room temperature. equilbratlon. In order to make meaningful measurements the middle phase is separated from the remaining oil drop. ' c Figure 11: Variation of IFT with temperature. As one increases the temperature of the sample tube a third phase (microemulsion) begins to develop as the equilibrium is disturbed.log /litre. This was achieved with some difficulty especially in the case of colourless oils.

0 Equilibrium concn. by which time. Samples of reservoir rock were taken from cores and crushed in a ball mill 2 5 g of sample were taken and until the powder passed 180 sieve. l O g /litre lo-? 10-4 15 20 25 30 Figure 13: Adsorption of Surfactants Temperature .OC Variation of IFT with temperature Tests the results obtained for of Hostapal BV on ..g /1 Figure 1 4 : Adsorption of Hostapal BV Figure 15: Adsorption of Hostapal BV Onto reservoir rock.Sample 2. amount abstracted by the solids was calculated. The aqueous portion ws then decanted and centrifuged for 30 minutes.This section describesstates of disaggregation. .5 1 . onto reservoir rock. hours 40OC. Sample 1. adsorption work on Hostapal reservoir material in various Although BV was terminated. (because optimal salinity falls outside the range of our interest).A e o o 5 1 0.5 1. c.o 0.c.16 d y n e s . the results showed some of the-limitationsof static adsorption tests.g / I Equilibrium concn. equilibrated with 50 cm3 o various concentrations of Hostapal BV in The suspension was stirred constantly for 4 distilled water or sea-water. the supernatant liquid was clear. : 10-2 1 J Oil : Dead crude Brine : 30 g /litre NaCl Surfactant : Arkopal N 6 0 . r AdsorDtive soo51 I Adsorptive capacity. Analysis of this Hence the liquid then gave the remaining concentration of Hostapal BV.

0018 0.0080 1.25 0. Final (equilibrium) surfactant concentration.Results and Discussion The tests performed are listed below and the results presented in Figs 14 to 18. c.0046 0.5 1 .o Equilibrium concn..00048 0.0059 0.3 50 . 3. .0033 0.Sample 3.0078 0.50 10 .957 Adsorptive capacity A.164 0.033 0. g surfactantlg rock 0. er a .gr /I.00092 0.C g /litre 0.352 0. Table 2 shows a typical sets of results. " water 0. TABLE 2 (SAMPLE 3) Initial surfactant concentration g /litre 0.0036 0. 25 .184 0.75 20 . Figure 16: Adsorption of Hostapal BV onto reservoir rock.193 0.098 0.

003 I i I 0.18 Adsorptive capacity A.5 1 . sample 4. g lg .o Equilibrium concn. -4 - 1nA -5 - -6 - -7 - 1 I I I In c Figure 18: Data from Figure 16 plotted logarithmically . c. g 11 Figure 17: Adsorption of Hostapal BV onto reservoir rock..

There are indications in Fig 1 4 and 15 that adsorption may be proceeding in layers.79 All the figures show the tendency for A to tend towards a constant for a given Portions of Sample 3 (Fig 16) sample as the equilibrium concentration increases. Table 3 A g lg <180~ 'fine powder 'coarse grains' cm size pieces 0. It should then be possible to use cyclic hydrocarbons to lower the EACN of stock tank crude to that of reservoir crude for use in partitioning and phase studies at ambient pressure. however. A qualitative test of the effect of particle size on the equilibrium adsorption of surfactant was performed in a similar manner to those described above. CONCLUSIONS Phase inversion temperature can serve as a guide to the conditions under which a non ionic surfactant will give an interfacial tension minimum.0028 0. were also equilibrated with the surfactant in seawater and the results appear to Sample 3 has similar permeability and show a much higher adsorptive capacity. If the way in which the parameters which affect surfactant properties change during the course of a flood can be assessed it should be possible to design a flood which will maintain it's properties. requires detailed knowledge of the reservoir.0036 0. stock tanks crudes. Acknowledgements We would like to thank the British Gas Corporation for their permission to publish this paper. . and aliphatic and aliphatic cyclic hydrocarbons. More useful data would be obtained from core flood experiments.0002 This would appear to limit the usefulness of static adsorption and calculation of a 'worst case' total adsorption capacity of a reservoir. This. The results obtained are shown in Table 3.0022 0. Using the linear relationship between EACN and PIT it is possible to assign an EACN to. porosity characteristics to the other samples. The curves have the same general form as the classical adsorption isotherms. A= adsorptive capacity A = Kc where c= final concentration K and n are constants or In A = In K + In C The data in Table 2 (sample 3 ) are plotted as In A against In C in Fig 18.

and Kosswig. 1 (1979). 2. Jour. Jour. pp 275-283. Somasundaran. and Wade.V. Phys. B. (1980). 68 (1964). S.. Adsorption of sulphonates on SOC.H.. Salager. M. Reid.L. pp 256-261. L. 4 (1967). pp 3485-3490. Pet. R.S. reservoir rocks. Tenside Det.Colloid Interface Sci. 12. A . K . anionic active detergents by two-phase titration. Determinaton of . value system to emulsifier selection. Phase behaviour of alcohol-hydrocarbon-brine mixtures. K . Schechter.80 BIBLIOGRAPHY . (1965).F. correlation for phase behaviour of nonionic surfactants. and Davis. - Comptes rendus du 5 eme Congress International de la Detergence. and Heinerth. emulsion and clould point in solution of nonionic emulsifier. August 979. pp 221-232.E. P. Balzar. pp 292-304. The correlation between phase inversion temperature in . E. 6 5. Barcelona. Bourrel.. and Hannah.. The phase inversion temperature as a criterion for the selection of survace active agents in the tertiary production of mineral oil. J. Pesheck.. The comparison between the PIT system and the HLB . Shinoda. Shinoda. H. Knickerbocker. K and Arai. Longman. W. Scriven.M.T. Chem.. Jour. G. pp 451-461. V W . M. C. Eng. Tenside Det. D..

CHEMICAL FLOODING 81 THE PROVISION OF LABORATORY DATA FOR EOR SIMULATION C. In this paper we will limit our discussions to the topic of laboratory core data. and its extension to low tension immiscible displacement assessed. since this often forms the basis for enhanced oil recovery methods. There are many problems on the way from core tests to reservoir performance prediction. Miscible flooding will not be considered. We will work bn the principal that if successful6. .efficient displacement cannot be obtained from a core sample. BP Research Centre. we will review the more conventional waterflood case. 2. The current status of these concepts is discussed with reference to oil slug propagation along a core. Exploration and Production Division. to use these data for the prediction and analysis of field trial performance. and how this may be used to examine the physical processes involved in oil recovery. and ultimately for the prediction of full field performance. and how the predictions are affected by the relative permeability input data. immiscible floods only. Sunburyen-names Abstract Laboratory core tests are important in the development and assessment of EOR processes. A brief discussion of methods of assessing potentially useful surfactant systems and possible artifacts associated with core tests on the laboratory scale is included. BROWN and G. E. 1. LANGLEY Petroleum Engineering Branch. It is vital that the core data obtained characterise the physical processes relevant to the field. Laboratory displacement tests are usually confined to one dimension. Waterflood tests on core samples can be used to gain information on the efficiency of displacement of oil by water from actual reservoir rock. In this paper we will discuss low tension. Introduction Our overall aim is to obtain data from laboratory tests on core samples. Laboratory core waterflood tests Before considering low tension flooding. The key parameter of relative permeability is discussed. and are appropriate to their use in field simulators. 0. one of the biggest being reservoir description and identification of reservoir heterogeneity. We will consider the theoretical aspects associated with oil bank propagation. then there will be little chance of obtaining success on the field scale.

residual oil is trapped behind the waterflood front as insular globules of oil usually occupying the larger pore spaces (1). the relationship being influenced by pore structure. at high fractional flow of water. It assumes that it is valid to extend the Darcy equation for single phase f l o w to the multiphase case. Relative Permeability The basic concept of relative permeability is limited to movement of t w o o r m o r e continuous phases in the same direction through porous media at a steady saturation level. For strongly oil-wet rock the displacement does not appear to be efficient. The scope for tertiary oil recovery may be high in this case. and large amounts of oil are recovered after water breakthrough. This residual oil level is dependent on the initial oil saturation. The concept of relative permeability was introduced to fulfil this role. Early breakthrough of water often occurs. The contrast in the oil-wet and water-wet case is shown in Figure 1. However. which has now become central to conventional numerical simulation of oil recovery processes. 1 I n order to predict waterflood performance.82 For stron&ywater-wet rocks the efficiency of displacement is good in the sense that the displacement appears piston-like. The scope for tertiary oil recovery may be' low in this case. FIG. . practically all of the oil is recovered before breakthrough of the flooding water.. 3. it would be convenient to generate a data set which could be used for the prediction of flood performance and which is a unique property of the rock. However. Residual oil is less well defined in that oil is by-passed rather than trapped and occupies smaller pores and surface grooves. oil may continue to be produced until very low oil saturation is obtained ( 2 ) .

using the Johnson. . It is noted from Figure 2 that the relative permeability to oil increases quite rapidly.the oil-wet case Parallel arguments apply to a waterflood of an oil-wet rock.83 The concept is most likely to hold in the case of a sample which has initially 100% saturation of the wetting phase and where the saturation of the non-wetting phase is increasing. so sw 3. the waterflood performance (and hence the relative permeability obtained from waterflood data). the relative permeabilities should not depend on interfacial tension. Moreover. Provided that capillary dispersion and end-effect problems are not involved. Relative permeability data can be generated either by a steady-state method (where both phases are injected simultaneously into the core sample and permeability measurements are made at a steady-state fractional flow).g. Figure 3 shows typical relative permeability curves for the oil-wet case. Figure 2 shows typical relative permeability curves for an oil flood of a water-wet rock. Bossler method (3) based on the Buckley-Leverett theory ( ) . The sum of the relative permeabilities througbuttheir range is very much less than unity indicating interference between the phases. provided that the phases remain continuous and immiscible then. even though the recovery performance may be significantly different.1 Waterflooding . or by analysing flood data (e. 4) Some validation of the relative permeability concept comes in two ways: 1) 2) The steady-state and flood derived data are similar Data produced fr6m floods using different viscosity ratios give data which lie on the same curve. does not appear to depend on flood rate ( 5 ) . and that to water drops rapidly. in concept.

apart from end-point data. and might not agree with that obtained from a flood. This view often results from the study of miscible or partially miscible displacement tests where diffusion processes act to distribute the fluid components equally over the pore structure of the rock. Possible shapes of the relative permeability curves are shown in Figure 4. Theoretically. but the data can vary depending on the test method. Perhaps a true steadystate test should avoid such saturation gradients in the core sample. tending towards straight lines where the sum of the relative permeabilities is unity at all saturations ( 6 ) . the basic concepts of relative permeability are not strictly adhered t o . As mentioned earlier. . The residual oil values obtained by the two steady-state methods are not necessarily the same. and that endeffects are negligible. the flood efficiency can be improved by straightening the oil relative permeability curve to reduce the rate of change of fractional flow of water. This situation can be approached by gradually changing the fractional flow over a long period of time. Reduction of interfacial tension will therefore have little effect on residual oil. the recovery performance of a waterflood on a water-wet core usually appears piston-like. As mentioned earlier. More experimental work is neede& before definite conclusions can be drawn as to which relative permeability behaviour is relevant to low tension immiscible displacement. Steady-state tests using miscible systems are likely to produce straight line relative permeability data since a mixture of the components slows in all conductive flow channels. Support for this view comes from the fact that increasing the capillary number (VL/y) has little effect (assuming constant viscosities) on recovery performance. Diffusion processes have the effect of improving the displacement efficiency. principally because the non-wetting oil phase does not remain in communication. A common method is to change the fractional flow in steps which results in a saturation front travelling down the sample.2 Waterflooding . provided that enough water has passed through the rock. in any case it is very low. an alternative viewpoint for immiscible systems is that the oil drainage relative permeability curves will not change as interfacial tension is reduced. In addition. Therefore no relative permeability data can be generated using conventional analysis of the flood performance. 3. provided that capillary dispersion and end-effects are negligible ( 5 ) .a4 It is often suggested that the relative permeability curves will change shape as the interfacial tension is lowered. This would require that the oil recovery performance is independent of interfacial tension.the water-wet case For the case of waterflooding of a water-wet rock. Relative permeability curves can be generated from steady-state tests. for oil-wet systems residual oil saturation is not a well defined quantity. irrespective of interfacial tension. The conceptual reasoning for the argument that relative permeability does not change as the capillary number increases is that the displacing fluid will preferentially occupy pore channels with higher flow capacity. The resulting relative permeability curves for these miscible displacements are pseudo-functions with little predictive capacity.

but this is an artifact of the model and is usually overcome by empirically altering curves of type 1 to be more like type 2 . the sum of the relative permeabilities approaching unity at all saturations. As the interfacial tension is lowered during steady-state tests it is more likely that an emulsion of oil-in-water or water-in-oil will form. there may be differences of predicted flood performance when using coarse grid finite difference numerical models.8).1 mN/m large changes in the water-wet curves occur. Data obtained by the steady-state method (9) indicate that below an interfacial tension of 0. it is to be expected that significant changes will occur. thus suppressing viscous fingering on the reservoir scale. The tendancy is for the relative permeability curves to become straight lines. It is noted from Figure 4 that the relative permeability to water at residual oil saturation is low.85 The shape of the relative permeability curves may be considered to be academic in the water-wet case. 10 . The absence of imbibition in the case of very low interfacial tension can also make a water-wet sample look oil-wet. The truth of the matter is that we do not have a satisfactory theory to combine viscous and capillary flow effects. this can be advantageous since water mobility following the flood front is kept low. The distribution of residual oil is also dependent on flood rate (1). The residual oil level is dependent on initial oil saturation and on flood rate. indicating that trapped oil occupies the largest pore channels. since both sets of relative permeability curves shown in Figure 4 will probably predict plug displacement if the Buckley-Leverett theory is used. In addition. The residual oil saturation obtained after a waterflood is a definite value. From the point of view of a field flood. in that oil flow completely ceases (unlike the oil-wet case). However. 5 krw sw The residual oil saturation is also dependent on interfacial tension. high flood rate can make a weakly ' water-wet rock appear oil-wet (7. . FIG. Considering the changes to the water-wet relative permeability curves as the interfacial tension is lowered. and we usually resort to some form of pseudo functions to match observation.

Figure 5 ) . It has been suggested that the secondary drainage curves retrace the primary imbibition curves (10. As the interfacial tension was lowered the relative permeability to water increased and that to oil decreased.01 mN/m) the relative permeability curves resembled oil-wet data. it is the mobilised oil which collects and mobilises residual oil at the front of an oil bank. Theoretical predictions using this type of data will be discussed in the next section. If we work on the principal that the interfacial tension at the back end of the oil slug is low enough to prevent retrapping of oil. Application of laboratory data to enhanced recovery prediction Reduction of residual oil saturation is. as trapped oil is mobilised (curve B . This supports the proposal that oil drainage curves could be used as a first approximation for the back end of an oil bank even for a water-wet rock. At low interfacial tension (0. ii) Having established a residual oil saturation at low values of V/y we can increase V/y by increasing V or decreasing The residual oil saturation will eventually decrease y. Prediction of oil flood performance obtained using steady-state relative permeability data can be compared to oil flood data starting from residual oil saturation after the primary waterflood. since it will depend on how the primary imbibition curves were obtained. Figure 5). 4. then as suggested earlier we might use oil-wet type relative permeability curves to describe fractional flow at the back end of the bank. Curves of residual oil saturation versus V/y viscosity) can be generated by two methods: (for constant i) Where each flood starts from the same initial oil saturation (curve A.86 As the interfacial tension decreases to very low levels the drop size is likely to become extremely small. Surfactant at the back of the bank prevents retrapping of the oil. We need to consider generation of secondary drainage relative permeability data.21). 11. In this situation straight line relative permeability data might be expected. Relative permeability data obtained from displacement tests have shown somewhat different results ( 9 ) . . although the basic concepts of relative permeability in fact no longer apply. Once an oil bank is developed. starting from the residual oil saturation left after primary imbibition. The relative permeability set would then look like those shown in Figure 7. a necessary but not sufficient condition for a successful tertiary recovery flood. Curves of the type shown in Figure 6 can be generated from steady-state tests. for application at the leading edge of the oil bank. A situation will be approached where the same fluid system is flowing in all conductive flow channels. Although many people have looked at the problem of generating relevant relative permeability data for low tension floods (13 . It may be essential to develop and maintain an oil bank. 12) but this needs confirmation. there is still a need to obtain more data to clarify the position.

a7 FIG. 9 Water Arrival . 6 s w AT WATER BREAKT!lROUGH FIG.

i. it cannot overshoot this point. unless simplified systems can be studied. The front of the oil bank progresses with a well defined front. The corresponding oil slug behaviour as simulated is shown in Figure ll(a). certainly the correct analysis of such phenomena is crucial to core testing. For each simulation run. a 0. The corresponding fractional flow curves are shown in Figure 1 ( ) 0b. The following outlines some features of oil bank propagation in two component. However. This result is as expected. as can be achieved with Rankine-Hugoniot shock front criteria as employed in a simple fractional flow analysis. the principal of an oil bank stabilised by relative permeability hysteresis can be demonstrated by simulating injection of an oil bank followed ld. divided (22). This result highlights the sensitivity of oil bank propagation behaviour to the relative permeability curves. in this the oil bank is not recognisably propagated since it is quickly dispersed. They also show that numerical dispersion is not dominant in the present examples. but the size of the bank degrades as oil is 'lost' into the following tail.88 5. OIL BANK PROPAGATION In EOR processes the development and propagation of an oil bank is considered to be of importance. without invoking other mechanisms. two phase (oil/water) displacement in cores as given by 1-D numerical simulation. Use of the same relative permeability set. analysis can only be carried out on the basis of assumptions which are difficult to validate. This is shown in Figure l ( ) The importance of an oil bank induced by hysteresis effects extends not only to the valid identification of "enhancedt1 oil. This can be compared with the experimental results and Buckley-Leverett analysis of Gladfelteqand Gupta The test system simulated consists of a 45 cm core. . For highly complex EOR systems. but with the hysteresis directions reversed gives the behaviour shown in Figure ll(b). the presence of a transient bank may in fact involve changes in the direction reverse to that overall. since the numerical model moves incrementally up the relative permeability curves to the imposed injection composition. Curiously little work has been reported which attempts to do thin A recent paper by Gladfelter and Gupta ( 2 2 ) is valuable. injection of an oil and water mixture (in the present case.e. but without the presence of a bank of increased oil saturation as-experimentally observed. The additional feature proposed to explain this behaviour was hysteresis of the relative permeability curves. in that it sets some experimental evidence against which current views of oil bank propagation may be weighed. The relative permeability hysteresis set first used was as shown as curves A and B in Figure lO(a). with the arrows indicating which limb corresponds to which directional change in water saturation. by injection of an oil/water mixture. 25% oil) into a core at residual oil saturation gave the results depicted in Figure l ( ) lc. showing a series of oil bank profiles as it progresses along the core. into forty grid blocks. One of the key findings is that a region of increased oil saturation (an oil bank) can be generated from the fractional flow properties. a sharp oil front is formed. with an oil/water ratio of 3:l.0475 pore volume slug of oil was injected at high rate into the core. but also to the correct assignment of water-increasing or water-decreasing steady-state relative permeabilities. Following the procedure of Gladfelter and Gupta ( 2 2 ) . which was initially set at residual oil saturation.

89 fw 0.3 .0 .s c (x/L) 1.-- 0.

90 .

If the immiscible displacement of oil by surfactant acts as a high rate water flood as suggested earlier (since in both cases capillary forces are dominated by viscous forces) the aqueous phase will preferentially travel through the larger pore channels. those curves which originate at initial conditions of complete saturation by a single phase.e. IRREVERSIBILITY AND HYSTERESIS IN CORE TESTS In this section some modelling approaches to describe non-identity are outlined. in considering the Buckley-Leverett theory. with an increased krw. the low tension flood does not have the imbibition which in a water-flood case ensures the oleic phase preferentially occupies the larger pores of a water-wet medium. Jones and Rozelle (23) consider that irreversibility results from the common "S" shaped fractional flow curve if the conventional tangent constructions are applied to waterflooding and oil flood respectively.1 The best documented tlhysteresisl' effects in relative permeability curves are associated with the irreversibility of the primary curves i. using curves A and C of Figure lO(a). However. straight line extrapolations) may artificially predict stable bank propagation.e. The use of assumed Ilideal" relative permeability data (i.e. This hysteresis gives rise to residual (or irreducible) phase saturations. However. Core tests can display a wide range of irreversible characteristics. i. This could result in a relative permeability behaviour as shown in Figure 7. It is commonly assumed that essentially irreducible saturations can be established for both . it can be noted that two solutions to the problem are to be expected.91 More drastic disparity between the relative permeability curves used at the leading and trailing fronts of an oil bank can of course give stabilised bank propagation without an oil/water mixture being subsequently injected. l ( ) lf) once more show degradation of an injected oil slug. The straightening of curve A to give curve C follows the commonly assumed functional change in the relative permeability curve induced by reducing the interfacial tension. The simulation results for this case (Fig. 6 . but with a kro curve falling below the prior curve at high oil saturations. experimental verification of the relative permeability behaviour in eor systems is still limited. and show that under certain conditions oil bank propagation will not occur. and the assumed curves may not offer a good description of the physical processes involved. 6. A more fundamental problem is whether the fractional flow curve itself is a unique function of saturation. These simulation runs indicate the need to take proper account of relative permeability variations in the analysis and simulation of transient processes. since the material balance may be applied for either the waterflood or the oilflooding directions. Figure ll(e) shows this. The conclusion that the direction appropriate to the particular stage of an EOR process must be used still holds. RELATIVE PERMEABILITY HYSTERESIS Primary Hysteresis 6.

swi) Soi/(l. In the absence of a scientific approach.Swi) swi with C being a constant for a given system. For strongly wetting phases. empirical relationships are generally developed on the basis of plausibility and utility. although the distribution and properties of these phases are by definition functions of wettability. A currently favoured parameterisation of relative permeability curves uses a scaled power law relationship. 6. This region is (in principle) fully accessible reversibly. Equations of this form have been used by Killough (24) to estimate hysteresis sets descending from the primary non-wetting relative permeability curve. the Land relationship (taking water to be the wetting phase and oil the non-wetting phase) is + (Sor) -1 = = + - + (Swi) -1 = c (1) where Sor + Sor/(l. In fact.92 wetting and non-wetting phases. The irreducible water saturation and the residual oil saturation values are expected to be dependent on the initial saturation established prior to reduction to residual. and for relative permeability curves the lack of experimental precision and the probable dependence of data on experimental design (25) have so far prevented detailed analysis. Thus it is usual to treat wetting and non-wetting phases asymmetrically. kro krw where = kro(Swi)*(S. so this effect is of little consequence. it is only within a reversible range that the term hysteresis can be truly applied. the residual saturation tends to be ill-defined.)"o krw(Sor)*(S.2 Secondary Hysteresis The term secondary refers to those curves which start from the end points of the primary curves. For the non-wetting phase. The characterisation of any hysteretic system can prove complex. It may be questioned whether the initial formulation of the multiphase flow relationships in the empirical Darcy-analogue form itself ensures that all subsequent analysis can be no more than empirical. that is in the saturation range Sw = Swi to (1-Sor). Land (12) has proposed a semiempirical relationship which correlates initial and residual non-wetting phase saturations.)"w = I So * = = (l-Sor-Sw)/(l-Swi-Sor) (Sw-Swi)/(l-Swi-Sor) * sw .

For surfactant flooding. A hysteresis parameter can be defined as and similarly for the water exponents. The variation of the relative permeability curves for an EOR process requires detailed description if a numerical simulation of the process is to be made. as used by Evrenos and Comer (26) variation of the with an alternative different parameterisation scheme. both types of behaviour being reported from experimental observations. 2 FIG. 12 . It can be noted that dependent on the value of Hi the hysteresis scan can occur in either direction. where the superscript arrows denote increasing and decreasing water saturation. it is to be expected that the hysteresis becomes less pronounced as the residual Sor = 0 .93 The secondary hysteresis can then be conveniently described by exponents. the paramaterised relative permeability curves are often modified in a systematic manner dependent on the new value of residual oil saturation (Sorc). Following this procedure with the hysteresis parameter.

Unfortunately.n - Sorc Sor (5) 6. this can at best be an approximation. Attempts to correlate primary capillary pressure curves with fractional flow curves have not to date proven sufficiently satisfactory to enable mapping of scanning curves from capillary pressure to fractional flow data. ( Sorc 1 a 1 . However. In order to estimate how scanning from intermediate points of the secondary bounding curves will progress in the absence of definitive experimental evidence. The most general theory available to describe hysteresis systems is Independent Domain Theory. as propounded by Everett ( 7 . In addition. the existence of domains and their independence. Enhanced oil recovery processes aimed at mobilisation of post waterflood residual oil necessitate changes in the forms of the relative permeability curves and in the directions in which a process description scans them. and the fractional flow can be considered to be a state variable which offers more hope of comparability with other hysteresis phenomena. but does provide a qualitative framework in which empirical relationships can be set. the commonly made approximation is also present in the comparison of equilibrium capillary pressure values with dynamic systems. The most direct consistency relationship between capillary pressure scans and fractional flow scans would be if one set were directly mapable onto the other. Paralleling other dependencies on Sorc. Topp and Miller ( 2 9 ) ) .Hi ( Sor I Two of a family of relative permeability curves using this relationship isshown in Figure 12. or more simply a power law expression. The other hysteretic process which is well established in petroleum engineering experience is the capillary pressure as a function of saturation. the theory has not proved quantitatively 2) applicable (due to the prime assumptions i. A less direct appeal to consistency can be made by application of similar functional forms to describe both capillary pressure and fractional flow hysteresis sets.Hi (Sorc) = (4) (1 1 .no (Sorc) 1 . . .94 oil saturation is decreased (since any conceptual model of the hysteresis mechanism centres on irreversibility following loss of hydraulic connectivity during phase trapping).3 Scanning Hysteresis Conventional oil recovery processes are usually described in terms of monotonic changes of saturation. since the fractional flow setis rigorously bounded whereas the capillary pressure set is asymptotically bounded. it is to be expected that mutually consistent descriptions should be possible.e. with the dependence of the no exponent on Sorc being 1 . Since the same physical processgives rise both to relative permeability 2nd to capillary pressure hysteresis. the relative permeabilities need to be expressed as a single variable. parallels can be sought in other hysteretic systems. In order to compare the two sets of phenomena. a logarithmic relationship between Hi and Sorc can be used.

these studies have concentrated on liquid/vapour systems of interest to hydrologists rather than on liquid/liquid systems and the hysteresis relationship examined has been between the saturation and flow potential.: + a. this equation is not convenient in that the relative permeability hysteresis set cannot be explicitly separated from this representation.(fi .)/(fi+fj) J J . and f. . However. A more flexible approach is to follow that which Killough ( 24) applied to capillary pressure hysteresis.-l/a S )' .F(kri . are fitting parameters which can be identified from sufficied data.f . Extending this to the relative permeability and fractional flow sets gives relationships of the type c where F where the subscript refers to a scanning curve from point Si. and scans within those curves which accord with the expected hysteretic behaviour as described by independent domain theory.)/(fi+f I . . if wf kri 0 = I cosh(Sw/S1)a cosh(SJS')a .(fi .l/a (Simax . kri fi (Si . However. f.F(fi 0 + + + (7a) (7b) + fi .kri) . a.Si + a)- 1 .95 In recent years a number of studies have been carried out in which unsaturated flow in porous aedia has been analysed numerically using empirical analytic hysteresis functions.fi) . The functional form of this relationship can reproduce fractional flow curves.. by which scanning curves are described by a scaled combination of the bounding curves. = = = . Applying an analogue of the empirical hysteresis function as recently used by Pickens and Gillham (29) to the fractional flow/saturation system gives (6) J where S'. and a is a fitting parameter. the superscript arrows refer to the bounding curves. Examples are shown in Figure 13.f.) .

The recovery of oil from cores during laboratory waterflood tests is affected by capillary forces on the sample scale. such as core end effects and oil saturation residual to surfactant flooding. This is essential to ensure that oil left in the core is "trapped oil" and not oil retained by core end-effects. the next step is to conduct core tests. 1 ft/day advance rate) to establish a residual oil saturation.96 The above discussion highlights the empirical way in which relative permeability hysteresis can be handled at present. Our present procedure involves the flushing of live crude oil through the core at reservoir temperature. the wetting conditions of a core sample play a large part in the displacement characteristics.1 Any potentially useful surfactant system must be able to reduce interfacial tension between oil and water by a significant amount (e. down to lo-' mN/m). It is therefore of considerable importance to march reservoir wetting conditions when conducting core tests. In order to do this we use preserved core material which is carefully prepared in order to change the surface wetting as little as possible. If successful. As we have discussed. - - 7. Surfactant is then injected into the core at 1 ft/day. Before injecting surfactant solution.2 Core scale artifacts It is well known that laboratory tests on core samples can produce misleading information due to the small scale of the core plug in relation to reservoir scale. important factors. the brine flow rate is increased to 200 ft/day. This is so even if the core plug represents a homogeneous reservoir rock. The core is then left to ltageItin contact with crude oil for several weeks. . Careful monitoring of the recovery performance enabled theoretical predictions to be tested. The flood front is dispersed by capillary forces (Figure 8). the position is even less well defined . 7. tests on short cores result in early water breakthrough (5).g. For oil-wet cores flooded at low rate. Having found a surfactant capable of doing this. The theoretical work may show up reasons for lack of success in core flood tests or possible reasons why successful core tests may not necessarily lead to a successful field test. DISPLACEMENT TESTS FOR ASSESSING POTENTIALLY USEFUL SURFACTANT SYSTEMS Screening tests 7. the next step is to conduct screening tests using glass bead or sand pack columns. A low rate waterflood is then conducted (e.g. For other.

However. If a low rate flood test is carried out on a core sample.e. 7. A similar situation arises in the case of residual oil saturation (i. In particular.97 If the core length or flood velocity is increased or the interfacial tension is decreased. We have also considered an ideal situation where clays within the porous media do not'effect the fluid flow characteristics. after sufficient water has been passed through the core so that oil flow ceases). The representation given in Figure 9 only applies over a small region of flow rate and interfacial tension. viscous forces will start to take over as the dominant factor controlling fluid distribution. together to create a long core. In order to improve the situation. then the oil recovery at water breakthrough may show an increase as the interfacial tension is decreased. Turning to the water-wet case. In our discussions we have tended to consider the extreme case of oilwet and water-wet rocks. where capillary forces completely dominate fluid distribution on the pore scale. The above discussion highlights the importance of careful interpretation of laboratory data if valid information is to be produced. to reduce the influence of end-effects. Increase of sample length. Careful is necessary in order that relevant data long cores are sometimes used Small core plugs may be butted arrangement of the small plugs is produced. particularly if the flood is carried out at rates representing reservoir flow rates. Adsorption of surfactants on to pore and clay surfaces has not been discussed. according to the curves in Figure 8. on the reservoir scale.g. . In practice rocks may have a variety of wetting conditions.3 Simulation aspects Preliminary analysis of core tests of EOR systems can be carried out with linear scaling relationships. An additional role of the simulation of core floods is provided by the greater degree of experimental accessibility. Water may arrive early at the outlet end of the sample but capillary forces delay water-breakthrough (Figure 9). In all cases the data must be considered in terms of its relevance to its intended use on input to a field simulator. Indeed not all of the pore surface may have the same wetting. e. will have no effect on fluid distribution on the pore scale. then the oil recovery at water breakthrough is improved. Capillary forces cause retention of oil at the outlet face of the core sample giving the impression of high residual oil saturation. Investigation of the sensivity to particular system properties in the optiiisation of a system requires the use of more detailed simulation methods. Reduction of interfacial tension can result in significant production of this end-effect oil which would not be representative of the reservoir scale. Above a certain flow velocity or below a certain interfacial tension. the separation of core scale effects as discussed above demand careful study if the experimental results are to characterise the physical processes relevant to the field. however. the recovery of oil at waterbreakthrough is less affected by sample length than in the oil-wet case (8). with the attendant changes in scale which can only be treated mathematically. Full consideration of these interesting effects is outside the scope of this paper. the reduction of interfacial tension would have no effect. to match reservoir type flow rates.

Despite the benefits of validation of a simulation model on the basis of laboratory results. List of Symbols kr S f V y L x relative permeability fractional saturation fractional flow superficial velocity interfacial tension core length distance along core it a. there is no guarantee of success in the field (although. irreversibility and system changes is of great importance to the correct modelling of an oil bank. great care must be exercised when attempting to extend its area of applicability. S+. 3. 0. The authors wish to thank Mr. defined in the text. Ltd.98 and the ability to repeat laboratory work. S . Berry and Dr. The simulation provides the language by which the dialogue between the laboratory and the field testing can proceed. conversely. C. as contrasted with the field situation. they cannot be ignored in the provision of data from core tests. complete with hysteresis. J . F . carried out the simulation work described in this paper. The availability of a correctly defined set of relative permeability curves. they require development and experimental validation. n. Although in field use. the concept of relative permeability remains empirical. F. I White who . the relative permeability functions which are appropriate to low tension immiscible displacement are as yet uncertain. 2. CONCLUSIONS 1 . such concepts may be of secondary importance to the heterogeneity. fi. . Numerical schemes which essentially scan relative permeability curves demand that these curves are defined over the full saturation interval. The behaviour of an oil slug injected into a core provides a prototype to assess many of the assumptions which are present when core data are prepared for simulation of EOR processes. Acknowledgements Permission to publish this paper has been given by the British Petroleum Co. As with any empirical quantity. failure at the laboratory scale cannot easily be reconciled with confidence in subsequent field application). In particular. the behaviour of transient saturations associated with multiple shock fronts may not be adequately modelled. - parameters. Despite its extensive use. Currently available simulation schemes are empirical. and it is therefore important that it should be able to describe the real physical processes rather than aim at mere plausibility and utility. The sensivity to these curves of oil bank stability (and presumably all multi-front systems) is such that plausibility alone is too weak a criterion for-acceptability. Hi. In addition.

3 C. JPT. 7.116. Eng.R. and Leverett. Land. J. M. Eng. 13* 1 4 15 * .G. 9.P. The effect of interfacial . 107 . Sept. Eng. tensions on relative oil-water relative permeabilities of consolidated porous media.425. Mechanisms of fluid displacement in sands.E.4 .L. SPE preprint 5810. 370-372. Bossler. page 99 .C.The reservoir engineering aspects of waterflooding. L. Craig Jr. Rapoport and W.L. 3.A. 1958. Pet. Pet..A. Experimental and calculated relative permeability data for systems containing tension additives.1 0 1. SOC. Sippel .F. 1971. Trans AIME (1958) 213. 12. D. J. 36 .148.99 Subscripts o w r i oil water residual irreducible References 1 .A.F. 423 ..A. 139 . 8. Oct 1980 page 391. R. Trans AIME (1942) 146. Owens. Trans AIME (1959) 216. and Rapoport L. W. Trans AIME (1953) 198. E. Gournay and R. J (Dec 1971) 419 . SOC.426. N.S. Pet. L. Parrish and R.. SPE Preprint 9783 T. Trans AIME (1939) 132. Trans AIME ( 9 1 .R. F. 0 11. spe Monograph Vol. The effect of fluid-flow rate and viscosity on laboratory determinations of oil-water relative permeabilities. 3 H. Kyte J. 247 . Leas Properties of linear waterfloods. D. SOC. vol. C R Sandberg.. Bardon and D. Handy and P. Datta Fluid distributions during immiscible displacement in porous media. J.S. 1.(1966) 6. Eng. C. Calculation of relative permeability from displacement experiments. Longeron Influence of very low interfacial tension on relative permeability. 5. 1966 page 261. Pet.O. J. 6. Comparison of calculated with experimental imbibition relative permeability SOC. Mungan Interfacial effects in immiscible liquidliquid displacement on porous media. 2.C. M. V O .W. Linear waterflood behaviour and end effects in water-wet porous media. Leverett Flow of oil-water mixtures through unconsolidated sands. 1 9 4. L.J. Page 24. Amarefule and L L Handy 1981.F. Johnson. SPE preprint 4104.. Doherty Series. Buckley. S. Geffen.253. Talash 1976. Morse Experimental investigation of factors affecting laboratory 15) relative permeability measurements. 4.W. A. and Naumann. 192. Salathiel Oil recovery by surface film drainage in mixed wettability rocks.M.

Effects of Fractional Flow Hysteresis on Recovery of Tertiary Oil. S.817. 17. Application of a general multiphase multicomponent chemical flood model to ternary. Hysteretic Moisture Characteristics and Hydraulic Conductivities for Glass Bead Media.48. G. and Rozelle W. SPE 2693. A. and Gillham. 1975. 1981. Killough.G. Trushenski Micellar flooding compositional effects on oil displacement.E.C. D. Canada (June 13-16. and Slattery.C. of 9th World Pet. Topp. Reservoir Simulation with History Dependent' Saturation Functions.W. Pickens. Three-Dimensional. 1976) 37 .1078. 27 * 28. SOC. J.100 16. G. 19781. Randomised Network Model for Two Phase Flow Through Porous Media.K. two-phase surfactant systems. Batycky and F. Asar Influence of interfacial tension on gas-oil relative permeability in gas-condensate systems. 1980) 508 . 1969. Numerical Simulation of Hysteretic Flow in Porous Media. J. A General Approach to Hysteresis. Everett. Calgary. Gupta and S. SPE 9803. SPE J (Feb. Pet. SPE J (Dec. Congress.R. W. H. Trans.D. and Miller.A.162. Soil. SOC.P. SPE J. Conley. .O. PhD dissertation University of Southern California (May 1980) C. S. Gilliland and F. Gladfelter. and Gupta.C. McCaffery Low interfacial tension displacement studies. Surfactant waterflooding Proc. (1954) 50.P.202. 24 * 25. Proc. J. (1966) 30. 22. Faraday SOC. Sci.I. Eng. 156 . F. Res. 29th Annual Tech.P. R. 26. and Comer.P. Thomas.E. C. 16. Tokyo May 11-16.E. 187 ... (May 1978) 807 .Y.H.128. Graphical Techniques for Determining Relative Permeability from Displacement Experiments.520. E. J. Lin. Finite Element Analysis of Solute Transport Under Hysteretic Unsaturated Flow Conditions Water Resour. Eng. paper 78-29-26. S.W. SPE preprint 6727. SOC.1071 . J. (June 1980) 191 .E. and Smith. Evrenos.G. Amer. 20. 21. 18.P. Pope The application of fractional flow theory to enhanced oil recovery. of CIM. SOC. (1980) .197. H. Flemings 1977. Pet.F. Jones. Winter and P. J April 1979 116 . R. meeting of the Pet. A.

development of a p r a c t i c a l enhanced oil recovery (EOR) technology depends on how much surf a c t a n t can be economically s a c r i f i c e d i n recovering a d d i t i o n a l crude o i l from a r e s e r v o i r .l i q u i d i n t e r f a c e . An example of how experimental cond i t i o n s may a f f e c t t h e measured values of s u r f a c t a n t l o s s e s is shown by t h e e f f e c t of s u r f a c t a n t s l u g s i z e on t h e apparent l e v e l of r e t e n t i o n and adsorption. The p r i n c i p a l o b j e c t i v e of t h i s paper is t o analyze and e v a l u a t e experimental procedures f o r determining s u r f a c t a n t a d s o r p t i o n and t o t a l s u r f a c t a n t r e t e n t i o n . Since s u r f a c t a n t s are mre expensive than t h e crude o i l . E f f e c t i v e o i l recovery by s u r f a c t a n t s is not a q u e s t i o n of p r i n c i p l e but r a t h e r a q u e s t i o n of economics. NOVOSAD Petroleum Recovery Institute CNgary. petroleum s u l f o n a t e (TRS 10-80). s u r p r i s i n g that a s u b s t a n t i a l r e s e a r c h e f f o r t has been devoted t o l a b o r a t o r y e v a l u a t i o n s of s u r f a c t a n t r e t e n t i o n i n porous media. Generally. It is not. c o s u r f a c t a n t . It i s shown that t h e i n t e r p r e t i o n of experimental d a t a is n o t s t r a i g h t forward.Alberta. water. e l e c t r o y t e . CaMda T2L 2A6 ABSTRACT T o t a l r e t e n t i o n of s u r f a c t a n t s i n a r e s e r v o i r d u r i n g chemical f l o o d s is probably one of t h e most important parameters i n t h a t i t determines the economic success o r f a i l u r e o f t h i s enhanced o i l recovery process. and s y n t h e t i c s u l f o n a t e (FA 400) are presented. and o i l . t h e e f f e c t o f . i t is q u i t e clear why s u r f a c t a n t a d s o r p t i o n has always been considered c r i t i c a l t o t h e success OT f a i l u r e of t h i s EOR process. Laboratory d a t a on r e t e n t i o n of p u r e s u l f o n a t e (Texas # l ) . t h e r e f o r e . an& no less t h a n f i v e components: s u r f a c t a n t . Therefore. INTRODUCTION Adsorption of s u r f a c t a n t s considered f o r enhanced o i l recovery a p p l i c a t i o n s has been s t u d i e d e x t e n s i v e l y i n t h e last few years1’6 as it has been convincingly shown that i t is p o s s i b l e t o develop s u r f a c t a n t systems which d i s p l a c e o i l from porous media almost completely when used i n l a r g e q u a n t i t i e s .CHEMICAL FLOODING 101 EXPERIMENTAL STUDY AND INTERPRETATION OF SURFACTANT RETENTION IN POROUS MEDIA J. . p h a s e behavior on t o t a l s u r f a c t a n t r e t e n t i o n fs discussed and experimental procedures are o u t l i n e d so that a d i f f e r e n t i a t i o n can be made between l o s s e s of s u r f a c t a n t due t o unfavorable phase behavior and t h o s e due t o a d s o r p t i o n a t t h e s o l i d . S p e c i f i c a l l y . These show t h e importance of experimental c o n d i t i o n s as similar experiments may y i e l d s u b s t a n t i a l l y d i f f e r e n t r e s u l t s when c o n d i t i o n s are v a r i e d . t h e systems s t u d i e d are complex as they c o n t a i n a minimum of tw l i q u i d phases. and that extreme c a u t i o n must be exercised before any i n t e r p o l a t i o n o r e x t r a p o l a t i o n of a d s o r p t i o n o r r e t e n t i o n data is attempted.

i t should not come a s a s u r p r i s e t h a t measured a d s o r p t i o n isotherms d i f f e r s u b s t a n t i a l l y from t h o s e p r e v i o u s l y observed f o r simpler s u r f a c t a n t systems. by d e f i n i t i o n . Numerous e x p l a n a t i o n s have been o f f e r e d f o r t h e s e peaks.102 It w a s recognized earlier t h a t physico-chemical a d s o r p t i o n may be only one of a number of f a c t o r s which c o n t r i b u t e t o t o t a l s u r f a c t a n t r e t e n t i o n . Other physico-chemical mechanisms may i n c l u d e s u r f a c t a n t entrapment i n a n immobile o i l phase5. a formation o f mixed micells6. no xo i x n o = -(xi m = = = 0 . such a s . a n a d s o r p t i o n maximum has been observed when t o t a l s u r f a c t a n t r e t e n t i o n has been p l o t t e d a g a i n s t t h e c o n c e n t r a t i o n of i n j e c t e d s u r f a c t a n t . f o r a d s o r p t i o n a t t h e s o l i d . P r i m a r i l y . I t has been shown r e p e a t e d l y t h a t . s u r f a c t a n t p r e c i p i t a t i o n by d i v a l e n t i o n s 6 s u r f a c t a n t p r e c i p i t a t i o n caused by a s e p a r a t i o n o f c o s u r f a c t a n t from s u r f a c t a n t ' . an a t t e m p t is made t o d i f f e r e n t i a t e between t h e a d s o r p t i o n of s u r f a c t a n t s a t t h e s o l i d . Also. S p e c i f i c a l l y . i t should be r e a l i z e d that most experimental procedures do not y i e l d a n amount of s u r f a c t a n t adsorbed b u t r a t h e r t h e s u r f a c e excess. Which of t h e s e e f f e c t s a r e r e s p o n s i b l e f o r t h e peaks i n a p a r t i c u l a r s i t u a t i o n and t h e i r r e l a t i v e impartance i s d i f f i c u l t t o asses.Xi) of component i i n t h e bulk phase b e f o r e (fraction). and t h e p r e c i p i t a t i o n and consequent r e d i s s o l u t i o n of d i v a l e n t i o n s 2 . i t seems t h a t . t o t a l mass o f t h e l i q u i d system ( g ) . of component i i n t h e bulk phase a f t e r is a t t a i n e d ( f r a c t i o n ) . I t i s shown n e x t that a n a d s o r p t i o n isotherm expressed i n terms o f t h e s u r f a c e excess a s a f u n c t i o n of a n e q u i l i b r i u m s u r f a c t a n t c o n c e n t r a t i o n must. i n view of t h e number of processes t h a t are t a k i n g p l a c e s i m l t a n e o u s l y and t h e l a r g e number of components p r e s e n t i n most o f t h e systems s t u d i e d . = . yrecipi. and s u r f a c t a n t t a t i o n due t o chromatographic s e p a r a t i o n o f d i f f e r e n t s u r f a c t a n t s p e c i e s When complications a r i s i n g from ion-exchange phenomena u s u a l l y involved i n s u r f a c t a n t f l o o d i n g are i n c l u d e d . r e l a t i v e concentration adsorption takes place r e l a t i v e concentration adsorption equilibrium i ne i m = excess of mass of component i i n t h e adsorbed phase p e r mass u n i t o f adsorbent (g/g). c o n t a i n a maximum if t h e d a t a are measured over a s u f f i c i e n t l y wide range of c o n c e n t r a t i o n s . t h e e f f e c t s o f " s t r u c t u r e forming" and " s t r u c t u r e breaking" c a t i o n s 8 . the s u r f a c e excess i s t h e only c o n s i s t e n t l y defined experimental v a r i a b l e which should be used i n d e s c r i b i n g t h e p r e f e r e n t i a l uptake of one component over a n o t h e r i n t o t h e adsorbed layer'. A p r i n c i p a l o b j e c t i v e o f t h i s work is t o e v a l u a t e t h e experimental techniques t h a t c a n be used f o r measuring s u r f a c t a n t a d s o r p t i o n and t o s t u d y experimentally two mechanisms r e s p o n s i b l e f o r s u r f a c t a n t r e t e n t i o n . b w e v e r . The s u r f a c e excess is defined by Equation (1)L: ne i where.l i q u i d i n t e r f a c e and 'the r e t e n t i o n of s u r f a c t a n t s due t o t r a p p i n g i n t h e immobile hydrocarbon phase which remains w i t h i n t h e c o r e following a s u r f a c t a n t flood . one should n o t expect smooth monotonically i n c r e a s i n g isotherms t h a t a r e p a t t e r n e d a f t e r a d s o r p t i o n isotherms f o r only two p u r e components.l i q u i d i n t e r f a c e . mass of adsorbent (g). PLUSCRMENT OF ADSORPTION AT THE SOLID-LIQUID INTERFACE Previous a d s o r p t i o n measurements of s u r f a c t a n t s considered f o r enhanced o i l recovery produced a d s o r p t i o n isotherms of unusual shapes w i t h unexpectel f e a t u r e s .

This problem is even more complicated when d e s i g n i n g experiments w i t h surf a c t a n t mixtures which are considered f o r s u r f a c t a n t f l o o d i n g as i n d i v i d u a l components of t h e mixture are d i f f i c u l t t o s e p a r a t e and. however. consequently. t h e i r e q u i l i b r i u m c o n c e n t r a t i o n s become d i f f e r e n t from t h o s e o r i g i n a l l y p r e s e n t so t h a t they w i l l depend on s p e c i f i c experimental c o n d i t i o n s such as t h e a d s o r b e n t / s o l u t i o n r a t i o . Equation (1) a p p l i e s a l s o t o multicomponent systems. However.103 and a n a d s o r p t i o n isotherm is d e f i n e d a s t h e s u r f a c e e x c e s s dependence on e q u i l i b r i u m c o n c e n t r a t i o n o f component i i n t h e b u l k phase: It should be clear from Equation ( 1 ) that t h e s u r f a c e excess must be equal t o z e r o f o r p u r e components ( x i = 0. c o n s i s t s of measuring . s o l u b i l i t y l i m i t s of most s u r f a c t a n t s i n r e s e r v o i r b r i n e s a r e reached a t low c o n c e n t r a t i o n s . x i = 1 ) and. This a p p l i e s t o f u l l y m i s c i b l e systems i n which a d s o r p t i o n isotherms are meaningful over t h e e n t i r e concentrat i o n range between pure component 1 and pure component 2 . It is not p r a c t i c a l t o perform a d s o r p t i o n experiments w i t h multicomponent systems i n such a way that t h e a d s o r p t i o n s u r f a c e is c u t i n a pre-determined way s i n c e i t i s n o t u s u a l l y known a p r i o r i w h a t t h e b u l k composition of i n d i v i d u a l components is going t o be a f t e r a d s o r p t i o n has taken place. i t has been p r e v i o u s l y shown that isotherms f o r i n d i v i d u a l components may d i f f e r s u b s t a n t i a l l y from a n o v e r a l l i ~ o t h e r m . t h e r e f o r e . The f i r s t . These are t h e main r e a s o n s why experimental d a t a from d i f f e r e n t l a b o r a t o r i e s are so d i f f i c u l t t o compare and why a g r e a t d e a l of c a u t i o n should be e x e r c i s e d i n i n t e r p r e t i n g t h e shapes of isotherms which were determined i n experiments i n which t h e i n i t i a l r e l a t i v e c o n c e n t r a t i o n of each i n d i v i d u a l component was held c o n s t a n t w h i l e t h e t o t a l c o n c e n t r a t i o n w a s varied. a b a t c h method. T r e a t i n g t h e mixture as a s i n g l e component b r i n g s a d d i t i o n a l u n c e r t a i n t y t o a d s o r p t i o n experiments. As t h e i n d i v i d u a l components a r e adsorbed t o d i f f e r e n t e x t e n t s . ~ This i s of importance f o r systems i n which each component w i t h i n t h e m i x t u r e s e r v e s a d i f f e r e n t f u n c t i o n . The apparent a d s o r p t i o n l e v e l may vary S u b s t a n t i a l l y from one experiment t o a n o t h e r a s e q u i l i b r i u m c o n c e n t r a t i o n s may be moving on t h e a d s o r p t i o n s u r f a c e i n d i f f e r e n t d i r e c t i o n s depending on t h e s p e c i f i c c o n d i t i o n s of t h e experiment. a presence o r a n absence of a maximum i n a n a d s o r p t i o n isotherm is dependent upon s u r f a c t a n t s o l u b i l i t y i n a b r i n e o r o t h e r continuous phase. METHODS FOR WXjUREMENT OF ADSORPTION ISOTHERMS There a r e t w o d i s t . The overa l l a d s o r p t i o n isotherm i s t h e n n o t s u i t a b l e f o r p r e d i c t i n g t h e system performance d u r i n g t h e f l o o d because d e p l e t i o n of i n d i v i d u a l components is not r e p r e s e n t e d by t h e o v e r a l l isotherm. a non-zero adsorpt i o n isotherm must c o n t a i n a t least one peak. Therefore. such adsorption isotherms cannot be expressed g r a p h i c a l l y i n a two-dimensional form without s p e c i f y i n g a d i r e c t i o n i n which t h e a d s o r p t i o n s u r f a c e ( f o r three-component systems) is c u t f o r viewing i n two dimensions 4.l i q u i d i n t e r f a c e . i n c t l y d i f f e r e n t approaches f o r measuring a d s o r p t i o n a t t h e s o l i d . Even i f i t is assumed that a chosen a n a l y t i c a l method c a n determine t h e sum of s e v e r a l s u r f a c t a n t s a c c u r a t e l y . such a s e i t h e r a c h i e v i n g a n u l t r a low i n t e r f a c i a l t e n s i o n o r improving t h e s o l u b i l i t y of o t h e r components. t h e whole mixture is u s u a l l y t r e a t e d as a s i n g l e component. and measurements of a d s o r p t i o n above such c o n c e n t r a t i o n l e v e l s are meaningless a s s u r f a c t a n t p r e c i p i t a t i o n t a k e s p l a c e .

The previous paragraphs were intended t o show t h a t thermodynamically v a l i d a d s o r p t i o n measurements can be best performed i n batch experiments. and a d s o r p t i o n is c a l c u l a t e d from Equation (1). i f t h e s u r f a c t a n t c o n c e n t r a t i o n a t t h e c o r e o u t l e t does not r e a c h t h e i n j e c t e d c o n c e n t r a t i o n . t h e p o s s i b i l i t y of s u r f a c t a n t being d i s t r i b u t e d between t h e b r i n e and hydrocarbon phases d u r i n g t h e flood. I n t h e second method. mst be considered. i n a way t h a t d i f f e r s from t h e i n j e c t e d s o l u t i o n . the r e t e n t i o n v a l u e s obtained from t h e s e two experiments would be r e l a t e d t o d i f f e r e n t c o n c e n t r a t i o n r e g i o n s . I n t h i s c a s e . t h e meaning of each v a r i a b l e i n Equation (1) is w i t h o u t ambiguity and t h e c a l c u l a t e d s u r f a c e excess is a v a l i d thermodynamic v a r i a b l e . t h e n t h e a d s o r p t i o n determined from a material balance is a t t a i n e d over a c o n c e n t r a t i o n range t h a t extends from t h e i n j e c t e d l e v e l a t t h e c o r e i n l e t t o t h e maximum e f f l u e n t c o n c e n t r a t i o n s measured a t t h e c o r e o u t l e t . A similar argument c a n be made when a d i f f e r e n t degree of s e p a r a t i o n between a c o s u r f a c t a n t and a s u r f a c t a n t o c c u r s i n experiments i n which v a r y i n g s l u g sizes are used. There are two a d d i t i o n a l c o n s i d e r a t i o n s concerning measurements o f adsorpt i o n i n displacement experiments i n which a n o i l phase i s p r e s e n t and when i t i s d i s p l a c e d from a c o r e by a s u r f a c t a n t s o l u t i o n . t h e s u r f a c t a n t flow v e l o c i t y becomes dependent n o t only on a d s o r p t i o n but also on two-phase flow c h a r a c t e r i s t i c s . For example. However. It should be r e a l i z e d t h a t . When t h i s o c c u r s . A l a r g e r s l u g may g i v e a maximum o u t l e t c o n c e n t r a t i o n o f 90% r e s u l t i n g i n t h e average c o n c e n t r a t i o n w i t h i n t h e c o r e being about 95% of t h e i n j e c t e d c o n c e n t r a t i o n . t h e pore volume a v a i l a b l e t o the s u r f a c t a n t is changing d u r i n g t h e flood a s o i l i s d i s p l a c e d . An example of such a s e p a r a t i o n i s shown i n Figure 2. i f t h e amount of s u r f a c t a n t i n j e c t e d is so l a r g e that t h e e f f l u e n t c o n c e n t r a t i o n r e a c h e s t h e i n j e c t e d c o n c e n t r a t i o n . t h e d e l a y i n t h e breakthrough curve may be caused by e i t h e r of t h e two phenomena. Unless t h e s u r f a c t a n t d i s t r i b u t i o n is known p r e c i s e l y a t each s t a g e of t h e food. a dynamic one. o r from t h e material balance when a smaller amount of s u r f a c t a n t is i n j e c t e d (Figure 1). i f the i n j e c t i o n of a 20% PV s u r f a c t a n t s l u g r e s u l t s i n a maximum o u t l e t c o n c e n t r a t i o n corresponding t o 10% of t h e i n j e c t e d c o n c e n t r a t i o n . t h e average s u r f a c t a n t c o n c e n t r a t i o n w i t h i n t h e c o r e would be approximately 55% o f t h e i n j e c t e d l e v e l . Therefore.104 s u r f a c t a n t c o n c e n t r a t i o n s i n t h e b u l k phase b e f o r e and a f t e r a d s o r p t i o n takes p l a c e . and a d i s t i n c t i o n between them cannot be made. A main disadvantage of u s i n g t h e b a t c h method l i e s i n i t s poor accuracy a t h i g h e r s u r f a c t a n t c o n c e n t r a t i o n s . Secondly. It has been shown w i t h s u r f a c t a n t s y s t e m s considered f o r enhanced o i l recovery that exceedingly a c c u r a t e a n a l y t i c a l methods a r e r e q u i r e d f o r measurement of a d s o r p t i o n from s o l u t i o n s when s u r f a c t a n t c o n c e n t r a t i o n s exceeds one p e r c e n t . a n d a g a i n r e t e n t i o n r e s u l t s should be d i f f e r e n t s i n c e i t has been shown p r e v i o u s l y that alcohol concentrations a f f e c t s u r f act an t r et en tio n su b sta n tia lly . s i n c e t h e i r s e n s i t i v i t y i s o f t e n n o t s u f f i c i e n t f o r measurements o f a d s o r p t i o n . The measured change i n b u l k s u r f a c t a n t c o n c e n t r a t i o n due t o a d s o r p t i o n becomes small and a d s o r p t i o n i s obtained by s u b t r a c t i n g two numbers of s i m i l a r s i z e . and t h e r e s u l t s would not be d i r e c t l y comparable. F i r s t l y . and t h e l o s s e s of s u r f a c t a n t are c a l c u l a t e d e i t h e r from t h e d e l a y of t h e s u r f a c t a n t breakthrough c u r v e . S i n c e a l l measurements a r e performed i n t h e b u l k phase. This makes t h e i n t e g r a t i o n i n d i c a t e d i n Figure l ( a ) mre d i f f i c u l t s i n c e t h e amount o f o i l recovered must be known a t each p o i n t i n t h e flood. the r e t e n t i o n of s u r f a c t a n t s i s determined from a flow-type experiment. a d s o r p t i o n cannot be determined d i r e c t l y from t h e d e l a y of t h e breakthrough curve f o r two reasons.

I . s i n c e it mav a f f e c t r e t e n t i o n of s u r f a c t a n t s so s u b s t a n t i a l l y . a d e c r e a s e .o C/% 0. Alcohols propagate through t h e porous mediua a t d i f f e r e n t '.0 PORE VOLUME Figure 1: Determination of S u r f a c t a n t Retention from Displacement Experiments It h a s been p r e v i o u s l y shown that t h e r e i s a n o r d e r o f magnitude d i f f e r e n c e i n t h e r e t e n t i o n of s u r f a c t a n t s from d i s p e r s e d s o l u t i o n s and from s o l u t i o n s i n which t h e s u r f a c t a n t s are t r u l y d i s s o l v e d Even though mst i n j e c t e d s u r f a c t a n t s o l u t i o n s used i n a d s o r p t i o n s t u d i e s c o n t a i n a l c o h o l s a s c o s u r f a c t a n t s i n order t o keep s u r f a c t a n t s f u l l y d i s s o l v e d . displacement experiments must be used f o r such systems. It should. It f o l l o w s that even minor v a r i a t i o n s i n t h e displacement experiments can produce s u b s t a n t i a l l y d i f f e r e n t r e s u l t s i n terms of s u r f a c t a n t r e t e n t i o n and a d s o r p t i o n . S u r f a c t a n t s o l u b i l i t y i n r e s e r v o i r f l u i d s could p o s s i b l y be grouped w i t h i n t h e phase behavior categorv b u t . It is e n t i r e l y p o s s i b l e that t h e d i f f e r e n t experimental techniques and procedures could have been r e s p o n s i b l e f o r such d i v e r g e n t r e s u l t s . o r no e f f e c t had been observed. i t is d i s c u s s e d s e p a r a t e l y . Therefore. i t is i m p e r a t i v e t h a t e v e r y a d s o r p t i o n measurement be d e s c r i b e d i n d e t a i l so t h a t t h e r e s u l t s from d i f f e r e n t l a b o r a t o r i e s c a n be r e a l i s t i c a l l y compared.105 from s u r f a c t a n t s o l u t i o n s of h i g h e r c o n c e n t r a t i o n s . t h i s r a y n o t be t h e c a s e i n t h e l a t e r s t a g e s of a f l o o d .0 0.0 4. The adsorption v a l u e s are also averaged over t h e range o f c o s u r f a c t a n t f s u r f a c t a n t ratios which depend on t h e s p e c i f i c characteristics o f t h e s u r f a c t a n t system. Another phenomenon a f f e c t i n g t h e r e t e n t i o n of s u r f a c t a n t s should be t r e a t e d s e p a r a t e l y .5 Continuous o r Large Slug I n j e c t i o n 0.0 2. however. This is n o t p r e s e n t l y t h e case. be r e a l i z e d that such tests measure averaged v a l u e s of a d s o r p t i o n over c o n c e n t r a t i o n s that r a n g e from t h e maximum e f f l u e n t c o n c e n t r a t i o n t o t h e i n j e c t e d c o n c e n t r a t i o n . . a s manifested i n t h e r e c e n t paper by Meyer and S a l t e r ' who surveyed t h e l i t e r a t u r e t o determine t h e e f f e c t s of a n o i l presence o n s u r f a c t a n t r e t e n t i o n and found that a n i n c r e a s e .

106 v e l o c i t i e s than s u r f a c t a n t s because they d i s t r i b u t e themselves between t h e o i l and t h e b r i n e d i f f e r e n t l y t h a n do t h e s u r f a c t a n t s (Figure 2 ) .5 2. Chenical s Three types of s u r f a c t a n t s were used d u r i n g t h e c o u r s e of t h i s study. a l c o h o l s were used i n e x c e s s q u a n t i t i e s thereby e l i m i n a t i n g o r s u b s t a n t i a l l y reducing t h i s p o s s i b i l i t y . in most experiments d e s c r i b e d in t h i s work.0 PORE VOLUME C o s u r f a c t a n t / S u r f a c t a n t Ratio a t t h e Core O u t l e t (10072 3’I n j e c t i o n of 2% 1 / 6 Texas ! l !/ n-Propanol i n 1 5 . N a C l Brine) . 00 Figure 2: 0. p r e c i p i t a t e s . TXS 10-82 served a s a n example of a commercial q u a l i t y petroleum s u l f o n a t e which . i t is l i k e l y that t h e s u r f a c t a n t l o s e s its: s o l u b i l i t y i n t h e b r i n e . a l c o h o l c o n c e n t r a t i o n s i n t h e e f f l u e n t were monitored so that a p o t e n t i a l problem o f poor s u r f a c t a n t s o l u b i l i t y could be a s s e s s e d a t t h e end of eack flood. I n a l l c a s e s .5 1. Should t h i s happen. This w i l l r e s u l t i n an a p p a r e n t i n c r e a s e i n s u r f a c t a n t r e t e n t i o n which cannot be e a s i l y d i s t i n g u i s h e d from e i t h e r a d s o r p t i o n o r t r a p p i n g in t h e hydrocarbon phase.‘ The main o b j e c t i v e o f t h i s work is t o determine s u r f a c t a n t r e t e n t i o n in & r e a c o r e s w i t h t h e main emphasis being t o d i s t i n g u i s h physico-chemical a d s o r p t i o n of s u r f a c t a n t s from l o s s e s of s u r f a c t a n t s due t o t r a p p i n g i n t h e i m o b i l e hydrocarbon phase t h a t is l e f t i n t h e c o r e a f t e r a flood. and l o s e s i t s a b i l i t y t o propagate through t h e core. 6b 1 - INJECTED SOLUTION ! 5i G i? a in \ 4- 3 I - z 3- 8 0 i? a in 2- 3 8 1- ‘. Therefore.0 1.

and t h e n waterf l o o d e d t o a r e s i d u a l o i l s a t u r a t i o n u s u a l l y i n t h e range o f 30 t o 35% o f pore volume. According t o t h e s u p p l i e r . - B e r e a s a n d s t o n e c o r e s w i t h a i r p e r m e a b i l i t y r a n g e o f 100 t o 1 .5 cm2 c r o s s . monoethanol amine s a l t o f a l k y l o r t h o x y l e n e s u l f o n a t e s u p p l i e d by Exxon Chemicals. t h e s t r u c t u r e o f which was p a t t e r n e d a f t e r typical molecules found i n petroleum-based f e e d s t o c k s . P M 337 D TRS 10-80 Octane - Orthoxylene . i t is 84% a c t i v e w i t h a median a l k y l c h a i n s i z e of around C12. 2 0 0 md. s u p p l i e d by Cleveland Q u a r r i e s .144.107 is produced by a d i r e c t s u l f o n a t i o n of petroleum-based f e e d s t o c k s . Texas.l i q u i d i n t e r f a c e .I o n i c Flame Spectrophometry . They were t h e n s a t u r a t e d under vacuum w i t h degassed b r i n e . simplied by Secondary Butyl a l c o h o l b o i l i n g p o i n t r a n g e 98" Eastman KDdak Company. This s u r f a c t a n t was used a s r e c e i v e d .s e c t i o n ) were c u t t o 30 CGI l e n g t h s and d r i e d i n a vacuum oven a t ll0'C f o r 24 hours.a c e t o n i t r i l e .5"C - s u p p l i e d by Matheson. . PDM 337 w a s a n example o f a s y n t h e t i c s u l f o n a t e . The f o l l o w i n g methods have been e x t e n s i v e l y t e s t e d and employed f o r c o n c e n t r a t i o n d e t e r m i n a t i o n i n t h i s work. According t o Frances e t a1. .m e t h a n o l . Brine Adsorbent - N a C l s u p p l i e d by F i s h e r S c i e n t i f i c d i s s o l v e d i n deionized water. and Texas 81 was a w e l l .W Spectrophotometry ( V a r i a n ' s Super Scan 3) . Houston. Texas # I - sodium s a l t o f 8-phenyla-hexadecyl-p-sulfonate w a s obtained from P r o f e s s o r Wade o f t h e U n i v e r s i t y o f Texas and has been used as r e c e i v e d .sodium phosphate s o l v e n t s ' (Waters A s s o c i a t e s I n s t r u m e n t ) . o i l flooded t o a connate water s a t u r a t i o n .boiling p o i n t r a n g e 143. Divalent Ions .5' Coleman and B e l l Co.5 x 2.l1 t h e p u r i t y o f t h e sample exceeds 98%.12 t e c h n i c a l g r a d e s u p p l i e d by P h i l l i p s Petroieum Company. 100°C.Chelatometric t i t r a t i o n (Perkin-Elnrer Instrument) E i T e r i n e n t a l Procedures Berea a r e s (2. A n a l y t i c a l Methods The p r e c i s i o n o f a n a l y t i c a l methods is o f u t m s t importance i n a l l s t u d i e s of a d s o r p t i o n a t t h e s o l i d . Surfactants .d e f i n e d pure s u r f a c t a n t . 3ihydro- Octanelo-xylenefSecondary Suty: slcohol-GC emplo>-lng Chromosorb S packing and t h e thermal c o n d u c t i v i t y d e t e c t o r (Hewlett-Packard Instrument).HF'LC u t i l i z i n g w a t e r . petroleum s u l f o n a t e s u p p l i e d by Witco Chemicals. Samples were d e s a l t e d and d e o i l e d accord-ins t o t h e procedures d e s c r i b e d by Shah e t a1.

and c o s u r f a c t a n t ceased. however. and c o s u r f a c t a n t c o n t e n t . s u r f a c t a n t . I n some f l o o d s . b r i n e . even c h i s small amount of adsorbed s u r f a c t a n t recovered by o i l is s u f f i c i e n t t o q u a l i f y t h i s method f o r d e t e r m i n a t i o n of trapped s u r f a c t a n t as q u a l i t a t i v e . S u r f a c t a n t l o s s was determined a t 0.A s u r f a c t a n t s l u g w a s i n j e c t e d i n t o t h e cores a t r e s i d u a l o i l s a t u r a t i o n a t c o n s t a n t rates of 2 ml/hour so that t h e a p p a r e n t f r o n t a l advance rate of the f l u i d d i d not exceed 30 cmlday. a n d u s u a l l y t h e most i n a c c u r a t e results were obtained w i t h PDM 337. This r e q u i r e d i n j e c t i o n o f 5 t o 1 0 pore volumes and t h e material balance on s u r f a c t a n t c l o s e d u s u a l l y between 90 t o 100% of i n j e c t e d s u r f a c t a n t . o c t a n e w a s d i s p l a c e d by nonane o r decane so that a complete displacement of r e s i d u a l cil could be v e r i f i e d and a material balance 011 o i l c l o s e d . This assumption was v e r i f i e d i n t h e following way: X 75X P of 3% s u r f a c t a n t s l u g was i n j e c t e d i n a b r i n e . Synchronized movements of a f r a c t i o n c o l l e c t o r and t h e s y r i n g e n e e d l e were automated. A f t e r a l l s u r f a c t a n t s trapped i n t h e o i l were d i s p l a c e d . I n o r d e r t o eliminate e v a p o r a t i v e l o s s e s of v o l a t i l e components. t h e samples were c o l l e c t e d a t two-hour i n t e r v a l s which r e s u l t e d i n 5 t o 10% o f pore volume being c o l l e c t e d i n each sample. This i n d i c a t e s that a minor amount of adsorbed s u r f a c t a n t can be recovered by t h e o i l .s a t u r a t e d c o r e and V followed w i t h t h r e e a d d i t i o n a l pore volumes of b r i n e . It seems reasonable t o suggest t h a t a degree of s u r f a c t a n t s o l u b i l i t y i n a l c o h o l s o l v e n t s could e x p l a i n t h i s t r e n d . s e v e r a l pore volumes of a hydrocarbon phase were i n j e c t e d i n t o t h e c o r e i n a n a t t e m p t t o recover s u r f a c t a n t s trapped i n o i l remaining i n t h e c o r e . t h e b e s t m a t e r i a l balances were obtained i n f l o o d s w i t h TRS 10-80. o c t a n e was continuously i n j e c t e d and an e f f l u e n t was analyzed f o r s u r f a c t a n t s . The f l o o d i n g sequence d e s c r i b e d above a l l o w s a d e t e r m i n a t i o n of t h e o v e r a l l s u r f a c t a n t r e t e n t i o n ( i . This procedure i m p l i c i t l y assumes t h a t t h e hydrocarbon phase does n o t remove adsorbed s u r f a c t a n t from t h e core.06 mg of s u r f a c t a n t p e r one gram of rock w a s recovered. When t h e production of o i l . A f t e r w r e than 5 P P o f throughput o n l y 0. . Liquid produced by t h i s hydrocarbon flood was analyzed f o r a l l components and recovered s u r f a c t a n t s were considered t o be s u r f a c t a n t s trapped i n t h e hydrocarbon phase d u r i n g t h e s u r f a c t a n t flood. However. t h e c o r e w a s flooded w i t h a s t r o n g s o l v e n t such a s e t h y l a l c o h o l o r i s o n r o p y l a l c o h o l i n a mixture w i t h b r i n e t o remove a l l remaining s u r f a c t a n t s from t h e core. E f f l u e n t f l u i d s were t h e n analyzed f o r o i l . e . The hydrocarbon flood g i v e s a amount of s u r f a c t a n t trapped i n t h e o i l phase due t o unfavorable phase behavior.6 mg/g. t h e o u t l e t l i n e w a s f e d through a s y r i n g e n e e d l e p i e r c i n g t h e septum o f a c o l l e c t i o n tube. S u r f a c t a n t f l o o d s were performed a s follows. I n g e n e r a l . S u r f a c t a n t removed from t h e c o r e by a l c o h o l s o l v e n t s is considered t o be s u r f a c t a n t adsorbed on t h e rock d u r i n g t h e f l o o d . no measurements of s u r f a c t a n t s o l u b i l i t i e s have been made. During t h e s u r f a c t a n t flood and t h e subsequent b r i n e flood (no polymers o r v i s c o s i t y improving a g e n t s have been used i n t h i s work). and that t h e bulk of adsorbed s u r f a c t a n t w i l l n o t be desorbed. Then. s u r f a c t a n t . t h u s a l l o w i n g u n i n t e r r u p t e d f l o o d i n g i n experiments l a s t i n g s e v e r a l days. t h e amount o f s u r f a c t a n t l o s t d u r i n g t h e f l o o d ) from t h e d i f f e r e n c e between t h e amounts o f s u r f a c t a n t i n j e c t e d and recovered d u r i n g t h e s u r f a c t a n t and subsequent b r i n e i n j e c t i o n . and t h e adsorbed s u r f a c t a n t recovered i n t h e f i n a l s o l v e n t f l o o d completes t h e m a t e r i a l balance.

109 I n o r d e r t o avoid experimental complications due t o t h e p o s s i b l e p r e c i p i t a t i o n of s u r f a c t a n t s by d i v a l e n t i o n s . Berea c o r e s were p r e f l u e h e d w i t h 5 t o 7 pore volumes of sodium c h l o r i d e b r i n e s i n o r d e r t o d i s p l a c e most of t h e exchangeable d i v a l e n t ions. - RESLZTS Ah?) DISCUSSION S t u d i e s of o i l recovery e f f i c i e n c y and s u r f a c t a n t r e t e n t i o n i n d i c a t e that b e t t e r . i t Qas thought t o be of i n t e r e s t t o perform s e v e r a l series of similar experiments and then o b s e r v e how t h e s e i n d i v i d u a l c o n t r i b u t i o n s t o t o t a l s u r f a c t a n t r e t e n t i o n are affected . S e p a r a t e phase behavior experiments i n d i c a t e d that such low d i v a l e n t i o n c o n c e n t r a t i o n s a f f e c t e d t h e phase behavior o f s u r f a c t a n t s o l u t i o n s i n that a minor s h i f t toward upper phase microemulsions w a s n o t i c e d . e . i n s u r f a c t a n t systems i n which t h e s u r f a c t a n t d i s t r i b u t i o n c o e f f i c i e n t is not a t extreme l e v e l s ( i . performing p r o c e s s e s a r e u s u a l l y accompanied by lower s u r f a c t a n t r e t e n t i o n even though lower r e t e n t i o n does n o t n e c e s s a r i l y mean higher o i l recoverp. t h e s e l e v e l s have n o t exceeded 90 ppm.14 Since our experimental technique can d i s t i n g u i s h between s u r f a c t a n t l o s s e s due t o a d s o r p t i o n and l o s s e s due t o unfavorable phase behavior. t h e r e i s a n i n c r e a s e i n d i v a l e n t c a t i o n c o n c e n t r a t i o n i n t h e propagating s u r f a c t a n t s l u g (Figure 3). . loot OO 7 2 1 PORE VOLUME F i g u r e 3: DivaleDt Ions Content i n t h e E f f l u e n t ( I n j e c t i o n N of 75% PV of 2% 110. K = [ ( c s ) o i l / ( c s ) b r i n e l f o r upper phase microtends t o zero f o r lower phase microemulsions o r K + emulsions) t h e c h a s e b r i n e would b l e e d s u r f a c t a n t from t h e o i l phase and no s u r f a c t a n t would e v e r be found trapped i n the o i l . For example. sodium c h l o r i d e b r i n e s were used throughout t h i s study. Even w i t h t h e s e p r e c a u t i o n s . but no s u r f a c t a n t p r e c i p i t a t i o n was observed.5 TRS 10-8OlSBA i n 1% a C l ) It should be noted h e r e that t h i s procedure f o r d i f f e r e n t i a t i n g trapped s u r f a c t a n t i n t h e hydrocarbon phase from t h e adsorbed s u r f a c t a n t is not a p p l i c a b l e t o a l l s i t u a t i o n s . I n our experiments.

0 1/50 F i g u r e 4: Phase Behavior of 3X PDM 337 S u r f a c t a n t (80120 volumetric r a t i o of 1.5% k C 1 / octane f o r d i f f e r e n t surfactantlsecondary butyl alcohol r a t i o s ) S u r f a c t a n t l c o s u r f a c t a n t r a t i o s of l:O. . Typical examples of t h e d a t a c o l l e c t e d a r e shown i n F i g u r e s 5 t o 7 and t h e r e s u l t s of t h e s e f l o o d s are summarized i n Table 1. The e f f l u e n t s were analyzed f o r s u r f a c t a n t . e . 1:1. An a d d i t i o n a l c o s u r f a c t a n t helped t o d i s s o l v e t h e s u r f a c t a n t i n t h e b r i n e and t h e s u r f a c t a n t r e t e n t i o n w a s reduced by one o r d e r of m a g n i t ~ d e . poor s u r f a c t a n t solub i l i t y may r e s u l t i n very high s u r f a c t a n t r e t e n t i o n i n Berea c o r e s . This series of f l o o d s c l e a r l y shows a l l of t h e d i f f i c u l t i e s which can be encountered when an a t t e m p t is made t o compare a d s o r p t i o n d a t a obtained from d i f f e r e n t displacement experiments. ~ n systems I c o n t a i n i n g o i l . c o s u r f a c t a n t . SURFACTANT CONTAINING PHASE VO. An i n c r e a s i n g c o s u r f a c t a n t c o n t e n t makes t h e s u r f a c t a n t s l i g h t l y -re b r i n e s o l u b l e and t h e phase behavior changes from a n upper t o a middle phase (Figure 4 ) . and 1:s were i n j e c t e d i n f o u r f l o o d s on Berea c o r e s t h a t had been waterflooded t o r e s i d u a l o i l s a t u r a t i o n s .110 E f f e c t of C o s u r f a c t a n t on S u r f a c t a n t R e t e n t i o n I t has been shown t h a n .l. The P M 337 s u r f a c t a n t w i t h secondary b u t y l a l c o h o l . a s a c o s u r f a c t a n t w a s D s e l e c t e d f o r t h i s p a r t of t h e study. systems c o n t a i n i n g o n l y s u r f a c t a n t .1 1/0. i n systems c o n t a i n i n g no o i l ( i . c o s u r f a c t a n t and o i l c o n t e n t . 1:0.5 1/1. poor s u r f a c t a n t s o l u b i l i t y may n o t r e s u l t i n s u r f a c t a n t molecule a g g r e g a t i o n b u t may l e a d t o a change in phase behavior i n which c a s e t h e s u r f a c t a n t d i s s o l v e s i n t h e upper hydrocarbon phase. I n that case t h e s u r f a c t a n t r e t e n t i o n would increase even though s u r f a c t a n t a d s o r p t i o n may e i t h e r not change a t a l l o r may even d e c r e a s e . and b r i n e ) .5.

5% NaCl Brine) 6 - 0 MICROEMULSION f jOIL $ 0 BRINE 5 4 Y s J g 3 3 0 ' 2 I 0 0 PORE VOLUME Figure 6 : Effluent Phase Behavior (Flood 112) I 2 3 . 1:5 PDM 337lSBA i n 1.111 PORE VOLUME Figure 5: Surfactant and Cosurfactant Breakthrouzh Curves (Flood 112: 50% PV I n j e c t i o n o f 3%.

i t should be noted that.1 K LL 0.0 PORE VOLUME F i g u r e 7: F r a c t i o n a l Flow of O i l (Flood 1 1 2 ) Oil Recover).5 12 . l and 1:0. 07 . 0.3 w z u k I - 0.1 110. e s s e n t i a l l y no s u r f a c t a n t is contained i n t h e e f f l u e n t . Floods w i t h the 1:l and 1:5 s u r f a c t a n t l c o s u r f a c t a n t r a t i o s have l e d t o t h e production of some s u r f a c t a n t . 07 . 12 . - 12 .02 z 15 10 lJO. 1. 52 66 85 17 19 22 21 111 115 - 10 . T h i s means t h a t n o t enough s u r f a c t a n t was i n j e c t e d t o s a t i s f y t h e a d s o r p t i o n c a p a c i t y o f t h e r o c k and t h a t t h e s u r f a c e s n e a r t h e end of the c o r e are probably n o t completely adsorbed w i t h s u r f a c t a n t .10 0.112 0.2 10 . Surfacuat co-surfacunc u~uht ratio Surfacunr Refention losses Due to Adsorption ?hse BeUhvior (c/c0)- I MI? (SorIf- (foul2 4 1 ulr Ull z 0 .21 5 8 70 F i r s t . .5.2 LL 0 0 z 2 0. i n t h e f l o o d s w i t h s u r f a c t a n t r a t i o s of 1 : O .

even though i t may b e tempting t o suggest t h a t t h e r e is enough d a t a i n Table 1 t o a s c e r t a i n t h e dependence of s u r f a c t a n t a d s o r p t i o n on a l c o h o l c o n t e n t .0 PORE VOLUME 3.113 b u t t h e c o n c e n t r a t i o n peaks a t t h e c o r e o u t l e t s a r e s u b s t a n t i a l l y d i f f e r e n t from each o t h e r and.0 0 0. and Table 2 s u m a r i z e s t h e d a t a obtained 1 i n t h e s e f i v e floods. as Figure 8 shows. A l s o . I I I I I 0.5 1. t h e normalized r a t i o s of s u r f a c t a n t and c o s u r f a c t a n t concent r a t i o n s are q u i t e d i f f e r e n t f o r t h e two f l o o d s . The l a s t three c o l u m s of Table 1 c o n t a i n t h r e e i n d i c a t o r s o f t h e o i l recovery of each s u r f a c t a n t f l o o d . ' ~ I ? I 5. Typical flooding r e s u l t s a r e shown i n F i g u r e s 9 t o 1 . Results show that e f f i c i e n c y i n i t i a l l y i n c r e a s e s w i t h c o s u r f a c t a n t c o n t e n t .*olF.0 I I I 4. s u r f a c t a n t l c o s u r f a c t a n t r a t i o w a s v a r i e d from 1:O. Therefore. t h e a d s o r p t i o n v a l u e s f o r t h e two d i f f e r e n t average s u r f a c t a n t c o n c e n t r a t i o n s a r e n o t d i r e c t l y comparable.0 Figure 8: >!ormalized C o s u r f a c t a n t / S u r f a c t a n t R a t i o s a t Core O u t l e t s The E f f e c t of Slug S i z e on S u r f a c t a n t R e t e n t i o n The A similar series o f experiments w a s performed w i t h TRS 10-80.0 2. h w e v e r . consepuently. This confirms a conclusion r e p o r t e d p r e v i o u s l y that lower s u r f a c t a n t r e t e n t i o n does n o t n e c e s s a r i l y l e a d t o t h e b e s t o i l recovery e f f i c i ~ n c y .0 Y I I . a c l o s e r l o o k shows that a comparison o f s u r f a c t a n t a d s o r p t i o n f o r t h e f o u r d i f f e r e n t systems cannot be nade w i t h o u t conducting a d d i t i o n a l experiments. . t h e f i n a l f l o o d performs less e f f i c i e n t l y t h a n t h e p r e v i o u s one.S t o 1 : l O .

5 TRS 10-80/ SBA in 1.0%NaC1) 0 PORE VOLUME Figure 10: Effluent Phase Behavior 2 4 . 1:0.114 L 0 SURFACTANT A COSURFACTANT 1 PORE VOLUME Figure 9: Surfactant and Cosurfactant Breakthrough Curves (Flood 69: 75% PV Injection of 2%.

).2 0.15 0.6 0.55 .8 1.1 0.0 0.2 0.0 0.95 1.10 . x 1110 115 113 113 80 94 Wl 0.20 0.35 0.so 0.115 0.0% NaCl/Octane System Surf acUnc cesurfaeunt W Injected Retantien nyh Trapped Surfacunt =&It Adsorption (dc.3 0 0.2 LL 1 0 z 2 5 a 0.5 75 - 0.48 Ill 110.5 0.3 0. 0.50 0.85 150 75 75 01 .0 PORE VOLUME Figure 11: 4 Fractional Flow of O i l (Flood 69) Table 2: Summary of Flooding Results with 3% TRS 10-80/SBA i n 1.1s 0.4 .

t h e r e is no trapped s u r f a c t a n t . A 150% P s l u g w a s s u f f i c i e n t l y l a r g e t o enable t h e e f f l u e n t c o n c e n t r a t i o n t o r e a c h t h e l e v e l of t h e i n j e c t e d concentration. While t h e r e is a d i f f e r e n c e i n r e t e n t i o n .0 0 2 3 PORE VOLUME Figure 12: S u r f a c t a n t and Cosurfactant Breakthrough Curves (Flood 99: 100% PV I n j e c t i o n of 2X. Again. they do so a t d i f f e r e n t rates. 0 SURFACTANT A COSURFACTANT / 0. w h i l e i n t h e second about one-third of the s u r f a c t a n t l o s s i s due t o unfavorable phase behavior. even though b o t h o v e r a l l r e t e n t i o n and a d s o r p t i o n i n c r e a s e w i t h i n c r e a s i n g s l u g s i z e . This apparent discrepancy can b e explained i n terms o f t h e amount of o i l trapped i n t h e hydrocarbon phase.116 It is i n t e r e s t i n g t o compare t h e s u r f a c t a n t r e t e n t i o n v a l u e s observed i n f l o o d s using a s u r f a c t a n t / c o s u r f a c t a n t r a t i o o f 1 : 3 i n which d i f f e r e n t s i z e V s l u g s of i d e n t i c a l composition were i n j e c t e d .6 p' I \ \ \ c'co 0. t h e d i f f e r e n c e i n adsorpt i o n i s smaller. I n t h i s c a s e . In t h e f i r s t f l o o d .5% NaCl) . F i g u r e s 12 t o 1 4 show a n example of t h e experimental d a t a and Table 3 p r e s e n t s a summary of t h e r e s u l t s . Another series o f experiments w a s performed w i t h t h e pure Texas #I s u r f a c t a n t . 1:6 Texas <!l/ n-Propanol i n 1.2 0.4 0.8 I '*O 0. i t is t h e 106s due t o t h e phase behavior which is a f f e c t e d more by t h e s i z e o f t h e s u r f a c t a n t s l u g . T h i s example shows c l e a r l y that i n f o r n a t i o n r e f l e c t i n g o n l y o v e r a l l s u r f a c t a n t r e t e n t i o n may be very misleading. The i n j e c t i o n volume i n t h e o t h e r comparable f l o o d w a s halved so t h a t t h e e f f l u e n t c o n c e n t r a t i o n reached o n l y 85% o f t h e i n j e c t e d concent r a t i o n .

31 PORE VOLUME Figure 14: Fractional Flow of O i l (Flood 99) .4 E Y - g 3 J 0 > 2 1 0 0 I n n L I PORE VOLUME Figure 13: Effluent Phase Behavior (Flood 99) 0.117 6 0 MXROEMLILSION fQ OIL BRINE 5 .

Another i n t e r e s t i n g a s p e c t observed i n t h i s experiment was t h e s h a p e o f t h e s u r f a c t a n t breakthrough c u r v e s ( s e e F i g u r e 15).0 0. 0. A s e x p e c t e d . and a d s o r p t i o n of 1 . The procedure used i n Flood 82 was t h e same a s f o r Flood 85 except t h a t i n Flood 82 t h e one PV o f t h e b r i n e t h a t followed t h e s u r f a c t a n t s l u g cont a i n e d 3% secondary b u t y l a l c o h o l .25 0.8 0.0 followed by one PV of 3Z S M in brlne 83 a 85 bur at m SLoyEll INJJLCTION 0. t h e r e t e n t i o n and a d s o r p t i o n l e v e l s a r e both lower. 2 mglg w a s observed.15 mgfg s u r f a c t a n t due t o p h a s e behavior. Even though t h e f l o o d s were r u n a t t h e same i n j e c t i o n rates.60 1. t h e r e was a l o s s of 0. Table 4: Summary of Flooding R e s u l t s w i t h 3%.5% ?. 12 .24 0.5 0.50 0. b u t t h e l i m i t e d d a t a a v a i l a b l e do n o t a l l o w f o r a unique i n t e r p r e t a t i o n and . Table 4 summarizes t h e r e s u l t s . 1:6 Texas $l/n-Propanol in 1. t h e amount o f s u r f a c t a n t trapped in t h e oil phase d i d n o t change a p p r e c i a b l y . I n a d d i t i o n .20 03 .5 mgfg was determined.6/6 x 0. t h e shape of t h e curve in Flood 82 g i v e s t h e i m p r e s s i o n o f a much h i g h e r l e v e l of d i s p e r s i o n t h a n t h a t i n Flood 85.118 Table 3: Summary o f Flooding Experiments with 2 X .5 0.5% NaCl/Octane System Flood I Flood Oewription Retention WlK lasses D e co u M i o r p t i o n (clco)Phase k h v i o r ma18 ~ ~~~~~ w/a ~~~ ~ 86A oil i n the core ti0 12 .54 0.1 74 Experiments have beer? performed t o e v a l u a t e t h e e f f e c t o f c o s u r f a c t a n t p r e s e n c e w i t h i n t h e chase b r i n e o n t h e r e t e n t i o n o f s u r f a c t a n t s . 1:1.75 PDM 337lSBA i n 1.15 0. RATE 0.0 0.5 79 5 36 I sU . S e v e r a l e x p l a n a t i o n s are p o s s i b l e .aCl/Octane I 011 Recovery PV Injected Rerention bsses h e LO Phase Behavior d 6 Adsorption 4 8 (c/co)ux ZROIP (Sor)fLnal (fol* 2 .4 1.8 0.5 0.7 0.4 22 22 27 1. however. The o i l r e c o v e r y w a s b e t t e r a s t h e f i n a l o i l s a t u r a t i o n i s lowered from 1 4 % PI’ i n Flood 85 t o 5% PV i n Flood 82.14 0. 0.L 0.65 85 82 Surfacunr slug o n l y Surfacunr 0.5 1. Flood 85 contained o i l a t r e s i d u a l o i l s a t u r a t i o n and a d s o r p t i o n of 0.05 38 55 x 18 14 7 0.75 1.0 Flood 86A c o n t a i n e d no o i l .

It took more t h a n 1 0 d a y s f o r t h e surf a c t a n t s l u g t o p r o p a g a t e through t h e c o r e .4 . l An a t t e m p t has been made t o f i n d o u t i f similar phenomenon t a k e s p l a c e d u r i n g d i s p l a c e m e n t t e s t s i n Berea c o r e s . f o r s u r f a c t a n t systems which a r e n o t a t o p t i m a l f o r m u l a t i o n ( i . I t has been noted b e f o r e t h a t . i n d i c a t e d t h a n a n attainment of adsorption equilibrium required a l m o s t two weeks o f c o n t a c t between a s u r f a c t a n t s o l u t i o n and a s o l i d a d s 0 r b e n t . t h e time r e q u i r e d f o r a t t a i n m e n t o f phase e q u i l i b r i u m may b e s u b s t a n t i a l . i n t h e absence of o i l . T h e r e f o r e . 1 5 The r e t e n t i o n l e v e l w a s t h e s a m e b u t t h e l o s s o f s u r f a c t a n t by t h e phase t r a p p i n g mechanism i n c r e a s e d s u b s t a n t i a l l y w h i l e t h e a d s o r p t i o n l o s s w a s s l i g h t l y lower. Flood 85 w a s r e p e a t e d b u t a t a n i n j e c t i o n r a t e t h a t w a s t e n times lower and e q u a l t o a n a p p a r e n t f r o n t a l v e l o c i t y of 3 cm/day. t h e a d s o r p t i o n always reached e q u i l i b r i u m w i t h i n 24 hours.119 t h e r e f o r e none i s o f f e r e d . The o i l recovery was b e t t e r and a n a d d i t i o n a l 4% PV o f o i l was recovered which i s i n agreement w i t h t h e p r e v i o u s l y p u b l i s h e d d a t a o n t h i s t y p e of w p e r i m e n t . However. among o t h e r r e s u l t s . It t h e r e f o r e seems r e a s o n a b l e t o s u g g e s t t h a t t h e a d d i t i o n a l r e s i d e n c e time f o r t h e s u r f a c t a n t i n t h e c o r e allowed i t t o be more c o n c e n t r a t e d i n t h e o i l phase. n o t a t a middle phase c o n f i g u r a t i o n ) . Our experiments e n a b l e t h e s u g g e s t i o n that t h i s p r o c e s s ofs u r f a c t a n t r e d i s t r i b u t i o n among t h e p h a s e s may be more r e s p o n s i b l e f o r t h e time dependence o f r e t e n t i o n t h a n is t h e slow a t t a i n m e n t o f a d s o r p t i o n e q u i l i b r i u m a t the solid-liquid interface. b u t t h a n a n i n c r e a s e i n a d s o r p t i o n w a s n o t observed. e . t- I 0 I 2 3 4 PORE VOLUME F i g u r e 15: S u r f a c t a n t Breakthrough Curves A r e c e n t l y p u b l i s h e d paper d e s c r i b i n g s t a t i c a d s o r p t i o n e x p e r i m e n t s . T h i s s u g g e s t i o n i s supported by p r e v i o u s l y r e p o r t e d r e s u l t s o n a d s o r p t i o n measurements i n b a t c h experiments i n which. i t i s observed t h a t d a t a such a s t h e s e should be o f concern t o people d e a l i n g w i t h numerical models f o r chemical f l o o d i n g s i n c e t h e d a t a s u g g e s t t h a t t h e chemical composition of t h e s u r f a c t a n t s l u g may s u b s t a n t i a l l y a f f e c t t h e a p p a r e n t d i s p e r s i o n .

Nevada. S. For t h e t h r e e s u r f a c t a n t s s t u d i e d .. C a l i f o r n i a (May 28-30. 1980). and Recovery Efficiency". MANEV. "Surfactant Flooding I: The E f f e c t of A l k a l i n e Additives on IFT. Experimental procedures that permit d i f f e r e n t i a t i n g between s u r f a c t a n t l o s s e s due t o a d s o r p t i o n and t h o s c due t o unfavorable phase behavior have been developed and t e s t e d . and petroleum s u l f o n a t e (TRS 1 -0 @8) g i v e comparable r e s u l t s f o r r e t e n t i o n and a d s o r p t i o n i n Berea cores. 4. paper SPE 8263 presented a t t h e SPE 5 4 t h Annual F a l l Meeting. C. ACKNOWLEDGEMENTS The a u t h o r wishes t o acknowledge t h e a s s i s t a n c e and d e d i c a t i o n of L a u r i e Baxter and G a i l Parker who performed t h e p r e c i s e experiments necessary f o r t h i s paper. (September 23-26. A. Displacement experiments y i e l d s u r f a c t a n t r e t e n t i o n v a l u e s which i n v o l v e averaging s e v e r a l v a r i a b l e s .. and FALCONER. "The Role of S u r f a c t a n t P r e c i p i t a t i o n and R e d i s s o l u t i o n i n t h e Adsorption of S u l f o n a t e on Minerals". . 2. Texas (September 21-24. K. I f t h e o v e r a l l r e t e n t i o n i s h i g h e r .. KRUMRINE. Stanford. A. GOYAL. Pure s u r f a c t a n t (Texas # 1 ) . P. REFERENCES 1.120 SUIPlARY Based upon more than one hundred displacement experiments w i t h t h r e e t y p e s of s u r f a c t a n t s i n Berea c o r e s . S .. 1979). CELIK. t h e o t h e r causes of s u r f a c t a n t l o s s e s must be accounted f o r so o n l y a d s o r p t i o n d a t a are used. 1980). paper SPE 8989 presented a t t h e SPE 55th Annual F a l l Meeting.2 mg/g. T. s y n t h e t i c s u l f o n a t e (PDY 3 7 ) . 6. Thermodynamically v a l i d s u r f a c t a n t a d s o r p t i o n isotherms should be determined i n b a t c h experiments. . t h e following conclusions may be made: 1. S u r f a c t a n t Adsorption. MEYERS. paper SPE 8298 p r e s e n t e d a t the 5 t h Symposium on O i l f i e l d and Geothermal Chemistry. .. S. S i n c e r e l y acknowledged a r e Bev Moore and G a i l Donaldson f o r t y p i n g t h i s manuscript. "The E f f e c t of Oil-Brine R a t i o on S u r f a c t a n t Adsorption from Microemulsion". J. n-propano'.? 3. Dallas. M. J. 0. a d s o r p t i o n l e v e l s d i d not exceed 1. 3. E. 2. Las Vegas. I f any theory developed f o r a d s o r p t i o n is a p p l i e d t o r e t e n t i o n d a t a obtained from displacement experiments. and SOMASUNDURAN. 5. and SALTER. s u r f a c t a n t l o s s e s due t o unfavorable phase behavior o r some o t h e r mechanism should be suspected. CAMPBELL. Adsorption of s u r f a c t a n t s can be reduced by t h e s d d i t i o n o f low molecular weight a l c o h o l s (sec-butyl a l c o h o l . P.

Plenum P u b l i s h i n g . ZORNES. L. and HANNA. "Adsorption of S u l f o n a t e s on Reservoir Rocks"..F l u i d P r o p a g a t i o n . 183-193. W.Efficac. REED. S. I.. J. 293-305.. New York (1981). E. HEALY'. (June 1978) SPEJ 18.. and SANDVIK. . E. DAUBEN. I n t e r a c t i o n and M o b i l i t y " . Semi-Annual Report. TRUSHENSKI. S. 1 9 7 9 ) . H.(April 16-19. 13... 249-254. MAERTER. N. 1978). D. 1 12.1 21 4. . U n i v e r s i t y of F l o r i d a . 70. C. Anaheim. 7.. W. T. Systems". "Adsorption from Liquid SIRCAR. E. H. E. R. p a p e r SPE 7059 p r e s e n t e d a t t h e 5 t h Symposium on Improved Methods f o r O i l Recovery h e l d i n Tulsa. . (May 1572) I & EC Fundamentals ll. J. (1975) SPEJ 15. W.207-218. "Phase Behavior of a P u r e Alkyl A r y l S u l f o n a t e S u r f a c t a n t " . and CARPENTER. W. M i x t u r e s on S o l i d s : Thermodynamics of Excess P r o p e r t i e s and T h e i r Temp e r a t u r e C o e f f i c i e n t s " . of A d s o r p t i o n Maxima and Minima". . S. (1974) SPEJ l4. M. SOMASUNDARAN. GALE. F. 10. %Micellar Flooding .. F. I. W. D. R. R. 6. C. 9. (1973) SPEJ 2. WILLHITE.: Petroleum S u l f o n a t e Composition 191-199. P. DAVIS. 5. "An Experimental I n v e s t i g a t i o n I n t o t h e Use of HPLC f o r t h e D e t e r m i n a t i o n of Petroleum S u l f o n a t e s " . Oklahoma. 14. (June 1979) J. PUERTO. and CONLEY. L. GLOVER. FRANCES. M. 0. S. NOVOSAD.87-100. "Research on Chemical O i l Recovery SHAH. NOVOSAD. and SCRIVEN. "A h'ew I n t e r p r e t a t i o n TROGUS. D.. and WADE. (June 1979) SPEJ 2. " S u r f a c t a n t Phase Behavior and R e t e n t i o n i n Porous Media". (August 20-25. J . and WALKER. 8.. and MYERS. Nicroemulsion Flooding". C. MILLER. 1978). " S u r f a c t a n t Waterflooding". SCHECHTER. E.. I . R. .. C a l i f o r n i a (March 13-17. G a i n e s v i l l e (June 1977). P r o c e e d i n g s o f t h e 3 r d I n t e r n a t i o n a l Conference on S u r f a c e and C o l l o i d S c i e n c e s h e l d i n Stockholm. and MICHXICK. 633-644.. M. R. E. J. P . Sweden. J . H. G . R. L. A. R. . "A Laboratory Study of . " T e r t i a r v S u r f a c t a n t Floodinn: S t u d i e s " . J. 1 . and PARRISH. W. L. GILLILAND. G. D. P. "Adsorption of P u r e S u r f a c t a n t and Petroleum Sulfonate a t t h e Solid-Liquid I n t e r f a c e " . and SANDVIK.. C o l l o i d S c i . p r e s e n t e d a t t h e 1 7 5 t h ACS N a t i o n a l Meeting.. 15.

This Page Intentionally Left Blank .

Recently.S.. and butyl cyclohexane a l l exhibit "optimum" conditions. crude o i l composition. .Department of Energy Bartlesville Energy Technology Center ABSTRACT The EACN concept. i. I n t h i s paper. and the cyclohexane r i n g EACN = 3. minimum i n t e r f a c i a l tension (IFT) f o r the same combinations of surfactant. which allows the s u b s t i t u t i o n of a crude o i l by an alkane o r an alkane mixture f o r phase volume or i n t e r f a c i a l tension studies. Crude o i l . i t was shown t h a t such parameters as alcohol type. cosurfactant. and therefore. and t h a t t h i s equivalence always holds. i This concept was later found t o be applicable t o crude o i l s and pseudo crudes . heptylbenzene. whereby an alkane or alkane mixture can be found t o model the IFT behavior of a crude o i l . pseudocrude. Replacing the crude with hydrocarbon w i l l f a c i l i t a t e screening of surfactant formulation. Thus. heptane. Here. has been gene r a l l y accepted. and s a l t concentration. Interaction of surfactant with heavy crude o i l components was thought t o be another. o r hydrocarbon) is independent of the surfactant formulation. being dark i n color and usually q u i t e viscous. INTRODUCTION The term Equivalent Alkaf_e3Carbon Number (EACN). and water-oil-ratio could have an e f f e c t on the EACN of a crude o i l .up from University of Texas This concept a r i s e s from the observation t h a t the i n t e r f a c i a l properties of any o i l with a surfactant can be modeled by the behavi o r of alkanes. can make equilibrium attainment very slow and phase volume observat i o n d i f f i c u l t . Inl-3 addition. Experiments t e s t i n g the l a t e r hypothesis is i n progress. the benzene r i n g appeared t o have EACN = 0 .CHEMICAL FLOODING 123 THE EACN OF A CRUDE OIL: VARIATIONS WITH COSURFACTANT AND WATER OIL RATIO MIN KWAN THAM and PHILIP BOALT LORENZ U.e. also. ---(l) where X i s the mole f r a c t i o n of component i. the EACN of a crude o i l i s independent of the alcohols and surfactants i n the formulations. w a s coined by the researcn &--. I n general. extended t h e concept of equivalent optimal s a l i n i t y t o high concentration surfactant systems (> 2%). 4 . the EACN concept is a very valuable one. (Em)mixture = 11 i xi EACNi. The p a r t i t i o n behavior of the alcohol was traced as one of the causes f o r t h i s aberration. An important finding of t h e i r s is t h a t the EACN of an o i l (crude. the Texas group and Glinsmann5. the EACN of a mixture of hydrocarbons follows the simple mixing r u l e .

They were The The a l k a n e s were pure-grade hydrocarbon from P h i l l i p s Chemical Company. I . and shaken f o r one minute i n a mecha n i c a l s h a k e r (40 Hz). Procedure For phase-volume s t u d i e s . .) o i l from t h a t f i e l d . Table I. Amoco C o s u r f a c t a n t 122 i s a m i x t u r e of e t h o x y l a t e d a l c o h o l s . 6 Various s u r f a c t a n t systems were s t u d i e d f i r s t .C. d . i n which t h e o p t i m a l s a l i n i t i e s of a s u r f a c t a n t system w i t h a series of a l k a n e s w e r e d e t e r mined. Except where noted o t h e r w i s e . used w i t h o u t p u r i f i c a t i o n . Inc. An e x p e r i m e n t a l s u l f p g a t e (Sample No. s u r f a c t a n t s o l u t i o n s were mixed w i t h o i l i n g l a s s t u b e s ( p r e c i s i o n b o r e t o 0. Oklahoma . t h e e f f e c t of WOR w a s a l s o s t u d i e d . . 5 Glinsmann's method of measuring t h e EACN of a n o i l was followed. Usually. By comparing t h e o p t i m a l s a l i n i t y of t h e c r u d e o i l w i t h t h e same s u r CN w a s determined. phenyl dodecane w a s from Eastman Kodak Company. It w a s found t h a t t h e EACN w a s n o t a c o n s t a n t v a l u e This paper r e p o r t s t h e r e s u l t s i n our i n v e s t i g a t i o n on t h e p r o b a b l e c a u s e s f o r t h i s v a r i a tion. As p a r t of our s u p p o r t i n g reseagch program f o r t h e DOE micellar-polymer Y. w e determined t h e EACN of t h e Delaware-Childers (D.001 cm i . F i n a l l y . EXPERIMENTAL Materials The s u r f a c t a n t s used (and t h e i r p r o p e r t i e s ) a r e l i s t e d i n Table I. one week t o s i x months were r e q u i r e d f o r complete e q u i l i b r a t i o n . Of t h e d i f f e r e n t c r i t e r i a o d e f i n i n g f a c t a n t system. w i t h s p e c i a l emphasis on t h e e f f e c t of a l c o h o l type. and a t e r . optimal s a l i n i t i ~ ! " ~ t h e one used h e r e w a s t h e e q u a l s o l u b i l i z a t i o n from phase volume measurements.124 pilot test i n Nowata County.474 5 0. The t u b e s were k e p t i n a n a i r b a t h a t 30' f o r e q u i l i b r a t i o n . u s i n g s e v e r a l s u r f a c t a n t systems. becay&213tudies have shown t h e s t r o n g i n f l u e n c e of a l c o h o l s on phase behavior and IFT The e f f e c t of c r u d e o i l components w a s t h e n s t u d i e d . ) . Suntech Lot 768511) prepared by Suntech Tech. t h e WOR w a s set a t u n i t y .o i l r a t i o s (WOR). P r o p e r t i e s of S u r f a c t a n t s Floodaid 1 4 1 TRS 10-410(a) Suntech I (b) S u l f o n a t e s of mixed x y l e n e s and propylene tetramer 65 372 344-390 (92%) TY Pe Blend of petroleum s u l f o n a t e s mixed w i t h C o s u r f a c t a n t 122 93" Petroleum sulfonate 62 418 400-450 % Active 45 450 wide Equivalent WeAght Equivalent Weight Distribution (a) (b) (c) (80%) Witco Chemical Company. Some of t h e s o l u t i o n s .e s p e c i a l l y t h o s e w i t h h i g h viscosity--were shaken a second t i m e t o e n s u r e thorough mixing.

were obtained by weight measurements. (2) water solubility of the cosurfactant (the solubilities are in the order IBA < TAA < Amoco 122). Distillation of crude oil into distillates and heavy ends were done at 400°F and 10 -Egg. Some experiments were performed with crude oil components. 1.125 Phase volumes were measured with a cathetometer. The following assumptions were made in the calculations: (a) all the surfactant and cosurfactant is in the surfactant phase (this is an incorrect assumption as can be seen later. but the effect on the phase volumes is negligible). tiary amyl alcohol. Asphatene Analysis for acids and bases was by column liquid chromatography determination was by pentane precipitation. TAA = ter.' . surfactant and electrolyte concentrations refer to the concentration in the aqueous phase. I 000 Io o / 5 7 0 T i3s 10. I 80 7 * 8 I I I I I 9 1 0 II 12 A L K A N E CARBON NUMBER 13 I 14 I Figure 1 Optimal salinity of surfactant systems with normal alkanes. Solubilization calculations were fashioned after the work of Glinsmann. Alcohol concentrations were measured with a gas chromatograph. - . (b) the volumes are additive. Vacuum.410. In the present work. Standard correction for the round-bottom end of the glass tubes. the optimal salinity increases with increases in (1) hydrocarbon chain length. RESULTS AND DISCUSSION . and for the oil and water menisci. 0 TAA (A) %Amoco 122 (D) !i i400 c I z U v) YZYO FA 141 ( F ) U I 200 I I ooc.8% Amoco 122 ( E l ' z 0 \ U 4% Suntech I. 2% 600 8 G . and (3) concentration of the water soluble cosurfactant. Optimal Salinities The optimal salinities for a number of systems with normal alkanes are plotted in The observed behavior is the same a8 that reported in the literature that is. IBA = isobutyl alcohol: Amoco 122 Amoco Cosurfactant 122.

17. EACN values.9 6. o i l w a s determined by comparing i t s optimal s a l i n i t y with t h a t of t h e alkanes f o r a given s u r f a c t a n t formulation. and 0. The v a r i a t i o n i n EACN observed above.7 D E 1.3 10. o i l with d i f f e r e n t s u r f a c t a n t formulations Optimal s a l i n i t y meq/l N a C l 680 EACN Surfactant system A 4% Suntech I 2% T e r t i a r y amyl alcohol (TW 5% TRS 10-410 3% TAA 5% TRS 10-410 3% I s o b u t y l alcohol 5% TRS 10-410 1% Amoco 122 5% TRS 10-410 9.0 percent cosur-. * - E f f e c t of S u r f a c t a n t Formulations on t h e EACN of Delaware-Childers O i l The EACN of D. which is i n good agreement with h i s values.8% Amoco 1 2 2 F 12% F 141* A * Sulfonate content equivalent t o 7. The EACN of an o i l w a s determined by comparing t h e optimal s a l i n i t y of t h e o i l with those of alkanes.16 5 0. Table I1 shows t h e results. i n Systems D and E .7 u n i t s i n t h e values i n d i c a t e s t h a t EACN as u s u a l l y determined i s n o t a constant quantity. The s l o p e s f o r t h e Suntech and Floodaid s u r f a c t a n t were 0. From Figure 1 and Table 11. g i v e n e a r l y i d e n t i c a l Also. Systems A and B y with f a i r l y s i m i l a r s u r f a c t a n t s . Thus. This i s i n c o n t r a d i c t i o n t o t h e p r e d i c t i o n of S a l a g e r ' s equation . which p r e d i c t s s l o p e independent of t h e alcohol. . a twofold v a r i a t i o n i n alcohol concent r a t i o n has no i n f l u e n c e on EACN. The value f o r t h e 5 percent TRS 10-410 3 percent i s o b u t y l alcohol (IBA) system obtained by least square f i t is 0.11 f o r t e r t i a r y amyl alcohol and 1. But t h e t r a n s i t i o n from C t o B t o D (with a s i g n i f i c a n t i n c r e a s e i n water s o l u b i l i t y of t h e c o s u r f a c t a n t a t each s t e p ) shows t h a t t h e cosurfactant s p e c i e s has a major i n f l u e n c e on t h e r e s u l t s . C.01. However. There are two p r o p e r t i e s of System F t h a t could c o n t r i b u t e t o i t s d i f f e r e n t EACN value: a wide d i s t r i b u t i o n of equivalent weight and a d i f f e r e n t type of alcohol.126 The observed l i n e a r re1 t nship of I n (S ) versus alkane carbon number (ACN) w a s reported by SalagergYfg. i t is necessary t o conclude t h a t t h e c u r r e n t l y accepted concept apply only over a narrow range of conditions.6% TRS 10-410 o r Suntech I The spread of 4. The s e v e r a l s u r f a c t a n t systems w e r e used t o determine t h e constancy of i t s EACN.12 and 0. Table 11.28. f a c t a n t 122. who found t h t t t h e s l o p e s f o r a l l t h e s u l f o n a t e systems were 0. The optimal s a l i n i t y and EACN of D.2 7.5 B C 197 193 410 590 222 9. C. It is t h e r e f o r e of i n t e r e s t t o study t h e e f f e c t of s u r f a c t a n t formulation on t h e EACN of a number of oils.15 6. n e c e s s a r i l y r e f l e c t s d i f f e r e n c e s i n p r o p e r t i e s between t h e o i l and alkanes.14 with 1. t h e s l o p e s were 0.8 percent f o r c o s u r f a c t a n t 122. r e s p e c t i v e l y .

C . The dist i l l a t e s show a downward s h i f t i n EACN as compared with t h a t of t h e whole crudes.7 5. System F S* E l Dorado o i l EACN S* 261 425 197 EACN 169 196 130 10. C. and D C . C.8) 7.0 8. with r e s p e c t t o t h e s u r f a c t a n t Systems C and P disappeared. o i l s .9 6. Table 111 lists t h e optimal s a l i n i t i e s and EACN of t h e s e systems with a number of o i l s . respectively. o i l s .2 f o r t h e whole Bradford and D.7 615 222 132.5) 295 8. I n t e r e s t i n g l y . C. pounds are known t o complex with t h e s u l f o n a t e s 5*artna.4 (15. Both show a d i f f e r ence of 0.2 8. Table V p r e s e n t s some f u r t h e r d a t a on t h i s . as expected. heavy- ends + decanea 185 10. as compared t o 0. These heavy corn. i t has been recognized tQat t h e equivalence between There are d e v i a t i o n s i n t h e alkanes and o t h e r series of compounds i s n o t exact a l k y l benzene and a l k y l cyclohexane series t h a t are g r e a t e r . These v a r i a t i o n s among t h e various crude o i l s may be compared w i t h t h e d i f f e r e n c e s i n t h e crude o i l composition (Table I V ) .0 11. The mixing of benzene with phenyl dodecane ( t o g i v e an EACN of 8) shows normal EACN with System C. even though t h e d e v i a t i o n i s n o t as l a r g e as t h e D. . Y e t . similar t o t h a t . phase behavior and EACN of most of t h e o i l s s t u d i e d with t h i s system followed a "regular" p a t t e r n .9 and 3.3 10. A downward s h i f t i n EACN w i t h System E. 238 193 103 12. o i l contains a l a r g e r q u a n t i t y of heavy b28es and a c i d s .1 7. digtillate D. showing t h a t t h e d e v i a t i o n can be q u i t e l a r g e when less conventional materials are used. t h e l a r g e d i f f e r e n c e s between Bradford and D C.6) (a) Equal weight r a t i o of heavy ends and decane The behavior of t h e components of t h e s e crude o i l s is q u i t e revealing. which are t h e standards. Optimal s a l i n i t i e s and EACN of crude o i l s and crude oil f r a c t i o n s System C . .127 EACN of Several o i l s With Systems C and F S u r f a c t a n t s Systems C and F were used t o compare t h e b h of d i f f e r e n t o i l s . Oil D.8 c.3 Bradford o i l Bradford D i s tillates Bradford heavyends + decanea D.9 u n i t s w i t h t h e two systems.5 11. System C was chosen because i t has been widely s t u d i e d and because t h e optimal s a l i n i t i e s . Actually. System F w a s chosen f o r its "irregularities". o i l case. On t h e o t h e r hand. The E l Dorado o i l r e s u l t s show d i f f e r e n c e s i n t h e EACN between t h e two systems. Bradford o i l i s high i n p a r a f f i n .4 (20) 10.6 (12.7 (4. t h e f a c t t h a t t h e d i s t i l l a t e s having d i f f e r e n t EACN with d i f f e r e n t s u r f a c t a n t systems i n d i c a t e s t h a t t h e r e are c e r t a i n components i n t h e d i s t i l l a t e s behaving d i f f e r e n t l y from t h e alkanes. t h e f a r t h e r one m o v e s away from EACN of 8. Bradford o i l shows an even smaller difference. Table 111.

9 9.46 34. On t h e o t h e r hand.0 X Acids17 17 X Bases X T o t a l Asphaltenes 55. - Alcohol P a r t i t i o n i n g and Its E f f e c t on EACN W e have shown earlier t h a t r e p l a c i n g i s o b u t y l alcohol w i t h Amoco 122 i n a s u r f a c t a n t formulation causes a downward s h i f t i n EACN.3 E l Dorado o i l Delaware-Chllders o i l G r a v ity "AP I 31.13 0.3 0. an assumption was necessary on t h e molecular weight (MU). i n t e r f a c i a l tension. The d i s t i l l a t i o n temperature suggested a M of 450.04 2.9 5. even f o r "regular" System C. (b) Cut temperature 392°F (corresponding t o molecular weight of 150).82 0.07 5. The weight f r a c t i o n s of d i s t i l l a t e s and heavy ends . It is obvious t h a t heavy ends are n o t equivalent a t a l l t o alkanes even with System C.5 f o r those from D C. phenyl dodecane should have an EACN of 12. phenyl dodecane does n o t observe the simple s c a l i n g l a w g s g f o r both s u r f a c t a n t systems. 32.58 1. and optimal s a l i n i t y of a s u r f a c t a n t . To g e t t h e EACN of t h e heavy ends by themselves from No v a l u e Equation ( l ) . (corresponding t o molecular weight of 280).01 3.07 4. The observed EACN d i f f e r s g r e a t l y from this value and cannot be explained t o t a l y by t h e smaller d e v i a t i o n previously reported f o r t h e case without alcohol €. from D.17 1.9 - 110 145 360 557 with crude o i l s was ob e ved. Table V.System C EACN 7. of MU could be found t h a t was c o n s i s t e n t with t h e EACN v a l u e s . Optimal s a l i n i t i e s and EACN of a k y l benzenes s . b u t gave EACN 20 f o r t h e Bradford heavy ends and 12.o i l system ' . *System E EACN 1. which corresponds J t o an alkane of carbon No. i t was necessary t o c u t t h e v i s c o s i t y by mixing with equal weights of decane. Crude o i l p r o p e r t i e s Bradford o i l 44. . 1:2 molar r a t i o Phenyl dodecane .3) f o r heavy ends. Under t h i s l a w . because alcohol p a r t i t i o n i n g is known t o be t h e prime determlnant of t h e phajts-&h2yior. percent X Aromatic through f r a c t i o n lza 0.3 X P a r a f f i n through c u t 7 (a) Cut temperature 437°F a t 40 nun Hg.0 0.0 s .50 Nitrogen. It is t h e r e f o r e of i n t e r e s t t o study the p a r t i t i o n i n g behavior of these alcohols.02 64.1 28 Table I V . Benzene phenyl dodecane mixture.9 0.6 36. For p r a c t i The d a t a on t h e heavy ends in Table 1 1 are a l o t more "irregular". The EACN of t h e decane mixture obeyed equation (1) only w i t h MW = 160 (EACN 11. C. and t h e d i s crepancy between Systems C and E are very large. 1 cal experlmental purposes. o i l (which l o s t only 4 wt-X I n d i s t i l l a t i o n ) required MW 254 (EACN 18) €or consistency with t h e whole-oil EACN.

129 Since determination of alcohol concentrations in crude oil poses considerable problem due to its wide boiling range--choosing the right column is difficult-only the partition coefficients in hydrocarbons were measured. Amoco 122 in octane and phenyl dodecane is striking. the presence of surfactant did not affect the alcohol partitioning behavior for the systems studied. a = alcohol originally in alcohol originally in b c = alcohol originally in d in the presence of 5% - - Table VII. % Phase Upper Phenyl Middle dodecane Lower Component 1 0.69 <o 01 Component 3 04 .25 08 . Partition coefficient - 0. The results of the partitioning experiments are listed in Table VI.8% Amoco 122 0. The numbers are relevant only at optimal salinity.6aaSd Phenyl dodecane 0. 3.15 0.69 <o.3ga 0. the aqueous alcohol concentration in phenyl dodecane is lowered.26 06 .5 .44 0.9% NaC1. 01 5. Distribution of alcohol components. Partition coefficients of alcohols ~ ~~~~~ Octane 3% IBA 1. It was suspected that the large differences in behavior of alkanes and allcyl benzenes would be reflected in the alcohol partitioning and suggest one cause for the difference between alkanes and crude oils.53 0. however. That is. System E Alcohol concentration.Cd deionized water.32b'd 55' .24 component 3 0. TRS 10-410 at optimal salinity. there is a strong preferential partitioning of the heavier alcohol compounds (components 2 and 3) into the oleic phase of the phenyl dodecane system (Table VII). the table shows that the same value was obtained for Co-surfactant 122 in pure water and in System E at optimal salinity of 3 4 percent. it is not possible to modify the definition of EACN to account for thig2 change in alcohol concentration.3% NaC1. In agreement with the findings of Tosh. 0. 0. % Phase Octane Upper Middle Lower Component 1 0. Thus. It is not known what will be the effect of this change of alcohol composition and concentration on the optimal salinity and EACN. 03 . s & . .84 0. Table VI. but data under other conditions are given for illustration.19 Component 2 0. such changes will make the effort to estimate a "true EACN" impossible. Partition coefficients are not very sensitive to salinity up to 3 percent. in comparison with octane. It is certain.25 0. Component 2 0. The differences between the partition coefficients of . In addition.26 05 . Partition coefficient - Alcohol concentration.

O \ .C. Under optimal condition (vo/vs) = (vw/vs) (v/vs)S* The value decreases with increases i n alkane carbon = Hsieh and Shah” density.C. Within experimental e r r o r . 10-410 I 2. i s an i n t e r e s t i n g property t o examine further.0 +-s >” v) Y \ 5 % TRS. oil Bradford light ends 0. This shows t h a t with System C . according t o Klevens . which shows t h a t t h i s relationship is q u i t e general.130 Solubilization a t Optimal S a l i n i t y The g&)ilization a t optimal s a l i n i t y . Such l i n e a r i t y w a s @served a l s o i n the s o l u b i l i z a t i o n of alkanes i n micelles . Systems D and E a l s o give t h i s l i n e a r r e l a t i o n with alkanes (not shown). Reed and Healy had a l s o noted some p a r a l l e l developments between o i l and water s o l u b i l i t y and s o l u b i l i zation i n micelles.O 0 o 0 0 6 . I n f a c t . 0 Alkane Bradford oil D. being r e l a t e d t o the i n t e r f a c i a l tension . the crude o i l s and d i s t i l l a t e s do not f a l l on the line. Puerto and Gale” found a correlation between (V/Vs)s* and alkane related (V/Vs)s* and the s i d e chain lgngth of an a l k y l orthoxylene sulfonate. I n t h i s work. o i l s modeled by optimal s a l i n i t y are a l s o modeled by the degree of solubilization. * 7 . and other c h a r a c t e r i s t i c s of the solubizates. I f the EACN values- . structure.0- > \ .0 3. I n t h i s case.O 4.3% IBA I I \ I I I I 4 Figure 3 I s a similar p l o t f o r System F.-light ends - 5 . the p l o t is l i n e a r f o r alkanes. Again a l i n e a r relationship f o r alkanes w a s observed. who studied s o l u b i l i z a t i o n as a function of molar v o l v s . we found t h a t the data f o r crude o i l s and crude o i l d i s t i l l a t e s a l s o f a l l on the alkane line. Figure 2 is a p l o t bf (V/V ) versus alkane carbon number s$ for System C.

0.C. 2." 8. t h e f i t is much b e t t e r . 4.would Since increase. perhaps due t o i t s high content of a c i d s . with the d a t a of Table V I I . D. 7 . C.0. It is noted t h a t by i n c r e a s i n g t h e WOR from 1 t o 2. o i l does n o t f i t t h e c o r r e l a t i o n very w e l l . o i l l i n e s are interchanged. by simply i n c r e a s i n g t h e WOR. t h e p o s i t i o n of t h e octane and D. That is t h e EACN of crude o i l changed from 7. 0. the i n c r e a s e i n EACN is as expected.0. oil Bradford . which would mean t h a t less of component l w o u l d b e e x t r a c t e d from t h e aqueous phase. I I I I I I I EACN Figure 3. \ \ 6. and asphaltenes.light ends 0 D. t h e r e i s no such f r a c t i o n a t i o n with octane. This suggests t h a t degree of s o l u b i l i z a t i o n might give mre c o n s i s t e n t values of EACN than optimal s a l i n i t y . -light ends o D \ \ \ \ \ \ \ 3. C. . It is p l a u s i b l e t h a t t h e WOR e f f e c t is r e l a t e d t o alcohol p a r t i t i o n i n g .131 determined with System C are used. Even so. bases. ". Consider t h e case of phenyl dodecane. Alkane Bradford oil A D.7 t o higher than 8. 0- --- \ +'. According t o Figure 1. and presumably w i t h o t h e r alkanes.0. The proportion of water-soluble component i n t h e aqueous phase.0. Effect of Water-Oil-Ratio (V/Vs)s* versus equivalent alkane carbon number Q on EACN The e f f e c t of water-oil-ratio (WOR) can be seen from Figure 4. t h i s should l e a d t o an i n c r e a s e i n S 4 .0' 12% FA 141 I. Increase of WOR reduces t h e proportion of t h e o i l phase. 5. C.

Optimal s a l i n i t y f o r phase behavior. non-hydrocarbon components of the crude o i l might have contributed p a r t i a l l y t o t h i s "abnormal" behavior. Volume of o i l solubilized i n the surfactant phase. EACN S* vO 0 vs vW . Volume of water solubilized i n the surfactant phase.C. W e would l i k e t o advise caution when applying the EACN concept.132 30C * . Lacina f o r the analyses on crude o i l s components. NOMWCLATURe A m Alkane carbon number. I WOR 2 3 Variation of optimal s a l i n i t y with water-oil-ratio CONCLUSIONS The EACN concept was found t o be i n e r r o r +hen s y s t e m involving ethoxylated alcohols and/or aromatics were used. B. Volume of water o r o i l solubilized per u n i t volume of surfactant a t optimal s a l i n i t y . This is under investigation. ACKNOWLEDGMENT The authors wish t o acknowlege the help of J. The higher boiling. Volume of surfactant i n the surfactant phase. oil I oc 0 Figure 4. Green and J. " 20c A Octane o D. Alcohol p a r t i t i o n i n g is found t o be an important f a c t o r causing t h i s deviation. meq/l NaC1. Equivalent alkane carbon number.

W. Scaling Rules t o Binary Hydrocarbon Mixtures. and I n t e r f a c i a l Tension of OilfBrinefSurf actantfAlcoho1 Systems.. J. H: . Schechter.: Micellar Polymer Flood P r o j e c t i n Nowata County.. S . J. R S . Morgan. W. J. " I n t e r f a c i a l Tension and Phase Behavior of S u r f a c t a n t Systems. "Cosurfactants i n Micellar System Used (10) Jones. . W. S. 3944 (1977).: SPE J. R S. R N. R. J.. Tension o r Phase Behavior. H: mum Formulation of SurfactantfWaterfOil Systems f o r Minimum I n t e r f a c i a l SPE J. J. 2. and Wade. 74. R.. H . Okla. B. Sci. and Schechter. R." J. R. L. September 23-26.. .. on Microemulaion Composition.. ." (11) Salteli. C. L. P.: "The E f f e c t of Chain Length of O i l and Alcohol As W e l l Aa S u r f a c t a n t t o Alcohol Ratio on t h e S o l u b i l i z a t i o n . and Shah. Cayias. 6 f o r T e r t i a r y O i l Recwery..: "Surfactantflooding w i t h Microemulsions Formed I n s i t u E f f e c t of O i l Characteristics." Paper 5813 presented a t t h e SPE Symposium on Improved oil Recovery. J.. . Cayias. R. J. J. Burtch. C. R. 59.. l . Cash. G.. H. D. 351-357 (1976). eds. B. J: .. Fournier. Sci. l . 6068 (1976). 1979. f o r Low I n t e r f a c i a l Tension. held i n Las Vegas. Coll.. L. and Wade. and Tham. W. (13) Wade. Schares. Academic P r e s s (1977). T. Schechter.: .. Schares. C. la. La J o l l a . 107-115 (1979). &266-279 (1981)... "The Influence of Type and Amount of Alcohol on SurfactantOil-Brine Phase Behavior and Properties. K.. . R. Shah.133 REFERENCES Cash.. "Modeling Crude O i l s f o r Low I n t e r f a c .. W. Salager." J.. R S . presented a t t h e 52nd Annual F a l l Technical Conference and Exhibit of t h e SPE-AIME." i n Improved O i l Recwery by S u r f a c t a n t and Polymer . Jacobson. M. "Calcium E f f e c t i n t h e DOE SurfactantLorenz. and Wade. I n t . Morgan." Paper SPE 8326 presented a t t h e 54th Annual F a l l Technical Conference and Exhibition of t h e SPE-AIME." .. To be published. Cayias. K.. Fournier. J. L.. C. i n Denver. "ERDA's Walker. held i n Tulsa. and Wade.. H : "The Application of Low I n t e r f a c i a l . Nevada." Paper SPE 6594. Thomas. M K. H : "The U t i l i z a t i o n of . Coll. W. Schechter. Petroleum Sulfonates f o r Producing Low I n t e r f a c i a l Tensions between Hydrocarbon and Water. . Schechter. Colorado. and Wade.." O i l and Gas J. L. MaCalllster. Schechter. W. S. 1977. Jacobson.: Flooding A Review. S. 31-38 (1977). 0. I n t . Phase Behavior. Schechter. L. F. D. I n t . and Schechter. Coll. K D. L. R. and Healy." SPE J. . . G. 6 Glinsmann. L. L. J. Cayias.. Salager. 1977. presented a t t h e SPE-AIME I n t e r n a t i o n a l S p p o sium on O i l f i e l d and Geothermal Chemistry. October 9-12. C a l i f o r n i a . J. 242-252 (1978)." SPE J.. . March 22-24. T. - "Opti. P. Sci. 0.. R. "The Influence of Alcohols (14) Baviere. S: ." O i l and G a s J. and Dreher. R.. Flooding. Polymer P i l o t Test. W." . & .. . B : "Modeling Crude O i l s . June 2728. and Lorenz. D. 161-167 (1976). 1976. (12) Hsieh. i a l Tension. J. C. D. 59. S .." Paper 6843. W. "Physicochemical Aspects of Microemulsion Reed. G. and Wade..

" SPE J. (20) Clementz. The University of Texas at Austin. Tech. Preparation and Related Studies. Jones. 4 . - (18) Salager. . D.. . G P. R L. . C and Gale. Coll. Virginia (1979). "Multiphase Microemulsion (24) Huh. E W : "Large-Scale Samples of Sulfonates for Laboratory . in a Micellar Solution System. .. National Technical Information Service. 111. (16) Malmberg. Chrom.. (21) Puerto. J G. Interfacial Tension in Multicomponent Systems. 2. Willhite..134 (15) Dominguez. 1-74. J L : "Physico-Chemical Properties of Surfactant-Water-Oil . J E.. C : "Interfacial Tensions and Solubilizing Ability of a Microemul. Int." in Solution Chemistry of Surfactants." Report No. . . . P.. 17.. 8 . W E : "Deactivation of Petroleum Sulfon. 147-160 (1976). r : "Low Interfacial Tension and Miscibility Studies for Surfactant Tertiary Oil Recovery Processes.231-250 (1981). 193-200 (1977). (23) Healy." Chem. D. D G: Systems. sion Phase That Co-exists With Oil and Brine. 1091-1093. Sci.D. . .. (25) Fleming.15261531 (1980). pp. Virginia 22161 (1979). . Int. Commerce. K . Reed. 2.. New York. Mixtures---Phase Behavior. W : "Phase Behavior of . Dissertation. ." J Coll. Rev. H. Studies in Tertiary Oil Recovery." J. S C. 1977. C A. 1950.. B. . Microemulsion Systems with Emphasis on Effects of Paraffinic Hydrocarbon and Alcohols.. National Technical Information Service. Springfield. R. .." J Phys.. . - (22) Tosch. Department of Commerce. G R : "Theory of . . . Mobile Phases Containing Aliphatic Carboxylic Acids I1 Applications in Fossil Fuel Characterization.. J B and Hoff. 2. M. . A W : "Distribution Equilibria .. T J . (17) Green. Vol. . M and Gerbacia.." J Pet. . 673-697 (1979). 31. and Green. 408-426 (1979). Microemulsion Formation and Interfacial Tension. Chem. (19) Miller. and Fort." Ph. U S Department of . Vinatieri. and Stenmark. (26) Klevens. L.. ed. . W W : "Estimation of Optimal Salinity and Solu.. ates by Crude Oils. 2.: "Solubilization." SPE J. Sci. . N. Gttal. and Adamson. 297-306 (1969). Springfield. . . DOE/BC/10007-4. Plenum. U S. bilization Parameters for Alkylorthoxylene Sulfonate Mixtures." J. September 1977. and Glinsmann. FE-2605-20. pp. R J : "Liquid Chrometography on Silica Using . ." Report No. D. . W C.

This would have a very adverse effect on oil bank formation. oil ganglia which have been trapped at small pore throats are released by lowering the interfacial tension. The application of these measurements to EOR systems is illustrated with examples of surfactants which give widely different oil displacement profiles. Sunbutyen-Tharnes. . BP Research Centre. increasing interfacial shear viscosity can reduce coalescence rates for oil drops in water almost to zero. JONES The British Petroleum Company Limited.CHEMICAL FLOODING 135 DYNAMIC INTERFACIAL PHENOMENA RELATED TO EOR J. In all except the initial release it could be argued that it is the dynamic properties of the interface such as the dynamic interfacial tension and the interfacial rheology which will govern each individual and hence the overall process. The effect of interfacial dilatational rheology on coalescence phenomena is illustrated with data for water-in-oil demulsifiers. the "Drop Volume Dynamic Tensiometer" allows dynamic interfacial tension to be determined over a wide range of rate of fractional area change. L. Techniques will be illustrated with results for pure oils as well as crude oils. Middlesex. CLINT. Chertsey Road. The behaviour of aqueous surfactant systems against crude oil is very different for fresh systems compared with systems where the phases have been pre-equilibrated. J. For crude oil films in water. which in turn is controlled by film drainage from between colliding droplets. This is the pulsed drop experiment which has experimental advantages over the interfacial trough method and allows parameters to be determined over a wider range of frequencies. A new method for the measurement of interfacial dilatational rheological parameters of oil/water interfaces is described. E. INTRODUCTION In an enhanced oil recovery process. and the significance of these data for EOR processes will be discussed. A technique developed by BP. NEUSTADTER and T. TWI 6 7 WV ABSTRACT The relevance of dynamic interfacial tension and interfacial rheology to EOR is discussed. This paper reports some novel methods for measuring dynamic interfacial tension and interfacial dilatational rheology which work very well for crude oil-water systems. increasing interfacial shear viscosity greatly reduces the rate of thinning. For the reverse system. H. prevented from being retrapped by maintaining a low tension (dynamic) and encouraged to coalesce to form an oil bank. The ease of oil bank formation is influenced by the kinetics of coalescence.

DROP VOLUME DYNAMIC TENSIOMETER Oil from a syringe pump is pumped at an accurately known volume flow rate to a syringe needle inserted through a septum cap into a small glass cell surrounded by a water jacket. For convenience of observation an image of the tip and drops formed is obtained using a microscope and TV camera and displayed on a monitor screen. If n is the number of drops per unit time then the volume of each drop I v = - Q n ..136 DYNAMIC INTERFACIAL TENSION This technique is essentially an extension of the drop volume method for interfacial tension and is illustrated in Figure 1... WATER JACKET ( I 'k SYRINGE PUMP SEPTUM CAP FIGURE 1 . (1) The interfacial tension y can then be calculated using the usual formula .. The experiment consists very simply of measuring the number of drops formed in a fixed period of time and repeating at a whole range of volume flow rates Q. (2) . The tip'of the syringe needle is ground flat and the inside and outside diameters determined accurately.

All aqueous solutions were made up in filtered sea water. There were large differences in the results depending on whether the oil/water systems were preequilibrated or whether they were fresh.. The latter may be the inside or outside tip radius depending on the wetting conditions.. Figures 2 and 3 illustrate the type of results obtained using Forties crude oil against two different surfactant systems. d I E FIGURE 2 . (3) Hence we are able to estimate both the interfacial tension and the rate of fractional area change simply by measuring the rate of formation of drops at a known volume flow rate.137 where P P ' is the density difference between the oil and water phases. If we make the assumption that the drops are spherical then the rate of fractional area change at the time when the drop detaches can be shown to be - .DYNAMIC INTERFACIAL TENSION -FORTIES CRUDE/SOOO PPM SURFACTANT "A" AT 7OoC . The crude oil used was a well head sample free of any additives such as demulsifiers or corrosion inhibitors. and R is the radius of the tip to which the drop is attached. Figure 2 shows the dependence of dynamic interfacial tension on rate of fractional area change for a surfactant system "A" at 7OOC.

The interesting point about these two systems is that they give totally different oil removal profiles when tested in a model sand column test.DYNAMIC INTERFACIAL TENSION I FORTIES CRUDE/50OO SURFACTANT "B" AT 70OC PPM This time the pre-equilibrated system gave interfacial tensions which were very small and at time9 unmeasurably so (only the one which could be measured is shown). For surfactant "A" which gave low fresh tension but high equilibrium tensions. The tensions for the fresh system showed the normal dynamic effect rising rapidly with modest rates of area increase. In contrast to this is the behaviour of the surfactant system "B" shown in Figure 3. which gave high fresh tensions but very low equilibrium tensions. removal of residual oil was Complete but required a very large number (15) of PV.138 The difference between fresh andpre-equilibrated systems is immediately apparent. removal of oil was rapid but incomplete. The preequilibrated tension rises rapidly at moderate rates of area increase whereas the tension of the fresh system stays remarkably low until very high rates of area change are reached where the area is roughly doubling every second. 6 5 d I E 0 1 FIGURE 3 . . The dashed line indicates that the tension remains low even at high rates of area change. About 35 per cent of residual crude oil was removed in less than 2 pore volumes (PV). For surfactant "B".

and tension is calculated by measuring the excess pressure inside the drop with a sensitive pressure transducer. However. and dilatational viscosity. (c) The method uses large quantities of oil with a large area exposed to air allowing loss of light ends.139 Admittedly the shape and duration of the oil displacement curve will be dependent on more than just the dynamic tension behaviour. the distinction between the two systems above is clear and the oil displacement behaviour is logically related to the dynamic tension properties. Surface wettability and the degree of adsorption will also be important factors. at each frequency. INTERFACIAL DILATATIONAL RHEOLOGY For the measurement of interfacial dilatational rheology the method employed in the past has been that of dilatational modulus measurements at various frequencies using an interfacial film balance (1). (b) Good results depend on the rapid response of the Wilhelmy plate and the maintenance of a well defined contact angle. These changes. allow calculation of Ed# the dilatational elasticity and nd. The experimental arrangement is shown in Figure 4. CHART RECORDER SENSITIVE PRESSURE TRANSDUCER SYRINGE PUMP WATER FROM THERMOSTAT FIGURE 4 - PULSED DROP METHOD FOR INTERFACIAL DILATATIONAL RHWlLOGY . The method involves propagation of longitudinal waves of the frequency of interest and measuring changes of interfacial tension with a Wilhelmy plate. Also the apparatus is not easily used at temperatures much above ambient. Area changes are calculated from drop diameters and the tip diameter. This technique suffers from a number of disadvantages including (a) Measurements are reliable only at fairly low frequencies where the wavelength of longitudinal waves is long compared with the distance between oscillating barrier and Wilhelmy Plate. We have developed a new technique which uses a small drop of oil pulsed in water. together with the phase differences between them and the area changes.

TRANSIENT PRESSURE INSIDE DROP FOLLOWING SUDDEN EXPANSION The equilibrium pressure after the experiment is lower than that at the beginning because the drop radius is larger. the final radius r ~ . TIME/MINS E'fGURE 5 . All of the pressure trace after the rapid rise is assumed to take place at a constant drop radius. The radius of tip needed depends on the region of interfacial tension being investigated. the single pulse Fourier transform method was used (2).. The shape of a typical pressure trace is shown in Figure 5. When the cell containing the aqueous solution of interest is sufficiently well thermostatted the drop radius (rl) is measured. Then the interfacial tension at any time Y (t) is given by . type SE 1150/WG. Output from the transducer is displayed on a chart recorder. (4) . Instead of the conventional oscillatory method for dilatational modulus measurements. the excess pressure inside the drop was measured using a transducer from SE Labs (EMI) Ltd. a fixed volume pulse is injected from the syringe pump over a short period of time which increases the radius to r2 and then the variation of pressure with time is followed on the chart recorder..140 The oil drop is formed at a ground glass or stainless steel tip.

.. (8) The real part gives us the dilatational elasticity:- = Ed(w) = [ : AA/A Ay(t) sin wt dt . (9) The lmaginary part gives the dilatational viscosity:- E" = *wnd(w) = w J: W A m AY(t) cos wt dt . It is convenient to take approximately 100 readings ffom the.. .141 The interfacial modulus is usually written:- t* = dy/dlnA = t' + it" ... A desk top microcomputer is adequate afthough a little slow... (6) For a perfect step function (instantaneous area change):- . Results are shown in Figure6 for the real (elasticr damponent of the modulus and in Figure 7 for the imaginary (frequency x viscosity) component. Equations 9 and 10 can be used to calculate td and n at any frequency from the decay curve.5 to prevent ionisation of the acid. (10) where w = angular frequency (radians per second).. to the frequency function. The method was evaluated using a model system of 10 ppm stearic acid dissolved in n-decane against distilled water adjusted to pH 2. Therefore:t*(w) = (7) ~ A / A1. (5) Taking Fourier transforms of the numerator and denominator coverts the perturbation time function AA(t)/A and the response time function y(t)....decay curve for use in these computations.. Thus:- €*(W) = . iw f- Ay(t) [cos wt - i sin otldt .

FILLED CIRCLES DROP METHOD - 10 8 6 d I E a W 4 2 0 10-1 1 FREQUENCY/Hz FIGURE 7 - IMAGINARY PART OF INTERFACIAL DILATATIONALMDDULUS. SYSTEM AND SYMBOLS AS FOR FIGURE 6 .TROUGH METHOD.REAL PART OF INTERFACIAL DILATATIONAL MODULUS FOR 10 PPM STEARIC ACID IN n-DECANE/DISTIUED WATER pH 2.5 AT 25OC.142 20 - I I I I1111 I I I11111~ I I111111 I I iiiirr - 15 d 10 E - I \ W 2 5 - 010-3 10-2 10-1 1 FREQUENCY/Hz FIGURE 6 . OPEN CIRCLES .

The shapes of the curves of E ' and E" are very close to those expected for a single relaxation mechanism. FIGURE 8 10 PPM STEARIC ACID IN n-DECANE/DISTILLED WATER pH 2. 4 I E 0 5 10 E ' / ~ N 15 20 m-l COLE-COLE PLOT FOR INTERFACIAL DILATATIONAL MODULUS. The relaxation time ' = I l/w = 64 sec. 0 0 2 5 Hz which is an angular frequency w = 2nu = 0. This is illustrated more strikingly in Figure 8 where a Cole-Cole plot ( E " against E l ) is shown.0157 s ' . .5 x 10-5 mol am-31. Agreement between the drop and trough methods is very good over most of the frequency range except possibly for the values of E " at intermediate frequencies. The implication is that the techniques measure real dilatational parameters and not artefacts. also using the Fourier transform method. This is the characteristic frequency of the relaxation process. FILLED CIRCLES DROP METHOD - - - The single relaxation mechanism implied by Figures 6 .5.143 In each case the results are shown in comparison with data obtained previously using the interfacial trough technique. The maximum in C'' which corresponds to the inflection point in E ' occurs at U = 0 . 7 and 8 is presumably diffusion of the stearic acid from the interface into the bulk decane phase. Each set of data is the average of three separate runs. OPEN CIRCLES TROUGH METHOD. This would seem to be a very reasonable relaxation time for a diffusion controlled mechanism in a dilute system [c = 10 p p = 3. A single relaxation mechanism has a semi-circular Cole-Cole plot and the data from interfacial trough experiments clearly follow a semi-circle quite closely. Again agreement with pulsed drop data is encouragingly good considering the great difference between the two techniques.

EFFECT OF INTERFACIAL RHEOLCGY ON COALESCENCE PHENOMENA The pulsed drop method has not yet been used to investigate coalescence phenomena. 6 0 5 - FORTIES/FORMATION WAT +10 PPM DEM 1113 +10 PPM RP 968 A 0 4 rl Ei 3 - - + 5 PPM CC 6601 w 2 1 0 1 FREQUENCY/Hz FIGURE 9 .144 The main advantages of the drop method over the trough method are (a) The system can be enclosed so that loss of light ends from crude oils is avoided. Results are shown in Figure 9 for E " as a function of frequency and as a Cole-Cole plot in Figure 10. dilatational parameters for the Forties crude oil/formation water interface can be quoted which were determined by the trough method. The influence of various water-in oil demulsifiers was investigated. (b) The system can easily be thermostatted at high temperatures. as an illustration of how interfacial dilatational rheology is involved in coalescence processes which are essential to oil bank formation.EFFECT OF VARIOUS DEMULSIFIERS ON IMAGINARY (VISCOUS) COMPONENT OF INTERFACI~RILATATIONAtMODULUS FORTIES CRUDE/FORMATION WATER AT 25OC . However. (c) The system is compact and very small quantities of materials are used.

Further work on this topic is planned. At this stage the mechanistic implications of these observations are not fully understood.5 22 9 +10 ppm RP 968 +10 ppn DEM 1113 The major effect of the &emulsifiers is to remove the relaxation process characterised by a long relaxation time. frequencies the dilatational viscosity can be greatly increased. These are compared with relaxation times for systems with low concentrations of three water-in-oil demulsifiers in the table below. Relaxation Time (Seconds) Forties crude/formation water 87 4 1 + 5 ppn CC 6601 4. The interface without additives gives two separate peaks indicating two different relaxation mechanisms are involved. However. normally higher.145 0 2 4 6 E'/~N 8 m-l 10 12 FIGURE 10 .COLE-COLE PLOT FOR SYSTBMS IN FIGURE 9. These demulsifiers are also found to reduce the interfacial shear viscosity of the crude oil/water interface. . SYMBOLS AS IN FIGURE 9. From the positions of the peak maxima we can calculate relaxation times for the two processes of 87 sec and 4 sec. from Figure 9 it can be seen that at some frequencies the dilatational viscosity is reduced whereas at other. Shorter relaxation times are expected to mean more rapid film drainage (3) and therefore more rapid coalescence.

is relevant to oil bank formation. 4 0 0 0. However. The reverse system. These implications are borne out by measurements of interfacial shear viscosity at the crude oil/water interface. Results for Iranian Heavy crude and for Forties crude in distilled water are shown in Figure 11.146 MEASUREMENT OF DRAINAGE RATES FOR SINGLE OIL FILMS IN WATER Direct evidence for the influence of interfacial shear rheology on the kinetics of drainage of thin films has been obtained by measuring the thickness of crude oil films in distilled water. . In contrast the Forties crude in water film drainage curve is not a straight line and indicates much more rapid drainage of the film than can be accounted for by the lower bulk viscosity of Forties oil.o (t/min)-4 1. there is clear evidence in the literature for the reduction of coalescence rates for crude oil drops in water when interfacial shear viscosity is increased ( 4 ) . drainage of water films from between colliding oil droplets.0 FIGURE 11 - FILM DRAINAGE - CRUDE OIL FILMS IN DISTILLED WATER AT 25OC For the Iranian Heavy case the thickness is proportional to t-+ in accordance with the Stephan-Reynolds equation indicating that drainage is essentially from between two rigid interfaces. The technique was the same as that used to measure thickness of oil films in air ( 3 ) . This implies that the Forties crude/distilled water interface is much more fluid compared with the Iranian Heavy case. Because crude oil is opaque it is not possible to perform experiments analogous to the single oil film drainage measurements outlined above. Measurements of the intensity of light reflected from the single oil film were used to calculate film thickness as a function of time. Over a period of hours the shear viscosity of the Iranian Heavy/distilled water interface builds up to quite high values whereas that for the Forties/distilled water interface remains low.5 2. Using the biconical bob shear rheometer the results shown in Figure 12 were obtained. but having the whole cell filled with water.5 1 .

2. For different surfactant systems which have markedly different oil removal profiles from sand columns. Both systems can be used with the single step pulse Fourier transform method.5. 6.6 E 8 H . SHEAR VISCOSITIES AT 25OC Clearly an important quality of an EOR surfactant will be the maintenance of low interfacial shear viscosity as an aid to oil bank formation. 5. Drainage of crude oils films in water can be followed by reflectance measurements of thickness. 4. two separate relaxation processes are detected. Water in crude oil demulsifiers remove the mechanism with the longer relaxation time. For a pure system of stearic acid in n-decane against distilled water at pH 2. For a Forties crude/oil formation water interface. The results agree well with those determined using an interfacial trough. dynamic interfacial tension behaviour can be completely different.4 3 3 w X v) a V . A pulsing drop method has been devised which can measure the 3. interfacial dilatational rheological parameters for oil/water systems.CRUDE OIL/WATER INTERFACIAL. CONCLUSIONS 1. . presumably diffusion to and from the interface. Drainage rate depends critically on interfacial shear viscosity.2 H 2 a B FORTIES I I z H 0 1 2 3 4 5 INTERFACE AGE/HOURS FIGURE 12 .4 E \ P * v) v) . A dynamic drop volume technique can be used to determine dynamic interfacial tension in crude oil/water systems as a function of rate of fractional area change. the complex dilatational modulus gives a semi-circular ColeCole plot indicating that relaxation at the interface is due to a single mechanism. presumably diffusion and molecular rearrangement.

TESEI.~ ) Number of drops per unit time (s-l) Excess pressure inside drop (Nm-2) Time (s) Interfacial tension (Nrn-l) -1 Complex interfacial dilatational modulus (Nm ) Real part of dilatational modulus (Nm-1) Imaginary part of dilatational modulus (Nm-l) Interfacial dilatational elasticity (Nm-1) Interfacial dilatational viscosity (Ns m-l) Frequency (cyclic) (Hz) Density (kg N 3 ) Relaxation time ( 8 ) Angular frequency (s-1) n AP t Y €* E ' € " Ed Ild U P ' I w ACKNOWLEDGEMENT Permission to publish this paper has been glven by The British Petroleum Company Limited. K. SAMPATH.C.E. R "Spectral Data of Surface Viscoelastic Modulus Acquired Via Digital Fourier Transformation" J. SHAH. AND NEUSTADTER. N. E.P.M. Colloid Interface Sci.T.148 NOMENCLATURE A Q R V 9 Area of interface (ma) Volumetric flow rate (m3 s-1) Tip radius (m) Volume of drop (m3) Acceleration due to gravity (m s . "Foaming of Crude Oils: A Study of Non-Aqueous Foam Stability" Chemistry and Industry. WASAN. 181. AND WHITTINGHAM. I. K. G.L. AND CINI. 1979.L.. D. JONES. McNAMARA. REFERENCES GRAHAM. 316. LOGLIO.J. S.. 3rd International Conference on Surface and Colloid Science. . 17. U.. AND ADERANGI.. 71. D. E. Plenum Press. (1979).1. T. p 53. J.. Rheology. NEUSTADTER. "The Role of Coalescence Phenomena and Interfacial Rheological Properties in Enhanced Oil Recovery: An Overview" J.81. "Interfacial Rheological Properties of Crude Oil Water Systems". CALLAGHAN. (19791.J. 23.. Stockholm. in the press.

CHEMICAL FLOODING 149 BEHAVIOR OF SURFACTANTS IN EOR APPLICATIONS AT HIGH TEMPERATURES LYMAN L. Various organic chemicals have been investigated for use with steam as flow diverters to minimize gravity override. Steam injection is the most frequently used process. Anionics appear to form precipitates with rock minerals. This has introduced additional requirements with respect to the temperature compatibility of chemicals used in these reservoirs. INTRODUCTION Much of the unrecovered oil in the United States occurs in heavy oil deposits. The results vary with the surfactant. At 180°C half-lifes for petroleum sulfonates varied from 1 to 11 days. Adsorption is temperature dependent although the experimental results for the anionics were obscured by precipitation. have all s h a m an order of magnitude reduction in interfacial tension at temperatures in excess of 120%. This problem increases with increasing temperature. Adsorption of nonionics were observed to decrease with increasing temperature at low concentrations but to increase with temperature at high concentrations. . Although the temperature requirements for chemicals to be used at steam tempiratures are much more rigorous. Surfactants are being evaluated as possible additives which will reduce the residual oil saturation in that portion of the reservoir which is flooded only with hot water during steam drive. Of particular interest is the thermal stability. therefore. mostly in California. One of the problem with steam is that it tends to finger through the formation and to override the oil. Those surfactants we investigated decomposed by first order kinetics. The only feasible way to accomplish this objective is to heat the oil in-place. be quantitatively expressed in terms of the half-life of the surfactant. This can be done by either steamflooding or in situ combustion. high temperatures are also encountered in the deeper reservoirs which are currently being considered for enhanced oil recovery. Activation energies were measured and these data can be used to predict half-lifes at other temperatures. Solubility of nonionics is known to be affected by temperature. Large accumulations of heavy oil are also known to occur in Venezuela. To recover this oil the viscosity must be reduced by orders of magnitude. Interfacial tensions have a l s o been measured as a function of temperature. Canada and elsewhere. HANDY Department of Petroleum Engineering University of Southern carifornrb ABSTRACT Temperature sepsitive properties of some anionic and nonionic surfactants used in EOR operations have been measured. Mixtures of sulfonates. The stability can. This has given rise to the investigation of various chemical additives which will improve the process. At the cloud point they dehydrate and become less soluble. Mexico. however.

A modification of the Epton titration has been proposed by Mukerjee which is reported to be more quantitative than the original method. The anionics included sodium dodecylbenzene sulfonate. The decomposition rate for a reaction following first order kinetics is - dC/dt = kt . The petroleum sulfonates included TRS 10-80 manufactured by Witco and Petrostep 465 manufactured by Stepan Chemical Corporation. These results indicate that the decomposition reaction for the anionics is first order. therefore. The most extensive of these is that of Handy et al. TRS 10-80.150 In the present paper we are concerned. In comparing our work with that of others. the absorption at these characteristic wavelengths is decreased. plot of the logarithm of concentration versus time was linear. The most common procedure is the Epton titration. The decomposition reaction for the petroleum sulfonates is the following: ArSO. Particular care was taken to eliminate air from the bombs. + H30= It would be possible. Dowfax 240 was from Dow Chemical Company.’ Data have also been reported by others for the petroleum sulfonate. a major factor is the method used for chemically analyzing for the active surfactant. Disubstituted aromatic rings have a characteristic absorption wave lengths at 220-240 nm and 260-280 nm. an acidic Dowfax sulfonate and several petroleum sulfonates. pH versus time was also observed to be linear. The surfactants were mixed at various concentrations without salt and aged at elevated temperatures in Teflon containers in Parr Acid Digestion bombs. THERMAL STABILITY A limited number of studies have been reported in the literature on the stability of surfactants suitable for oilfield operations at temperatures in excess of 100°C. but problems are also encountered with polymers at high reservoir temperatures. finally.2 In our earlier report results were presented for anionic and nonionic surfactants. The bond which ruptures during high temperature aging is the sulfur-aromatic ring bond. The decrease in the concentration of the active surfactant can be measured quantitatively from the change in the peak heights. The other anionic surfactants gave similar behavior. the effect on adsorption onto the solid matrix. The nonionic was an alkylphenoxypolyethanol manufactured under the trademark of Igepal CO-850 by GAF. primarily. which involves a dye transfer between two phases. to monitor the reaction from a measurement of the pH. When the sulfur-aromatic ring bond ruptures. Representative data from reference 1 are given on Figures 1 and 2 The . with surfactants. We used instead W spectrophotometry. We have not tested that procedure. We found the end points difficult to detect in this procedure. Four aspects of the effect of temperature are considered: the effect on the stability of the surfactants. + 2H20 ArH + SO. Concentrations were determined from a comparison of peak heights with those observed for solutions of known concentration. The alkylphenoxypolyethanols could also be analyzed by W absorption because these compounds also have a disubstituted aromatic ring. Long term aging tests were conducted in sealed borosilicate glass vials. but no temperatures were stated for those experiments. the effect on water-oil interfacial tensions and. the effect on solubility.

From the slope of the plot one can determine the activation energy.151 c = c0c-kt or log c = -+ -kt 2. 1-Concentration of TKS 10-80 as function of heating time at 149°C and 204°C Fig. . With the activation energy one can determine rate constants and half-lifes at other temperatures. the elapsed time is equal to the half-life of the surfactant. 2 . This is particularly useful in estimating the stability of surfactants at lower temperatures for which the decomposition rates are low and long times would be required to measure the half-lifes. = 243 x IU3M I 0 99 OI 144 Im s HEATING TIME (HRS) 240 Fig. In the equation log -E a 2.303 log co In these equations C is concentration in moles per liter. Figure 3 is a plot of the log of the rate constant versus the reciprocal of the absolute temperature for TRS 10-80.987 cals and T I s the absolute temperature in OK. R is 1. C is the initial 0 concentration.303 RT + a Ea is the activation energy in cals/mole. - TRS 10-a0 70 C. One can also show that when C/Co 4. t is time in days and k is the rate constant in dy-. This plot is typical of those obtained for the surfactants which were tested.pH of TRS 10-80 as function of heating time at 140'C If one has rate constants at several different temperatures one can determine the activation energy for the reaction. as' The rate constant is determined from the slope of the semilog plot.

2 25. 28 of temperature on the solubility of f IO~PK-IJ petroleum sulfonates have not been reported.0 24.5 180 130 157 180 130 180 t$(days) 6.75 . Because of its high activation energy relative to the other surfactants. These results would be expected to be representative for aryl sulfonates. Wt.9(pH) 17.4 12. Table 1 At 180°C Petrostep 465 is the most stable of the surfactants we investigated.0 2 1 ?A? 2.0 26.0 70 . Reed has measured a significant increase in the solubility of rock minerals at steam temperatures.22 Ea(kcals) 24. apparently.21 I 1 1 I I Quantitative data on the effect 2. I 1 I 1 TRS 10-80 I w 0.6 1. this surfactant would have a half-life of about 16 years at 100°C.13 .22 13.2 Dowfax 2AO TRS 10-80 500 415 Petrostep 465 Igepal CO-850 465 1100 8.2 25.6(W) 6.The rate constant (k) by several authors that precipitation 1 of the sulfonates occurs at the a s tunction of -(OK-') for T higher temperatures in natural sand'IKS 10-80 stones.5 TABLE 1 SUMMARY OF DECOMPOSITION DATA FOR SURFACTANTS Surfactant NaDDBS Mol. "C 130 180 150 180 177 149 204.0 NA 12. 444 108 11 .. the presence of salt in the solutions decreases the solubility of the sulfonates. 348.152 A summary of decomposition data for several surfactants is given in . None of the surfactants have adequate stability for use at normal steam temperatures. These precipitates are likely to decrease in solubility with increasing temperature.75 5.0 26.84 .84 8. This occurs not as a result of a direct temperature effect on the solubility of the surfactants but. SOLUBILITY 0.5 Temp. but evidence has been cited Fig. In general. * s 5 .4 3.3 24 . as a result of an interaction with minerals in the porous media. but better stabilities have been informally reported for alkyl sulfonates.' The petroleum sulfonate ions form precipitates with divalent cations.7 0. 3 .4 12.4 25.

41 1.5x10-' 95 . At very specific temperatures called the cloud points.72 3. brine and an increase in temperature to 10C 8' did redissolve sulfonate still precipitated at 40°C after the distilled waterflood.78~ 2.09 x l o' 7. 4-Desorotion curve for NaDDBS with distilled wate.Ziegler observed turbidity in the produced fluid from a Berea sand pack when sodium dodecylbenzene sulfonate solutions were injected at a concentration of 1400 pmols/liter. However. I) -40.705 0. In this experiment the sand pack was flushed with 1374 pmols/liter . finallv.100 CMC = 100 w l / L Cloud Points cn trunOl/L) 73 Cloud Point ("C) >180 113 106 366 640 Sorption Properties Temperature ("C) 45 70 Keq (dm'/pmol) 5.831 kl (dm'/pmol. The cloud point is a function of the molecular weight of the surfactant. the solubility of the alkylphenoxypolyethanols shows a marked sensitivity to temperatures. Distilled water redissolved sulfonate precipitated out of. The solubility of nonionic surfactants is not as sensitive to salt concentration as that of the anionic surfactants. Fig.2 molar salt solution could be redissolved when distilled water was injected and when the temperature was increased. Cloud points as a function of concentration for Igepal CO-850 are shown in Table 2 .2 x 1.2 (hours-') 0.34 x l o' A (pmol/m2) 0. the ethoxy groups in these compounds lose associated water and the solubility decreases abruptly to form precipitates. TABLE 2 SUMMARY OF PHYSICAL AND SORPTION PROPERTIES FOR IGEPAL CO-850 Molecular Weight - 1. surfactant In 0 2 M NaC1.h) k2 AHo (d. data in Figure 4 show that surfactant precipitated out of a 0. On the other hand. Then the pack was flushed with salt solution only.524 0.5 x 2. the electrolyte composition and the concentration of the surfactant.21 0. with distilled water and. at 10C 8'.

-24Dynamic and static adsorption data were obtained for sodium o 20dodecylbenzene sulfonate x (NaDDBS) and Igepal CO-850. From chromatographic transport theory the quantity of surfactant adsorbed at this concentration could be calculated. The water transporting the surfactant is likely to be moving primarily in a heated region immediately below the steam zone. Consequently. packed in a core holder. The pore volumes of solution required to move the surfactant through the core were measured. ' The surfactant concentration in the injected solution was increased stepwise and the volumes required to move each concentration step through the core was measured. fired Berea sandstone. These data are shown at 25OC and 9SoC for concentrations up to 70 pmols/L on Figure 5. however. adsorption isotherms will be required for the prevailing temperature at which the surfactant is being transported. the NaDDBS has a low solubility in 0. Surfactant transport could be combined with heat transport through the reservoir. Surfactant solution was injected. various were solid much more Ziegler et al. The surface area of the sand had been measured by a variation of the BET method. If this is the case.2 molar NaCl and also tended to precipitate at the higher temperatures when in contact with the Berea sandstone. The core was saturated with brine or distilled water and placed in an oven to maintain the temperature at the desired value. starting at low concentrations. An abundance of data exists in the literature for adsorption of surfactants onto various substrates at room temperature. The results show - I I I I I I I I I 9 $ NoCl = 02 Y - ' 0 . we have made an initial effort to determine adsorption Isotherms as a function of temperature for an anionic and a nonionic surfactant. Consequently. 16: As discussed earlier. The porous medium was a disaggregated. one would prefer to have the surfactant front remain in the heated portion of the reservoir. Measuring adsorption isotherms at steam temperatures is a difficult problem. obtained data using a dynamic. Data will be presented later which shows that interfacial tensions are reduced at higher temperatures. In that case the surfactant will be moving in a hot portion of the reservoir under isothermal conditions. the effect of temperature on adsorption becomes a matter of considerable importance. In steamflooding. chromatographic transport procedure. These data normally obtained by equilibrating the surfactant solutions with the surfaces. Whichever mechanism prevails in the reservoir. From these data the adsorption isotherm isotherms were also measured E 28' by the conventional static method at 25OC and 95OC. The surfactant concentration shock could either lead or trail the temperature shock.154 EFFECT OF TEMPERATURE ON SURFACTANT ADSORPTION If low concentration surfactants are to be used in combination with steamflooding or hot waterflooding in a reservoir. it is well-established that the steam overrides the oil. only adsorption isotherms obtained by the static method are reported for NaDDBS.

0 (v ' - - - < -+kl The dynamic adsorption isotherms for Igepal are . Static results are given in Figure 6 . oa +25% centration of 67 pmols/L 2 -9.. the O0 0 2 0 40 60 80 m pore volumes of injected SURFACTAM CONCENTRATION. 70°C and 95°C. Surfactant was LEGEND injected at an initial con. Consistent with a 2 1 1 1 1 1 1 1 ~ Langmuir-type isotherm. The experiments with Igepal CO-850 were complicated by the cloud point. 6 . Data obtained in the absence ot salt show less temperature dependence.Breakthrough curves f o r Igepal CO-850 . This would be consistent with the limited static data showing a decrease in adsorption with temperature.h T=95% NaCI = O O M 02 00 o 4 a 1 2 x) 24 28 32 PORE VOLUMES INJECTED Fig.i ence. Dynamic data were obtained at 45°C. The slug produced after increasing L i temperature. and by the instability of this surfactant at high temperatures.CO = 331uM. 28 Equilibration time for the ' 1 1 1 1 1 1 1 1 95OC curve was limited to E three hours. Degradation NeCl :O O M was a serious problem if significantly longer times were used. : a I8 v = 139 m/h LhMOLE. but the slugs of surfactant being produced after reducing Lhe salt Concentration or after increasing the temperature had been explained earlier as being more the result of dissolving precipitated surlactant than desorption or adsorbed surfactant. given qn Figure 8 At low concentrations adsorption decreases with temperature. v = 139 m/h ture only while moving -c. may have resulted le in part from decreased adsorption at elevated temperatures. Data were not obtained at higher I 1 I I I I I I I temperatures because of the 0 0 0 0 0 limit established by the cloud point's. Figure 7 is an example I 1 ! 6 of results obtained by the dynamic method for Igepal CO-850. 'Thc results o t desorption experimenLs are shown in Figure 4. no dynamic data are reported for NaDDRS.Static adsorption isotherms f o r in concentration decreased Igepal CO-830 with increasing concentration. This effect is also associated 16 : 597uM. v 1 137 m/h through the core. Degradation of Igepal is not a problem in the dynamic procedure because the surfactant is LEGENO at an elevated temperaCo = 6?uM. which is characteristic of this class of surfactants. however.rations in excess of about 200 pmols/L. Because of the precipitation problem.95% and at two incremental coni~ centration higher than the initial. M I 1 O6 surfactant required to produce the incremental step Fig. 7 .155 that adsorption decreases with increasing temperature as one would expect. but adsorption increases with temperature for concent. The results show a slight temperature depend.

The epoxy can be easily drilled out of the tubes to permit using them again. On the other hand. Although the spinning drop would appear to be the preferred method. The Langmuir constants for Igcpal are given in Table 2. An important factor in measuring interfacial tensions by either the pendent drop or the spinning drop method is the density difference between the water and the oil.Dynamic adsorption tants in reservoirs but. If time is an important factor in establishing equilibrium between the surfactant solution and the oil.oil0 loo 300 500 surfactant solutions as functiohs of SURFACTANT COWCLNTRMIOII. In the pendent drop method a drop can be suspended at the most for one-half hour. but it does not have the versatility of that developed at the Technical University of Clausthal. the pendent drop procedure is also less suitable than the spinning drop. The density of water decreases more rapidly with .1 mN/m and that is with low precision. As the ethoxide groups lose their associated water. These data are required for any process using surfacFig. An epoxy was found that was effective in sealing the capillary tubes at the above temperatures and pressures. This becomes particularly critical when measuring interfacial tensions at elevated temperatures.156 with the cloua point. Some reports have indicated that equilibrium for these systems has not been established even in matter of days. Our equipment is easy and inexpensive to build. Their apparatus operates at higher pressures and temperatures and permits the exchange of fluids in the rotating capillary during the experiments. EFFECT OF TEMPERATURE ON INTERFACIAL TENSIONS Few results have been reported giving interfacial tensions of. a drop can be maintained indefinitely in a spinning drop apparatus. H I temperature. '*'*' We have modified the spinning drop equipment of Gash and Parrish for use at temperatures to 2OO0C and preseures to 30 The design of the spinning drop equipment is such that it is a simple matter to construct an air bath around the capillary tubes which contain the spinning drop. We have used two methods for measuring interfacial tensions as functions of temperature and pressure. The spinning drop reportedly gives data below 0. we have obtained data by both procedures. isotherms for Igepal CO-850 for the high temperatures associated with steamflooding. No bearings need to operate at thermostat temperatures.001 mN/m. Normally. particularly. The problem arises with caustic solutions and crude oils. the surfactant becomes less soluble and would be expected to separate out onto the solid phase more readily. These are the-pendent drop and the spinning drop methods. but the surfactants. The minimum interfacial tension that can be measured on the pendent drop equipment is about 0. 8 . equilibrium times for surfactant solutions and refined oils are small. were not suitable for testing the procedure at temperatures in excess of 100°C because of solubility problems. Igepal and NaDDBS. The results of the dynamic method with Igepal indicate that the method is suitable for determining adsorption isotherms at elevated temperatures.

Since the cloud point indicates a decrease in the surfactant solubility. become greater than that of the water.7 1 0 ' 50 ' API CWOE 6rP. ' 10 0 5."' To correct crude oil densities for temperature. For some crudes the density of the oil m y . The interfacial tension minima for the nonionics coincided with the cloud point for the particular surfactant concentration. 9 The interfacial tension between the nonionic surfactant.Effect of salt concentration and temperature on interfacial tension of 0. Group 1. * Kepresentative data obtained by the spinning drop method are shown on Figure 10. Density data are readily available for water and can be generated easily for the surfactant and brine solutions. The interfacial tensions shared little dependence on temperature up to 18OoC.4PIFRUqE loo I50 x)o TEMPERATURE. it is not surprising that the interfacial tension . 9-Effect o f temperature on density of water and crude o i l s Fig. 250 0 3 o m s o l 2 o ~ m o ~ TEMPERATURE.157 temperature than that for oil. "c . Interfacial tensions were measured between mineral oil No. the density difference between water and oil can become quite small at higher temperatures. The lowest interfacial tensions were observed at salt concentrations of 5. .0 fo-doa$4 vs'. 9 and TKS 10-80 in various concentrations of salt. The spinning drop equipment cannot be used under those circumstances.9OAF'I California crude oil showed a marked minimum when plotted versus temperature as shown on Figure 1 . Densities for repined oils and pure hydrocarbons were determined using data from "Petroleum Refinery Engineering. A small error in estimating these densities can have a significant effect on the calculated interfacial tensions. but they are affected substantially by the salt concentration. Igepal DM-730 and a 15. . 1" No salt was present in this example but similar data were obtained with the surfactant in presence of salt.0 g/L. . were used. Consequently. The concentration of the TRS 10-80 was kept constant at 0. Interfacial tensions for TRS 10-80 against a representative crude oil also showed little effect of temperature. The lower curves on this figure are duplicate runs and show reasonable agreement. 10.5 g/L.5 g/L TRS 10-80 versus mineral oil No. 'Ic Fig. Density data for water and three crude oils taken from El-Gassier et al. are shown in Figure 9. w&u NQ9 in^ z ID r/L MOCI *Lo1) fi WaCl -X- 0. This problem becomes particularly acute when the oil phase is a crude oil. the volume correction factors from ASTM D-206-36. in fact.

C factants against yure hydrocarbons and mineral oils. these mixtures stiow an abrupt decrease in interfacial tension at temperatures in excess of 120°C. The equivalent alkane carbon number. 100. the trend of the effect of these variables. 1 1 .0 are shown on Figure 12. E salt concentration and temperature on 10 . for the surfactant mixtures was 0.Effect of surfactant indicate an optimum surfactant and concent rat ion and temperature salt concentration at 177°C to obtain on interfacial tension between a minimum interfacial tension. At 93°C the minimum intertacial tension was 0. Representative data 0.005 nN/m Jacobson et a1. The experiments are being extended to obtain data for several hydrocarbons and.I 1 . at least qualitatively.9".Ridge crude with an API gravity of 15. 12 Interfacial tensions as functions of NaCl and TKS 10-80 concentrations at 177OC - . 0.I5 As shown in c. FACN. interfacial tension against a crude E oil was made with the pendent drop b equipment. ?h 'e 1 0 50 100 I50 200 TEMPERATURE.1 mN/m as compared t o 0.5 ID 20 30 NoCl CONCENTRATION (WT %) CONCLUSIONS The results of our experimental work and data reported by others suggests conclusions about Fig. evaluate the relation between the change in interfacial tensions with temperature and the FACN 1 I 1 1 1 1 concept > 0. IHS 10-80 and criltle otl.2 not capable OP measuring the ultra low tensions it can show.02 0. O C results are interesting in that they Fig. Vigure 13. The temperature was 177OC. Additional data have been obtained using a mixture of surINTERFACIAL TENSION CONTOURS I77. 'L'lie decrease in transparency 01 Llie aqiieous phase at the cloud point was a limiting factor in measuring interfacial tensions of nonionics as a I'unction 0 1 temperature by either tlic spinning drop or the pendent drop met hod.5 calculated as recommended by 0 v = 0.158 decreases at tliis temperature.005 mN/m at 177OC. thereby.0 g / L temperatures but the minlmum interfacial tension increased with decreasing temperature. Similar minima were observed for lower NaCl = 0.0 500 Y A more detailed study of the efrects of surfactant concentration. Although this equtpment is 0. The surfactant in this case was TRS 10-80 and the oil was CaliPornia Wheeler.

Some ol these conclusions are quite specific and dependable for the systems to which they apply. A substantial amount of data is being accumulated relating interfacial tension and temperature. but the effect is not as substantial as one might have expected. Stabilities can be estimated from these energies at 00. The surfactants investigated 02 SURFACTANT were observed to decompose by Y TRS 10-80 first order kinetics. 4 Dynamic and static methods were used for evaluating the temperature effect . Igepal DM-730. The data suggest that adsorption decreases for both sodium dodecylbenzene sulfonate and for Igepal CO-850. The nonionic. 13. Other results show a decrease in interfacial tension with temperature for mixtures of sulfonates against pure hydrocarbon or mineral oil. This limits the concensa tration at which these surfactants can hr u c d at higher temperatures. The n. Evidence suggests that the sulfonates may be precipitated at steam temperatures as a result of an interaction with solubilized rock minerals which show limited solubility at elevated temperatures. Activation energies were rn determined for several surfactants. That temperature appears to be related to the cloud point. Certainly more work is required to extend the til nuinber of sureactants w i c a liavc been evaluated at high tempcratures. I . stability of a surfactant at a f oo( NeCI CONC g/L given temperature is its Iialf0 2 0 life. a quantitative measure of the E PETROSTEP 450 . Therefore.159 t h e behavior or surfacLants a t elevated temperature. 3. & I)I PETROSTEP 465 . 5. Many are tentative. On the other hand.o 1 I 1 1 1 I as MIXTURE OF ANIONIC SURFACTANT VS N-OOD€W 4a2 5 a 05 1.Interfacial tensions as Ciinc2 .dodccanc stabilitv of the best sulfonate would be only marginally acceptable at temperatures to 180°C but other surfactants need. The solubility of the nonionics decreases abruptly at the characteristic cloud point. FIR. higher or lower temperatures than those used in the experiments. . on adsorption. Additional data are required with other surfactants in consolidated sandstones. The anionic petroleum sulfotions o f temperature and salt concennates were observed to be more tration for lsurfactant mixtures against stable than the nonionics. pendent drop data do suggest a significant decrease in interfacial tension with temperature for optimum salt and surfactant concentrations. For specific types of petroleum sulfonates some data indicate little effect of temperature on interfacial tensions. All of the surfactants tested would be adequately stable at normal reservoir temperatures E i 1 - .to be evaluated. showed a sharp minimum in the int'erfacial tension at a specific temperature.

I. E. 13. 12. SOC. .160 REFEKENCES 1 . "Measurement of Interfacial Tensions and Contact Angles at High Temperature and Pressure". and ERSHAGHI. M.. . HANDY. Calgary. McCAFFERY. 14. . Analytical Chemistry (May 1956) 2 (5) 870. MORGEN. S A. 10. "Characterization of .. W. . Petroleum Sulfonates". CELIK. Report of Group d'Etude IFP-IMF Sur lee Milieux Poreux Toulouse (1977). Additives in the In Situ Recovery of Bitumen from Oil Sands".. 6. EL-GASSIER. SOC. M. GOPALAKRISHNAN. 5 ." J. GASH. Jour. D. Personal Communication. . "Interfacial Tension . 4 . 9 . P. D R.. C. L. C . "Gravel Pack and Formation Sandstone Dissolution During Steam Injection. L . Tenside Detergents (1978) 15 (5) 247-251. F. . "An Enhanced Oil . 2. ZIECLER. Interfacial Tensions Involving Mixtures of Surfactants". J. M. paper SPE 7867 presented at SPE Fourth Intl. and HANDY.. SANDVIK. presented at the 32nd Annual Technical Meeting of the Petroleum Society of ClM. SvaDosium on Oilfield and Geothermal Chemistry. "Use of Ionic Dyes for the Analysis of Ionic Surfactants and Other Ionic Organic Compounds". Pet. MANEV. P. R S . May 28-30. 1980.. Stanford University. A. JACOBSON. Recovery Method -. "Effect of Temperature on Surfactant Adsorption In Porous Media". and MAX. "Low . 11. J. 81-32-13. Engr. 0 . (1976) 122-128. L L. Pet. "Applications for the Spinning Drop Technique for Determining Low Interfacial Tension". W. and PARRISH. Tech. V. (June 1977) 184-192. . ZIECLER. M. E . PROWSE. . and DENEKAS. M. . Paper No. 1979.. . Jour. "Petroleum Refinery Engineering". 8 . . G. M. (June 1980) 941-949. G . H. R.. E I.. BURKOUSKY. of Canadian Petroleum Technology (July 1972). . (April 1981) 21 (2) 218-226. P . L. (1958) 157-161.. 1981. and CAMBARNOUS. REED.. May 3-6. . and SOMASUNDARAN. W. paper SPE 9003 presented SPE Fifth Intl. J. B. V. Sept. Pet. Properties of Surfactant-Oil Systems Measured by a Modified Spinning Drop Method at High Temperatures". Houston. GOYAL.. P. paper SPE 8263 presented at the SPE 54th Annual Technical Conference and Exhibition. Tensiometer". 22-24.. L. "Thermal Stability of Surfactants for Reservoir Application". I. 7 . J. W. . 3. and RANKINE. 1979. SOC.. . SCHECHTEX. M .. GALE. ZEKRI. Jan. "The Role of Surfactant . and ERSHAGHI. and WADE.Injection of Steam with Surfactant Solutions".. BOREIS. Las Vegas. O. Engr. AMAEFULE. NELSON.. E. Pet. Pet. Jour. MUKERJEE. Symposium on Oilfield and Geothermal Chemistry. "A Simple Spinning-Drop Interfacial . L. Technology (January 1977) 30-31. J K . M. Engr. HANDY.. 23-26. 15. A . "The Role of Surfactant Precipitation and Redissolution in the Adsorption of Sulfonate on Minerals". J. ISAACS.

London SW7 2AZ ABSTRACT The e f f e c t s of layering within porous material with regard t o basic flow mechanisms and chemical dispersion have been investigated.CHEMICAL FLOODING 161 SURFACTANT SLUG DISPLACEMENT EFFICIENCY IN RESERVOIRS. TRACER STUDIES IN 2-D LAYERED MODELS ROBERT J. WRIGHT. and the behaviour of non-unit mobility r a t i o displacements i s predictable using r e l a t i v e l y simple conceptual/mathema t i c a l models. f l u i d v i s c o s i t y and flow r a t e . DAWE and COLIN G. gravity and c a p i l l a r y pressure influences were eliminated by using model f l u i d s of matched density and complete miscibility. . Experiments The variables controlled were layer permeability and dimensions. Imperial College. The dispersion of chemical t r a c e r s between layers has a l s o been modelled mathematically and t h e r e s u l t s have been applied t o laboratory t e s t s on heterogeneous cores. The importance of channeling and crossflow e f f e c t s a r e emph- asized by t h e r e s u l t s . RICHARD A. WALL Petroleum Engineering Section. have been performed within unconsolidated g l a s s bead packs.

a high permeability channel ( a ) and a less permeable channel ( b ) . W e have attempted t o view miscible and immiscible displacement mechanisms on a common b a s i s s i n c e the two concepts merge i n ultra-low-tension systems. O f p a r t i c u l a r consequence are the non-random v a r i a t i o n s i n permeability to be found within porous rocks. however the fundamental problems a r e ocnmnon to a l l E. Experimentally. phase equilibrium and emulsification characte r i s t i c s have been excluded i n t h e present work. Figure l ( i )i l l u s t r a t e s two p a r a l l e l channels composed of homogeneous and continuous porous media. FO PATTERNS I N LAYERED MEDIA LW I t has been found useful t o consider simple two-channel conceptual m o d e l s i n order t o account f o r crossflow behauiour i n multilayered and s t r i a t e d media. Layering s t r u c t u r e s a r e a common feature of sandstones and their e f f e c t s have been reviewed i n recent l i t e r a t u r e w i t h reference to f l u i d flow (1) and dispersion mechanism ( 2 ) .O. I t is well known that t h e n a t u r a l heterogeneity of petroleum reservoir material is one of the major problems i n chemical E. W a r e taking the e approach t h a t the gross f l u i d flow and dispersion e f f e c t s w i t h i n heterogeneous media shoald be b e t t e r understood before laboratory core-flood r e s u l t s and data from l i n e a r homogeneous packs can be applied t o the reservoir system. The displacement of f l u i d (1) by f l u i d ( 2 ) within this model ( i n t h e x direction) has a resulted i n two displacement boundaries ( a t X and q). based on simple two layer-channel interactions. This approach follows contributions within the l i t e r a t u r e on dispersion ( 2 ) & (3) and crossflow ( 4 ) & ( 5 ) i n such model systems. The experimental work discussed here involved idealized layered models of packed Ballotini. W have investigated layered models. CrosSflow d i r e c t i o n s and approximate magnitudes can be demons t r a t e d mathematically by considering t h e v a r i a t i o n of flow p o t e n t i a l along the axes of the channels. The flow mechanics of displacements a t various (favourable and unfavourable) mobility r a t i o s were recorded by photographing dye t r a c e r boundaries under conditions of flow r a t e f o r which diffusion/dispersion e f f e c t s were small. The efficiency of s u r f a c t a n t slugs is probably the most l i k e l y application of these considerations. flow mechanisms and dispersion e f f e c t s have been monitored using dye tracers. Displacements have been of an i d e a l miscible type and therefore represent p e r f e c t microscopic displacement efficiency. and w e describe here numerical predictions w i t h one example application. processes. To quantify dispersion phenomena we have considered equiviscous miscible displacements.162 INTRODUCTION . . displacing f l u i d t h e more viscous i n F i g .R. l ( i i i ) . The properties peculiar t o surfactants such as adsorption. l ( i i ) and the l e s s viscous i n F i g .O. both conceptual/ e mathematical and physical ( v i s u a l ) . Conceptual models were developed.R. The instantaneous pressure p r o f i l e s a r e p l o t t e d f o r two d i f f e r e n t viscosity r a t i o s . processes.

This assumes no c a p i l l a r y p r e s s u r e . p2 P1 >1 Location.163 Y t (i) Displacement i n dual channel m o d e l 1 ii P (ii) P r e s s u r e P r o f i l e s f o r !J2 > p1 0 1 P - R (iii) P r e s s u r e P r o f i l e s f o r p2 < p 1 0 X+ x4 L Figure 1. no crossflow between t h e channels (as i f s e p a r a t e d by a n i m p e r m e a b l e b a r r i e r ) . a u s e f u l method for r e p r e s e n t i n g local croltsflow tendencies as i n d i c a t e d by pressure drops ( a t f i x e d x) between the channel axes. C r o s s f l o w would therefore be s t r o n g e s t around t h e displacement f r o n t s a n d o c c u r s i n t h e d i r e c t i o n s i n d i c a t e d i n Table 1. I t is. g r a v i t y o r c c m p r e s s i b i l i t y e f f e c t s . Fig. Table 1. Fluids Crossflowing Prom Channel:a b b Into : Channe 1 - 1(ii) "b X (2) b a a b (1) (1) E ( 2 ) l(iii) <1 "b X (1) E (2) a . however. d i s p e r s i o n . also f o r t h e moment.

t h e length of t h e flow model was 20 a .8. and p o r o s i t y was approximately 38%. Displacements w e r e performed i n a v i s u a l model composed of g l a s s beads packed t o form a c e n t r a l (high permeability) l a y e r surrounded by l e s s permeable packing. the permeability r a t i o s between l a y e r s was 2. l ( i i )and ( i i i ) . Smaller s c a l e differences a r e apparent around t h e main displacement boundary. penetration i n t o t h e high permeability channel is retarded. Channel widths were around 1 cm within a t o t a l model n width of 10 cm. Constant flow r a t e s were used and matched density aqueous s o l u t i o n s were employed.0 ( P l a t e 21. For "favourable" v i s c o s i t y r a t i o s these conditions a r e reversed. when displacing f l u i d is less viscous (more mobile) the leading "finger" within the high permeability channel advances r a p i d l y due to crossflow i n t o t h e f i n g e r a t its base and o u t 06 t h e channel around i t s leading t i p . . 1. t h e advancing Cusp being squeezed i n a t t h e f r o n t (with some t r a c e r dispersed ahead as a t h i n plume) and widened a t i t s base where it j o i n s t h e main displacement f r o n t . The photographs ( P l a t e s 1 . and flow is i n a l l cases from l e f t t o r i g h t . Hence. b u t t h e dominant influence is crossflow a s d e t a i l e d i n Table 1.164 Experimental. and 3 . Roughly 0 . P l a t e 1: unfavourable mobilities P l a t e 2: equal m o b i l i t i e s P l a t e 3: favourable mobilities The d i f f e r e n t displacement boundary p a t t e r n s and t h e consequent differences i n displacement e f f i c i e n c y can be explained to sane e x t e n t by t h e a x i a l pressure gradients of Figs. l o c a l f i n g e r i n g i n t h e forme<>case and a sharp s t a b i l i z e d boundary i n t h e latter favourable mobil+&y case.3 ) i l l u s t r a t e displacement p a t t e r n s with fluorescent t r a c e r s under t h r e e d i s t i n c t conditions of v i s c o s i t y r a t i o . Crossflow is seen t o a l t e r t h e shape of t h e finger by swelling i t s f r o n t and squeezing i t s base. p2/p1 = 0. O ( P l a t e 3 ) ..22 ( P l a t e 1 1 . 2 pare volumes of displacing f l u i d has been i n j ected. Glycerol/water mixtures and s o d i m chloride o r sodium sulphate s o l u t i o n s were used.

0 0. 0 2 ) and estimation of pressure gradients.0.1 1. For equal width s t r a i g h t layers of isotropic media.02 These span the extremes of practically no crossfhw (curve 1) t o near m a x i m u m crossflow (almost zero resistance to flaw between channel centres.06 0.1 (Fig.- kt ka . hence displacement velocities (as a function Of frontal position). kt is the effective cross permeability between the channel axes.165 Quantitative Results: Unfavourable Mobility Ratio Continuous Displacement Crossflow can be quantified i n the two channel conceptual model described above by means of a numerical crossflow index:= A.- (Ax)2 d ' where A is the crossflow boundary surface area per u n i t volume of channel ( a ) . .02) layer z ( k b 2. 2 i l l u s t r a t e s calculated instdtaneous channel (a) pressure profiles for xa = 0.5). For various values of "a" displacemant tracks were calculated for viscosity ratios of 0 . Figure 2.d is equal to the layer width and A = - d * -kt ka 2 1 + ka/kb Fig. A is the numerical inter-node fractional distance and d is the separation of the channel axes. 4) and 0. x ka Is the permeability of channel ( a ) .5. curve 4 ) .3). 0 1 Numerical cdkculations of distance/time tracks have been performed based on incremental advances of a displacement front (Ax = 0 . 5 (Fig. o.Ooo1 3 4 0.01 0. Possible pressure distributions along channel ( a ) . xb 0 when p2/ p l = 0. for the mossflow indexes and layer aspect ratios given below:Curve 1 2 d & a (Ax = 0.1.2 4.01 (Fig.0 0.

1 E l I Figure 5. f o r mobility n a t i o s near u n i t y crossflow is unimportant and displacement v e l o c i t i e s a r e constant with t i m e .5 Figure 4. 3 - 5 : Time/Distance t r a c k s f o r displacement f r o n t within high permeability ChdMel. . lJ2/lJ1 = 0. The l i m i t i n g l i n e a r t r a c k s f o r high a are approached q u i t e c l o s e l y f o r moderate values of The behaviour is only s e n s i t i v e v i s c o s i t y ratios and l a y e r aspect-ratio.166 L 02 ' Figure 3.01 Figs. Clearly.1 and p2/p1 < 0. a s a general guide. P 2 h 1 = 0.1. numerical c a l c u l a t i o n s .1 * = 0. lJ2/Pl 0. to crossflow index when d/L > 0.

. .22 .5 ) which is n o t expected t o be a good approximation i n a l l cases. t h e s e p r e d i c t i o n s are based on equating xb/L to t h e dimensionldss t i m e (of Figs.approxi m a t e l y s t r a i g h t l i n e t r a c k s were obtained. When p l o t t e d v e r s u s time. The r e s u l t s i n terms o f t h e l e a d i n g f r o n t displacement (xa) and are t h e main f r o n t (a) p l o t t e d on Fig. Flow v i s u a l i z a t i o n experiments were conducted with matched d e n s i t y f l u i d p a i r s having " adverse" v i s c o s i t y ratios.33 0.3 . 6.8 kakD d/L 1' ' / 2 } 0. 45 a a "Mean f r o n t " p o s i t i o n s were e s t i m a t e d from c o l o u r photcgraphs t a k i n g i n t o account d i s p e r s i o n and local f i n g e r i n g .I R e l a t i v e Front Positions. The c o r r e l a t i o n o f experiment and c d l c u l a t i o n s is encouraging. a d e t a i l c o n t r a d i c t e d by t h e n m e r i c a l r e s u l t s . However.: 2. along w i t h t h e numerically p r e d i c t e d curves using t h e parameters given i n Table 2. S i m i l a r f i n d i n g s are d e s c r i b e d by Peaceman and Rachford (6) f o r vtscous f i n g e r i n g i n randomly v a r i a b l e porous media.167 Experimental Results. The packed bead models were as described above.IS L xb Figure 6. I t is n o t i c e a b l e t h a t t h e r e is a s i g n i f i c a n t dependence on v i s c o s i t y ratio: An i n t e r e s t i n g f e a t u r e o f m o s t experiments is the r e l a t i v e l y f a s t i n i t i a l p e n e t r a t i o n i n t o t h e high p e r m e a b i l i t y l a y e r . data scatter being n o t too serious. Experiments were d i s t i n g u i s h e d by t h e parameters given below:- Experiment 0 * * 0 1 3 2 4 1 1.

33 and 12.9 f o r a These r e l a t i v e r a t e s a r e about a f a c t o r of two viscosity r a t i o of 0. Relative permeabilities. 8 . These a r e b e t t e r described a s synthetic functions since they a r e derived by adding together t h e e f f e c t s of the individual layer properties. I t is useful t o consider a t t h i s point the effectiveness of an a n a l y t i c a l solution method based on 1-dimensional flow theory and "pseudo" r e l a t i v e permeability functions ( 7 ) . The r e l a t i v e permeabiliw t o displacing ( 2 ) and displaced (1) phases a r e p l o t t e d versus s a t u r a t i o n o phase ( 2 ) on f Fig. A useful feature of t h e present displacements is t h a t they should give r e s u l t s which a r e similar t o p e r f e c t ultra-lowtension displacements (having negligible c a p i l l a r y pressures and 100% microscopic displacement e f f i c i e n c y ) .5 t 5. Crossflow is the p r i n c i p l e mechanism by which a displacement f r o n t may be s t a b i l i z e d against the influence of l o c a l permeability variations. 7. g r e a t e r than those indicated i n Fig.0) flowing regimes the f a s t (S2 = 0 0. 8 .0 0. .5 f o r a viscosity r a t i o of 0. however they can be applied to miscible processes when the e f f e c t of dispersion is negligible. 0 X/L 0.0 a r e predicted t o be 8. I t may therefore be possible. m t i u t i v e & s u i t s : Favourable Mobility Ratio Continuous Displacement. I t is thought therefore t h a t 1 dimensional flow theory exaggerates the e f f e c t of mobility r a t i o f o r reasons concerning crossflow mechanism. These extended d i s t r i b u t i o n s a r e not found i n p r a c t i c e even when l o c a l fingering is taken i n t o account: however i t is only t h e averaged displacements w i t h i n 1. - 1.to obtain predictions f o r miscible and low tension displacements w i t h i n layered media which a r e s i g n i f i c a n t l y b e t t e r than those provided by a n a l y t i c a l 1-dimensional methods. The r a t i o of displacement r a t e s - - 1.168 Discussion of Analytical Methods. Predicted saturation/distance p r o f i l e s based on the above functions using the v i s c o s i t y r a t i o s of i n t e r e s t a r e given on Fig. 7 f o r the model parameters of experiments 3 and 4. Use of these functions is i d e a l l y r e s t r i c t e d to immiscible (no diffusion) processes.14 which w i l l be considered (dotted l i n e s ) . 1 Fig. The dual-channel pressure p r o f i l e s discussed above can be used to explain t h i s flow mechanism and t h e "shock front" concept of 1-dimensional displacement theory ( 5 ) . 6 .22.5 Fig. Theoretical saturation distributions. using convenient approrimations.14) and s l o w (S2 = 0. 0 0 s.

Fig.Ap = 0. The geometry of t h i s region can be approximated by a t r i a n g l e enabling an expression t o be derived f o r a s t a b i l i z e d q). Ap = 0. while on Fig. 9.91 xlO-’an/sec) and the e f f e c t of t h i s is t o compensate t o some e x t e n t f o r the e f f e c t of a higher v i s c o s i t y r a t i o . 10 Shock f r o n t formation is c l e a r l y not observed. which tends t o discourage channeling i n t o the high permeability zone) depends upon the crossflow which i t s e l f i s governed by the region between t h e two p r o f i l e s . Phenanenon. I t has been found t h a t the basic c h a r a c t e r i s t i c s of such s t a b i l i z e d displacement p a t t e r n s can be approximated by considering dual-channel 1 pressure p r o f i l e s . 10 are the observations f o r p2/p1 = 10. Pressure p r o f i l e s f o r favourable m o b i l i t i e s with crossflow. assuming the v e l o k i t i e s of the two f m n t e are separation “6”( = x a - . Figure 1 i l l u s t r a t e s the form of such p r o f i l e s when viscous crossflow (but not gravity) is allowed f o r . I n t h e case of t h e higher v i s c o s i t y r a t i o displacement t h e r e is l i t t l e change i n t h e f r o n t a l shape with time. The s u p e r f i c i a l flow r a t e was greater i n t h e l a t t e r case (1. The r e s u l t s r e f l e c t a considerable influence of g r a v i t y since t h e displacing f l u i d was more dense and was flowed v e r t i c a l l y upward.149 9/cm3.113 g/cm3. 9 i l l u s t r a t e s traced displacement f r o n t s ( f u l l l i n e s ) i n r e l a t i o n to t h e layer bounda r i e s (dashed) f o r three s t a g e s ( f r a c t i o n a l pore volrrmes i n j e c t e d indicated).169 I n sane preliminary work we used a packed bead model containing four f a s t flow channels (permeability r a t i o 13:l) of d i f f e r e n t width. Predictions based on s y n t h e t i c r e l a t i v e permeabilities f o r t h i s model lead t o the s i n g l e shock f r o n t s shown ( d o t t e d ) . The o s c i l l a t i o n s of f r o n t a l boundary appear t o increase i n amplitude with increase i n channel diameter ( t h e f a r r i g h t channel is r e a l l y a half-channel since there is a no-flow boundary a t i t s s i d e ) . a s compared with 0. Fig.11 = 5.12/l. Pig. Here l.8 x 10-3cm/sec. The S t a b i l i z a t i o n 1 Figure 1 .

. . . 202 5. outlines the outstanding problems and the types of research and development which are under way. . . . 197 (5). as a subject of research in semiconductor thermionics. . Effects of Products from Nearby Cathodes. . . . . . . . . . . and presents those rather fragmentary theoretical developments which research workers have succeeded in achieving a t the present writing. . I. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 206 210 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Decay of Enhanced Emission. . . . . . . . . . . . . . . . . . . . . . 199 a. . . . OXIDEAND PRACTICAL ELECTRONICS 11. . . . . . . . . . . . . . . . . . Electrical Conductivity of Thorium Oxide. . . . . . . . . . . . . . . . . . . . . . the thorium oxide system is more amenable to quantitative understanding than is the barium oxide emitter.. . . . . . .. . . . . . . . . . . . Comparing the thorium oxide situation with that of barium oxide one finds that less work with theoretical intent has been done in the former case. . . . Even more than with barium oxide. . . . . . . . . . . . . . . . . Powdered or Sintered Specimens. . . . THORIUM 1. . . Thermal Activation. . . INTRODUCTION The appearance of thorium oxide on the scene of high-powered tube engineering has been gradual over the past three decades. . . . . . . . . Actually it may appear that. . . . . . . . . . . .’ As a refractory material. . . . . . . . . . . . . Three “Stateu of Activation”. . . .. . . . . . . . . . . Electrolysis. . . 201 4. . . . . . . Activation by Reverse Current. . . . and decisive experiments are lacking. . . . . . . vapora at ion . . . . . . . . . . . . . . Optical Phenomena in Crystalline Thorium Oxide. . . . Crystalline Specimens. . . . . . . . . . b. Preliminary The introduction of thorium oxide into the field of practical electronics came about because of its metallurgical as well as its thermionic properties. . 199 b. . and was naturally accelerated by World War 11. . . 200 d. .. . . . . . . . . quantities of the order of one percent are added t o tungsten t o control recrystallization . . . . . . . . . . . . . . . . . . . . . . . . . 204 a. . . . . . . . . . 193 (3). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . DANFORTM I’aye 193 (2). . . . . . . . .. it may be said t o be used in applications where barium-strontium oxide falls short in some aspect of ruggedness and where the extra heating power required by the thoria cathode is not impracticable. . . . . . . . . . .. . . . . . . . . E . . . . . . . This is due to the fact that the latter has occupied a far more important commercial position for over a quarter of a century. . . .170 IT. . . . . . . . Mechanisms of Disappearance. . the theory is still only semiquantitative. . . . . . . . . . . . . . . . . . (1). . . . . . . . .. 198 3. . . . . . . . . . . I n general. . . . . . . . . . . . . . . . . . . . . . . . 196 (4). . . . . . . . . . . . . . . . . . . . . . . . . . . S p u t t e r i n g . . . . . . . . . . . . . . . . . . . . . . . . General . insoluble in tungsten. . . . . . . . 199 c. . . . . . . . . . The present paper describes certain practical applications of thorium oxide emitters. . . . . . .

Fig. 6 = 0. 08. This should be viewed i n r e l a t i o n to the p r e d i c t i o n s o f 1-dimensional flow theory based on t h e s y n t h e t i c r e l a t i v e permeability f u n c t i o n s f o r t h e s e The a n a l y t i c a l models. T h e o r e t i c a l curves applying to o u r model. + 7. 5 b b b 8 + b 3 I . 0 6 .e. These g i v e t h e e q u i l i b r i u m s t a b i l i z e d f r o n t a l s e p a r a t i o n s p r e d i c t e d using t h e above equations. and o t h e r p e r m e a b i l i t y ratios ( i n d i c a t e d on t h e curves) are included. Figure 13. . experimental ( b a r s ) and theoretical ( l i n e s ) . 0 5 . V i s c o s i t y R a t i o E f f e c t . b m b. Figure 12. i.along with crosses i n d i c a t i n g t h e s e p a r a t i o n s based on t h e p o s i t i o n a t which t h e channel is completely occupied by d i s p l a c i n g f l u i d . F r o n t a l S e p a r a t i o n s . 7 shows t h e s e f o r experiments 7 and 8 . s o l u t i o n i n d i c a t e s a s i n g l e shock f r o n t through t h e whole system f o r v i s c o s i t y ratios g r e a t e r than 2. b . Assuming t h e b a r s t o be a c c e p t a b l e a s an experimental estimate of t h i s parameter (remembering t h a t no p i s t o n .171 maxima covering t h e scatter of d a t a are p l o t t e d on Figure 13.82. Our experimental r e s u l t s c l e a r l y demonstrate t h a t t h i s is n o t t h e case and w i l l be of more serious consequence t o displacement e f f i c i e n c y as Layer (or other channel) diameters i n c r e a s e . Experiment NOS. .l i k e f r o n t is observed i n t h e high p e r m e a b i l i t y channel) then the u s e f u l n e s s of t h e mathematical approximation is supported.

exhibiting a t y p i c a l equiviscous displacement. Although the volumes of these s l u g s a r e about 20% of t h e pore volume. The low v i s c o s i t y slug ( P l a t e 6) is continuing to be squeezed from the low permeability medium i n t o t h e f a s t flow channel. p a r t i c u l a r l y near the f r o n t o f t h e slug. P l a t e 4 shows a low v i s c o s i t y s l u g f i n g e r i n g and channelling ahead i n a similar way t o the displacements discussed abbve involving continuous i n j e c t i o n . since mobilized o i l banks w i l l be p a r t l y o r wholly composed of discontinuous o i l whose flow w i l l be highly non-Newtonian. The high v i s c o s i t y slug ( P l a t e 7) has been s p l i t by t h e chase f l u i d which has channelled through and is crossflowing o u t of the high permeability layer. This is a p o s s i b i l i t y which w i l l be i n v e s t i g a t e d i n f u t u r e modelling work.6:4. Capillary pressure e f f e c t s could a l s o be important even though i n t e r f a c i a l tensions may be low. F r s u r f a c t a h t s l u g s the f l u i a r e d i s t r i b u t i o n s discussed above w i l l be o combined oith considerable adsorption. added t o t h e chase fluid.172 SLUG DISPLACEMENTS I t has been found t h a t continuous i n j e c t i o n t e s t s model well t h e development of displacement boundaries a t f r o n t and rear of a "slug" up t o t h e time when overtaking occurs. dispersion. The breakdown of s l u g i n t e g r i t y could possibly be r e s i s t e d by chemicals. l o s s of slug i n t e g r i t y occurs. Plate 4 Plate 5 The permeability r a t i o was a s before (2.8) and the v i s c o s i t y r a t i o s involved i n t h e displacement of f l u i d (1) by f l u i d ( 2 ) by f l u i d (3) were f o r P l a t e 4.6:1. Behind t h e slug w have a favourable e r m b i l i t y r a t i o displacement of t y p i c a l pattern. mass-transfer and g r a v i t y e f f e c t s .s i t u g e l l i n g polymers which a r e s e n s i t i v e to s a l i n i t y environment ( 8 ) . p3:p2:p1 = 3:1:3. Plate 7. however the slug is 'near being divided i n t o t h r e e portions. 4. and f o r P l a t e 5 . The high v i s c o s i t y slug of P l a t e 5 shows a s t a b i l i z e d form a t i t s f r o n t . ( 9 ) .without dye. An example could be the i n . .designed s p e c i f i c a l l y t o r e s i s t c e r t a i n crossflow processes and t h e mixing of out-of-sequence f l u i d s . P l a t e 6. I t is pushed by a s i m i l a r liquid.

2 and 5 . however these integral solutions are very complex because they are intended t o cope w i t h a large time range. Expressions of s i d l a r form are applicable t o other channel geanetries (e.g. which could apply t o comon f i e l d conditions i f d is of the order of lm and to laboratory core tests i f layers of a few mrn width are present within the porous m e d i u m . Figure 15 shows numerical points and analytical curves representing the distribution of average concentration w i t h distance i n the flow direction (normalized for t = 0 .following the approach of Lake and IiraSaki ( 2 ) and Koonce and Blackwell ( 3 ) for chemical dispersion and Satman and Zolotukhin (10) for the analogous problem i n heat transfer. i t is generally found to be of the order of the molecular diffusion coefficient (11). The approximations derived f o r the heat transfer problem (10) involve also a square root of time dependence. the upper one (between Y values 0. Figure 1 4 shows computed isoconcentration contours .independent of concentration. 2 . 2 ) The analytical approximations were derived wing solutions t o the zeroconcentration-boundary-condition case ( 1 2 ) . Our approximation is:- .1 intervals) within a two layer system. position and flow rate. Eere w consider a two-layer mode1. evaluated for short dimensionless times. between 0 . cylindrical) provided times are short. NTD 14 d2 where T is the absolute time from the s t a r t of the displacement. To scale these effects it is useful t o define a transverse dispersion number ( 2 ) : 14G. W examine here the intermediate e range of NTD. N~~ = 3 .(at 0. W consider flow parallel t o the layers and tracer dispersion normal t o e t h i s direction (longitudinal dispersion coefficient is zero). For reservoir r a t e s . d2 V where L and d are the length and width of the system. V is the superficial flow rate i n the high permeability layer. The l a t e r a l dispersion coefficient has been taken to be constant. Tracer injected a t u n i t concentration is dispersed as shown a t three values of the dimensionless t i m e : t . It is of i n t e r e s t to obtain convenient analytical approximations to the mean tracer concentration within a given cross-section of the flow channel (and of the non-flowing matrix). while when N T D > ~ composition is practically constant over any cross-section through the system and the behaviour can be represented by a single effective longitudinal dispersion coefficient ( 2 ) .173 DISPERSION I N LAYERED MEDIA The s t a b i l i t y of chemical slugs w i t h i n channelled porous media can be e strongly affected by diffusion/dispersion processes.0) flowing from l e f t t o right.. Kt is the Mansverse dispersion doefficient.T Kt = _ .5 and 1. the lower is stagnant but receives injected tracer by l a t e r a l dispersion from the permeable layer. Lateral dispersion is insignificant when N T D < 0 .

a FRRCTIONAL OISTRNCE I N FLOY 0.-. ?- I I I I 1 -m 0-1 12 .4 0. m 0. 0.9 0.4 0.T I t l E = O -05 I? S I I .5 0.7 0-1 0.) I OlRECTlON Figure 14.9 0. Isoconcentration contours i n two layer system.2 0 .1 > m t Z O Z a 0. .8 0. s 0.1 0 .) I FRRCTIONAL OISTRNCE I N FLOY OlRECTlON T I t l E = O 10 - I I I? S I * 3 ‘ 25 N I 5 0.

? 0 A 1. A similar method can be used t o approximate t h e averaged Concentrations w i t h i n t h e non-flowing matrix (Em) f o r the t w o equal-capacity l a y e r s h e r e considered:- Em = 2 .6 0. 0 t + (1 - (1 . f i r s t . layers) within a heterogeneous core sample displacement flow tests of d i f f e r e n t rates have t o be compared.v i o l e t a b s o r p t i o n monitoring techniques.(141. Generally a non-zero flow rate w i l l apply to t h e e x t e r n a l matrix i n c o n t r a s t t o t h e s t a g n a n t case as above. such l a b o r a t o r y Systems can model similar problems on t h e reservoir scale. 0 t+ (1 - x% . Laboratory tracer tests on layered r e s e r v o i r materials are o f i n t e r e s t f o r t w o reasons. Handy (17) has used d u a l tracers to e v a l u a t e d i f f u s i o n e f f e c t s and w e have begun tests on l a y e r e d sands t o n e s using u l t r a .? 0. .10 1=0.175 COnPUlEO 0 T=O.15 + f 0 l u 0 .0. X is t h e f r a c t i o n a l d i s t a n c e e q u a l t o x/V. second.i . .1-2l'"I 1-1 I . Tracer e f f l u e n t p r o f i l e s have been analysed i n terms o f v a r i o u s models intended t o account f o r heterogeneity (13) . E = 1 - 2 .1 O. Applications to a Multichannel Problem. This n e c e s s i t a t e s c o n s i d e r a t i o n of the problem as one of r e l a t i v e flow rates using moving co-ordinate methods.x I I 1 11-3 I I I I a. FITTED _I Figure 15.3 0.1 04 0.1 0. One approach t o a multichannel problem is to consider each i n d i v i d u a l channel as i n t e r a c t i n g with a surrounding matrix which possesses t h e s u i t a b l y averaged p r o p e r t i e s o f t h e rest of t h e porous body. very l i t t l e r a w d a t a of this kind is to be found w i t h i n t h e petroleum l i t e r a t u r e . To estimate mass t r a n s f e r rates f o r t h e channels (e. -1 I c.g. Our main source is t h e high q u a l i t y r e c e n t work o f Spence and WatJcins (16). Unfortunately.OS .a I I 0-S FRRCTIONRL OISTRNCL Cross-sectional averaged concentrations i n flow channel. (15). conventional methods f o r c h a r a c t e r i z i n g d i s p e r s i o n coe f f i c i e n t s f o r miscible displacement and r e l a t i v e p e r m e a b i l i t y f u n c t i o n s for l o w t e n s i o n immiscible d i s p l a c e m n t may be u n r e l i a b l e .XI+) where is t h e average i n j e c t e d tracer concentration w i t h i n t h e cross s e c t i o n ( a t X) of t h e flowing channel.

ten t o twenty steps for each effluent curve have been found t o be satisfactory. Asatisfactory analysis can be performed with a proQramable calculator (a program suitable for an "BP 41C" Is available from the authors). Subsequent tracer measurenmnts have t o be processed to allow for the ( t i m e dependent) contitbutions from a l l the faster-flowing channels. . This characterizes the f a s t e s t flow channel(s) of the sample. ignored i n the present analybls w i l l become impartant. we can estimate the fractional cross section of the t channel (6jS) and i t s effective mass transfer coefficient (M ) thus:j = Vj L/T. and the assumption t h a t flow within the matrix surrounding any given channel-"i" is approximated by the mean velocity oif the whole displacement (v). while for T > L f i s u p s are taken backward I n time redbfining F as the concentration of displaced fluid. a r e known from t w o experiments (1 and 2) performed : difgerent rates. These expressions can be applied to effluent eanposition values measured shortly a f t e r the f i r s t detected breakthrough of displacing phase from the multichannel system.176 Using the above approximations. (f (1) and f j ( 2 ) ) .) and the individual ("breakthrough") channel "j" contribution can be obtained using the following h1gorithm:- The effluent profile is analyzed forward i n time as presented above for T C Lfi . L being the length of the test core. Small t i m e steps should be avoided since errors due mainly to intralayer longitudinal dispersion. W T j (1) M t = 4Kt/d2 = v j ( 2 P j (2) for a l a y e r .. Thus i f the fractions of displacing fluid a t its moment of breakthrough a t the effluent end of the single channel j". a method of effluent curve analy s i s has been derived (18). The gross composktion measured i n one experiment is F(a functibn o'f.

Velocity d i s t r i b u t i o n s . coNcLus10Ns Surfactant E. Predi c t i o n s of mobility r a t i o e f f e c t s could therefore be made using conventional 1-dimensional displacement theory. Figure 1 7 gives the "no dispersion" velocity p r o f i l e s of the porous media. 0 cm and 0 .177 Example r e s u l t s based on some of the t r a c e r composition p r o f i l e s of Spence and Watkins a r e indicated on Figures 16 18. f o r Righly heterogeneous media allowance f o r crossflow e f f e c t s .R.O.4 could be n i n t e r p r e t e d i n terms of layers of & o u t 2 . Miscible type r e l a t i v e penneabilities . Diffusion/disper$ion e f f e c t s can be large. Figure 17. l o s s of i n t e g r i t y due to flow mechanism has been observed in slugs of around 20% pore volume. as discussed above. Experimentally. should be included. 5 a width respectively. . For s h o r t dimensionless times it is possible to model khese phenamena a n a l y t i c a l l y to match numerical simulations and to analyze tracer t e s t data. Mass t r a n s f e r coeffkcient d i s t r i b u t i o n s ~ ~ lines:"Sandstone SS2" 1 1 Dashed l i n e s : "Carbonate B 17" Figure 18. However. - - Figure 16. depending on the width of layers. Mass t r a n s f e r coefficient d i s t r i b u t i o n over the cross section of t h e samples is given on Figure 16 f o r a sandstone and a carbonate. The l a t t e r can be represented as r e l a t i v e permeability functions (Fig. 18) applying to the i d e a l "no dispersion" case o r t o the i d e a l near-zero-interfacial tension immiscible displacement case. Mathematical approximations have been found which a r e capable of modelling the channelling and crossflow e f f e c t s present i n non-unit mobility ratio displacements. M values of lo-' and 1 0 . slugs w i l l be susceptible t o layer and streak permeabi l i t y heterogeneities found within reservoirs due t o disturbance of flow p a t t e r n s and increased dispersion.

2.. & DAWE. Paper 9229. & JOHNSON. S.. (N0.R. ( J u l y 19711.. R. a l . R.E. Numerical C a l c u l a t i o n of Multidimensional (Dec 1962) 2.Pet. .E. 1965) 2 . & RACZiFORD.1024. Q u a n t i t a t i v e Measurements Of Flow Heterogene i t y I n Laboratory Core Samples And Its E f f e c t On F l u i d Flow Characteris t i c s . Flow Heterogeneity I n Reservoir Rocks. WEBBER..P.. L. S T A . S.178 AK O LD E E T C N WE GMNS Dr. S0c. T. n WRIGHT. S. Paper 10115. S0c. 4. (Dec. 1975) I .6) 1011 .E. 67 . et. M. A..J.Pet. A Miscible Displacement Model. KCONCE.71. PERKINS.1979) 77. (Dec. WRIGHT. R.4.Eng.(Oct. R.. & HIRASAKI. S. 11. H.E. €EARN. Application of the Time-Dependent O v e r a l l AM N Heat T r a n s f e r C o e f f i c i e n t Concept t o Heat T r a n s f e r Problems I n Porous Media. E. & BLAQCWELL. & D W . 16. 318 .. ROSMAN. . A E Porous Media. . & SWEENEY. WRICBT. AIME (1959) 216.E. W e are g r a t e f u l t o t h e Department o f Energy f o r f i n a n c i a l support.Eng.. Heterogeneous Porous Media. 1975.Eng. A Model Of O i l Ganglion Movement I n EGBOGAII..Inst. CRANK. 467 s 476. R. A Review o f Diffusion and Dispersion i n POrOUS Media.. D. 70 . J.J.3. S.E. Paper 8436. A. M.t o be submitted f o r p u b l i c a t i o n . 11..84. HAWTWORNE. 9. g.Pet. R. 5.145-154. S. 6 1 - 18. The Mathematics o f Diffusion.Fr. & Z O L O T W H I N . The E f f e c t o f Miscfoscopic Core Heterogeneity On Miscible Flood Residual O i l S a t u r a t i o n ..P. Process Technology Improves Oil Recovery. 327 340. 14. Taylor's Dispersion I n S t r a t i f i e d Porous Media.du P e t r o l e (Nov-Dec 1980) 35. 7. S0c.J. - MACK..Sec. 65.P. The E f f e c t of C a p i l l a r y P r e s s u r e I n a Multilayer Model of Porous Media. . - 8.P. R. PEACEMAN. Oxford Univ. A. 805 813. 3. O i l & G a s J. 10. Press. Paper 5631.E. ( 6 ) . & WATKINS.J. KOVAL. Miscible Displacement.Eng. E.. S .J. Rev..Pet. T. S0c.P. 40. 17. Paper 9247 LAKE. K. Allmen is thanked f o r performing t h e d i s p e r s i o n computations and Mr.Pet. - Simulation O f S t r a t i f i e d Waterflooding By Pseudo Relative Permeability Curves. SPENCE. . A Examination Of The Multiphase Darcy Model Of F l u i d Displacement I n Porous Media. An Evaluation O f Diffusion E f f e c t s I n Miscible Displacement. Hughes f o r t e c h n i c a l help. J.328.. S. 13. HANDY. A Method For P r e d i c t i n g The Performance Of Unstable Miscible (June 196312. Paper 8909. S0c. 12.P. I d e a l i z e d Behaviour of S o l v e n t Banks i n S t r a t i f i e d Reservoirs. REFERENCES L. JOHNSON. C. (March 1963) 2. Displacements I n Heterogeneous Media. Paper 3610: 15.. C. R. Trans. & SIMON. No.. O. A.Eng.P. - 6. Influence On F l u i d Flow of Conrmon Sedimentary S t r u c t u r e s I n Sand Bodies. G.

On the basis of typical reservoir data. HEC. which is not satisfactory at very low shear rates and at relatively high shear rates. . conclusions are drawn for the selection of polymers according to their rheological properties. An improvement of the mathematical description was achieved by using the Carreau viscosity equation and deriving a filter law for porous media. BPS) used in enhanced oil recovery.CHEMICAL FLOODING 179 SOME ASPECTS OF THE INJECTIVITY OF NON-NEWTONIAN FLUIDS IN POROUS MEDIA PETER VOGEL and GUNTER PUSCH Institut fur Tiefbohrkunde und Erdolgewinnung. Technical University Clausthal. the behaviour of an injection well during polymer injection is investigated by calculating the pressure profile around a wellbore. The validity over a wide range of shear rates was proven by experimental results obtained from flood tests in sand packs with one typical product each of the three polymer classes (PAA. From these data. the rheological behaviour of polymer solutions is commonly described by the power law. West Germany ABSTRACT In existing numerical models.

The importance which is at present attached to this field of research is thus evident. To begin. that is.180 I NTRODUCTI ON Flooding with viscous media has aroused increasing interest in the field of enhanced o i l recovery. The liquids employed are aqueous solutions . the solutions exhibit pseudoplastic behaviour. that is. varies as' a function of the shear rate. 2 / . The investigations were initiated by the following two questions: - - How can the viscosity values indicated in a rheogramme be applied to flow processes in porous media? Can these polymer solutions be injected into the reservoir without exceeding the fracturing pressure of the rock? In the following. In general. CHARACTERIZATION OF THE POLYMERS EMPLOYED Information about the flow behaviour of non-NEWTONian fluids is provided by their rheogramme. the rheogrammes of the polymer solutions used here are presented. a method which allows a calculation of the injectivity of polymer solutions on the basis of the rheogrammes and of the knowledge of the characteristic reservoir data is presented. In the field of enhanced oil recovery. Numerous pilot projects are currently in progress or have already been terminated / 1 . All of the considerations discussed in the following are based exclusively on the information gained therefrom. the plot of the viscosity as a function of the shear rate. the viscous behaviour of polymer solutions in porous media has become of vital importance as far as their injectivity is concerned. A characteristic feature of these polymer solutions is that the decisive parameter for the description of their flow properties. Chiefly aqueous polymer solutions are employed as viscous flooding media. a decrease of the viscosity with augmenting shear stress. the viscosity. this is both important and experimentally easy to obtain.

181 Figure 1: Viscosity behaviour of a polysaccharide solution Figure 2 : viocosity behaviour of a hydroxyethylcelluloae solution .

a linear decrease is observed at higher values. A double logarithmic scale has been chosen for the graphic representation. Polymer solutions which yield a mutually comparable additional oil recovery (p' of additional oil per m 3 of polymer solution consumed) in flooding tests were thereby selected. For the calculation of the flow behaviour of these nonNEWTONian fluids. The three curves display characteristic features in common: A plateau occurs in the range of low shear rate. reservoir temperature of 5OoC) of a typical. an analytical expression for the dependence of the viscosity on the shear rate. representative product in each of the three classes of polymers used in enhanced oil recovery. is of special importance. which represents the experimental values of the rheogramme over a wide range of shear rate. Figure 1 shows the rheogramme for a polysaccharide. The preceding figures show that the four-parameter equation found by CARREAU /3/ . and figure 3 that for a polyacrylamide solution. figure 2 that for a hydroxyethylcellulose.182 100 I Viscosity behaviour of a polyacrylamide solution Figure 3: (original brine with a salt concentration of 1 0 0 g/l.

The power law frequently employed in previous publications is considerably simpler to handle analytically. n o denotes the viscosity at the shear rate 0 = 0. The significance of the parameters in the CARREAU equation. the essential steps in the development of a filter law for CARREAU fluids m d e s c r i h e d . For a wide range of shear rates. n-1 is the slope of the linearly decreasing part of the curve. The procedure common to their derivations is as follows: First the capillary flow is treated analytically for the liquid in question.are obtained. and can be determined directly from the horizontal portion of the curve in the range of very low shear rates. are briefly explained. sizable errors can result in the description of the flow processes in porous media. A F I LTER LAW FOR CARREAU FLUIDS Filter laws for non-NEWTONian fluids are known only for a few special cases / 4 . This procedure is adopted in the following as well. values indicative of rl. By means of supplementary measurements performed in the range of high shear rates. as will be shown by means of an example. as described in this work with respect to the viscosity model. Consequently. in order to obtain a filter law with the use of an appropriate capillary bundle model. 6. For the polymer solutions investigated in this work. a filter . as well as a simple method for determining them. and is therefore preferred for the treatment of concrete problems. however. a power-law dependence of the viscosity on the shear rate does not describe the experimentally observed behaviour with sufficient accuracy.183 (1) provides a good fit to the experimentally determined rheogramme for the polymer solutions under investigation here. 7/. The plateau for the range of low shear rate and the linearly decreasing part of the curve intersect at a point whose abscissa is approximately equal to 1/X. an extension. 5. In the following. is indispensable.

0 Figure 4: Straight capillaric model of a porous medium The simplest capillary model of a porous medium /a. Figure 4 illustrates this concept.184 law is thereby derived for CARREAU fluids. On the basis of can be derived. For the derivation. it is no empirical corrections are required. 9 / consists of a bundle of circular cylindrical capillaries of equal radius R. In the remarkable that this theory. A comparison of the DARCY filter law with the law of HAGEN- POISEUILLE yields the "hydraulic equivalenceradius" for this simple model: By means of this concept. the flow through a porous medium is related to the capillary flow of the liquid in question and can be treated accordingly. and the porous medium is replaced by a capillary bundle which is hydrodynamically equivalent with respect to porosity and permeabi 1ity . for this purpose the velocity profile and the averaqe velocity of the capillary flow must be known. results. It is necessary first to calculate the flow behaviour of CARREAU fluids in capillaries. a filter law for CARREAU fluids The procedure is justified by experimental following considerations. a circular cylindrical capillary .

The introduction of the wall shear rate 0 . This differential equation is transcendental in the derivative of the function being souqht. the direction of .185 .flow is taken to be that of the positive z-axis. this fact proved to be a considerable problem in the further course of the calculations. this expression can be easily transformed to a filter law. 1 Figure 5: Flow throuqh a circular tube of radius R and lensth L is considered . as a parameter is decisive for the solution of this problem. The calculation / l o / finally yields an analytical expression for the average velocity during capillary flow. The differential equation for the radial velocity distribution v(r) is (3) whereby po . the one-dimensional filter law takes the following form : .and a cylindrical coordinate system is introduced. The z-axis and the capillary axis are identical. v(r).pL denotes the applied pressure difference. In the case of the capillary bundle model used. By means of the hydraulic equivalence radius.Figure 5 .

the following substitutions have been made : .186 In order to save space.

In order to carry out the required flood experiments. Sand packs of 50 percent porosity and 5 D permeability. served as porous media. compacted by vibration. the result is (7) With the use of the present results. on the other hand. as well as the permeability and porosity of the porous medium have been determined. COMPARISON OF THEORETICAL AND EXPERIMENTAL RESULTS The theoretical results are verified by experiment. the effective viscosity in the porous medium was determined directly from the measured data according to ( 7 ) on the one hand. and on the maximal shear rate 9 occurring in the capillary bundle model.With the exception of a correction factor. no empirical correction factors are thereby required. an apparatus similar to that already used by DARCY was employed. The maximal shear rate is obtained from the transcendental equation which admits an iterative solution according to the BANACH fixed-point theorem. and by means of the previously derived filter law. and the corresponding filter velocity is determined from ( 4 ) . This factor depends on the parameters of the CARREAU equation . If the DARCY equation is solved for the viscosity. . The algorithm necessary for the numerical solution of equations (4) and ( 6 ) requires the following steps: After the parameters of the CARREAU equation. the external form of this filter law is identical to that of the DARCY law. q R is calculated from ( 6 ) for predetermined values of the pressure gradient.

A 10080.'s . were treated as power-law fluids.C A R R E N MODEL ----. \ t L u w K 100 Figure 7 : VELOCITY ( m/dl .z t i 1:s * Figure 6: Effective viscosity for flow of polysaccharide solution in porous media' 500.C . polysaccharide and hydroxyethylcellulose.CARREAU MODEL CALCULATED -POWER -LAW-MODEL '. decreasing range in their rheourammes. D VELOCITY Im/d) .2 - Effective viscosity for flow of hydroxyethylcellulose solution in porous media .Z .e 1'. 1. i'.4 . t A \ \ 400.Z .188 For comparison.8 1.6 . OBSERVED VALUES CALCULATE0 .CALCULATED-POWER -LAW MODEL - o I c u 10 .--\ \ \ \ 0 OBSERVED VALUES CALCULATED . which exhibit a dominantly linear. 300. W .

and 8 show the dependence of the viscosity on the filter velocity and compare the experimental and theoretical results. The power-law model describes the dependence of the viscosity on the filter velocity with sufficient accuracy in the case of polysaccharide. These examples demonstrate the advantages of the new filter law for the questions under investigation. CALCULATION OF THE INJECTIVITY BEHAVIOUR During enhanced oil recovery. the pseudoplastic behaviour of the polymer solutions used exerts a pronounced influence on their injectivity.CALCULATED -CARREAU HODEL 0 OBSERVED VALUES Figure 8 : Effective viscosity for flow of polyacrylamide solution in porous media From the filter law for power-law fluids.189 100 t . Hence the agreement between theory and experiment can be regarded as qood. whereaa considerable deviation occurs for hydroxyethylcellulose. Once the questions concerning filtration . For the CARREAU model. the effective viscosity in a porous medium was likewise calculated. . 7. and less than 15 percent for the hydroxyethylcellulose solution. Figures 6. the deviation between the experimental and theoretical results is less than 10 percent for the polysaccharide and polyacrylamide solutions.

etc. A t this juncture. thus a criterion for decision must be established. The multitude of influential parameters necessitates a restriction to a typical case encountered in practice. . The following. with the use of the filter law just presented. For a depth of 1000 m and under the assumption that the average reservoir pressure corresponds to the hydrostatic pressure . In the following. the flowing pressure and radial distribution of pressure around the injection well are calculated for an injector in a radially symmetric reservoir and for a predetermined injection rate. this is a vital cirterion because of the high financial risk involved. the question of the injectivity of the polymer solution involved remains to be answered by the reservoir engineer. stability. realistic. have been clarified for a given reservoir in the course of the product selection procedure. The predetermined injection rate and the average reservoir pressure also affect the decision.069 m rW = Injection rate q = 100 m3/d Depth = 1000 m FORMULATION THE SELECTION C R I T E R I O N OF From the standpoint of reservoir engineering. the essential criterion for the injectivity of a polymer solution is that the fracturing pressure of the rock must not be exceeded during the injection. qeometrical and physical reservoir data are employed for the model calculations: Reservoir: Permeability Porosity Effective reservoir thickness K = 1000 d = 0.24 h = 4 mD m Well : Cased with 7" diameter and ideally perforated in the reservoir zone Wellbore radius 0. A method must be provided for predicting the behaviour in the field on the basis of laboratory data. an important decision of whether or not a selected product is suitable for field application must be made.190 adsorption.

The ranue of influence of the injector is selected at re = 200 m. the reservoir pressure of 100 bar is assumed to prevail at the outer boundary. a radially symmetric reservoir is thereby assumed. Thus. This function is subsequently integrated. With reference to / l l / . this results in a maximal bottom-hole flowing pressure of 180 to 240 bar: hence the bottom hole flowing pressure may exceed the average reservoir pressure by a maximum of 80 to 140 bar durinq polymer injection. This provides a possibility of determining the distance from the well corresponding to given values of the pressure gradient by means of the filter law and equation ( 8 ) .18 and 0.24 bar per metre of depth. CALCULAT IONAL PROCEDURE The object of the calculation is to determine the relationship between the pressure gradient and the distance from the well. the filter law just developed allows only the determination of the corresponding filter velocity for given values of the pressure gradient. As a result of its structure. the following procedure is adopted for determining the locally prevailing pressure gradient. the entire calculation is performed numerically.191 a value of 5 = 100 bar results. The calculation starts with the determination of the pressure . the following criterion for decision is obtained: The polymer solution is injectable provided the pressure drop over a distance of 200 m from the bore hole does not exceed 80 to 140 bar. For the injector under consideration here. Furthermore. The order of magnitude of the fracturinu qradient typical for sedimentary rocks lies between 0. From the equation of continuity the following expression is obtained for the radial velocity distribution: whereby r denotes the distance from the wellbore axis. Because of the complicated structure of the filter law previously derived.

DISTANCE (M 1 180. The value of the pressure gradient corresponding to the radius thus determined is then taken as the pressure gradient at the well. and the corresponding values of the radius are determined from the filter law and equation ( 8 ) . 150. Calculated pressure profile during polymer injection . 1 2 3mi 360.192 gradient at the bore hole. Subsequently. of which one is smaller and one larqer than that prevailing at the well. a tabular representation of the pressure gradient as a function of the distance from the well is obtained. 90. 400. By nesting of intervals a sequence of pressure gradient values is constructed in such a way'that the values of the radius determined from the filter law and equatisn ( 8 ) converge toward the wellbore radius.. 120. Figure 9: 30. 60. 210.3 POLYSACCHARIOE POLYACRYLAMIDE HYDROXYETHYLCELLULOSE 32 0- - 200: 2(O: 0. For this purpose. are initially assumed. .The calculation of the total pressure drop is subsequently performed by means of numerical integration. two values of the pressure gradient. this value is decreased stepwise. The procedure is truncated as soon as the wellbore radius has been approached with the required accuracy. Thus.

In contrast. The pressure difference. the concentration of the polymer solution to be used must be reduced. the radial pressure distribution around the injection well during the injection . the model calculations indicate that the maximal permissible injection pressure will be exceeded. and partially fulfilled for the polyacrylamide in this case. is plotted as a function of the distance from the well for the three polymer solutions under investigation.the use of the new filter law. the hydroxyethylcellulose exhibits a decidedly deviating behaviour. the polymer solutions are suitable for injection provided the pressure difference remains less than 80 to 140 bar over a distance up to 200 m from the injection well.193 RESULTS OF THE MODEL CALCULATION In figure 9. and increases to more than 350 bar over a distance of 200 m. According to the criterion formulated here. If. It must be emphasized that this is a model calculation. Moreover. The parameters of the CARREAU equation are then determined from the rheogramme. CONCLUSIONS The rheological behaviour of aqueous polymer solutions is well described by the CARREAU model. the maximal values of 80 and 140 bar for the injection overpressure are indicated. A further possibility is the purely theoretical plotting of rheogrammes for injectable fluids by the variation of parameters in the CARREAU equation. as referred to the well. This condition is fulfilled for the polysaccharide. already amounts to 140 bar at a distance of about 20 m. whereby the effects described are attributed solely to the dependence of the viscosity on the filtration velocity. and the calculation is repeated with the use of these values. as referred to the pressure at the injection well. in a practically relevant case. With. the pressure difference occurring during injection. A filter law derived for such fluids is described and experimentally verified. the viscosity is thus decreased.

Pet. h d Formation thickness Permeability Length Power-law index Average pressure Pressure drop Injection rate Radial coordinate External boundary radius Wellbore radius Radius of the tube Velocity Filtration velocity Shear rate Shear rate at the tube wall viscosity Zero-shear-rate viscosity Infinite-shear-rate viscosity Time constant Porosity REFERENCES 1 CHANG. the polysaccharide solution fully. and the polyacrylamide solution conditionally satisfies this criterion under the given conditions.. H. Among the products investigated here. NOMENCLATURE h K L n P q PL Po r re rW R V V f P YR ‘ 1 ‘10 1 ‘. Polymer Flooding Technology Yesterday and Tomorrow J. Tech. A polymer solution is judged as suitable for injection as far as the bottom hole flowing pressure does not exceed the fracturing pressure of the rock at the bottom of the hole. 197818 1113 1128 - - . L.19 4 of polymer solution is calculated. (Aug.

Eng. Univ. R. Germany" Erdoel-Erdgas-Zeitschrift 94 (July 1978) 7 . E. SOC. STEWART. 1953) 10. VOGEL. SCHAEFER. 9 ( 1 9 6 5 ) 2. BIRD. 761 - 8.E. 19721. P.J. LIGHTFOOT.. J.J. (1975) 15. PARK. R. . Madison 1968 BIRD. Sons.F.. Thesis. . R B . HIRASAKI. J .. A E . TU Clausthal 1980... "The Flow of Non-Newtonian Solutions Through Packed Beds" 773 Polym. W. New York ( 1 9 6 0 1 . West Germany 1 . 'Rheological Equations from Molecular Network Theories" Ph. HAWLEY. . 9 ( 1 9 6 5 ) 2 . Thesis. BONDOR. Pet. W.P. T. Wiley a. 252 259 - 3. Scie. "Mathematical Simulation of Polymer Flooding in Complex Re servoirs" SOC. 206 4. J. SADOWSKI. Rheol.N. 17 - 23 9. Rheol. "The Physics of Flow Through Porous Media" University of Toronto Press ( 1 9 6 3 1 .195 2.. (Oct. "Non-Newtonian Flow Through Porous Media" Trans.B...E. SADOWSKI. 251 - 6. of Wisconsin. "Theoretical Models of Porous Matter" Producers Monthly 17 (Aug. SOC... T. A. "Transport Phenomena" 207 J. GRODDE.L..C. Eng.. 115 117 "Untersuchungen zur Berechnung des FlieSverhaltens wlSriger Polymerl6sungen in Sandpackungen" Ph.D.H. . . G J r T H A M r M .. H. P. - 5. SCHEIDEGGER.. K.D. "Non-Newtonian Flow Through Porous Media" 271 Trans.. 0 11. 243 250 - 7. 369 - 382 . CARREAU. BLANKS. SCHEIDEGGER.. "Experience with the Application of Polymer to Improve Water Flood Efficiency in Dogger Reservoirs of the Gifhorn Trough. M a c .

This Page Intentionally Left Blank .

I n a l l types of porous media. This i s explained by t h e e x i s t e n c e near t h e pore w a l l of a depleted l a y e r i n which polymer concentration and thus v i r c o s i t y i s smaller than i n t h e bulk. The r h e o l o g i c a l c h a r a c t e r i z a t i o n of s o l u t i o n s h a s s h a m t h a t xanthan macromolecules behave l i k e r i g i d rods i n the s a l i n i t y conditions selected. no d i l a t a n t behavior was detected even a t the highest flow r a t e s . 2) they can be ured i n l e s s permeable format i o n s than previously claimed. A l l microgels were c a r e f u l l y removed from s o l u t i o n s i n order t o study t h e behavior f a r away from i n j e c t i o n wells. 3) very good i n j e c t a b i l i t y i s expected f o r microgelfree solutions. CHAWETEAU and A. The model's p r e d i c t i o n s a r e i n agreement with experimental r e s u l t s . such as packs of g l a s s beads. The r a t i o between mobility reduction and r e l a t i v e v i s c o s i t y decreases a s pore s i z e decreases and polymer concentration increases. This d e p l e t i o n i s due t o s t e r i c e f f e c t s and does not depend on chemical n a t u r e and pore shape. ZAITOUN Institut Francais du Pktrole. Water permeability was not reduced a f t e r polymer flow. The mobility reduction i r l e s s than i n l a r g e c a p i l l a r i e s and decreases with pore size. I n f i n e c y l i n d r i c a l pores. mobility reduction a t low shear r a t e s was found t o be constant and lower than the Newtonian v i s c o s i t y a t low shear r a t e s . The p r a c t i c a l a p p l i c a t i o n s of t h i s study f o r EOR a r e 1) w t h a n s o l u t i o n s a r e b e t t e r sweeping f l u i d s i n heterogeneous r e s e r v o i r s than conventional f l u i d s h v i n g the same average v i s c o r i t y . Moreover. the depleted l a y e r e f f e c t decreases with shear r a t e u n t i l i t vanishes a t high f l o w r a t e s .P.France ABSTRACT The b a s i c r h e o l o g i c a l behavior of xanthan polysaccharide s o l u t i o n s has been extensively i n v e s t i g a t e d by varying polymer concentration. t h e same behavior i s observed. 92500 Rueil Malmaison . carborundum p a r t i c l e s and. showing t h a t the r h e o l o g i c a l behavior was not influenced by r e t e n t i o n o r adsorption phenomena.sand g r a i n s . I n various unconsolidated porous media.CHEMICAL FLOODING 19 I BASIC RHEOLOGICAL BEHAVIOR OF XANTHAN POLYSACCHARIDE SOLUTIONS IN POROUS MEDIA: EFFECTS OF PORE SIZE AND POLYMER CONCENTRATION G. B. except f o r pore diameters smaller than macromolecule length. . A model based on t h i s physical hypothesis i s proposed f o r c a l c u l a t i n g mobility reduction a s a function of pore s i z e and polymer s o l u t i o n p r o p e r t i e s . pore s i z e and t h e chemical n a t u r e of porous media. The experiments c a r r i e d out i n Fontainebleau sandstones having d i f f e r e n t permeab i l i t i e s confirm t h i s observation and show t h a t pore t h r o a t diameters i n consolidated porous media a r e l a r g e r than predicted by the usual c a p i l l a r y models. A comparison between flow curves and rheograms gives an estimation of e f f e c t i v e shear r a t e s i n pore t h r o a t s of porous media a s a f u n c t i o n of average velocity. 31 I .

In the conditions chosen (salinity = 5 g/l NaC1. due to the screening of charged groups. particularly for fermentation broths. recent improvements in manufacturing processes. hydrolyzed polyacrylamides have undoubtedly been more extensively studied in the laboratory and used in field applications. Its molecular weight should be close to 0. mainly at the lowest shear rates in the Newtonian regime. All solutions were obtained by dilution with salted water. and neither dilatant behavior nor mechanical degradation has been observed in oil recovery conditions. most of the oil to be recovered which is located far from the injection well will be swept by a polymer solution without microgels. the poor quality of most industrial products available on the market has excluded xanthan polysaccharides from many field applications. The rigid rodlike conformation of xanthan polysaccharide molecules in most reservoir conditions enables the problems mentioned above to be avoided. However. POLYMER SOLUTIONS The xanthan polymer used is a sample manufactured in a fermentation-broth form by RhBne-Poulenc laboratories with a fermentation process specially designed to avoid microgel formation. 3 ) The mechanical degradation which occurs when hydrodynamic forces on the stretched molecules overcome the strength of carbon-carbon bonds ( 3 ) . showed that this phenomenon was related to the existence of a depleted layer near the wall. pH = 7. However. - . is well documented (5). the macromolecular flexibility of this type of polymer causes several detrimental effects (1): 1) The viscosity decreases sharply as salinity increases. the hydrolysis of acrylamide groups at high temperatures. performed with a well-characterized polymer solution and well-defined porous medium. The present investigation aims to study the influence of polymer concentration and rock permeability in order to estimate the effects of this depleted-layer phenomenon on the sweeping properties of xanthan solutions. So their use is limited to low salinity and temperature reservoirs. The first experiments carried out in porous media with a microgel-free solution (7) showed that the apparent viscosity or mobility reduction is less than the viscosity determined in a viscometer. up to now. Moreover. These newly manufactured polymers contain so few microgels that they will be adsorbed or retained at a short distance from the injection well.198 INTRODUCTION Both hydrolyzed polyacrylamide and xanthan polysaccharide solutions are candidates for enhancing oil recovery. In these conditions. can lead to precipitation in the presence of calcium ions. Thus knowing the basic rheological properties of such a solution in porous media is very important from a practical point of view. so that xaqthan solutions could be widely used in the near future. particularly in powders. The influence of these microgels on their flow behavior has been extensively studied in well-defined porous media (6). The addition of 400 ppm NaNg protected solutions against bacterial attack. particularly in the presence of bivalent ions. reduce to a great extent the risks of well plugging. The viscosity is almost insensitive to salinity.8 x lo6. except in a very low salinity range. This behavior isdue to the coil-stretch transition of macromolecules in converging zones of porous media ( 2 ) . But. So this polymer is potentially very attractive. 2 ) The dilatant behavior at high flow rates which decreases injectability. Further experiments. due to steric effects (8). observed even in neutral conditions ( 4 1 . Up to now. particularly for high salinity reservoirs. The poor solubility of some products and the existence of both microgels and cellular debris. clarified and filtered at very low shear rate to remove any possible remaining microgel with a method previously described (6).

a t very low shear r a t e s . over a wide shear r a t e range ( 0 . BULK RHEOLOGICAL PROPERTIES Shear flow V i s c o s i t y measurements were performed with a s e r i e s o f g l a s s c a p i l l a r y viscometers. 1 t o 3000 s-1) f o r various polymer concentrations (25 t o 2400 ppm) using Rabinowitch-Mooney c o r r e c t i o n f o r power-law f l u i d s . t h e polymer molecule was shown t o behave l i k e a r i g i d rod having a 0.199 8 = 30°C). Viscosity-shear r a t e curves f o r v a r i o u s polymer concentration 2 + . which i s t h e r a t i o between polymer s o l u t i o n and b r i n e v i s c o s i t i e s i s independent of shear r a t e and equal t o q r o .62 pm length and 16. The following can be observed: 1) A Newtonian regime. 2) A t r a n s i t i o n zone. i n which r e l a t i v e v i s c o s i t y decreases with shear r a t e according t o a power l a w whose exponent is 2 m.1) show how s o l u t i o n s behave i n pure shear flow. t h e experimental d a t a f i t very w e l l with the Cerreau model A ( 9 ) . 3 ) A shear-thinning regime. c h a r a c t e r i z e d by a c r i t i c a l shear r a t e . The p l o t s of shear v i s c o s i t y versus shear r a t e i n log-log coordinates (Fig. '1 r o I 7r = [1+ ( TrX Y) '3" 100: 50- 20- l: o 5- 'L 1 Figure 1.5 1 diameter (8). previously d e s c r i b e d . Over t h e shear r a t e range t e s t e d . i n which r e l a t i v e v i s c o s i t y 7. equal t o the inverse of a r o t a t i o n a l r e l a x a t i o h time ir.

times t h e i n i t i a l c o i l diameter) explains t h e magnitude of t h e viscous f r i c t i o n i n c r e a s e with polyacrylamide. However. For shear r a t e s l e s s than a c r i t i c a l value 600 s-i. Figure 2. The high O s t r e t c h i n g degree ( t h e s t r e t c h e d length may be . This increased viscous f r i c t i o n occurring i n converging flow near the entrance t o t h e c a p i l l a r y is explained by t h e s t r o n g o r i e n t a t i o n of the rods i n t h e flow d i r e c t i o n when t h e product of r e l a x a t i o n time by elongat i o n r a t e is s u f f i c i e n t l y high (2) (10). t h e apparent v i s c o s i t y becomes g r e a t e r than shear v i s c o s i t y . meaning t h a t r e l a t i v e v i s c o s i t i e s a r e equal i n both converging and shear flow. t h e apparent v i s c o s i t y i n the model was found t o be equal t o the shear v i s c o s i t y . t h e polyacrylamide molecule is both with t h e same flow conditions ( 2 ) . s t r e t c h e d and o r i e n t a t e d i n the flow d i r e c t i o n by t h e converging flow. ForY > yf. this i n c r e a s e i n apparent v i s c o s i t y is very s m a l l . compared t o t h a t obtained with polyacrylamide s o l u t i o n Indeed. Influence of converging flow zones on apparent v i s c o s i t y The c a p i l l a r y r a d i u s was chosen s u f f i c i e n t l y small so a s t o avoid any i n e r t i a e f f e c t i n our expe imental conditions. thus involving d i l a t a n t behavior. o*= .200 Converging flow A estimate of viscous f r i c t i o n i n converging flows can be made by measuring the n apparent r e l a t i v e v i s c o s i t y i n a model c o n s i s t i n g of successive s h o r t c a p i l l a r i e s separated by c y l i n d r i c a l expansionsfm! which the geometry is shown i n Figure 2 .

201 W L EFFECT I N F O THROUGH FINE CAPILLARIES AL LW The e f f e c t s of pore s i z e on apparent v i s c o s i t y were f i r s t i n v e s t i g a t e d i n a very simple system. in order t o make t h e i n t e r p r e t a t i o n e a s i e r . was used t o o b t a i n s u f f i c i e n t pressure drops measured by oil-water manometers. Bulk r e l a t i v e v i s c o s i t y versus shear r a t e i s p l o t t e d a s a dashed l i n e . so that t h e c a p i l l a r y length t o r a d i u s r a t i o l / r given i n Figure 3 depends on pore diameter. Results and discussion The r e s u l t s of flow experiments a r e shown i n Figure 3. each one containing f i v e membranes separated by nylon g r i d s . 4--+ 3- -3 2- -2 -Nuclepore Membmnes _____ 500prn Copillories A s e r i e s of six f i l t e r holders. Experimental f a c i l i t y Nuclepore membranes were s e l e c t e d f o r these experiments because t h e i r pores have a well-defined c y l i n d r i c a l shape.l i n e curves show the v a r i a t i o n s of r e l a t i v e apparent v i s c o s i t y measured during flow through membranes having d i f f e r e n t pore diameters. The most importcmt r e s u l t is t h a t i n t h e Newtonian regime the apparent v i s c o s i t y in f i n e pores is found t o be lower than i n bulk s o l u t i o n s and decreases with pore . The s o l i d . The thickness of the Nuclepore membranes is constant and approximately equal t o 10 p m .4 and 12 pm) were determined by e l e c t r o n microscopy (81. and the average diameters a r e given i n Figure 3. namelywith a well-characterized r o d l i k e polymer s o l u t i o n flowing through f i n e c y l i n d r i c a l c a p i l l a r i e s . The average diameters and a r e a l pore d e n s i t i e s corresponding t o nominal diameters (ranging between 0.

t h e polymer concentration w i l l i n c r e a s e from zero a t w a l l c o n t a c t up t o bulk c o n c e n t r a t i o n a t a d i s t a n c e c l o s e t o h a l f t h e l e n g t h of a macromolecule. In t h i s l a s t case t h e macromolecules a r e r e t a i n e d on the upstream s i d e of t h e membrane. It must be noted t h a t a comparison between apparent v i s c o s i t i e s i s v a l i d i n t h e Newtonian regime even f o r c y l i n d r i c a l pores having d i f f e r e n t length to-radius r a t i o s . except f o r t h e s m a l l e s t one whose diameter (0. This deplet i o n i s due t o t h e s t e r i c hindrances which reduce t h e p r o b a b i l i t y t h a t t h e macromolecular c e n t e r of mass may be a t a d i s t a n c e less than one macromolecular h a l f length from t h e w a l l a s shown i n Figure 4.62 pm). t h e entrance e f f e c t s can i n c r e a s e apparent v i s c o s i t y i n r e l a t i v e l y s h o r t c a p i l l a r i e s (111. A c o a x i a l two-fluid flow model has been proposed t o schematize polymer s o l u t i o n flow (Fig. The bulk s o l u t i o n with a r e l a t i v e v i s c o s i t y q r b flows i n t h e A depleted s o l u t i o n center of the c a p i l l a r y i n s i d e a r a d i u s equal t o ( r . A t t h e h i g h e s t shear r a t e s . Thus. and t h e shear r a t e dependence must be s t u d i e d with models having s i m i l a r geometric shapes such as g l a s s bead packs ( s e e below). causing an extra-pressure drop and thus a curve upturn i n low shear range. 4 ) . the i n c r e a s e i n v i s c o s i t y due t o t h e polymer i s l e s s near t h e w a l l than i n t h e bulk. A t higher shear r a t e s . The v e l o c i t y i s zero a t t h e w a l l and equal i n both Allowed hsitiom of Rods Ir Figure 4 Schemetic v i e w of polymer s o l u t i o n flow through f i n e pores w i t h a depleted l a y e r e f f e c t . A s a consequence of t h i s d e p l e t e d l a y e r .28 pm) i s l e s s than molecule length (0.5 ) . t h e macromoleby hydrodynamic f o r c e s and can e a s i l y pass through t h e memcules are oriented branes. causing a lower o v e r a l l apparent v i s c o s i t y i n f i n e pores than i n t h e bulk. showing t h a t flow p r o p e r t i e s were not d i s t u r b e d by a d s o r p t i o n o r r e t e n t i o n phenomena. This e f f e c t i n c r e a s e s a s pore diameter decreases. and i t p h y s i c a l l y explains t h e apparent s l i p a t t h e w a l l p r e d i c t e d f o r concentrated s o l u t i o n s (14). t h e water permeability was unchanged a f t e r polymer flow. Such a depleted-layer h a s been t h e o r e t i c a l l y p r e d i c t e d f o r both c o i l polymers (12) and r o d l i k e p a r t i c l e s (131. This decrease i n apparent v i s c o s i t y a s pore diameter decreases has been i n t e r preted ( 8 ) by t h e e x i s t e n c e of a depleted l a y e r near t h e pore.wal1.202 diameter. In a l l experiments. having a r e l a t i v e v i s c o s i t y 'Iw flows i n an annulus having t h i c k n e s s 5 surrounding t h e bulk s o l u t i o n .

75 1 1. From this model. 3) and the predictions of this model was found in choosing the following values for depleted layer characteristics: 8 = 0.31pm).6/r ) 4 (2) p ' Trb' 1 .7 I 43 I 8. the relation between the diameter dependence of apparent viscosity and the depleted. an analytical equation has been derived to calculate apparent relative viscosity 'Irp as a function of pore diameter 2 r: rlrw rlrFJ = where 1 - (1.175 0.080 1 bead pack to of permeability reduction after polymer flow was checked for .6 from the axis.4 0. the same behavior was recently observed with polyacrylamide solutions when there are no effects of adsorption on flow properties ( 1 ) .77 The value of 6 is close to half the length of a macromolecule (L/2 = 0. and the absence TABLE I Permeability k km2) 137 200-250 36 Apparent viscosity rP 0.160 0.40 3.41 diameter tY 0 I I 3.97 &I I 20-3.185 0.40 0.21 0.4 40-50 2.203 the bulk solution and the depleted solution at a distance r . and the value of qrw is consistent with the physical hypothesis proposed. Good agreement is found between their calculations and our experimental findings. Thus. Moreover. Moreover W B E R T and TIRRELL (15) have recently proposed a calculation based on the finitely extendable nonlinear elastic dumbbell as a molecular model and the exclusion of all molecule configurations intersecting the walls.1/p ) (1. FLOW THROUGH UNCONSOLIDATED POROVS MEDIA Pore size dependence Calibrated glass beads having different diameters (see Table 1) were packed to obtain porous media having similar pore shapes but different pore sizes. ' 1.90 index 0.3 p m I .130 0.110 0. The flow experiments were performed with a 400 ppm xanthan solution.very ascertain the absence of any adsorbed layer effect.0 10-20 1 0.layer phenomenon seems to be very well established. . ' Very good agreement between the experimental apparent relative viscosity in the Newtonian zone (Fig.11 2.180 0.66 0.

.204 The flow-experiment r e s u l t s a r e q u i t e s i m i l a r t o those observed i n flow through c y l i n d r i c a l pores (Fig. . -1 I.'' Figure 5. and apparent v i s c o s i t y becomes independent of pore s i z e a t high shear r a t e s ( y 7 3000 s-1). . As expected.3 10 111 2- . 2).5 (3 1 Where a is a shape parameter c h a r a c t e r i s t i c o f t h e pore s t r u c t u r e .. -Apparent (k) Permeobilily m pn r' 8 \ '. . Due t o the s t a t i s t i c a l homothety of bead packs. The apparent v i s c o s i t y i n the Newtoqian zone is XP5gA NoCl 4.. For porous media.... The a value was found t o be equal t o 1 .-.. t h e rod o r i e n t a t i o n with $hear decreases t h e depleted l a y e r e f f e c t a s the flow r a t e i n c r e a s e s .. a * .. . . the value of P i s experimentally determined as being t h a t which gives t h e same c r i t i c a l y c corresponding t o the onset of shear-thinning behavior f o r both the shear viscosity-shear r a t e curve and the apparent viscosity-shear r a t e curve i n the porous medium under consideration. It decreases with the pore size-molecule length r a t i o and i n c r e a s e s a s pore s t r u c t u r e heterogeneity increases. The value of a should be one f o r a bundle of c a p i l l a r i e s having the same diameters. 5 ) . . 7 f o r packs of l a r g e spheres having the same diameter (8). . This i s the case when t h e bead-diameter d i s t r i b u t i o n becomes wider o r when t h e consolidation degree of sands giving sandstones i n c r e a s e s (8). Pore s i z e dependence of apparent v i s c o s i t y i n flow through glass-bead packs found t o be lower than the bulk v i s c o s i t y and decreases with average pore s i z e evaluated by pack permeability a s shown i n Figure 5.5 ? = 4 x 4 v (8 k 0 . A t t h e highest flow r a t e s . the apparent v i s c o s i t y overcomes t h e shear v i s c o s i t y .Bulk Sheor Vscosity Viscosity -2 . l n c r e a s e i n viscous f r i c t i o n i n converging zones of porous media where the macromolecules a r e o r i e n t a t e d i n t h e d i r e c t i o n of flow (Fig. t h e shear r a t e dependence of t h e depleted-layer e f f e c t can be deduced from flow experiments i n t h i s type of porous media. The maximum w a l l shear r a t e i n the average pore t h r o a t diameter was c a l c u l a t e d by: -0. This can be explained by t h e . . 3' 1lSl 1361 1841 12 41 -4 Glass Beod Pocks I11 06 IOZ1 .

5 for homogeneous bead packs. polymer concentration effects The influence of polymer concentration was systematically studied by using a Carborundum pack having a permeability equal to 0.1 p m 2 . and the absence of permeability reduction was checked after every polymer flow experiment. .6pm. ' . y( e -1 's C' The depleted-layer effect as a function of shear rate at different polymer concentrations incarborundum packs 1 l o I '"%2 .48 and an effective pore diameter of 2. the effective pore-throat diameter deduced ore diameter calculated from the polymer apparent viscosity is plotted versu from the simplest capillary model.205 As shown by results obtained with polymer flow through Nuclepore membranes. This effective diameter corresponds to an average hydrodynamic diameter of pore throats where polymer flows. The polymer concentration was varied from 200 ppm to 1600 ppm. Equation (2) gives the relation between apparent viscosity and pore diameter.' . Thus an effective diameter can be calculated for each glass bead pack from the apparent viscosity measured. 2 r = 2 ( 8 k 0-1)8". 10 -0 1 7rP r)r Figure 6. So the average hydrodynamic diameter of pore throats is approxirmrtely equal to 2 ( 8 k 0-1)0. In Figure 9.. a porosity equal to 0. Additional points deduced from experiments carried out in sand packs (8) are also lined-up on the same curve. On the other hand the mean pore size is proportional to the square root of the permeability for homothetic porous media. All the points corresponding to experiments performed with glass-bead packs are lined-up on the first bissectrix.

t h e apparent v i s c o s i t y = 17.5) i s less than one t h i r d of bulk shear v i s c o s i t y ( 'Irb = 62).206 Both shear v i s c o s i t i e s i n dashed l i n e s and apparent v i s c o s i t i e s i n s o l i d l i n e s a r e p l o t t e d i n Figure 6. t h e r a t i o between apparent v i s c o s i t y and shear v i s c o s i t y . Moreover. i . Equation (2) can be w r i t t e n : . The depleted-layer e f f e c t appears t c be insens i t i v e t o t h e pore shape and chemical n a t u r e of porous media. 1200 1600 C ( ppml The i n f l u e n c e of polymer concentration on t h e magnitude of depleted l a y e r e f f e c t A t the h i g h e s t concentration t e s t e d ( c = 1600 ppm). a t low shear r a t e s (Fig. e . The f i r s t observation i s t h a t t h e general behavior i s q u i t e s i m i l a r t o t h a t observed i n glass-bead packs. a f t e r both d i v i s i o n by V r b and i n v e r s i o n . i n c r e a s e s sharply with t h e polymer concentration. Indeed. 7 ) . This r e s u l t i s c o n s i s t e n t with t h e s t e r i c o r i g i n of t h e phenomenon. ( tlrp This polymer concentration e f f e c t could a l s o be predicted. t h e magnitude of the e f f e c t . '7rP I I I 1 '7r 60- XP 5911 NaCl pH=7 8 = 3 0 ° C Corborundum Packs 40- -40 / / 1 20- 0 0 400 800 Figure 7.

075 0.0033 004 .] constant diameter I I I I I 8 0 . FLOW THROUGH SANDSTONES Permeability effects As shown above. the depleted-layer effect depends only on pore size for a given polymer solution.5 1.0. the Cb/Cw ratio is expected to be constant (13).5 (6) 2rc= 2 (8 k 0 is valid only for homothetic unconsolidated packs.1 ~ ~ 5. the thickness of the depleted layer 6 is expected to be constant so that: (5) qrb’ qrp = k + (1 k ) P - where k = (1 l 4 is a positive constant. . always less than 1 for a given porous medium.4 21 15.8 0.165 0.087 2. 2 m = 0.3 6.0096 0. 9 = 4. TABLE I1 -6 Flow through sand packs and sandstones (XP solution.5 0. As a consequence.056 Sand:tone I I I I 9.0373 0.060 0.6 I I 0. electron microscopy observations (16)hsve shown that pore throat diameters generally larger than those calculated by Equation (6). rlw In the concentration range tested. For and are the natural porous media such as sandstones.83 2. Since the viscosity of these polymer solutions is roughly an exponential function of polymer concentration (81. influence of permeability cannot be predicted by a simple model.165 2. Grain diameter Dg(pLm) Permek ( p porosity Apparent viscosity rP Shearthinning 1 ~ ~~ abi1i3 Shear-rate Pore-thros.119 Ssnd.49 0.tone 0. the depleted-layer effect incresses iinearly with p = ‘ r lb. the qrb/f)r.95 2.22) .69 2.0 Sand Sand 2 I I 1 I I 0. &As a consequence .207 For dilute solutions. the well-known relation between pore size and permeability deduced from the simplest capillary model -1 0. thus explaining the concentration dependence observed for the depleted-layer effect.7 0.1 14.062 0.056 I I I I I 4. However.2 I I 13.38 I I I 3.056 0.0206 0.ratio increases very sharply with polymer concentration. this relation is no longer valid.0 I - 44 .

The apparent viscosity t s l e s s than in the bulk i n the Newtonian zone. There remulta are high as 3 in very low permubility range (lO'3<k<!~1O'3pd). an inaea8e i n a and the use O f valuer is obrerved a8 permeability docreares (Table 111. Equation (2) giver the effective diameter of pore throats 2 r a a fuoction of m permability for Fontainebleau sandstones (Table 1 ) A8 e d c t e d . t U a e f + r-tive 1. diameter 2 rp is always larger than 2 rc. and the r a t i o rp/rc increaaes a s $-me. MOS. The r a t i o rp/rc should be no l y one for Fontainebleau rand Pack8 having the same grain diameter (10<k~2Opm but reacher values a s 1. deduced from experimantr vfth Nuclepore membranes. This trend is w r u i s t e n t with the secondary crystallization process which explains the decrease in permeability for Fontainebleau sandrtoner.4p1d (Table 11) were selected t o obtain a quantitative evaluation of the depleted-layer effects.3 p and q d l . After polymer flow experiments. 103 104 y(sec-1) The depleted layer e f f e c t s in flow through Fontainebleau Sandrtoner 1 using polymersolution characteristics in the bulk ( ' b = 4) and near the wall ( 8 = 0. A l l the cores were preflushed by a hydrochloric acid solution t o remove the slight quantity of iron contained i n the sample i n order t o avoid possible interaction8 with the polymer. which indicater a depleted-layer effect that increases as permeability decreases. as shown in Figure 9.3 x 10-3)Id). a s could be expected from the pore d i m e t a r heterogeneity increase with the consolidation pSOC888. P .rie8 Of smdstoner.OVOX'.208 some cores of quartzitic Fontainebleau sandstones having permeabilities ranging from 3x10'3 t o 0. even for the l e s s permeable sample (3. % 4 7)r 4 3 3 2 2 1 loo 1 10' Figure 8. in which experimental point8 corresponding t o 8andrtones are plotted as solid circler.b i l i t y &creaser. 7 7 1 . but are =re accurate because of the mathod homogeneour 8. The experimental r e s u l t s shown i n Figure 8 are similar t o those observed i n unconsolidated porous media. the i n i t i a l permeability of each core was exactly restored. c o n r i r t m t w i t h Dullien's observations.

due t o molecular alignment are too small to induce large concentration differences between the different zones . Ddrodvnamic retention The f i r s t type of hydrodynamic retention. I 20 5‘ 10 0 2r or 2&mi Figure 9 . As theoretically expected. which is related to thermodpamic effects (17) and thus does not depend on pore-molecule r e l a t i v e dimensions. i .209 I I I I - --I Fontainebieau Sandstones 2r~=35(2r. The lowest limit for use of such xanthan solutions should correspond t o pore throat diameters equal to_ycromolecular length ( k O .11) Sandpacks I 1. the entropy difference.4) I I 20-1 0. the use of such polymers is never limited by polymer dimensions. t o permeability much lower than 10 p d f o r sandstones having a structure similar t o Fontainebleau sandstones.3601. Comparison of pore throat diameters detenrdned by polymer injection method with measured or calculated pore diameter i n various porous media From a practical point of view. / p . e . was found t o be almost negligible for these u n t h a n solutions having rodlibe molecules.)~~’ / (8. these r e s u l t s show t h a t xanthan solutions can pass very e a s i l y through even the low permeability zones of reservoirs. Practically. Sandstones I (k)Permeability in p m 2 5 I 1 V i 2 16 . 0 ) i 30. 6 pm) .

e. glass bead packs. 4 ) The depleted-layer effect is also observed in Nuclepore membranes. which depends on pore-molecule relative dimensions end which can be explained by the slow accumulation of polymer molecules in the zones of porous media where pore-throat diameters can be smaller low permeability sandstones than molecular size. was observed in very (k = 3 and 9 x 10-3 pro2). comparison in with those observed in dilatant regime with coiled polymers . the microgels contained in injected solutions are retained in a zone located around the injection well. as predicted by our model. so that w dilatant behavior is observed even at the highest flow rates such as those existing around the injection well. Carborundum and sand packs. Thus. F l o w experiments performed In well-calibrated cylindrical pores established the validity of this equation and provided the characteristics of the depleted layer. so that this effect becomes very significant from a practical point of view.and Sandstones. the concentration differences observed after sudden flow-rate changes (from Newtonian to shear-thinning regimes. In high permeability sandstones. as they are in reservoirs far away from the injection wells. Indeed. The injectability of microgel-free xanthan solutions should be excellent. This is consistent with the steric origin of this phenomenon. particularly. this effect seems to be independent of the pore shape and chemical nature of porous media. with a small apparent viscosity and without any permeability reduction. 5) The magnitude of the depleted-layer effect increases sharply with the polymer concentration. 2 ) An analytical equation derived from a schematization of polymer solution flow as a two-fluid concentric flow is proposed to predict apparent viscosity as a function of pore size and polymer solution characteristics. large concentration decreases in the effluent have been noted and the apparent viscosity was found to increase slowly as indicated by the arrows in Figure 8.namely from 6 to 700 sec’l) werE very small (m/q<I%). . when the flow rate has been suddenly increased and kept constant in the shear-thinning regime. This phenomenon is interpreted by the existence near the pore wall of a depleted layer where average polymer concentration and viscosity are lower than in the bulk. The second type of hydrodynamic retention. the apparent viscosity becomes independent of pore size. its thickness close to the half-length of the macromolecule is consistent with our interpretation of the origin of the depleted layer. at the highest shear rates. the rodlike conformation of xanthan molecules minimizes viscous friction in zones of converging flow inside the porous structure.2 10 of the porous medium. 6) The magnitude of the depleted-layer effect increases as sandstone permeability decreases: Microgel-free ranthan solutions can pass easily. Moreover.(la) (19) (20) (1). 3 ) The depleted-layer effect decreases as shear rate increases so that. through sandstones having very low permeabilities. The main conclusions of these investigations are the following: 1) The apparent viscosity of polymer solutions flowing through fine pores is always less than bulk shear viscosity at low shear rates in the Newtonian regime and decreases as pore size decreases. - - CONCLUSIONS The basic rheological behavior of xanthan solutions in porous media has been studied with solutions without microgels. i. Indeed.

J.C. and Rh6ne-Poulenc Industries provided d'Exploitarion du Petrole ( ARTEP the polymer sample. 7. CHAUVETEAU.. and MOAN. 349-358. 1980. Acknowledments This research was supported by the Association de Recherches sur les Techniques . Sci. J. Rheol. Oct.. 4. Paris. 4-7. 1981. CHAUVETEAU. M.. CHAUVETEAU. CHAUVETEAU. 2. 5.. 8. G. S. . Aced. . (19721. 9.211 7) The average hydrodynamic diameter of pore throats in a given sandstone can be deduced by measuring the apparent viscosity of a well-known polymer solution in the Newtonian regime. enables xanthan solutions to sweep. "Mechanical Degradation of Semi-Dilute Polymer Solutions in Laminar Flowr". C. San Antonio. Polymer Bulletin (to be published). MOAN. paper SPE 10 060 presented at the Annual Technical Conference and Exhibition. 107-110. G. Pol. Techn. . G. (19801. and CHAUVETEAU. 19791. "Rodlike Polymer Solution Flow through Fine Pores: Influence of Pore Size on Rheological Behavior". "Rheological Equations from Molecular Network Theories". MULLER.accepted for publication in Journal of Canadian Chemical Engineering. which decreases apparent viscosity mainly in low permeability zones. . G. G. (Feb.. Journal de Physique-Lettres.. G. C W U . Pet. Dallas. FENYO. (Feb. "High Molecular weight Hydrolyzed Polyacrylamides. Sept. J. 42 (1981) L-201 L-204. Submitted for publication in Journal of Rheology.R. 9'Molecular Interpretation of the Different Properties of Coil Polymer Solution Flow Through Porous Media in Oil Recovery Conditions". 25. I11 Effects of Temperature on Chemical Stability". - 3. Delaplace and R. REFERENCES 1. 288. Sci. "Influence of Microgels in Xanthan Polysaccharide Solutions on Their Flow Behavior Through Various Porous Media". and KOIILER. "Xanthan Polysaccharide Plugging Behavior in Porous Media: Preferential Use of Fermentation Broth". 1981) 23. G.. . M. "Ecoulement laminaire en milieu poreux de solutions de macromolecules de taille non negligeable devant les dimensions des pores". CHAUVETEAU. GHONIEM. CHAUVETEAU. Trans. Tabary who performed laboratory experiments. . A w l . J. C.. and SELEGNY. 627-633. 8 ) The effect of the depleted layer.. MULLER. P. Thus polymer injection is a new method for investigating pore structure. 16. "Thermal stability of High Molecular weight Polyacrylamide Aqueous Solutions". and WOLFF. E. G. KOHLER. 99-127. "The Onset of Dilatant Behavior in Non-Inertial Flow of Dilute Polymer Solutions through Channels with Varying Cross Sections". N.. The authors wish to acknowledge the contribution of Ph. G. . Paper SPE 9295 presented at the 55th Annual Technical Conference and Exhibition. o i l better in heterogeneous formations than conventional fluids having a viscosity that is independent of pore size. N. ... 6. soc. 1981. 21-24.

(19791.L. presented at the Improved Oil Recovery Meeting. (19791. AUVRAY. N.. "Correlation between Pore Structure of sandstones and Tertiary Oil Recovery". . Sci. METZNER. J . of Pol. CHAUVETEAU.K. (Oct. B y 219-220. CHAUVETEAU. . . 18. 19811..in Jmroved Oil Recovery by Surfactmt and Polymer Floodinq. and KOHLER. 19791. "Effective Viscosity of Dilute Polymer Solutions near Interfaces". Acad. Acad.C. Paper S R 4745. WILLHITE.T. M..B.G. L.. .. GAUDU.R. sc. 19731. Houston. Pet. Sci. Fluid. CHAUVETEAU . 288. SOC. 79-95.A. "Entrance Effects in Capillary Flow of Dilute and Semi-Dilute Polymer Solutions". . 439-451. "Ecoulements viscosimetriques de polylderes enchevetres". New-York (1977). Press. H . AUBERT. "The Effects of Rheological Properties and Polymer-Rock Dimension-Sensitive Interactions on Polyacrylamide Solution Flow through Porous Media". 11. G. VOl. J. and GHONIEM. on colloid stability". . "Polymer Flooding: the Essential Elements for Laboratory Evaluation". 17. G. 12. S. Acad. C. D. 82-83. and TIRRELL. M. 22-26. CHAUVETEAU. Press. and DULLIEN. (April 9. April 22-24 (1974). G. . . 19. "Solutions de macromolecules rigides : Effete de paroi. HOA. L. in "Improved Oil Recovery by Surfactant and polvmer Floodinn". 13. 16. Meeting. ACS Polymer Preprint (1981) 22. BATBA. Erin.R. 1073-1084.G.511-553. &&. Rheol. %Mechanisms of Polymer Retention in Porous Media". 1. LEIBLER. 13. 256-258. G. . 14. 463-474. P. New-York (1977).. soc. . "Effects of Polymer Solutions J. Non. J. JOANNY. MOAN. 15. J. 17. Acad. de confinement et d'orientation uar un dcoulement". DE GENNES . V. F. "Flow of Polymeric Solutions and Emulsions through Porous Media".. N. and ANNE-ARCHARD. .F. "Relation entre le champ de vitesse dlelongation et l'apparition d'un comportement dilatant d'une solution de pol-re diluee dans un Ccoulement convergent non-inertiel". (1981). Journal de Physique (Janv. . 5. and DE GENNES. Submitted for publication in C. (1978) Oct. . and DOHINGUEZ. 42. J.. Paris. . Inc.2 12 10. R. .. . Inc. G. P. A.Newt. 20.P.

. The oil is paraffinic and has a viscosity of 40 cPo at reservoir tenperkture (30OC). LABASTIE and L. and the oil eone extends over an area of 20 km2. The pilot is developped with one injector and seven producers. Several commercial products have been tested. mainly for viscosity and injectivity . The water being almost fresh (0. But the seventh intermadiate producer.CHEMICAL FLOODING 213 THE CHATEAURENARD (FRANCE) POLYMER FLOOD FIELD TEST A. found in the southern part of the Paris Basin . 5 meters thick. is part of the Neocomian (Lower Cretaceous) oil reservoirs. outlined in Fig.4 g/l TDS) it has been decided to use hydropolyacrylamides. Field deBcriDtion The Chateaurenard field. The injection has been started in 1977 and on account of the quantities injected so far. this response is due to the effect of nobility control. the 44 ha pattern enclosea a very important pore volume (700 000 m3). drilled at a shorter distance (280 m) from the injector. a liquid polymer. we have not yet seen any response in the sir main producers (which are at a distance of 400 to 500 m from the injector). 1. maybe amplified by a local reservoir heterogeneity. dissolved in produced water. The water must be carefully treated before dissolution to avoid polymer degradation and formation plugging. VIO E fAquitaine ( M u c t i o n ) l Abstract A polymer flood is operated by Elf Aquitaine in the Chateaurenard (France) field. is presently being used for the pilot. it is located 100 km SSE of Paris. at a depth of 600 m . in a layer of unconsolidated sand. located in the Paris Basin. has shown vexy interesting results with a sharp decrease of the WOR (WOO tons of tertiaq oil have been produced) .

2 14 Fig. 1 - Map of Chateaurenard field .

2 - Type l o g . because the clay layer between them is discontinuous and does not form a tight barrier. this sedimentation type gives massifs with sharp lateral variations of facies.2 15 Geol o m There are three distinct structures. The reservoir forms a roughly triangular monocline whose closures are a fault in the east and the wedgeout of the sands in the south. 2 . belonging to the Hauterivian stage of the Neocomian) of the central structure . these two levels can be considered as a single reservoir. - The deposit of these sands is in the form of submarine channels . Situated at a depth of about 600m. CHUtLLLS FIELDS - . the reservoir is formed of three layers of unconsolidated sands separated by shale as depicted on the type log of Fig. FIRMIN. the dip of these layers is very slight.- Gamma-Ray Laieroloa Fig. about l o . The reservoir concerned by the polymer injection is formed by the two upper levels (R1 and R 2 . separated by North-South faults with throws of 15 to 20 m. T Y P E LOG C H A T E A U R € N A R D St.

Peak oil production . aa.s (CPO) Water salinity (TDS). but with a wide g r a i n sine distribution (80 to 3 5 0 p Y The average permeabilitg ) 1 ( ). 3 the relative permeability curve3 water appeared early in the production and water cut increased quickly . This sandstone is relatively fine ained (average 150k). mD Clay content. cumulative m b) production was 26 $ of OOIP. with some amount of clay(2 to 15 k). $ Permeability.Relative permeabilities . but with rather large and unforeseeable variations on is 1 Darcy m account of the channel type sedimentation system. mainly through the action of an edge water drive. its average value for the field was 89 $. $ Residual oil saturation. OC ( O P T Oil gravity. $ Current field avera e oil saturation. The average total thickness in the pilot area is 5 metres. the relevant characteristics are indicated in the folhing table : Depth. with a porosity of 30 $.reached 267 000 m3 in 1964 . The fluids are a relatively viscous oil (40 cPo at 30OC. m (ft) Porosity. ppm 600 (1970) 30 10 00 2 to 15 30 30 55 30 (86) 089 (27) 4 (40) 0 400 7 0 Production history The field was discovered in 1958. $ Temperature. I n 1980. g/cm3 (OAPI) Oil viscosity. p m Water hardness (Ca + Mgp. and an almost fresh water . 1 8 production was 90 95 0 0 m3. with half f o r R1/R2 reservoir. in 1980. Because of very small dip a le and adverse mobility ratio (see kn Fig. % Initial water saturation. 0 .2 16 Characteristics of R1/R2 reservoir : fluid properties The reservoir is formed by uadonsolidated sand. reservoir temperature) of paraffinic type and without dissolved gas. initial oil in place was estimated at 1 1 M 3 (69 millions b l .

To dissolve polymer in produced water was the easiest for field operations. This pattern is outlined in Fig. Polymer choice and slug design The water being almost fresh and the temperature low.CR 6 CR 12 CR 16 .Polymer pilot pattern The surface area of this pattern is 44 ha (110 acres). but it was necessary that the two layers were not separated at the polymer injectca: well. a seven spots pattern has been selected. and it encloses a pore volume of 700 000 m3. So we have studied the degradation of polymer in presence of oxygen and iron (little amounts are present in produced water). and for this polymer pilot it has been decided to use this spacing. maybe responsible of poor sweeping efficiency .CR 19 Bis . the oil saturation was still 55 $. After several interference tests.CR 2 bis) at distances of 400 to 500 meters from the 1 injector . with one injector (CR 9 bis) and six mpin producers (CR 3 . because of the two layers with a discontinuous separation. after many years of waterflooding. The mobility ratio is very adverse and before polymer flood. 4 . It has shown that we must avoid the presence of both iron and oxy en. . The R1/R2 was intergsting for this test. hydrolyzed polyacrylamides were selected. - - -Clay isopachs (separation R1 . a seventh intermediate producer (CR 56) has been drilled at a shorter distance from the injector. 4. in order to get an earlier response. but that little amounts of one of them is not detrimental (see Fig.2 17 POLYMER PILOT DESIGN Pattern selection The Chateaurenard field has been developped with an average well spacing of 400 m. 5 .6 7 .R2) Fig. but it can Be detrimental for polyacrylamides stability () 1.

6 A f t e r 2 hours Chemical ctegraduation of polyacrylamides : v i s c o s i t y ($ of i n i t i a l v i s c o s w y f u n c t i o n of Fe and 02 c o n t e n t - .2 18 Fig.

64711 . l i l i r i 5. oxygen (<0. In this example. After treatment.01 ppm) and iron (< 5 p m .2lk ritlltticl) G~(pmhr.33 PV with 700 ppm polymer has been chosen. Surface installations The polymer solution is prepared with produced water. . with p) low amounts of oil (< 3 ppm). residual oil (1000 ppm). available in great quantities but not clean : it is contaminated by iron (Fe" and Fe+++).ii ritcitiw) 0 5 m 16 ZO 'tiw. with viscosity decrease designed to prevent deleterious effects of viscous fingering (3). the water is $ood for polymer dissolution. oxygen ( < 1 ppm) and clay particles. product A is used for this operation. products B and C2 are not satisfactory (plugging behaviour) . This is important because a plugging behaviour can be hidden if the fbw rate is to high. and the flow rate is chosen to give same shear rates as in the field flood. The injectivity test is a constant flow rate test through a 5. even if sandpacks floods may be necessary for further investigation. 7 as an example.umilliporeR filter with pressure drop measurement . This test seems a good screening procedure f o r comparison of products. A I (liqiil) I (liqiil) Sl(prlrr. clay and insoluble iron is carried out in a flotator using nitrogen. . products A and C1 are good. A bacteria killing agent is added to the water. The polymer slug will be followed by water injection. due to microgel deformations in impoktant pressure gradients ( 2 ) . Removal of oil. AP millibrn I f .2 19 Commercial products were tested mainly for viscosity and injectivity. a slug of 0. Some results are given in Fig. a nitrogen blanket prevents anay oxygen entry. the pore size ( p is similar to the one of 5 ) Chateaurenard reservoir. Polymer concentration and slug size have been determined by performance predictions with a reservoir model .h It has been found that dry polymers (powder) need a retention of several hours after dissolution to become satisfactory on injectivity test. which also strips the water of oxygen traces . that is filtered before polymer dissolution.

without any problem. since 1980. The time of oil bank breakthrough is in accordance with our predictions. 9). which is only 55 $ now (increasing). the seventh internlediate producer (CR 56) . has been put into production in 1978. but it is not possible to explain such an oil cut assuming an homo eneous repartition of permeabilities and saturations before polymer flodd fsee predicted and observed production in Fig. the water cut was about 90 $. followed by water. Until 1979. dry powder polymer was used. drilled at a distance of 280 m from the injector CR9 bis. The fluid flow diagram of Fig. and then final dilution .7 $ pore volume?. However. polymer solution is 5rfiltered at the wellhead before injection. the flow rate was 135 m3/day (850 bbl/day) . Untgll mid 1979. as in the other well of this area. which is normal on account of quantities injected so far. has given very interesting ) results (see-fig. 9. Up to 'une 15. A slug of 235 000 m3 is to be i n j e M . then we observed a sharp decrease to 20 $ of the water cut. with residence time between dissolution and injection to get good injectivity characteristics. . It has been changed for liquid polymer (emulsion). no injectivity problems were encountered with polymer solution. with decreasing concentration at the rear front to prevent fingering. easier to handle and good for injection immediately after dissolution. lil 151 I* (IWW TIM t t t t litqcc PILOT PERFORMANCE : FIRST RESULTS The polymer injection has begun in 1977. 8 outlines the equipment. it has been increased to 250 m3/day (1 600 bbl/day) . 9000 tons of tertiary oil have been produced. The six main producers have not yet shown any response. This well.2 20 At the beginning of the operation. 172 700 m3 of polymer solution have been injected (24. that is designed for first dissolution of polymer in a concentrated master solution.

due to mobility control effect. A. which can be swept by polymer flood in good conditions.221 A very poorly waterflooded zone (of lower permeability) has been reached by polymer flood. b b . which is carefully treated. 07 0s b 0.2 b b b b * b 0 CONCLUSIONS 1 . some other poorly w t r d tp aezones may exist. a poorly waterflooded gone has been swept by polymers. We have now to wait for the response of other producers to have a good idea of polymer flodding performance in this field. due to mobility control effect..The polymer solution is prepared without 2 - 3 problem using produced water. However. this effect has been very important because 6f a local Eeservoir heterogeneity. This very good result would have not been possible without mobility control. so such a result cannot be generalized. But with this channel r y e sedimentation system. h -. so it must be attribuBed to the effect of polymer injection. . go00 tone of tertiary oil have been produced.significant decrease of WOR has been observed in the closest producer well . The solution is easily injected and does not show any plugging behaviour.

International Symposium on hydrocarbon exploration. 1975).222 REFERENCES 1 - 2 - 3 - G. 17) E. Chauveteau . drilling and production technics (Paris.Paper SPE 6848 presented at 52nd SPE Annual Fall Meeting (Oct. Claridge . Chauveteau et N. Kohler .Conditions de stabilit6 des solutions de polymkres lors d'une injection sup champs. 10 .The effect of rheological properties and polymer rock dimension sensitive interactions on polyacrylamide solution flow through porous media . G. 1 7 ) 97.A method of design of graded viscosity banks .12 dec. .49th Annual Meeting of the Society of Rheology ( 9 8 . L.

Signal H i l l . Additional o i l was recovered i n core f l o o d s when 1. Now w i t h X t r a Energy Corporation. . Laboratory i n v e s t i g a t i o n s conducted j o i n t l y by THUMS and the Department o f O i l Properties i n d i c a t e d t h a t Ranger Zone crude could be r e a d i l y emulsified i n the presence o f water containing as low as 0. C a l i f o r n i a . AND FIELD RESULTS VERNON S.CHEMICAL FLOODING 223 CAUSTIC FLOODING IN THE WILMINGTON FIELD. heterogeneous. The purpose o f the p i l o t demonstration i s t o evaluate t h e e f f i c i e n c y o f t h e caustic displacement mechanism i n t h e environment o f a s t r a t i f i e d . MAYER THUMSLong Beach Company JOHN D. high o i l v i s c o s i t y r e s e r v o i r where primary waterflood recovery i s r e l a t i v e l y poor. The r e s u l t s o f t h e l a b o r a t o r y core t e s t work and t e s t s o f t h e r e a c t i o n between a l k a l i n e s o l u t i o n s and r e s e r v o i r sands used i n r e s e r v o i r simulations i n d i c a t e d o i l r a t e response and t o t a l incremental o i l recovery are v e r y dependent upon t h e caustic concentration and caustic s l u g size. The p i l o t t e s t i n v o l v e s t h e i n j e c t i o n o f c a u s t i c s o l u t i o n i n t o a modified staggered l i n e d r i v e w e l l p a t t e r n c o n s i s t i n g o f e i g h t i n j e c t i o n w e l l s which surround eleven a c t i v e producers i n an area o f approximately ninety-three acres.1% b y weight sodium hydroxide. Wilmington Field. CALIFORNIA LABORATORY. The t e s t area i s located i n t h e Ranger Zone o f F a u l t Block V I I w i t h i n the Wilmington F i e l d . i n association w i t h the United States Department o f Energy. CARMICHAEL City of Long Beach Department of Oil PToperties' ABSTRACT A c a u s t i c enhanced waterflood t e s t i s being conducted i n the Ranger Reservoir o f the Long Beach Unit. MODELING. A l k a l i n e consumption calculated t o be very large. C a l i f o r n i a by t h e Department o f O i l Properties o f the City o f Long Beach and i t s f i e l d contractor.0 weight percent sodium c h l o r i d e was added t o the a1k a l ine s o l u t i o n . THUMS Long Beach Company. 1 . BREIT Scientific Software Cotporntion EDWARD H.

The p r e . high v i s c o s i t y and high organic acid content. c FIGURE 1 . INTRODUCTION The Wilmington F i e l d i s the l a r g e s t f i e l d i n C a l i f o r n i a .f l u s h amounted t o approximately 10 pore volume percent.0 weight percent s a l t i n softened water i s being i n j e c t e d . 1. Fig.4 weight percent sodium o r t h o s i l i c a t e and 1. The predicted f u t u r e performance o f the f i e l d f o r both continued waterflooding and a caustic f l o o d i s sumnarized. 1980. A l k a l i n e f a c i l i t i e s were completed and placed i n operation on March 27. It has seven basic r e s e r v o i r zones w i t h crudes t h a t g e n e r a l l y have a r e l a t i v e l y low g r a v i t y . A l k a l i n e s o l u t i o n containing 0.FIELD LOCATION MAP . Pre-flush i n j e c t i o n consisted o f 11.224 This paper sumnarizes the r e s u l t s o f the caustic core f l o o d s which were performed t o evaluate t h e entrainment mechanism o f o i l displacement and l a b o r a t o r y t e s t s t o evaluate t h e long term consumption o f hydroxide ions by t h e r e s e r v o i r sands. The past performance o f the f i e l d and the r e s e r v o i r simulation history-match o f t h a t past performance are discussed. The recovery e f f i c i e n c y f o r the waterflood i n the Ranger Zone o f the Wilmington F i e l d has been low due p r i m a r i l y t o a h i g h l y unfavorable m o b i l i t y r a t i o between water and o i l and s i g n i f i c a n t r e s e r v o i r s t r a t i f i a t ion.5 m i l l i o n b a r r e l s o f softened f r e s h water w i t h an average o f 0.96 weight percent s a l t .

I . wettability reversal. These t e s t s -- ! # I f .-@ Q r a t e varying from s i x f e e t per day p r i o r t o do P water breakthrough t o one $0 f o o t per day a f t e r water 1% Ir breakthrough (the 9i reduction was done t o I@@: avoid excessive pressure s : I A : l B gradients within the 0 .. z : H .cL IOU II to minimum water I OL I mv = s a t u r a t i o n using crude I I oil. I n t h e l a b o r a t o r y the plugs were placed i n a modified Hassler sleeve apparatus.. I. The enhanced I I o r a l k a l i n e water d r i v e I I tests were then . The cores were heated t o r e s e r v o i r temoerature o f 125'F and lQor . Included among these are emulsification and entrainment.. S*rllr using Ranger Zone crude --.1 waterf looded t o residual " .S*l"l..r minimum water s a t u r a t i o n A C.. . ..iory namuitm dynamically driven t o 0 Phi. ~. The long term a l k a l i n e comsumption t e s t s were performed w i t h sand packs * i c h were prepared i n L u c i t e columns and v a r i e d i n length from s i x t o twelve inches w i t h a diameter o f approximately one and a h a l f inches. The comparative core f l o o d s were performed w i t h preserved core m a t e r i a l which was c u t p a r a l l e l t o the core axis. 1 ' . i. o i l s a t u r a t i o n (see Fig.bor. Comparative Core Flood Studies Frozen preserved core samples were jacketed on coring i n p l a s t i c tubing. Several mechanisms have been postulated f o r t h e improved o i 1 recovery r e s u l t i n g from a l k a l i n e waterflooding. thawed and confined a t 1600 p s i overburden oressure.*.. 3' A 1 I # 2). although l i m i t e d t o r e s e r v o i r s w i t h s u i t a b l e crude o i l s . ' SIL s ia n n be u 6s re performed. The plugs measured approximately two inches i n diameter by f i v e inches long.CORE SAMPLt NO. The cores were 5 w0. The samples were then water driven at a A I. ' I I * I * 1 . - - a ' a 0 - .225 The concept o f a c t i v a t i n g the n a t u r a l surfactants present i n t h e crude o i l by contact w i t h a l k a l i n e water.c r u d e o i l r e a c t i o n which would take place i n the r e s e r v o i r . has p o t e n t i a l economic advantages over commercial surfactant flooding owing t o the high cost o f the s u r f a c t a n t s and t h e low cost o f a l k a l i n e materials.i. and emu1s i f i c a t i o n and entrapment1 The r e 1a t ionship between these p o s s i b l e mechanisms i s necessarily more complicated i n caustic waterflooding than surfactant i n j e c t i o n due t o the complexity o f the a l k a l i . w.21 oil.nti. . 9' again dynamically driven 'L' )*on m u s . A N. Then the cores were 1 I . LABORATORY STUDIES Laboratory i n v e s t i g a t i o n s have been performed f o r t h i s a1 k a l i n e p i l o t p r o j e c t t o provide comparison core f l o o d t e s t s between waterflood and a l k a l i n e f l o o d recovery and d e f i n e the extent o f caustic consumption by the r e s e r v o i r rock.. 3 0 1' 1 I' core). .

2.226 1. improvement i n o i l recovery n was approximately 10 pore volume percent. Figure 2 shows a t y p i c a l response t o t h i s t y p e entrainment mechanism a1 k a l i n e waterflooding. (Sixty-two comparative core flood t e s t s were performed. I n j e c t water t o breakthrough o r a pre-determined w a t e r / o i l ratio. a l l o f the core f l o o d t e s t s were combined and analyzed t o o b t a i n a more s t a t i s t i c a l l y meaningful average core response t o a l k a l i n e flooding. ( D i f f e r e n t a l k a l i n e concentrations were used i n the various t e s t s performed ) . No strong c o r r e l a t i o n was found between improvement i n o i l recovery and t h e concentration of a l k a l i injected. 3).) O t h e average.s o f t water s o l u t i o n containing sodium c h l o r i d e a t a concentration o f 1%. These r e l a t i v e p e r m e a b i l i t y phenomena were used i n r e s e r v o i r simulation matches o f i n d i v i d u a l core FIGURE 3 . I n j e c t a pre-flush containing 1%sodium c h l o r i d e b r i n e i n softened water. and Follow w i t h an a l k a l i n e . 3.OIL/WATER RELATIVE PERMEABILITY . Therefore. These t e s t s were used t o o b t a i n r e l a t i v e p e r m e a b i l i t y t o o i l and water f o r both the waterflood and t h e a l k a l i n e f l o o d performance (Fig.

S.2' FIGURE 4 . It was evident from t h e t e s t s t h a t t h e consumption o f a l k a l i n e m a t e r i a l i s a long term phenomenon. I n the l a t t e r type t e s t s sand packs were prepared.a l k a l i solutions w i t h 1% sodium c h l o r i d e were i n j e c t e d i n t o t h e sand packs f o r periods ranging between 30 and 104 days. r e v e r s i b l e adsorption chemical consumption tests. The number o f pore volumes o f i n j e c t i o n required f o r concentration o f output s o l u t i o n t o reach t h e concentration o f t h e i n j e c t e d s o l u t i o n ranged upward t o 38 pore volumes. and long-term f l o w t e s t s . The t e s t s conducted included s t a t i c e q u i l i b r i u m tests. studies were undertaken i n an attempt t o d e f i n e the magnitude o f the caustic consumption which can be expected t o occur i n an a l k a l i n e f l o o d i n t h e Ranger Zone o f F a u l t Block VII. The consumption l a b o r a t o r y work i s described i n more d e t a i l i n t h e "Fourth Annual Report" o f t h i s p r o j e c t prepared f o r the U. S t a t i c periods o f varying length followed a f t e r which the a1 k a l i n e i n j e c t i o n was resumed. s o f t w a t e r . Reducing the f l o w r a t e o f t h e i n j e c t i o n increased the number o f pore volumes required t o reach an e f f l u e n t concentration n e a r l y equal t o t h e i n l e t concentration. Long Term A l k a l i n e Consumption I n Reservoir Sands The e f f e c t o f a l k a l i n e consumption i s a c r i t i c a l economic consideration. Department o f Energy.227 tests. and f o l l o w i n g waterflooding t o breakthrough. 4 shows t y p i c a l r e s u l t s f o r long term a l k a l i n e consumption r e s u l t s where t h e amount o f consumption. i n terms o f mass per u n i t volume i s p l o t t e d versus t h e concentration residence t i m e product. The o u t l e t c a u s t i c concentrations were measured d a i l y during the e n t i r e f l o w t e s t . As a r e s u l t . They then were scaled f o r the two-dimensional s l u g size o p t i m i z a t i o n cases and f o r a three-dimensional model o f the p i l o t area f o r the caustic f l o o d p r e d i c t i o n cases. The upper curve o f Fig.LONG TERM ALKALINE COMSUMPTION RELATIONSHIPS . s e n s i t i v i t y o f caustic consumption t o f l o w rate.

Produci. These six subzones l i e at depths of 2. The e n t i r e s t r u c t u r e i s eleven miles long and t h r e e miles wide underlying approximately 13.400 f e e t w i t h a net thickness o f 305 feet. and 237) l i e between the depths o f 2. The p i l o t p a t t e r n area i s i n the eastern p o r t i o n o f t h e Long Beach U n i t o f Wilmington F i e l d between the Juniper0 and Temple Avenue F a u l t s i n t h e Ranger zone and i s shown i n Fig. The upper f o u r zones containing low g r a v i t y . The r e s e r v o i r rock i n a l l zones i s sandstone w i t h d i f f e r e n t degrees o f consolidation and varied s i l t and c l a y content.000 and 7. medium-to-f i n e grained sands.600 t o 3.228 RESERVOIR DESCRIPTION The Wilmington f i e l d i s located i n the south-western p o r t i o n o f Los Angeles.ng zones i n t h e F i e l d (Tar. Productive subzones underlying t h e p i l o t area contain crudes w i t h a wide range o f .PATTERN AREA SCHEMATIC The modified l i n e d r i v e c o n f i g u r a t i o n of t h e p a t t e r n represents a t y p i c a l waterf lood we1 1 p a t t e r n f o r t h e Ranger Zone o f the Long Beach Unit. Lower Terminal.000 acres. FIGURE 5 .000 f e e t subsea and range i n age from l a t e Miocene t o Pliocene. Ranger. high v i s c o s i t y crude are t h e major o i l reservoirs. The p r o p e r t i e s o f each zone are sumnarized i n Table 1 . The f a u l t s d i v i d e the r e s e r v o i r i n t o pools and have proven t o be e f f e c t i v e b a r r i e r s t o f l u i d and pressure comnunication. The f i e l d i s an asymnetrical a n t i c l i n e w i t h a north-west south-east axis broken by a series o f transverse normal f a u l t s . Upper Terminal. Union P a c i f i c . Ranger i s t h e l a r g e s t and most p r o l i f i c of the Unit's r e s e r v o i r s . C a l i f o r n i a as shown i n Fig. 6). Cumulative o i l production t o date i s i n excess o f 1 b i l l i o n barrels. 5. It consists of several distinct intervals or subzones separated by impermeable shale sections (see Fig. Dips rang: fo rm a maximum o f 20" on the northern f l a n k t o approximately 60 on the southern f l a n k . Each subzone i s an i n t e g r a t e d sequence of shales and unconsolidated to semi-consolidated. 1. It i s the l a r g e s t f i e l d i n C a l i f o r n i a and one o f t h e major f i e l d s i n North America. poorly sorted. Ford.

) . From subzone t o subzone the o i l g r a v i t y depends upon geological age w i t h the lower ( o l d e r ) subzones containing higher g r a v i t y crude and t h e upper subzones containing t h e lower g r a v i t y crude. pressure gradients existed across the p a t t e r n w i t h t h e average r e s e r v o i r pressure being approximately 85% o f hydrostatic. t h e o i l g r a v i t y w i t h i n a subzone depends upon s t r u c t u r a l p o s i t i o n w i t h t h e higher subzone g r a v i t y a t the h i g h s t r u c t u r e p o s i t i o n s and low g r a v i t i e s a t the lower s t r u c t u r e p o s i t i o n s . The general c h a r a c t e r i s t i c s o f these p r o p e r t i e s are: an o i l g r a v i t y range o f 1 ' 2 ' 4-7 API. physical properties.229 TABLE 1 Ranger Subzone t0 .274 . This phenomenon i s due t o comnunication between the p i l o t and o l d e r producing areas i n t h e v i c i n i t y . 7.000 STB. (Performance o f the confined p a t t e r n i s shown i n F i g .000. Waterflooding operations began concurrently w i t h development.246 .265 . A t t h e time o f the development.TYPE LOG * Based on 1600 P S I Confining Pressure Core Analysis Data and Special Logging Programs.000.490.000 STB. The modified three producing row l i n e d r i v e f l o o d p a t t e r n was aided by peripheral a q u i f e r injection.289 . PRODUCTION HISTORY I n i t i a l development i n t h e p i l o t area began irr August 1967. t h e p i l o t area had produced 38.RESERVOIR CHARACTERISTICS BY ZONE Net Pay (feet) Net Volume (Acre Feet) Porosity* Perm* ( f r a c t i o n ) (md) 260 270 109 52 41 48 29 26 9341 4578 3647 4339 2655 2331 F . O i l production f o r t h e p i l o t p a t t e r n area as o f September 30. The i n i t i a l development o f t h e p a t t e r n was completed i n e a r l y 1975.000 STB o f water. 1980 was 11.270 321 179 173 220 131 H X G SIMULATION YOWL LAYERS RANQEI ZONE PILOT AREA - 64 FIGURE 6 . O f t h i s amount. Cumulative water i n j e c t i o n by the e i g h t surrounding i n j e c t i o n w e l l s was 55.

2 30 FIGURE 7 . The p e r m e a b i l i t y changes r e s u l t i n g from t h e d i v a l e n t i o n p r e c i p i t a t i o n were not considered i n the simulator work f o r t h e Range V I I p i l o t . The primary displacement e f f e c t s o f a caustic f l u i d are represented by changes i n r e l a t i v e p e r m e a b i l i t i e s to o i l and water. t h e simulator accounts f o r t h e i n j e c t i o n and production o f up to s i x d i f f e r e n t a c t i v e agents i n an aqueous phase. The match o f t h e l a b o r a t o r y response f o r core experiment No.42 pounds and a maximum long term consumption o f 0.84 pounds per cubic foot. 21 using these parameters i s shown i n Fig. 2. The s i m u l a t i o n work was based on t h e use o f t h e r e l a t i v e p e r m e a b i l i t y curves produced i n t h e l a b o r a t o r y experiments shown i n Fig. This s i m p l i f i e d approach permits t h e modeling o f enhanced recovery p r o j e c t s without the necessity o f determining the exact mechanisms o f t h e displacement i n minute d e t a i l . e t al3. 2. and t h e long term k i n e t i c a l l y c o n t r o l l e d i n t e r a c t i o n between caustic and t h e rock m a t r i x i t s e l f . For enhanced waterflooding. t h e instantaneous o r e q u i l i b r i u m adsorption o f c a u s t i c solution. . The model also accounts f o r the consumption o f a c t i v e m a t e r i a l w i t h i n a caustic s l u g by three d i f f e r e n t mechanisms: The i n t e r a c t i o n between a caustic s l u g and formation water t o form p r e c i p i t a t e s o f d i v a l e n t cations. Any o r a l l o f these agents may be caustic o r polymer type f l u i d s o r a combination o f these types o f f l u i d s .PATTERN AREA OF PERFORMANCE HISTORY RESERVOIR SIMULATION HODELIN6 The formulation o f t h e caustic simulation model has been reported i n an e a r l i e r paper by B r e i t . an instantaneous c a u s t i c consumption o f 0.

the incremental o i l recovery increases approximately 50% when no long term consumption is assumed to be present.000 RB 1.42 4.340 STB 99.4 weight percent in the largest pore volume slug. produced the greatest amount.800 STB 52. In contrast. However.879. Injection surveys have confirmed that over half the injection water i n these two injectors has entered the Fo subzone leaving it at a much higher average water saturation than lower zones. The low alkaline concentration of 0.231.*) Preflush Injection 1.0% S a l t in Softened Water Solution (390 Days) % PV Start of Alkaline Injection 5-24-50 - Performance of t h i s area of the f i e l d and simulator was characterized by rapid water breakthrough followed by a gradual r i s e i n WOR to i t s current average value of 50.000 STB 562. the higher concentration. because those areas . A can be seen by the results of the three 0. discussed are sumnarized in Table 3. smaller slug volume cases produced a m r e rapid oil r a t e response as can be seen in Fig. 60%. of incremental o i l . 5. Fig. this.874.91 84. The r e l a t i v e success of an alkaline flood will be m r e dependent on the o i l recovery at wells f a r removed from the injection rows than of the wells d i r e c t l y adjacent to the injectors.5 PV%/Yr.000 STB 45. the o i l r a t e reaches a maximum value approximately 15% higher in the absence of long term consumption. Injection was scaled from the planned f i e l d injection r a t e of 34.75 Reservoir Pore Volume o f Study Area I n i t i a l Oil in Place (TOIP) Waterflood Oil Recovery: TO 5-01-79 % TOIP e PV Waterflood O i l Recovery 5-01-79 To economic limit of 150 WOR (11-01-85) Cumulative Waterflood Oil Recovery % TOIP % PV Injection Rate 5-01-79 on to End (7. TABLE 2 - ALKALINE SLUG OPTIMIZATION STUDIES BASE MTA F R STUDY AREA O 1.806 B/D 1. the r e l a t i v e permeability adjustment was made linearly between a1 kaline and waterflood behavior depending on the active alkaline concentration in each c e l l . increase i n production tended to be at low rates. In addition. including two injectors was selected for s l u g optimization studies using the N-HANCE reservoir simulation model.231 Caustic Flood Optimization Study A small area o f the FQ subzone.54 34.65 29. 8. continuing on to l a t e i n the l i f e of the producers being modeled. subsequently.8 weight percent s alkali cases. T h i s figure also i l l u s t r a t e s the effect of long term caustic consumption on the projected results.000 STB 646. The reservoir and waterflood characteristics of this area are sumnarized i n Table 2.000 STB per day. Results of the optimization study runs In a l l b u t one case.

5 Variable* Rate flinimum ’Threshold A l k a l i n e R e l a t i v e Permeability Adjustment % Accelerated 226.8% f o r 1.12 7.- r r .99 3.0 89.0% A l k a l i 6 PV % Accel e r a t ed 162. % TOIP % PV - 84.09 3.1 Rate 0.8% A l k a l i Rate None 159.8 7-01-86 75.0 year and 0.51 Pro-Rated A l k a l i n e R e l a t i v e Permeability Adjustment % Accelerated 114.7 0.0% f o r next 1.4% o r t h o s i l i c a t e f o r f i r s t 1.5 1.47 4.5 6-05-86 58.1 12-06-85 37.0% A l k a l i Rate 7TFv-TNone 142.8 11-01-85 - 7.12 3.0 0. 0.30 11.83 3.75 2.- * 0. Economic Limit (150 WOR) Date Incremental O i l Recovery Above Water Flood M Bbl.0 2-16-88 77. FIGURE 8 - OIL PRODUCTION RATES SLUG SIZE OPTIMIZATION .99 1.31 4.2 Rate 0.8% A l k a l i TFVT-Constant 143.9 9-01-86 53.3 10-21-85 57.8% A l k a l i Accelerated 122.3 1.77 6. 1.2 0.5 years.4% A l k a l i Accelerated 137.0 year. 0.03 4.0 year.0 years.67 6.0 141.25 4.05 4.2% f o r 2.4% f o r 1.232 TABLE 3 Alkaline s1 ug Description Waterflood - ALKALINE SLUG OPTIMIZATION S N D I E S O I L RECOVERY M T A O i l Recovery Long Term Consumption MST Bbl.4% A l k a l i Rate 3 PV s .

0 weight percent and then tapered back t o a concentration o f 0. The amount o f c a u s t i c t h a t can be t r a n s m i t t e d through these closer areas without being consumed i s o f considerable i n t e r e s t . 9). (Fig. Currently. Increased transmission r a t e s could also be expected f o r constant concentration s l u g i n j e c t i o n s at accelerated i n j e c t i o n rates. F i r s t . the cumulative production and the caustic moving o u t s i d e t h e area were both disappointing i n comparison t o the constant concentration i n j e c t i o n cases. S i m i l a r l y . This case d i d show a considerably greater o i l production and a rm s i g n f i c a n t l y higher o i l production r a t e than t h e l i n e a r s h i f t i n fo normal w a t e r / o i l r e l a t i v e p e r m e a b i l i t y curves t o caustic r e a t i v e p e r m e a b i l i t y curves.dependent upon both t h e concentration o f the caustic i n an area and i t s residence time. This long term consumption i s . The lower the long term consumption t h e greater i s t h e transmission o f t h e c a u s t i c through t h e area mear the i n j e c t o r s and out i n t o the r e s t o f the r e s e r v o i r . another run was made i n which the change t o the enhanced recovery r e l a t i v e p e r m e a b i l i t y curves w i t h i n t h e model was made at a minimum threshold a1 k a l i concentration which corresponded t o t h e decrease i n i n t e r f a c i a l tension from the l a b o r a t o r y experiments.233 near the i n j e c t o r s have been more completely swept. FIGURE 9 - ACTIVE ALKALI TRANSMITTED THROUGH OPTIMIZATION STUDY AREA Two a d d i t i o n a l v a r i a t i o n s from the constant concentration cases already discussed were run. Total a l k a l i i n j e c t e d was the same as i n the p r i o r runs however.2 weight percent.4 weight percent c u r r e n t l y being i n j e c t e d i n the f i e l d t o a maximum o f 1. Although t h i s run d i d show some minor acceleration i n t h e o i l r a t e response. we are unable t o determine which o f 9 . a v a r i a b l e concentration case was r u n i n which t h e a l k a l i n e content was v a r i e d from t h e 0.

The mI B II FIGURE 10 - OIL PRODUCTION RATES. WATERFLaOD AND CAUSTIG FLOOD CASE I AND CASE I1 .4 weight percent constant concentration cases were selected f o r p r e d i c t i o n o f the performance o f the e n t i r e p i l o t area. o u t l i n e d i n an e a r l i e r section. and one layer. i n each o f t h e remaining subzones. and by minimal changes i n the t r u n c a t i o n o f r e l a t i v e p e r m e a b i l i t y curves between the p a t t e r n and t h e areas t o the north and south. two w i t h i n the Fo subzone. Owing t o the f a c t t h a t t h e greatest cumulative o i l production was achieved f o r the continuous i n j e c t i o n o f 0. One o f these used t h e prorated a l k a l i n e f l o o d r e l a t i v e p e r m e a b i l i t y adjustment. The r e s u l t s of these p r e d i c t i o n s are sumnarized i n Table 4 and F i g .p e r m e a b i l i t y s h i f t techniques more accurately represents t r u e r e s e r v o i r phenomena. Predicted Performance o f t h e P i l o t Area O i l recovery p r e d i c t i o n cases were run f o r continued waterflood operation and f o r t h e two caustic f l o o d cases.5. and the other t h e minimum threshold a l k a l i n e r e l a t i v e p e r m e a b i l i t y adjustment. 7 was matched by c o n t r o l l i n g t h e amount o f f l u i d which migrated o f f t h e p a t t e r n area t o the n o r t h and south. The performance shown i n Fig. Fig. The model included areas t o both the north and south o f t h e pattern. Case 11.4 weight percent a l k a l i and the equipment l i m i t a t i o n s on i n j e c t i o n concentration and i t s r a t e i n the f i e l d . Performance Match o f t h e P i l o t Area The h i s t o r i c a l performance o f the p a t t e r n area was matched using a black o i l r e s e r v o i r simulation model containing 1770 g r i d c e l l s i n seven layers.234 these two r e l a t i v e . The i n j e c t i o n i n t o each subzone was s p e c i f i e d on the basis o f surveys o f the i n j e c t i o n wells. 10. the 0. Case I.

Late i n the l i f e o f the f i e l d . Minimization of t h i s consumption appears p o s s i b l e by i n j e c t i n g t h e a l k a l i n e s o l u t i o n a t a higher concentration and/or i n j e c t i n g at a higher r a t e t o minimize t h e residence t i m e i n t h e r e s e r v o i r . As a consequence. the increase o r incremental production i n t h e Fo subzone i s overwhelmed o r masked by t h e production from lower zones. However. the simulation experience has indicated the d e s i r a b i l i t y o f c o n t r o l o f t h e placement o f i n j e c t i o n f l u i d by subzone ( v e r t i c a l l y ) f o r a more e f f i c i e n t a l k a l i n e i n j e c t i o n p r o j e c t .235 TABLE 4 - PATTERN AREA PREDICTED PERFORMME THROUGH OECEMBER. s i g n i f i c a n t parts o f the o i f production from t h e w e l l s w i t h i n the p a t t e r n i s occurring from zones lower than t h e Fo subzone. throughout the h i s t o r y o f t h e f i e l d . Furthermore. these were w e l l swept by the i n j e c t i o n water p r i o r t o caustic i n j e c t i o n leaving l e s s o i l t o respond t o i n j e c t i o n o f caustic. The l a b o r a t o r y core t e s t s have confirmed the a p p l i c a b i l i t y o f a l k a l i n e f l o o d i n g t o Ranger Zone crude and r e s e r v o i r rock. . mLusIoIIs The f i e l d performance t o date and t h e predicted performance from t h e s i m u l a t i o n studies i n d i c a t e t h a t t h e r e are many complicating f a c t o r s t o t h e successful a p p l i c a t i o n o f a l k a l i n e f l o o d i n g i n a heterogeneous r e s e r v o i r . A d d i t i o n a l l y . 1994 ~~ ~ Subzone Waterflood Cum O i l MSTB A l k a l i Case I Cum O i l MSTB A l k a l i Case I 1 Cum O i l MSTB F0 H X ( i 64 TOTAL 4390 2871 1742 3560 1787 699 lnmQ 4497 2990 1751 3580 1791 706 1 m 4758 3048 1762 3622 1795 715 l??OU predicted performance of the a l k a l i n e f l o o d i n comparison t o continued waterflood i n j e c t i o n i s disappointing. the high degree o f consumption indicated b y l a b o r a t o r y work and simulation r e s u l t s can be a c o n t r o l l i n g f a c t o r i n t h e success o f any caustic i n j e c t i o n p r o j e c t . The runs i n d i c a t e t h a t those w e l l s i n the f i r s t row o f producers away from the i n j e c t i o n w e l l s do respond t o c a u s t i c i n j e c t i o n . I n t h e p r e d i c t i o n cases h a l f t h e caustic water continued t o f l o w i n t o t h i s F subzone. However. the Fo subzone has taken over h a l f o f the i n j e c t e d water. Actual f i e l d r e s u l t s are needed t o c a l i b r a t e t h e consumption parameters t o f i e l d conditions.

(January. Presented a t the 1979 C a l i f o r n i a Regional Meeting of SPE o f AIME.D.E. . A p r i l 18-20. 1976) 85-94. Wilmington F i e l d . Johnson.H. C. Ranger Zone. Ventura.: "Status o f Caustic and Emulsion Methods". 1980. Long Beach Unit.. Mayer.S. Pet.2 36 REFERENCES 1. DOE by t h e City o f Long Beach. E-AC-03-76ET-12047. Annual Report f o r t h e Prepared f o r Period June 1979 May. V. 2. E. - 3.. B r e i t . Tech. Department o f O i l P r o p e r t i e s and THUMS Long Beach Company under Contract No. C a l i f o r n i a . and Carmichael. SPE Paper No. J- "Caustic Waterflooding Demonstration P r o j e c t . 1979. J. C a l i f o r n i a .: "An E a s i l y Applied Black O i l Model o f Caustic Waterflooding". 7999." Report SAN/12047-4.

Its potential was generally recognized by the petroleum industry well over 50 years ago and several papers were published in the 1920's describing early research in this area. the enriching components in the injected gas transfer to the crude oil and generate a solvent bank consisting of a modified crude oil. carbon dioxide. In this process. following the end of World War 11. however. were also developed during the 1950's. nitrogen. there was a dramatic increase in research directed toward improving our knowledge of what might be called the "physics and chemistry of fluid flow in porous media" and toward the development of the three basic areas of enhanced oil recovery--thermal. methane and mixtures of many of the a ove were tested. History The concept of miscible flooding is quite old. a self sustaining solvent bank is formed le ing a small volume of denuded crude as residual in the reservoir. Investigations into the use of miscible flooding techniques to improve oil recovery was a significant part of this increased effort. In the mriched-gas process. propane.J. . Laboratory tests were conducted to determine which fluids could be used for miscible flooding. chemical. essentially all of the oil (possibly excluding some asphaltenes) is flushed from the region contacted. butane. ketones. BLACKWELL Exxon Reduction Research Company Miscible flooding continues to be one of the most intriguing enhanced oil recovery methods because of its potential for recovering all of the oil flushed by solvent. two multiple-contact methods of a hieving miscible displacements4 5 the high pressure or vaporizing gas method and the enriched gas process. However. and miscible. interest in enhanced recovery techniques was low. but is able to generate a solvent bank within the porous medium during the displacement process. LPG. During the 1930's and early 1940's. Some of the first research was on completely misciblel~q (frequently called f irst-contact miscible) systems in which all mixtures of the solvent and oil form a single phase fluid. Almost every available fluid including alcohols. In the high-pressure gas process. Both of the latter involve injection of a fluid which is initially not miscible with the crude. If the phase behavior of tht! gas oil system is favorable. and one of the most exasperating. the injected gas is enriched with intermediate and higher molecular weight components vaporized from the first crude contacted.M I S C I B L E GAS D I S P L A C E M E N T 231 MISCIBLE DISPLACEMENT: ITS STATUS AND POTENTIAL FOR ENHANCED OIL RECOVERY R. It was an exciting era. because only in rare instances have actual field performances come anywhere close to the high recovery efficiencies potentially possible from this process.

However.s that each must meet in order to have a miscible displacement. In order to do this. The principal reason for the poorer than anticipated recovery efficiencies was severe channeling of the solvent banks. the primary objective was to design miscible floods using the smallest amount of LPG or enriched gas possible. Tests were conducted in Hele-Shaw models. short models. Many believed that the remaining problems (such as viscous fingering) would soon be solved and that miscible flooding would usher in a new era of high enhanced oil recoveries. Nevertheless. However as confirmed later in many field tests. In some instances. f . a number of field tests were initiated. measurements of the amount of mixing that occurs as fluids flow through porous medin were needed and several companies including Exxon initiated work in this area. hence the incentib for developing other methods of controlling solvent channeling was and remains quite high. viscosity levels and fluid densities. viscous fingering and gravity overriding were invariably major factors--either causing solvent channeling or aggravating the channeling associated with reservoir heterogeneity. some people hoped that viscous fingering would turn out to be a "laboratory artifact" in the sense that it would be less of a problem in the field than it was in laboratory floods. it was viscous fingering or solvent channeling that would likely dominate the behavior of a miscible flood. It rapidly became apparent that. attention was turned (about 1954) to determining the conditions required for effective use of each method in field applications. However. Unfortunately. OOIP were obtained in many of these tests. the dominant cause of this channeling was reservoir heterogeneities such as permeability variations in different strata. In order to establish the physical principles responsible for viscous fingering and to quantify its effects. Because of this optimism. a number of floods were run in each model using fluid systems with different oil/solvent viscosity ratios. incremental oil recoveries of only 5 to 10%. etc. At about the same time. Hill in 1952 and Dietz in 1953 discussed gravity stabilization of displacement fronts and established the critical rate concept for the control of viscous fingering by gravity segregaticn. solvent channeling was a serious problem in field applications and at least as detrimental there as it was in laboratory floods. narrow models and wide models. practical production rates can be achieved in only a llmif d number of reeervoirs without exceeding the critical rate. we and other research laboratories began our first experimental and theoretical studies of viscous fingering. and in models containing various permeability heterogeneities. At the beginning of this period. although the effects of mixing must be considered in any attempt to predict miscible flood performance. It is perhaps worthwhile to point out that viscous fingering and the closely rela ed gravity overridg phenomenum had been recognized sometime earlier. Initially. Viscous fingering was studied in long models.238 Following the discovery of these three basic approaches and the development of methods for determining the cond1tior. in homogeneous sand or glass bead packs. The earlier optimism turned suddenly into pessimism. fractures. These incremental recoveries not only fell far short of original expectations but were far from being economically attractive. there was a general air of optimism during the middle 50's.

a zonal injection program was initiated to provide an improved distribution of the gas and water into all zones and thereby i m p r m e the overall sweep efficiency. purifying. The early breakthrough of C02 in the largest miscible-C02 flood in the United States. the WAG ratio was increased from its initial value of about 0. research in mobility control methods dropped to a low level. including the use of polymers. and with the growing disenchantment with miscible flooding. N rertheless. By default. none appeared particularly attractive at that time. C02 injection was initiated in this project in January 1972. I regret to say however. First. and then second. One of the methods was gas-ucter injection which later became known as the WAG process. In November 1972. but one with obvious deficiencies.1XOOIP) to 88 million STB (or 6. it became necessary to curtail injection when C02 production exceeded the capacity of the esisting gas plants to extract the C02 from produced gas.239 Several papers and patents were published during the late 50's and early 60's describing methods that might increase thz reservoir v o l y e swept by the solvent bank. provides a dramatic illustration of the problem. Although detailed flood performance information is available from only a few of these projects. Numerical methods for simulating process performance have been developed along with better techniques for arriving at the critically important reservoir descriptions used for analysis of field results and in making predictions of the performance and economic viability of a miscible flood in a specific field.7XOOIP). Nevertheless. but high volumetric sweep efficiencies are possible only if solvent channeling can be controlled effectively. and reinjecting the produced C02 were significant factors in the 1977 decision to reduce the volume of C02 injected from 20% HCPV as planned originally to about 12% HCPV.000 acre Sacroc project in the Kelly-Snyder field of West Texas. C02 channeling and production cont iued to be an exasperating but manageable problem throughout the flood. Two important steps were taken. no dramatic breakthroughs have occurred to date and our ability to control solvent channeling has not changed much since 1960. particularly in the use of C02. the extra investment and operating costs involved in recovery. several laboratories renewed their research activity in this area and several new patents and papers describing the addition of various surfactants and other modifications to the WAG process have appeared recently. d2 . Several other methods. were also investigated. the WAG approach became the generally "accepted" method for mobility control. the 30. In the middle 1970's. Results from both field applications and laboratory studies have provided additional insight into the dynamics of miscible displacement processes. - Field Applications Miscible gas projects have provided the industry with valuable field data and operating experience during the past 15 years. Breakthrough occurred in June of the same year after injection of less than 2% HCPV of CO and increased rapidly. Another was thc use of foaming agents. the results generally lead one to the same conclusions reached in 1950's--that is. miscible flooding technology has advanced significantly during the past two decades.5:l to 3:1. Unfortunately. that in my opinion. A paper by Kane describes in detail efforts to maintain control of CO 2 production. The corresponding reduction in the estimated incremental oil recovery was from 107 million STB (8. high displacement efficiencies can be achieved in the regions flushed by solvent. particularly in the area of C02 miscible flooding.

In recent years these research efforts have emphasized work on CO flooding. the potential precipitation of a solid (asphaltene) phase. crude oi?. S. although there has been a limited amount of work on the use of ogher gases such as nitrogen. and the relationship between oil remobilization and rock wettability. brine and classes of surfactants at pressures above and below the minimum miscibility pressure and temperatures spanning the range from 20' to 100°C. liquid-liquid-vapor and liquid-vapor equilibrium mixtures were observed.240 In general. It also describes their chromatographic analysis procedure which permits the characterization of the hydrocarbons present in the various fluid combinations described above throughout the CI1 C36 range (as well as the usual Cl Cl0 range). In recent years. The ideal computer program for modelling miscible C02 displacements. must simulate the generation of the miscible rolvent bank. Similar comprehensive studies are needed using several C02. less frequently in the 10 to 15% range.. there is a need for additional carefully scaled experimental and theoretical studies of the interactions of phase . there are a number of industrial and university laboratories actively engaged in miscible displacement research. viscous fingering. Current Laboratory Research At the present time. for simulating miscible floods have occurred during the past 15 years. I feel additional research is merited. As mentioned earlier. Research efforts to develop better mobility control techniques were mentioned earlier. and will rarely exceed 15% OOIP. flue gas and C02 enriched with intermediate hydrocarbons such as propane. possible by water flooding) will most often fall in the range of only 6 to 30 percent of the original-oil-place (OOIP). further improvements are greatly needed. gravity segregation. The use of foams is again being investigated and some encouraging laboratory and field test results have recently been reported. is the solution to the problem of viscous fingering. behavior. the greatest strides in micellar-polymer technology have been made as the result of fundamental studies of the basic mechanisms involved. butane. Even though I remain skeptical that foam injection. The report describes their comprehensive study of the complex phase behavior of a particular C02crude oil system in some detail and carefully delineates the reservoir conditions under which liquid-liquid. Department of Energy report by Orr 5 et a1 is an excellent example of this type of study. Equilibrated samples taken from PVT cells are also being analyzed and compared with he slim tube results. as such. etc. rock lithology and heterogeneity. density and viscosity of individual equilibrium phases of fluids produced during laboratory floods in slim tubes. A recent U. for example. Several laboratories are engaged in fundamental studies of phase behavior and C02 flood performance often including measurements of the composition. or through reservoir cores. Nevertheless. several laboratories are investigating different methods for improving volumetric sweep. and predict the amount and compositions of the various phases present in every - . the performance of Sacroc and other field experience obtained to date suggests that incremental oil recoveries (over that. - - Similar studies are being carried out in other laboratories using other C02 crude oil systems. Mathematical Simulation Although worthwhile improvements in our numerical technique.

These simplified simulators normally use a limited number of gas and pseudo components to represent the injected gas. A recent paper by Todd includes comparisons of the advantages and disadvantages of the principal types of "miscible" simulators. Consequently. and the natural gas and crude oil (including the asphaltenes). Effective use of these specialized simulators requires that the user understand the limitations of the various simulators since interpretation of results often complicated by the simplying assumptions used. This type of application is not only a good way to test the capability of the computer program. The number and composition of the required pseudo-oil components can be determined by comparing reservoir model results obtained by use of the simplified computer program with those obtained using a fully compositional simulator. nor how precisely we are able to model these phenomena mathematically. Any grid block may contain a number of mobile and immobile immiscible hydrocarbon liquid phases (each containing over 30 components). fs RESERVOIR DESCRIPTION The need for a reliable description of reservoir geology and other re !rvoir engineering data can hardly be overemphasized.241 grid block each time step. A particularly important application is their use in conjunction with laboratory tests. none provide all of the features that one might desire. and carbonated water. No matter how well we m w o the chemistry and physics involved in a miscible displacement. and correctly model the dispersion or mixing of the miscible components transfer of components tetween immiscible phases. comparison of different gas injection programs. fully compositional models are frequently not as practical for field wide reservoir engineering studies as they are for special studies such as the simulation of the flood performance of a cross section or a small reservoir pattern area. are also frequently useful for specialized applications. Other types of simulators. Numerical dispersion remains a serious problem for the grid block sizes typically required in most reservoir studies. Compositional simulators with sophisticated phase behavior packages and other simulators with specialized capabilities have been developed to model miscible gas processes. such as modified black oil simulators. Current compositional simulators can model most of the physical and chemical phenomena involved in a miscible flood. but it is also a good way to test our understanding of the chemical and physical processes involved in a miscible gas flood. viscous fingering is difficult to model and is usually approximated empirically using a mixing parameter model. it is not possible to make useful reservoir performance predictions of miscible processes without having a reliable reservoir description. etc. It must be recognized that a much better reservoir description is required for predicting miscible . The simulator must be able to model block-toblock flow of both miscible and immiscible phases. Greatly simplified compositional simulators are frequently used for reservoir performance predictions. unfortunately. optimization of the WAG ratios. and large scale or field wide stud ies . and computing costs are normally high because of the overall complexity of the simulator. Typical applications of the resulting simplified model include sensitivity studies of flood performance for various geological models of the reservoir.

Depositional environment data and information on subsequent diagenetic changes of rock matrix can be particularly valuable in predicting continuity of permeable zones. EOR Potential For Miscible Processes United States: During the past decade. Estimates of the incremental recovery over that possible from waterflooding. Surprisingly small changes in the reservoir description can lead to significant differences in prediction of miscible flood performance and project economics.4 billion per year in 1984 and should reach a level of about $2. Our own estimates for the incremental reserves that can reasonably be added by the year 2000 fall into the 3 to 5 billion range. includes a breakdown of their projections of expenditures for enhanced oil recovery in the United States during the 10year period 1979-88.5 billion per year (of which $2. Despite this apparent conversatism.242 flood performance than is normally required for a comparable study of a water flood in the same field. timing for this increased production will be critically dependent on near term investments and development of C02 supplies. but its acquisition and careful interpretation is absolutely necessary. Many of the same types of data are needed for predicting miscible flood performance. and quality of routinely available reservoir description data have been dictated primarily by reservoir engineering needs for conventional primary and secondary recovery processes. this is not what one finds primarily because of the uncertainties in volumetric sweep caused by inadequate reservoir description data. F6S predicts that expenditures for miscible gas processes will grow at a rate of about 25% per year from a level of about $0. type. these estimates will turn out to be far too conservative. Acquisition of this additional reservoir description data can be both difficult and costly. Since the basic displacement mechanisms and phase behavior concepts used to predict miscible flood performance are well known.7 billion per year in 1980 to $1. The total expenditure allotted to miscible gas processes during the *2 x 109 barrels A recent study by Frost and Sullivan" . there have been numerous studies of the future potential of miscible gas processes and other enhanced recovery processes in the United States.1 billion is for injected gases) in 1988. in the estimates of the incremental oil recovery possible over waterflooding and in the economic assumptions used. the amount. I believe that the United States and possibly Canada will begin to see significant production from miscible gas processes during this decade. one might assume that the incremental oil recovery from miscible flooding could be easily estimated and that the incremental oil volumes predicted by the various studies would be similar in niagnitude with perhaps some differences in timing. these same changes in reservoir description may be unimportant when predicting the performance and economics of a waterflood. However. shale deposits or tight streaks. and the frequency and distribution of the openings (or windows) through these impermeable layers. whereas. range from an "almost assured" 2 billion barrels* to "possibly optimistic" estimates of over 30 billion barrels. In the past. Useful geological input includes the depositional environment of the reservoir. Hopefully. Miscible processes have the most potential of the various enhanced oil recovery processes for near-term production of light But oil and could begin to make its contribution felt by the mid-1980's.

but new miscible gas projects are also being considered for several other regions of the U. This would more than double current U. and political problems can be resolved. a reduction of only 15% in the assumed recovery efficiency reduced the estimated recovery to 0. For example. it was estimated that miscible gas processes could increase oil recovery by 1. Consequently. However. S.75 billion or 36% of the $38 billion projected for all EOR processes. The study utilized the tax and royalty regulations of the federal government and the province of Alberta which were in place or announced in lY78. 1. Production of 100 k BID is not anticipated outside the Permian Basin of West Texas..352 from hydrocarbon miscible and 0. Most will employ CO although some projects will use nitrogen or methane enriched 2 with LPG. acquisition of adequate supplies of COP at a reasonable cost may not be possible and miscible hydrocarbon gases may be used despite their high cost.243 10-year period was $13. This reduction of almost 75% indicates that a significant volume of marginally profitable (high risk) oil is included in the base case estimate. very few projects with marginal screening study economics remain as economically attractive prospects after more detailed studies have been completed. operating costs. production (currently about 70 k B/D) from all miscible gas projects. The use of nitrogen will usually be limited to deep high temperature reservoirs containing high gravity crudes because of miscibility pressure restrictions. These estimates include projections for oil field equipment and services as well as the cost of the injected fluids. One reason is that early recovery estimates almost invariably drop as reservoir geology becomes better defined. Plans are gearing completion for three new pipelines which will bring over one billion(l0 scf/day of C02 to the Permian basin in West Texas from formations in Colorado and New Mexico.885 billion barrels. S. in some areas of the U. If we assume that injection of 10 k scf of CO will provide approximately one barrel of enhanced oil recovery production. including Louisiana and the Mid-continent area. Canada: In a recent (March 1980) study12 of the potential of enhanced oil recovery in Canada. 2' 8F 5 Jay and Black Jack Creek Fields in Florida. .476 billion barrels. economic. (e.g. Past experience dictates that without significant increases in oil price. the study will need to be updated when current negotiations betveen the federal and provincial governments have been completed. until the late 1980's or early 1990's.885 billion barrels is extremely sensitive to small changes in the values assumed for recovery efficiency. tfen C02 from these pipelines would result in an oil production rate of 100 k B/D.533 from CO This base case estimate was made for an assumed oil price of $20 per barref although higher prices ($25 and $100) were used in sensitivity studies. Most of the near term activity will continue to be concentrated in West Texas. S. Thus the potential for miscible flooding in Canada remains highly uncertain but it appears likely that an incremental production of about 1 billion barrels of oil could be achieved if current technical. etc. For example. in offshore reservoirs) or in Canada. The study found that the base case estimate of 1. nitrogen will be used by Exxon in the 15400 foot. . Current plans call for completion of the first two pipelines in early 1983.

However. We have a much better understanding of the capability and limitations of the various methods and the role that EOR can realistically be expected to play in our efforts to meet the world's energy needs. Miscible gas processes should make a significant contribution to this objective . I anticipate that several EOR projects will be initiated in North Sea fields before the end of the decade. Plans for a miscible C02 project onshore in a depleted East Midland oil field and a miscible hydrocarbon project offshore have already been announced. consideration of potential applications of EOR processes frequently starts when a field is still in the early stages of its productive life. Both are hydrocarbon miscible. the two largest miscible gas projects are in Libya and Algeria. Thus it is not surprising that the evaluations of the potential of various EOR methods in the North Sea have already progressed past the screening stage and more detailed engineering studies are currently in progress for a number of fields. However. The Intisar D project in Libya was started in 1969 and has been producing at approximately 100 k BID. Undoubtedly. the volume of C02 available is limited. use hydrocarbon gases rather than C02. I feel that the thousands of man-years of research that the industry has devoted to the development of enhanced recovery technology are finally beginning to bear fruit. The Hassi Messaoud project in Algeria was started in 1964 and has been producing at approximately 60 k B/D. The Industry is once again becoming increasingly optimistic about EOR potential. There are several other small project in various parts of the world with a total production rate of perhaps 5001000 BID. . But there is a difference between the optimism of the 50's and that of the 80's. Although several reservoirs should be good miscible C02 flood candidates. it is premature to estimate probable incremental recoveries that can be attributed to future use of miscible gas processes. but I am confident that the solution to many of these problems can be found and that substantial volumes of EOR production will become economically feasible in the future.244 North Sea: In recent years. The ultimate potential for the use of EOR processes in the North Sea remains to be determined. Interest and activity in applying miscible gas and other EOR processes are expanding rapidly throughout the world. Hence. the first miscible gas projects in the North Sea may. I am optimistic about the future potential of EOR. Currently there is a shortage of trained scientists and engineers in the area. the number is increasing rapidly and the outlook for solving the remaining technical problems and designing economically attractive projects is promising. Until the C02 supply problem is resolved. other projects will follow but the bulk of EOR activity in the North Sea will probably not occur until the next decade. I recognize that significant and chal1engit:g problems remain to be solved. in fact. EOR technology of the 80's is more mature than it was then. Similarly I understand that consideration is being given to a surfactant flooding pilot offshore. Other Areas: Although North America may have the largest number of existing or planned miscible-gag projects.

Sci.: Proc. H. (1953) 56. U. S. G. AIME (1956) 207. J." SPE 7998. April 1979. B: Gas-Water Injection. AIME 1 8 (1950) 215.." 8 . 105. M. "Improving Miscible Displacement by ." Trans. DOEfETf12082-9. J. May 1981. Feb. Energy Research Institute ISBNO-0-920522-09-2. J. V. AIME. N. . 12. H. 6. 5. Acad..245 REFERENCES 1.. Recovery by High-Pressure Gas Injection. M. 11." SPE 7091. et al. Presented at Fifth Symposium on Improved Methods for Oil Recovery in Tulsa.: "Enhanced Oil Recovery Techniques: State of the Art and Potential" presented at Seminar on Improved Techniaues For the Extraction of Primary Forms of Energy sponsored by United Nations Economic Commission for Europe. A.. J." Drilling and Production Pract. B. R. 1980.. 1981. 246. A.: "The Effect of Gas Composition Upon Oil Recovery by Gas Drive. 4.. NO. L. Philip.: "A Laboratory Investigation of Displacement From Porous Media by A Liquified Petroleum Gas" Trans. p. Austria.: "Genie Chemique. 2. 13. Whorton. Orr. B. Amst.. . 3. and Kieschnick. Chierici. W. 1950. "#38 Billion Projected for Enhanced Recovery in '80s": Engineer International pages 98-100.: "Modeling Requirements for Numerical Simulation of CO Recovery Processes. Taber. Everett. and Crump. P. F. Stone. Scie. ApI. et al: "Liquid-Liquid Displacement in Porous Media as Affected by the Liquid-Liquid Viscosity Ratio and Liquid-Liquid Miscibility. S.. Calif. L. J. Okla. Vienna. "Displacement of Oil by Carbon Dioxide. H... 10. 281. (1958). 83 7. (1952) I. "Performance Review of a Large Seale C02-WAG Project Sacroc Unit-Kelly Snyder Field. Hill. 213. April 1978. 6. AIME 198 (1953) 33. Eng. P. 9 Henderson. S. Dietz. Petroleum Canadian Prince. 9. D. Kane. presented at the 1979 Regional Meetlng of the SPE (AIME) in Ventura. Nov. "Enhanced Oil Recovery Potential in Canada"." Trans. F : "A Preliminary Report on Oil . Caudle. J. Todd. March 1980.. and Dyes." Chem. 247." Trans. L.

This Page Intentionally Left Blank .

ROBERT E. The immiscible results are slightly more attractive. For this purpose we have considered both immiscible and miscible CO2 drive in a conceptual reservoir simulation with properties akin to the Forties field. The first part of the paper is concerned with problems of predicting phase behaviour using equations of state based on the Peng-Robinson and Generalised Redlich-Kwong formulations.MISCIBLE GAS DISPLACEMENT 247 THEORETICAL ASPECTS OF CALCULATING THE PERFORMANCE OF COZ AS AN EOR PROCESS IN NORTH SEA RESERVOIRS DAVID S. results are presented of 3-D compositional simulations for a simplified reservoir model based on the Forties Field. JOHN D. Although experimental PVT data are available for C02-hydrocarbon systems. and their predictions for miscibility are reviewed for North Sea reservoir conditions. Some of the problems of pseudo-components and interaction coefficients are discussed in this context. HUGES. DT2 8DH ABSTRACT This paper examines some aspects of calculating the performance of COf as a prospective enhanced oil recovery agent in North Sea reservoirs. Three-dimensional results are obtained which compare the performance of the miscible and immiscible displacement processes. Current reservoir conditions are found to give an immiscible Cop-displacement. The Peng-Robinson and Generalised Redlich-Kwong equations of state are compared for existing experimental data. First the problems of predicting the phase behaviour of CO2 with reservoir oils are examined. Some preference for the latter is given because of its superior prediction of fluid densities. The few component equilibrium factors in this model are adjusted to match the equation of state implications discussed above.Dorchester. The use of the Redlich-Kwone . these are at lower pressures than prevail in the North Sea. MATTHEWS. This occurs because: (i) local displacement efficiency is good as a result of pi1 swelling and transfer of hydrocarbons into the gas stream and (ii) volumetric sweep is good with component exchange between gas and oil reducing viscosity and density differences. INTRODUCTION This paper examines some of the problems of predicting oil displacement behaviour by CO2 in the context of typical North Sea field characteristics. Dorset. The reservoir pressure is then increased to achieve an MCM displacement. Good sweep efficiencies are obtained in the watered-out reservoir from the immiscible COq-displacement calculations. Second. MOTT AEE Winfrith. but modelling approximations in both cases may be giving a false impression of the real comparability. The paper falls into two areas.

45724 ai 2 R 2 2 Tci/Pci . The f a c t o r s i n f l u e n c i n g t h e r e l a t i v e sweep e f f i c i e n c i e s under immiscible and m i s c i b l e drive a r e discussed. .37464 + 1 .07780 R Tci/Pci . w h i c h then d i c t a t e s t h e choice of an e q u i l i b r i u m K-factor c o r r e l a t i o n f o r use i n a compositional r e s e r v o i r s i m u l a t i o n code. f o l l o w i n g t h e p r e s e n t c o n v e n t i o n a l w a t e r f l o o d . A n e a r optimum p r o c e s s i s then e v a l u a t e d i n a f u l l three-dimensional model. The p r e d i c t e d immiscible COz-drive i s found t o b e more a t t r a c t i v e than expected due t o i t s o i l s w e l l i n g and mass t r a n s f e r behaviour. I n t h e case of t h e F o r t i e s f i e l d t h e minimum m i s c i b i l i t y p r e s s u r e i s p r e d i c t e d t o be j u s t above t h e o p e r a t i n g p r e s s u r e . (3) (4) ai = I + m ( 1 . e s p e c i a l l y i n a p p l i c a t i o n s t o C02 systems. M i s c i b l e c a l c u l a t i o n s f o r t h i s same three-dimensional model have a l s o been undertaken u s i n g t h e Todd and Longstaff mixing approximation i n t h e s i m u l a t i o n code. = 0. 2 6 9 9 2 ~ (5) . PVT PROPERTIES OF C02 AND RESERVOIR OILS Choice of Equation of S t a t e When p r e d i c t i n g PVT p r o p e r t i e s of r e s e r v o i r f l u i d s from a thermodynamic e q u a t i o n of s t a t e . 2 m = 0.248 e q u a t i o n i n a few pseudo-component f o r m u l a t i o n i s i l l u s t r a t e d . The d i f f e r e n c e s i n recovery e f f i c i e n c y between t h e s e two types of displacement a r e b e l i e v e d t o b e w i t h i n t h e u n c e r t a i n t y of t h e methods used. a r e t h e Peng-Robinson (PR) e q u a t i o n (Ref 2) and t h e G e n e r a l i s e d Redlich-Kwong (GRK) e q u a t i o n i n a form f i r s t proposed by Zudkevitch and J o f f e (Ref 3 ) . The Peng-Robinson e q u a t i o n t a k e s t h e form p = -RT \rb - a(T) 9 v'(v+b) + b(v-b) a. . which i m p l i e s o p p o r t u n i t y t o c o n s i d e r b o t h immiscible o r m i s c i b l e C02-flooding of t h e r e s e r v o i r . bi = 0. a t y p i c a l approach i s t o use a two c o n s t a n t c u b i c e q u a t i o n based on t h e Redlich-Kwong e q u a t i o n (Ref I ) which g i v e s a s a t i s f a c t o r y compromise between s i m p l i c i t y and accuracy. The o t h e r symbols have t h e i r c o n v e n t i o n a l d e f i n i t i o n s which a r e given a t t h e end of t h e paper.Tr 11. The two e q u a t i o n s which a r e most commonly used. 5 4 2 2 6 ~ 0 . Various a l t e r n a t i v e COpJwater i n j e c t i o n s t r a t e g i e s a r e compared f o r immiscible displacement based on a v e r t i c a l two-dimensional scream-tube s e c t i o n . The second p a r t of t h e paper examines both immiscible and m i s c i b l e o i l displacement by C02 i n a conceptual 5-spot p a t t e r n w i t h p r o p e r t i e s a k i n t o t h c F o r t i e s f i e l d .

The a l t e r n a t i v e procedures s u g g e s t e d by Yarborough (Ref 4) and Coats (Ref 5 ) give broadly similar r e s u l t s . The use of t h e %parameters i n t h e GRK e q u a t i o n overcomes t h i s problem ( a t t h e c o s t of a l o s s of s i m p l i c i t y ) and t h i s e q u a t i o n g e n e r a l l y g i v e s good p r e d i c t i o n s of l i q u i d d e n s i t i e s i f t h e %parameters are chosen a p p r o p r i a t e l y . I n t e r a c t i o n c o e f f i c i e n t s f o r p a i r s of h y i r o c a r b o n components are g e n e r a l l y z e r o o r v e r y smill (except f o r methane-heavy hydrocarbon p a i r s ) b u t non-zero c o e f f i c i e n t s f o r hydrocarbon-C02 b i n a r i e s are e s s e n t i a l i f a c c u r a t e p r e d i c t i o n s are t o be o b t a i n e d . and t h e c h o i c e of i n t e r a c t i o n c o e f f i c i e n t s i s a major. A second i n t e r a c t i o n c o e f f i c i e n t f o r C02-hydrocarbon b i n a r i e s was i n t r o d u c e d by modifying t h e mixing r u l e f o r t h e parameter ' b ' i n e q u a t i o n ( 2 ) . For example. I n t h e PR e q u a t i o n t h e c r i t i c a l Z-factor i s n e c e s s a r i l y 0. The i n t r o d u c t i o n of t h e %parameters i s t h e most s i g n i f i c a n t d i f f e r e n c e between t h e two e q u a t i o n s .249 I n t h e GRK e q u a t i o n p = .26. The vapour p r e s s u r e and s a t u r a t e d l i q u i d d e n s i t y are normally d e r i v e d from c o r r e l a t i o n s i n terms of c r i t i c a l p r o p e r t i e s . R a i and S i a r e & temperature dependent f u n c t i o n s which a r e c a l c u l a t e d f o r each component by f i t t i n g t o t h e vapour p r e s s u r e and s a t u r a t e d l i q u i d d e n s i t y of t h e component a t t h e g i v e n temperature. Interaction Coefficients I n b o t h e q u a t i o n s t h e mixing r u l e f o r parameter ' a ' employs b i n a r y i n t e r a c t i o n c o e f f i c i e n t s 6 i which must be determined e m p i r i c a l l y . w h i l e b i n a r y d a t a s u g g e s t r a t h e r l a r g e r c o e f f i c i e n t s (eg.20 and 0. whereas f o r l i q u i d hydrocarbons t h e c r i t i c a l 2 . 0. b u t t h e p r e s s u r e s i n t h e s e b i n a r y systems are u s u a l l y much lower t h a n found i n r e s e r v o i r s .10 f o r a l l C02-hydrocarbon b i n a r i e s g i v e s r e a s o n a b l e r e s u l t s i n t h e PR e q u a t i o n f o r t e r n a r y and multi-component systems.RT v-b T'v a (v+b) . problem when a p p l y i n g a n e q u a t i o n of s t a t e t o C02/hydrocarbon m i x t u r e s . ) xi x j 3 ' . W have found t h a t an i n t e r a c t i o n c o e f f i c i e n t o f 0. - A more s y s t e m a t i c approach f o r C02-hydrocarbon m i x t u r e s h a s been proposed by Turek e t a1 (Ref 6 ) f o r t h e GRK e q u a t i o n . a l o n g w i t h a l i q u i d d e n s i t y a t a s i n g l e r e f e r e n c e temperature. Thus t h e PR e q u a t i o n n e a r l y always u n d e r p r e d i c t s t h e d e n s i t y of hydrocarbon l i q u i d s . For s u p e r c r i t i c a l t e m p e r a t u r e s Qa and are assumed t o t a k e t h e same v a l u e s as a t t h e c r i t i c a l temperature.13 f o r b u t a n e ) . a t 100°C t h e d e n s i t y of decane is u n d e r p r e d i c t e d by 6% and t h e d e n s i t y of pentadecane by 12%.f a c t o r i s known t o be between 0. ( 1 + Dij) (bi + b . and t h e mixing r u l e s ( 2 ) a r e used t o c a l c u l a t e a and b. The c o n v e n t i o n a l approach t o t h i s problem i s t o d e r i v e t h e i n t e r a c t i o n c o e f f i c i e n t s from d a t a on b i n a r y m i x t u r e s . . and t h e r e i s some evidence t h a t t h e r e s u l t i n g e v a l u e s are n o t o p t i m a l f o r multi-component systems a t h i g h e r p r e s s u r e s . t o r e a d b = 1Z . normal b o i l i n g p o i n t and a c e n t r i c f a c t o r .307 f o r a l l components.

Comparison of P r e d i c t i o n s f o r C O ~ / S y n t h e t i cO i l Systems To i l l u s t r a t e some of t h e s e p o i n t s . I n t e r a c t i o n c o e f f i c i e n t s between C 1 . t h e PR and GRK e q u a t i o n s have been used 0 t o p r e d i c t s a t u r a t i o n p r e s s u r e s and d e r i s i t i e s of m i x t u r e s of Cog w i t h t h e I component s y n t h e t i c o i l whose composition i s g i v e n i n Table 1 . where a v a l u e of 0.06 4. C02-interaction c o e f f i c i e n t s and &parameters were t a k e n from Reference 6 .95 cIo ‘14 4.01 was used i n o r d e r t o match t h e observed bubble p o i n t s of t h e o r i g i n a l o i l . Experimental d a t a on t h i s system a t 48. i n e r r o r by 21: a t 48.6OC i s g i v e n by Turek e t a l .C I O and C I .21 5. C 0 2 4 e n s i t i e s p r e d i c t e d by t h e PR e q u a t i o n a r e a c c u r a t e t o w i t h i n 5%.2 50 This reduces t o e q u a t i o n (2) when D i j = 0. some compensating adjustment of t h e R’s i s needed. A consequence of t h e changes t o t h e C02-parameters from f i t t i n g t o b i n a r y d a t a i s t h e o v e r p r e d i c t i o n o f C02 d e n s i t i e s a t h i g h p r e s s u r e s .10 f o r a l l C02-hydrocarbon p a i r s . 13 3. The c a l c u l a t i o n s were c a r r i e d o u t u s i n g t h e VOLE phase e q u i l i b r i u m code (Ref 7) I n t h e c a l c u l a t i o n s w i t h t h e GRK e q u a t i o n t h e developed a t AEE W i n f r i t h .97 30. a t 100°C t h e d e n s i t y o f C02 i s o v e r p r e d i c t e d by 10% a t 200 b a r s and by 19% a t 300 b a r s . f o r example. e x c e p t t h a t t h e C I .if t h i s form of t h e GRK e q u a t i o n i s t o be used t o c a l c u l a t e d e n s i t i e s i n a compositional s i m u l a t o r . t h e p r e d i c t e d bubble p o i n t was t h e n of t h e o r i g i n a l o i l a t 65. even a t high p r e s s u r e s .95 4. and a l s o Ra and f o r s u p e r c r i t i c a l C02 were assumed t o be q u a d r a t i c f u n c t i o n s of temperature.06 . These i n t e r a c t i o n c o e f f i c i e n t s were assumed t o be r e s p e c t i v e l y q u a d r a t i c and c u b i c f u n c t i o n s of hydrocarbon a c e n t r i c f a c t o r . The PR c a l c u l a t i o n s used an i n t e r a c t i o n c o e f f i c i e n t of 0. and i n t e r a c t i o n c o e f f i c i e n t s f o r a l l hydrocarbon p a i r s were s e t t o z e r o e x c e p t f o r C I .9OC. C2 and C3 w i t h C6’ hydrocarbons were taken from Katz and F i r o z a b a d i (Ref 8). The polynomial parameters were t h e n determined by a r e g r e s s i o n a n a l y s i s u s i n g phase e q u i l i b r i u m d a t a on f i f t e e n C02-hydrocarbon b i n a r i e s . Thus.C ~ ~ c o e f f i c i e n t s had t o b e i n c r e a s e d by 0.67 Component ‘ 6 c7 c8 Mole p e r c e n t 3. TABLE I COMPOSITION OF SYNTHETIC O I L (Ref 6 ) Component Mole p e r c e n t 34.01 t o f i t t h e observed bubble p o i n t However.96 5.6W.C I O and Cl-Ci4. no d e n s i t y d a t a were used i n t h e r e g r e s s i o n a n a l y s i s .9OC and 65.04 c2 c3 c4 c5 3. These developments t o t h e GRK e q u a t i o n have emphasised a c c u r a t e p r e d i c t i o n s of phase behaviour r a t h e r than d e n s i t i e s .

The GRK equation is a c c u r a t e a t low C02 concentrations.0 0-2 0 .251 Predicted s a t u r a t i o n p r e s s u r e s a r e shown i n Figure 1 .4 MOLE FRACTION CO2 FIG.C 0 BUOBLE POINT 0 0 CALCULATED 0 ! i n L PR EOUAlION GRK EOUAllON -. This implies t h a t some f i t t i n g of parameters t o experimental d a t a i s needed i f a c c u r a t e p r e d i c t i o n s of d e n s i t y and phase behaviour a r e t o be obtained simultaneously. MEASURED DEW POINT 414'C 6M.6 0.SYNlHElIC OIL SYSIEM.0 0. who found t h a t while l i q u i d d e n s i t i e s a r e underpredicted by the PR equation. s i n c e the d e n s i t y of pure C02 i s overestimated. and by 3 o r 4% a t 65. and o t h e r c a l c u l a t i o n s on s y n t h e t i c o i l mixtures.9OC.6 MOLE FRACTION Cq 06 FIG. i t appears t h a t n e i t h e r equation i s capable of accurate p r e d i c t i o n s of oilIC02 mixtures across the e n t i r e composition range.6OC. From these r e s u l t s . b u t a t high C02-concentration the d e n s i t y i s overpredicted by lo%.4 0. DENBlllES OF SATURATED FLUID IN .HE SYNTHEIIC OIL SYSTEM I&8.9'C 1 COz - Density p r e d i c t i o n s a t 48. 1. This i s c o n s i s t e n t with the work of Sigmund e t a1 (Ref 9 ) .I 0. The PR p r e d i c t i o n s a r e r a t h e r less a c c u r a t e . SAIURAIION PRESSURES IN THE COz. t h e f r a c t i o n a l changes i n volume due t o the a d d i t i o n of COq t o o i l a r e represented q u i t e w e l l . The PR equation underpredicts d e n s i t i e s by 5 o r 6% a t 48. although some improvement could probably be made by a d j u s t i n g i n d i v i d u a l i n t e r a c t i o n c o e f f i c i e n t s .9OC a r e shown i n Figure 2.- 0 * 1 1 ' I I I 1. The GRK r e s u l t s agree q u i t e w e l l with the measurements except t h a t the p r e d i c t e d c r i t i c a l p o i n t s occur a t a r a t h e r higher C02 concentration. 700r MEASURED CALCULATED CREOUAIION CRK EOUATION 0 --- b \ t 0 I I I I 0. . 2.6 I 1. e s p e c i a l l y a t the lower temperature.

I n any case i t may prove s i m p l e r t o make a few PVT measurements on COZ/oil m i x t u r e s . depending on t h e c o r r e l a t i o n used t o c a l c u l a t e pseudo component p r o p e r t i e s . A s s e s s i n g t h e v a r i o u s methods f o r c a l c u l a t i n g PVT p r o p e r t i e s f o r C 0 2 / r e s e r v o i r o i l systems i s d i f f i c u l t because o f t h e l a c k o f e x p e r i m e n t a l d a t a o n o i l s f o r which a comprehensive a n a l y s i s o f t h e h e a v i e r f r a c t i o n s i s a l s o a v a i l a b l e . This r o u t e h a s been followed i n -the p r e s e n t work f o r t h e GRK e q u a t i o n . A comnon approach i s t o d i v i d e t h e heavy f r a c t i o n i n t o groups. A f u r t h e r problem i s t h e s e l e c t i o n of i n t e r a c t i o n c o e f f i c i e n t s f o r t h e s e pseudo components. Various c o r r e l a t i o n s have been proposed f o r d e t e r m i n i n g t h e s e p r o p e r t i e s . and i s p a r t i c u l a r l y s i g n i f i c a n t i n C 0 2 / 0 i l systems. w h i l e f o r t h e PR e q u a t i o n a v a l u e o f 0. I n b o t h c a s e s t h e i n t e r a c t i o n c o e f f i c i e n t between methane and t h e h e a v i e s t f r a c t i o n s (c](j+) w a s a d j u s t e d t o match t h e observed bubble p o i n t of t h e o r i g i n a l o i l w i t h o u t C02. average b o i l i n g p o i n t and molecular weight a r e normally determined f o r each ' c u t ' . t h e s e c o e f f i c i e n t s v a r i e d c o n s i d e r a b l y between d i f f e r e n t o i l s and d i d n o t f o l l o w any obvious s y s t e m a t i c t r e n d .05) f o r t h e GRK e q u a t i o n . b u t when a p p l y i n g t h e PR e q u a t i o n t o North Sea o i l s . This i s p a r t i c u l a r l y convenient i f a d i s t i l l a t i o n a n a l y s i s h a s been c a r r i e d o u t on t h e o i l . t h a n t o c a r r y o u t the d e t a i l e d compositional a n a l y s i s required f o r i n p u t t o a p r e d i c t i v e e q u a t i o n o f s t a t e model. The s e l e c t i o n of parameters f o r t h e s e pseudo components i s a s e v e r e problem when a p p l y i n g an e q u a t i o n of s t a t e t o r e s e r v o i r o i l s .1 was used f o r a l l C02-hydrocarbon p a i r s . The s y s t e m a t i c approach t o C02-hydrocarbon i n t e r a c t i o n c o e f f i c i e n t s d e s c r i b e d e a r l i e r h a s t h e advantage t h a t i t i s p o s s i b l e t o use t h e f u n c t i o n a l dependence on a c e n t r i c f a c t o r t o e x t r a p o l a t e t h e pseudc components. S p e c i f i c g r a v i t y . each pseudo component r e p r e s e n t i n g a group of components h a v i n g similar p r o p e r t i e s . and i t i s n e c e s s a r y t o d i v i d e t h e heavy f r a c t i o n s (normally Cg and above) i n t o pseudo components. i t h a s always been found n e c e s s a r y t o use l a r g e methaneheavy hydrocarbon i n t e r a c t i o n c o e f f i c i e n t s (up t o 0. b u t t h e e q u a t i o n of s t a t e model r e q u i r e s 'pseudo' c r i t i c a l p r o p e r t i e s as i n p u t . each of which h a s i n d i v i d u a l components whoge b o i l i n g p o i n t l i e s w i t h i n a c e r t a i n range.4) t o f i t t h e observed bubble p o i n t s . w h i l e t h e PR e q u a t i o n u n d e r e s t i m a t e s o i l d e n s i t y by between 10 and Z O X .252 Pseudo Component R e p r e s e n t a t i o n of R e s e r v o i r O i l s The a p p l i c a t i o n of an e q u a t i o n of s t a t e t o s y n t h e t i c o i l s is r e l a t i v e l y s t r a i g h t f o r w a r d as i n d i v i d u a l components can be i d e n t i f i e d and r e p r e s e n t e d as such i n t h e t h e o r e t i c a l model. w i t h o u t f i t t i n g t o e x p e r i m e n t a l d a t a ) of t h e s e p r o p e r t i e s . as one pseudo component can be a s s i g n e d t o each ' c u t ' i n t h e d i s t i l l a t i o n . This c o e f f i c i e n t w a s always small (around 0. F u r t h e r work is needed t o develop t h e e q u a t i o n of s t a t e method t o t h e s t a g e where i t can g i v e r e l i a b l e a p r i o r i p r e d i c t i o n s ( i e . f o r example t h o s e o f C a v e t t (Ref 10) and Whitson (Ref 1 1 ) which are most convenient as they use s p e c i f i c g r a v i t y and average b o i l i n g p o i n t as c o r r e l a t i n g parameters. However. t o o b t a i n a c c u r a t e d e n s i t i e s w i t h t h e CRK e q u a t i o n i t was n e c e s s a r y t o use t h e measured d e n s i t y of each ' c u t ' when c a l c u l a t i n g t h e 0-parameters. However. t h e heavy f r a c t i o n s of r e s e r v o i r f l u i d s c o n t a i n so many d i f f e r e n t isomers t h a t i t i s i m p o s s i b l e t o i d e n t i f y them i n d i v i d u a l l y . where t h e heavy f r a c t i o n s have a s t r o n g i n f l u e n c e on t h e phase behaviour a t h i g h p r e s s u r e . . C a l c u l a t i o n s on North Sea o i l s have shown t h a t t h e GRK e q u a t i o n p r e d i c t s t h e d e n s i t y of o i l w i t h o u t C02 t o w i t h i n a few p e r c e n t .

The choice of c o r r e l a t i o n f o r pseudo component p r o p e r t i e s appears t o be a t l e a s t as important a s the choice of equation of state. The two GRK c a l c u l a t i o n s p r e d i c t s i m i l a r s a t u r a t i o n p r e s s u r e s . G4K + WHITSON GRK i CAVETT PR + WHITSON -*-• PR + C A V E T T * * .8 MOLE F R A C T I O N C 0 2 FIG. However. Experilqental d a t a a r e needed t o resolve the position.--- 01 0 I I 1 1 0.72 f o r t h e PR c a l c u l a t i o n s .253 P r e d i c t i o n of PVT P r o p e r t i e s of CO?/Forties O i l Mixtures A t the p r e s e n t time. and 1.6 0. CALCULATED SATURATION PRESSURES OF FORTIES OIL C02 MIXTURES - . .58 and 1. I t i s t o be expected t h a t the GRK p r e d i c t i o n s a r e less s e n s i t i v e t o the pseudo component c o r r e l a t i o n because of the subsequent matching t o the measured density using the *parameters. and t o derive d a t a f o r a model with a small number of pseudo components with approximately the same e s s e n t i a l p r o p e r t i e s . I n the absence of such d a t a . Figures 3 and 4 show c a l c u l a t e d s a t u r a t i o n pressures and swelling f a c t o r s f o r COq/ F o r t i e s o i l . with values of 1. .50 and 1. The various methods give similar p r e d i c t i o n s f o r the o i l swelling f a c t o r s as a f u n c t i o n of Cop concentration. while the two PR c a l c u l a t i o n s give s i g n i f i c a n t l y d i f f e r e n t s a t u r a t i o n p r e s s u r e s . the swelling f a c t o r s f o r s a t u r a t e d o i l a t a t y p i c a l p r e s s u r e of 200 b a r s vary considerably..64 from the GRK equation.. These c a l c u l a t i o n s have shown some s i g n i f i c a n t d i f f e r e n c e s between d i f f e r e n t methods f o r p r e d i c t i n g t h e PVT p r o p e r t i e s of r e s e r v o i r oil/CO2 mixtures. using the a l t e r n a t i v e PR and GRK equations i n an 18-component model i n the VOLE code (swelling f a c t o r i s defined a s the volume of o i l p l u s CO2 a t s a t u r a t i o n pressure r e l a t i v e t o volume of o r i g i n a l o i l a t i t s s a t u r a t i o n p r e s s u r e ) . 3. no experimental phase behaviour d a t a a r e a v a i l a b l e f o r C02 and North Sea o i l s .2 04 0. depending on which c o r r e l a t i o n i s used f o r t h e pseudo component p r o p e r t i e s . Both the Whitson and Cavett c o r r e l a t i o n s were used f o r e s t i m a t i n g the p r o p e r t i e s of t h e twelve pseudo components t o r e p r e s e n t the C6+ f r a c t i o n . d e t a i l e d equation of s t a t e c a l c u l a t i o n s have been performed t o generate PVT information f o r COZ/Forties o i l mixtures.

the few component equation of s t a t e model becomes a s o p h i s t i c a t e d c o r r e l a t i o n with a wider range of v a l i d i t y than the K-value approach. CALCULATED SWELLING OF SATURATED FORTIES OIL CO2 MIXTURES The equations of s t a t e were used t o e s t i m a t e minimum m i s c i b i l i t y p r e s s u r e (MMP). i n a l l cases i t should be p o s s i b l e t o o b t a i n an accurate r e p r e s e n t a t i o n by tuning a p p r o p r i a t e parameters. but has f a l l e n t o around 180 b a r s . I n essence. using a s i m p l i f i e d model of the multi-step process leading t o m u l t i p l e c o n t a c t miscibility. However. The MMP f o r F o r t i e s w a s p r e d i c t e d by both PR and GRK models t o be between 210 and 220 b a r s . and i n t e r a c t i o n c o e f f i c i e n t s a r e a d j u s t e d u n t i l a good match i s obtained. i t i s e s s e n t i a l t o reduce the number of components so t h a t computing c o s t s a r e acceptable. when using a compositional s i m u l a t o r . I t i s u n l i k e l y t h a t such a coarse r e p r e s e n t a t i o n w i l l give accurate results unless c e r t a i n parameters have been a d j u s t e d t o f i t d a t a from experiments. t h e r e i s l i t t l e t o choose between d i f f e r e n t equations of s t a t e . . Thus t h e c a l c u l a t i o n s suggest t h a t C02-injection i n F o r t i e s would give r i s e t o an i k s c i b l e displacement.* * ---- 1 0. Few Component Representation of F o r t i e s O i l To r e a l i s t i c a l l y r e p r e s e n t the PVT behaviour of r e s e r v o i r o i l s some 15 t o 40 d i f f e r e n t components ( o r pseudo components) a r e needed.4 0. When mathematically s i m u l a t i n g a COP flood. o r from more d e t a i l e d c a l c u l a t i o n s . When used i n a few component model w i t h i n a compositional s i m u l a t o r . In the numerical procedure adopted i n t h e VOLE code a t Winfrith.8 a P u U 4 1. but t h e pressure i s only j u s t below the MMP.8 MOLE FRACTION COP FIG.2 0. the methane and C02 a r e normally kept a s s i n g l e components. This can be done by non-linear r e g r e s s i o n a n a l y s i s using an equation of s t a t e phase e q u i l i b r i u m code i n which t h e Q-parameters ( o r c r i t i c a l p r o p e r t i e s i n the PR equation).254 1. and between two and f o u r pseudo components a r e used t o r e p r e s e n t the C T . 4. compared with a f i g u r e of 240 b a r s from the l a t e s t c o r r e l a t i o n of Holm and Josendal (Ref 12).6 0.6 WHITSON CAVETT (RESULTS ARE SIMILAR FOR EACH EQUATION OF STATE) CRS CORRELATION * * . In the F o r t i e s f i e l d the r e s e r v o i r p r e s s u r e w a s i n t i a l l y 220 b a r s . the f l u i d composition a t each s t a g e i s determined by mixing o r i g i n a l o i l with the gas phase from the previous s t a t e i n a multi-step process.f r a c t i o n of t h e o i l .

c6-C1Os Cll-C19 and C20+. t h e 6-component and 18-component p r e d i c t i o n s of d e n s i t y . 5. so long as t h e parameters have been a d j u s t e d t o match d a t a from experiments o r more d e t a i l e d c a l c u l a t i o n s . One-dimensional calculat i o n s using an equation of s t a t e compositional simulator which has r e c e n t l y been developed have demonstrated t h a t these phenomena could be a c c u r a t e l y represented i n a few component model. The o i l composition used was s i m i l a r t o t h a t found i n the F o r t i e s r e s e r v o i r . The f i r s t used an 18-component r e p r e s e n t a t i o n of the r e s e r v o i r o i l .255 I n an i d s c i b l e C 0 2 f l o o d two main recovery mechanisms operate. C 1 . with o i l r e c o v e r i e s d i f f e r i n g by a t most 0. The assumed w e l l s e p a r a t i o n between an i n j e c t o r and a producer was a t y p i c a l value of 700 metres. and the s i x components s e l e c t e d were COP. A repeated 5-spot p a t t e r n has been assumed i n a 100 m t h i c k sandstone w i t h uniform p o r o s i t y of 27% and permeaD b i l i t i e s of 400 m i n t h e h o r i z o n t a l d i r e c t i o n and 40 mD i n the v e r t i c a l d i r e c t i o n . A f t e r some minor adjustments of the methane-heavy hydrocarbon i n t e r a c t i o n c o e f f i c i e n t s . These r e s u l t s suggest t h a t a s i x component model can give an adequate r e p r e s e n t a t i o n of t h e PVT p r o p e r t i e s of the f l u i d s .42 cp.0 1. They a r e the swelling of t h e o i l through s o l u t i o n of C02 and evaporation of c e r t a i n components from the o i l i n t o t h e produced gas stream. The r e c o v e r i e s of i n d i v i d u a l components a r e a l s o i n c l o s e agreement. PREDICTED PERFORMANCE OF ONE DIMENSIONAL C02 DISPLACEMENT OF FORTIES OIL AT 210 BAR THE DISPLACEMENT O OIL BY C02 I N A CONCEPTUAL RESERVOIR F C a l c u l a t i o n s of o i l displacement by C02 have been performed using a s i m p l i f i e d t h r e e dimensional conceptual r e s e r v o i r with p r o p e r t i e s broadly similar t o those found i n t h e F o r t i e s f i e l d (Ref 13).5%. The connate water s a t u r a t i o n was 23% and t h e r e s i d u a l o i l from waterflooding was 30%. There i s c l o s e agreement between t h e two cases. f o r t h e two d i f f e r e n t r e p r e s e n t a t i o n s . . C2-C5. and the second a 6-component r e p r e s e n t a t i o n i n which the parameters were obtained by an averaging procedure chosen t o give the same values of the equation of s t a t e ' a ' and 'b' c o e f f i c i e n t s .6 1. - 18 COMPONENT MOOEL loor 6 COMPONENT MOOEL u 4 0 z 2 O 90 LOO0 d > a 80 10 5 B 0 0 8 60 2000 a ~ > w ul 8 & 50 1000 0. Two c a l c u l a t i o n s were performed f o r a C02 f l o o d i n a one-dimensional geometry. Figure 5 shows the o i l recovery and GOR a s a f u n c t i o n of the amount of Cog i n j e c t e d . Water v i s c o s i t y was 0. v i s c o s i t y and bubble p o i n t s agreed everywhere t o w i t h i n I o r 2%.2 C O 2 INJECTEO ( I N I T I A L HVOROCARBON PORE VOLUME) FIG.

The broad adequacy of t h i s b a s i c mesh arrangement f o r the c u r r e n t comparative e x e r c i s e was examined by performing mesh refinement c a l c u l a t i o n s i n two dimensions. The c a l c u l a t i o n s of waterflood behaviour were performed using t h e compositional simulator described l a t e r . Calculations of waterflood behaviour were performed assuming the 5-spot p a t t e r n t o be produced a t a constant r a t e of 5000 m3/day (31000 BPD) with an equal i n j e c t i o n of water i n t o each i n j e c t o r . Some underriding of o i l by water had occurred so t h a t the top two l a y e r s contained about h a l f of the unswept o i l . 7 and 13. The s i x component r e p r e s e n t a t i o n of F o r t i e s o i l described e a r l i e r w a s used i n t h e study. Modelling Assumptions Adopted i n Immiscible and Miscible Displacement Calculations Displacement of o i l by Cop i n the model r e s e r v o i r has been s t u d i e d under i d s c i b l e and miscible conditions by performing c a l c u l a t i o n s f o r p r e s s u r e s j u s t below and j u s t above the minimum m i s c i b i l i t y pressure. To reduce mesh o r i e n t a t i o n e f f e c t s . o i l and s o l v e n t . The method r e l i e s on t h e assumption of an e f f e c t i v e o i l v i s c o s i t y and d e n s i t y . which has t o be defined by . and t o match measured o i l d e n s i t i e s and v i s c o s i t i e s .3m) being smaller than t h a t of the lower l a y e r s (20 m) t o allow g r a v i t y o v e r r i d e t o be followed w h i l s t economising on the number of meshes. R e l a t i v e p e r m e a b i l i t i e s were derived assuming t h a t t h e sandstone was predominantly water wet. s i n c e t h i s may be a t y p i c a l l i m i t i n g s i t u a t i o n f o r off-shore water handling. a c u r v i l i n e a r a r e a l mesh p a t t e r n was adopted. b u t t h e c a l c u l a t e d d i s t r i b u t i o n of water s a t u r a t i o n s a r i s i n g from the model i s expected t o be t y p i c a l of a p o s s i b l e condition f o r a CO2 displacement process. with a recovery of 51% of the o i l i n place. and an e f f e c t i v e gas v i s c o s i t y and d e n s i t y . CRS. based upon u n i t mobility p o t e n t i a l flow. The miscible c a l c u l a t i o n s were c a r r i e d o u t using the Todd and Longstaff mixing model (Ref 17). This condition when the water c u t reached 50% has been used a s the s t a r t i n g p o i n t f o r the CO2 displacement c a l c u l a t i o n s .f o l d symmetry t o be invoked and a three-dimensional r e p r e s e n t a t i o n of a symnetry s e c t o r was used i n the c a l c u l a t i o n s . w. i t was not thought necessary t o include the marginal e f f e c t s a s s o c i a t e d with the use of c a p i l l a r y pressure curve. This a r e a l mesh consisted of 4 stream tubes with 12 subdivisions along each tube the subdivisions being s e l e c t e d t o give equal a r e a s i n the s h o r t e s t stream tube (see Figure 7). A t t h i s condition. In the v e r t i c a l c r o s s s e c t i o n . With the r a t h e r coarse l e v e l of mesh d e f i n i t i o n described above. Breakthrough of water occurs a t t h e producer a f t e r 1800 days. with an equation of s t a t e f o r d e n s i t i e s . the thickness of the uppermost l a y e r s ( 6 . and the Corey approximation (Ref 1 4 ) was used t o o b t a i n water r e l a t i v e p e r m e a b i l i t i e s . system which assumes d i r e c t m i s c i b i l i t y of t h e phases. This i s a model f o r a two component. t h i s reduces the t i m e s t e p penalty which occurs i n e x p l i c i t c a l c u l a t i o n s of s a t u r a t i o n s . t h e parameters i n the K-value c o r r e l a t i o n being adjusted t o match t h e s a t u r a t i o n pressure and swelling f a c t o r s of oil/CO2 mixtures p r e d i c t e d by the d e t a i l e d equation of s t a t e model ( s e e Figure 4). (Ref 16) used f o r t h e multi-dimensional immiscible displacement c a l c u l a t i o n s employs e q u i l i b r i u m K-value c o r r e l a t i o n s f o r c a l c u l a t i n g phase behaviour. most of t h e unswept o i l was confined t o the two o u t e r stream tubes around the production w e l l . using a mixing parameter.256 The assumption of homogenetity i n the 5-spot p a t t e r n allows e i g h t . The water c u t rises t o 50% a t 2100 days when the recovery i s 59% of t h e o i l i n place. Before undertaking c a l c u l a t i o n s of o i l displacement by CO2 i n j e c t i o n . Imbibition o i l r e l a t i v e p e r m e a b i l i t i e s were derived by applying the Land concept of reduced f r e e s a t u r a t i o n s (Ref 15) t o the Corey drainage approximation. 6 v e r t i c a l l a y e r s were assumed. i t was f i r s t necessary t o define the s t a t e of the model r e s e r v o i r a f t e r waterflopding. The Intercomp Composit i o n a l Reservoir Simulator.

s* l 2 gr + 4(S* g - s*gr)/c s -s I A value of C = 1. Whilst Todd and Longstaff recommend a value of w = 0.(SgF) defined by Land i s f i r s t computed a s a f u n c t i o n of the c u r r e n t gas s a t u r a t i o n (S ) and the h i g h e s t value of gas s a t u r a t i o n previously reached w i t h i n each g r i d f l o c k (Sgmax).86 was obtained from Equation 10 by s u b s t i t u t i n g a trapped gas s a t u r a t i o n (S ) of 0. The r e l a t i v e permeability treatment i n the models f o r miscible and immiscible displacement a r e very d i f f e r e n t . In t h i s approach. Todd and Longstaff c a l c u l a t e d the o i l displacement f o r an a r e a l bead pack system which gave good agreement with experimental d a t a obtained by Lacey (Ref 18).67 f o r a r e a l s t u d i e s .0 t o a s s e s s the s e n s i t i v i t y of the r e s u l t s t o changes i n t h i s f a c t o r .5 and w = 1. In a gas d r i v e t h e gas becomes mobile a t an i n i t i a l s a t u r a t i o n of about 0.30 i s t y p i c a l (Ref 2 2 ) . Using a value of W = 0. The o i l r e l a t i v e permeability i s a f u n c t i o n of both water and gas s a t u r a t i o n s and is given by r lr 1 r - (9) Each of the r e l a t i v e p e r m e a b i l i t i e s i n immiscible displacement a r e s u b j e c t t o three-phase h y s t e r e s i s e f f e c t s a s the flow regimes change (Ref 21). I n the p r e s e n t s t u d i e s . Good agreement with Lacey's r e s u l t s was a l s o obtained i n the p r e s e n t work using the modified CRS model with a c u r v i l i n e a r g r i d and w = 0.67.whereas when gas i s displaced a trapped gas s a t u r a t i o n of 0.257 empirical means.05. The CRS code was t h e r e f o r e modified t o allow f o r gas h y s t e r e s i s e f f e c t s . Unmixed o i l and C02 p r o p e r t i e s were e n t e r e d i n t a b u l a r form.[s*g0.8 f o r v e r t i c a l cross s e c t i o n s and f o r 3-dimensional c a l c u l a t i o n s . no experimental evidence t h a t t h e mixing model i s c o r r e c t f o r a v e r t i c a l c r o s s s e c t i o n geometry and the e f f e c t i v e d e n s i t i e s assumed i n the model may not be v a l i d . a value W = 0. The "free" gas s a t u r a t i o n . the water r e l a t i v e permeability i s a f u n c t i o n of the water s a t u r a t i o n only and the gas r e l a t i v e permeability is a f u n c t i o n of the gas s a t u r a t i o n only. Gas r e l a t i v e p e r m e a b i l i t i e s were obtained by applying t h e Land method (Ref 15) t o t h e drainage curve c a l c u l a t e d from t h e Corey approximation (Ref 1 4 ) . For immiscible displacement.7 was used f o r most of the c a l c u l a t i o n s . The equations a r e s* where gF . Warner (Ref 19) has recommended a value of w = 0. however. There i s .5 S*gr +/(s*g . t h r e e phase r e l a t i v e p e r m e a b i l i t i e s a r e evaluated using Stone's second method (Ref 20). s i n c e l i t t l e gas i s trapped.67. . b u t some c a l c u l a t i o n s were a l s o performed using w = 0. The CRS code was modified t o provide a s u i t a b l e v e h i c l e f o r t h i s mixing model. This involved bypassing the phase e q u i l i b r i u m c a l c u l a t i o n and i n c o r p o r a t i n g the Todd and Longstaff equations. F a i l u r e t o account f o r the h y s t e r e s i s i n gas r e l a t i v e p e r m e a b i l i t i e s r e s u l t s i n o p t i m i s t i c r e c o v e r i e s .77 ( i e 1gr sWc): H y s t e r e s i s e f f e c t s i n the o i l and water r e l a t i v e p e r m e a b i l i t i e s were considered t o be small and were ignored.30 when the maximum gas s a t u r a t i o n had reached 0.

i f p r a c t i c a b l e . The s e p a r a t e o i l and gas r e l a t i v e p e r m e a b i l i t i e s were then obtained by making the l i n e a r as sump t ion. Two-Dimensional Studies A number of comparative imniscible displacement c a l c u l a t i o n s have been performed using a v e r t i c a l . i t being argued t h a t o i l and gas behave a s the same phase under miscible conditions. Case b Case c Case d Case e I n j e c t i o n of a s i n g l e s l u g of C02 over the f u l l h e i g h t of the s e c t i o n A l t e r n a t i n g 100 day i n j e c t i o n s of Cog and water a c r o s s the f u l l h e i g h t of t h e s e c t i o n Simultaneous i n j e c t i o n of C02 i n t o the lower h a l f of t h e s e c t i o n and water i n t o the upper h a l f A l t e r n a t i n g 100 day i n j e c t i o n s of C02 and water. A s a r e s u l t . For consistency w i t h t h e waterflood. Continuous i n j e c t i o n of C02 (Case a ) was used a s t h e b a s i s of comparison. s e p a r a t i o n of C02 and water w i l l considerably reduce the corrosion problems a s s o c i a t e d w i t h a l t e r n a t i n g C02 and water i n j e c t i o n . This implies a r e s i d u a l hydrocarbon/C02 s a t u r a t i o n a f t e r water d r i v e of 0 . with the COP i n j e c t i o n r e s t r i c t e d t o the lower h a l f of t h e s e c t i o n and water i n j e c t i o n r e s t r i c t e d t o t h e upper h a l f . with only 15% of t h e o i l being recovered when 0. With Case a. Cases d and e . two dimensional model t o examine various C02-injection s t r a t e g i e s a f t e r waterflood. so t h a t t h e amount of C02 trapping i s a l s o reduced. A l t e r n a t i n g water with t h e C02 i n Case c reduces t h e peak value of C02 s a t u r a t i o n a t t a i n e d i n any g r i d block. However. The p r a c t i c a l problems of dual completions have not been considered i n d e t a i l . more of t h e Cog can be mobilised during t h e subsequent chase water flood and t h e recovery i n c r e a s e s t o 39%. namely more of the swollen o i l i s d i s p l a c e d .22 PV COP i n Case b i n c r e a s e s o i l production t o 34%. the asymptotic value of o i l recovery is 45% of the t a r g e t .22 PV of i n j e c t e d C02 followed by chase water u n t i l the produced water reaches 90%. I n c l u s i o n of chase water a f t e r 0. would both r e q u i r e dual completions with the i n j e c t e d C02 flowing down a c e n t r a l pipe and water through a surrounding annulus. . The r e s u l t s of these c a l c u l a t i o n s a r e i l l u s t r a t e d i n Figure 6 . This v e r t i c a l model c o n s i s t e d of the second longest of the f o u r stream tubes i n the mesh scheme i d e n t i f i e d previously. kro = -krhc So S g S (13) Hysteresis e f f e c t s i n gas r e l a t i v e permeability were not included i n the miscible c a l c u l a t i o n s .258 The r e l a t i v e permeability treatment i n the Todd and Longstaff miscible model i s based upon the assumption t h a t o i l and s o l v e n t behave a s a s i n g l e phase. and some of the C02 i s moved t o become e f f e c t i v e i n o t h e r p a r t s of t h e r e s e r v o i r . 3 0 .22 PV of C02 i s i n j e c t e d . The following a l t e r n a t i v e s t r a t e g i e s were then examined f o r 0. t h i s s i n g l e hydrocarbon/C02 phase uas given the same r e l a t i v e permeability (krhC) a s t h a t of o i l i n water. f o r two reasons.

This r e s u l t i s a d i r e c t consequence of the d i f f e r e n t treatments f o r gas and o i l r e l a t i v e permeabilities i n the two models. followed by chase water. w h i l s t r e s t r i c t i n g t h e gas t o the lower h a l f of the r e s e r v o i r and water t o the upper h a l f . The immiscible and miscible displacements showed o i l recoveries of 61% and 54% of t h e t a r g e t o i l r e s p e c t i v e l y .22 PV of C02 i s i n j e c t e d . By i n j e c t i n g C02 i n t o the lower h a l f of the r e s e r v o i r with water i n t o the upper h a l f .5 COMPARISON OF OIL RECOVFRIES FOR DIFFERENT If4JECTION SlHATEGIES The c a l c u l a t i o n s f o r C02 i n j e c t i o n i n t o a v e r t i c a l two dimensional model r e s e r v o i r have been repeated f o r miscible displacement conditions. The c a l c u l a t i o n a l models described e a r l i e r f o r t h e two processes involve very d i f f e r e n t p h y s i c a l concepts. gas o v e r r i d e i s reduced and the o i l recovery i s increased t o 59% of the t a r g e t . A l t e r n a t i n g the gas and water i n j e c t i o n s .259 In each of the above c a s e s . a two dimensional a r e a l model was used with a s i n g l e mesh block i n the v e r t i c a l d i r e c t i o n . with a l t e r n a t i n g 100 day s l u g s of gas and w a t e r . 0 1ooo l l Y E (DAYS) FlG. A s i s discussed l a t e r . Case e . c u r r e n t l i m i t a t i o n s i n r e p r e s e n t i n g these physical concepts a r e thought t o be the reason f o r the somewhat lower c a l c u l a t e d o i l r e c o v e r i e s f o r the miscible condition. The m i s c i b l e displacement c a l c u l a t i o n s produced o i l recoveries s l i g h t l y below the immiscible values. which produce d i f f e r e n t values f o r t h e r e s i d u a l gas . with Case e again producing the h i g h e s t l e v e l of o i l recovery. This trend i n d i f f e r e n c e between miscible and immiscible displacement was observed i n a l l of the c a l c u l a t i o n s reported i n t h i s paper. For these c a l c u l a t i o n s . produces a f u r t h e r s l i g h t improvement i n o i l recovery t o 62% of t h e t a r g e t . The r e s u l t i n g a r e a l d i s t r i b u t i o n s of o i l f o r the two processes a r e i l l u s t r a t e d i n Figure 7. It can be seen t h a t i n the immiscible case the C02 has been e f f e c t i v e f o r a g r e a t e r d i s t a n c e from the i n j e c t o r than i n the miscible case. As shown i n Table 2 . the r e s u l t s showed the same trends as those f o r i m i s c i b l e displacement. Areal sweep e f f e c t s were examined f o r both m i s c i b l e and immiscible displacement of o i l when a t o t a l of 0. Case d. t h e r e i s a s i g n i f i c a n t amount of g r a v i t y override.

followed by chase water Simultaneous i n j e c t i o n of C 0 2 i n t o lower h a l f of r e s e r v o i r and water i n t o upper h a l f followed by chase water I n j e c t i o n of C02 i n t o lower h a l f of r e s e r v o i r and water i n t o upper h a l f with a l t e r n a t i n g . followed by chase water.260 TABLE 2 COMPARISON O IMMISCIBLE AND MISCIBLE DISPLACEMENT I N A VERTICAL. 7 and 1 0 f o r a l t e r n a t i n g 100 day i n j e c t i o n s of Cop and water over t h e f u l l h e i g h t . The e f f e c t of varying the Todd and Longstaff mixing parameter was examined i n the two dimensional s t u d i e s . a r e shown i n Table 3. This h i g h l i g h t s an important f a c t o r i n t h e mathematical r e p r e s e n t a t i o n of gas displacement processes which needs f u r t h e r t h e o r e t i c a development coupled with experimental information. The high a r e a l sweep e f f i c i e n c y obtained with the immiscible displacement i s caused by two e f f e c t s : ( i ) component exchange between o i l and gas reduces the i n i t i a l v i s c o s i t y r a t i o of 20:l t o 3. 5 .100 day cycles between C02 and water i n j e c t i o n . Results of c a l c u l a t i o n s f o r w = 0 .22 PV OF CO2 ~ ~~~~ O i l Recovery (2 of Target) Strategy Immiscible a b c Single s l u g i n j e c t i o n of C02 Single s l u g of C02 followed by chase water A l t e r n a t i n g 100 day i n j e c t i o n s of C02 and water. followed by chase water Miscible 15 34 12 29 39 39 d 59 51 e 62 52 - . s a t u r a t i o n s .2:1 ( i i ) up t o one t h i r d of the gas d i s s o l v e s i n the o i l keeping gas s a t u r a t i o n s low and hence maintaining low gas m o b i l i t i e s . TWO DIMENSIONAL F O CROSS SECTION F R A TOTAL INJECTION OF 0. I t might have TABLE 3 EFFECT OF VARYING TODD AND LONGSTAFF M I X I N G PARAMETER ON OIL RECOVERY F R WATER O ALTERNATING WITH GAS I N TWO DIMENSIONAL STUDIES ( X of t a r g e t ) V e r t i c a l Cross Section Areal Model 44 54 63 . 0 . of the s e c t i o n .

but decreased to a level of 50% for about 1000 days before it increased rapidly once more. compared with 51% for miscible displacement. The calculated oil production for immiscible displacement was 56% of the oil remaining after waterflood. These illustrate the effects of gravity on the recovery and show the additional penetration made by the C02 . These calculations were performed assuming the Cop-injection strategy which produced the highest oil recovery in two dimensional studies.33 RESULTING FIG 7 RESIOUAL HYDROCARBON OlSTRlBUTlON 2 0 0 0 OAY WAG I N A R E A L MODEL FROM been expected that the sensitivity to the value of w would be different in the horizontal and vertical planes. In both cases the water cut continued to rise immediately after initiation of CO2-injection.070.261 IMMISCIBLE MISCIBLE KEY EOUIVALENT HVDROCARBON SATURATION SO * 0 6 0. This has not occurred in an obvious manner.15 I 0. followed finally with chase water across the full height of the column. Three-Dimensional Calculations Calculations of immiscible and miscible displacement of oil have been performed using the three-dimensional conceptual reservoir model described earlier.21- 0. The results are presented in Figure 8. but not as high as would be incurred from continued injection of water without C02.16 0.24 - 0. although the direct link between effective viscosities and densities in the Todd and Longstaff model may not be realistic. The distributionsof hydrocarbon in the vertical cross section along the second longest streamtube areshown in Figure 9 . ie. injection of C02 restricted to the lower half of the reservoir and water injection restricted to the upper half with alternating 100 day injections of C02 and water. because in the latter the effective densities will influence the override behaviour. Thus both processes imply a need to handle high water-cuts.

262 FIC. COMPARISON OF IMMISCIBLE A 0 MISCIBLE DISPLACEMENTS FOR THE N THREE OlMENSlONAL RESERVOIR MOOEL (b) M I S C I I L DISPLACEMENT FIG. MSTWEUTIONS OF HYDROCARBON IN A VERTICAL CROSS SECTION OF THE THREE DIMENSIONAL MODEL (HYOROCARBON MASS AS PERCENT WATER FLOOD RESICUAL) .9. 8.

causing the oil to swell artificially and consequently increasing the mobility ratio. whereas a full treatment should consider all non-wetting phases. Similarly. should also change as the displacement process approaches miscibility. Gravity override is an important characteristic of miscible gas displacement processes and there is a need to develop and validate an independent mixing model for effective densities. These results are broadly similar for both immiscible and miscible cases with the previous two-dimensional cross section calculations. Oil recoveries with both processes are slightly optimistic because no allowance has been made for the solubility of CO2 in water. Immiscible viscous fingering may well be important. However. The method of calculating effective oil and solvent densities in the Todd and Longstaff model has not been validated. preventing the CO2 from contacting some of the oil has also been neglected. The relative permeability treatment for the immiscible case. Although hysteresis has been introduced into the gas relative permeabilities for the immiscible calculation. and there is no experimental evidence to resolve this problem. The mixing model used for the miscible calculations is designed to provide effective viscosities in the presence of viscous fingering. . and this indicates that selection of that model was a good basis for comparing general injection strategies. the model adopted has certain limitations.. which assumes distinct gas and oil phases. since it allows substantial amounts of C02 to dissolve in the oil ahead of the displacement front. the formation of free gas behind the front is inhibited by the coarse mesh mixing. Modelling Factors Influencing the Calcu1atio:is The higher oil recovery calculated for immiscible displacement compared with miscible displacement was unexpected. In the mixing model for the miscible cases. All of the above factors require more detailed quantatative analysis supported by experimental measurements to allow more definitive comparisons to be made. Gravity override may be a partially stabilising influence on fingering in the areal plane. there are a number of features in the simulation models which need further theoretical and experimental investigation. The assumption of instantaneous equilibration over the whole grid block with the multicomponent model causes immiscible predictions to be optimistic. whereas the effects of viscous fingering are ignored in the multicomponent immiscible model. particularly when miscibility conditions are approached. However. this introduces new problems in estimating the proportions of oil and gas that are trapped. only a single C02/hydrocarbon phase exists. The effect of water blocking. and there is no equivalent to the hysterises in trapping of CO2 assumed in the immiscible relative permeability model.263 during immiscible displacement. This phenomenon is particularly emFhasised in a coarse mesh arrangement. Predicted recoveries from the immiscible model are therefore likely to be optimistic. The treatment used has considered variations in trapping of the gas phase only.

264 Nomenclature a.b C Parameters i n equations of s t a t e Imbibition trapping constant Binary i n t e r a c t i o n c o e f f i c i e n t f o r the parameter 'b' Vapour-liquid equilibrium p a r t i t i o n coefficient D K Kr m P R R e l a t i v e permeability C h a r a c t e r i s a t i o n constant i n PR equation Pressure Universal gas constant Saturation S T V (S* e f f e c t i v e s a t u r a t i o n ) Absolute temperature Molar volume Mole f r a c t i o n Parameter i n PR equation Binary i n t e r a c t i o n c o e f f i c i e n t f o r t h e parameter 'a' Acentric f a c t o r . X a 6 w n Subscripts C r i t i c a l property Free gas c r i t i c a l gas hydrocarbon components maximum oi1 reduced property water connate water . empirical mixing par-ameter Parameter i n GRK equation.

G. 1649-1655. J . V. i n "Equations of S t a t e i n Engineering". A p r i l 1980.. (February 1949). and ROBINSON. D. V. 16. HOLM.B. t h e i r Computer Simulation". 182. October 1980. CAVETT.Pet. (October 1980).. "Reservoir Development Planning f o r t h e F o r t i e s F i e l d " . D r T P F i s h l o c k and >lr R I Hawes and t h e h e l p of M r I R Hawkyard i n u n d e r t a k i n g computations.S. 9. References I. P e t . "An Equation of S t a t e Compositional Model".H. PENG. 0. K. " E f f e c t of O i l Composition on M i s c i b l e Type Displacement by Carbon Dioxide". 2.. 6. and KWONG.E. p a p e r SPE 8814. San F r a n c i s c o . (November 1978). "Phase E q u i l i b r i a i n Carbon Dioxide . p a p e r PDlO(5) p r e s e n t e d a t t h e 10th World Petroleum Congress. p r e s e n t e d a t t h e 1st SPE/DOE Symposium on Enhanced O i l Recovery.Jn1.A. HILLIER.A. London. L. R. 1979. . 2 7 t h API Mid-year Meeting. - 8. A.N. MOTT. AEEW R 1331 ( 1 9 8 0 ) . COBB R. Dallas. 4.R. Eng. 12. COATS.Y.L. ZUDKEVITCH. e t a l . Tech. 363-376. "On t h e Thermodynamics of S o l u t i o n s . P. AIChE J n l (19731.. D. DIMMOCK.. 44.M. "A New Two-Constant Equation of S t a t e " .A.K. 13. 5.WI and JOSENDAL. SIGMUND. paper EUR 183 p r e s e n t e d a t t h e European Offshore Petroleum Conference and E x h i b i t i o n . Fundam. E. An Equation o f S t a t e .112-119. T u l s a . Chemical Reviews. D. D. KATZ. "Development and E v a l u a t i o n of a Method f o r C a l c u l a t i n g t h e Phase Behaviour o f Multi-Component Hydrocarbon Mixtures from a n Equation o f S t a t e " . - 1I. " C h a r a c t e r i z i n g Hydrocarbon P l u s F r a c t i o n s " . Advances i n Chemistry S e r i e s . Bucharest.H. Chem. 10. Proc. F u g a c i t i e s of Gaseous S o l u t i o n s " . September 21-24 1980. REDLICH. R. and FIROZAEADI. " P r e d i c t i n g Phase Behaviour of Condensatel Crude-Oil Systems u s i n g Methane I n t e r a c t i o n C o e f f i c i e n t s " . The a u t h o r s acknowledge t h e a d v i c e given by D r F J Fayers. . Washington DC. Ind. (1974) 15. " A p p l i c a t i o n of a G e n e r a l i s e d Equation of S t a t e t o Petroleum R e s e r v o i r F l u i d s " . Soc. P. e t a l p "Laboratory C02 Floods and.265 Acknowledgement The work r e p o r t e d i n t h i s paper h a s been funded by t h e Department o f Energy. 1962.H. TUREK. " P h y s i c a l Data f o r D i s t i l l a t i o n C a l c u l a t i o n s Vapour-Liquid Equilibrium". C. October 1978. Paper EUR 9 8 p r e s e n t e d a t European Offshore Petroleum Conference and E x h i b i t i o n .Eng. 233-244. 7. J. 3. J. London.Multicomponent Hydrocarbon Systems: Experimental Data and an Improved P r e d i c t i o n Technique".. . 59-64.M. p a p e r SPE 9231 p r e s e n t e d a t t h e SPE Annual F a l l Technical Conference and E x h i b i t i o n . American Chemical S o c i e t y .. and JOFFE... WHITSON. 385-435. L. YARBOROUGH. ( 1 9 7 9 ) .. " C o r r e l a t i o n and P r o d u c t i o n of Phase E q u i l i b r i a w i t h t h e Redlich-Kwong Equation of S t a t e " .

192. . Pet.. LACEY. COREY. 243. J . Presented a t the Second J o i n t Symposium on Enhanced O i l Recovery.R. 18. and LONGSTAFF.266 14. Tulsa. V. and GRAUE. . "Numerical Simulation of Compositional Phenomena i n Petroleum Reservoirs". Trans AIME ( 1 9 6 8 ) . 19. R. Tech.S. 20. J . 22.. 1339-1347. W . A p r i l 1981. 53-61. J. Tech. "The Development. Numerical Simulation o ( 1 9 7 3 ) .. BRINKMAN. Nov 1954. Tech (Oct 1977). 149.. Paper SPE/DOE 9804. P e t . . WARNER.T. J. "An Evaluation of Miscible C02 Flooding i n a Waterflooded Sandstone Reservoir". p 269-284. and MORSE.r e l a t i o n Between Gas and O i l R e l a t i v e Permeabilities". E . "Estimation of Three Phase R e l a t i v e Permeability and Residual O i l Data".H. Producer's Monthly Vol X I X . J. 12. J . (July 1972). W .S.L. I . (Oct 1973). 874-882. Recovery by Waterflooding" Trans AIME ( 1 9 5 1 ) . . Testing and Application of a Numerical Simulator f o r P r e d i c t i n g Miscible Flood Performance". FARIS.. NOLEN. "Scaling of Flow Parameters f o r Miscible Gas Flood Simulation Studies". STONE.156. M.Pet. LAND. J . "Calculation of Imbibition R e l a t i v e Permeability f o r Two and Three Phase Flow From Rock Properties". TODD. C. H. P e t .R. "Effect of Free Gas S a t u r a t i o n on O i l 135-140. C. SPE Reprint S e r i e s N I I ... A. 806-816. HOLMGREN. Can. F. 16. "The I n t e r . D .R.A. "Effect of Bank S i z e on O i l Recovery i n High Pressure Gas-Driven LPG-Bank Process". J . H. BREIT.Tech (August 1961). J. . 15. 21.S. 17.

SIAVASH GHARIB and RICHARD OYEKAN Department of Petroleum Engineering. i t s s w e l l i n g a n d c o n s e q u e n t i n c r e a s e i n m o b i l i t y . compared t o water. MAHGUIB EL ARABI. On t h e same p l o t is shown t h e v a r i a t i o n i n d e n s i t y and v i s c o s i t y of t h e carbon d i o x i d e as a f u n c t i o n of t h e p r e s s u r e . A 50 t o 100 f o o t l e n g t h of 3/8 i n c h t u b i n g i s packed w i t h sand. E x p e r i m e n t s have i n d i c a t e d t h a t n i t r o u s o x i d e . A t y p i c a l r e s u l t of a s l i m t u b e e x p e r i m e n t is shown i n F i g u r e 1. S c a l e d p h y s i c a l model s t u d i e s have been u n d e r t a k e n i n a n a t t e m p t t o d e t e r m i n e w h e t h e r t h e l a r g e r a t i o s of i n j e c t e d c a r b o n d i o x i d e t o produced c r u d e o i l . t h e r e s u l t s need t o b e q u a l i f i e d by n o t i n g t h a t t h e protot y p e r e s e r v o i r s u s e d i n t h e s e s t u d i e s are v e r y f a v o r a b l e f o r t h e u s e of CO2* The r e c o v e r y mechanism a p p e a r s t o b e c h i e f l y t h e s o l u t i o n of carbon d i o x i d e i n t h e oil. 11. DOSCHER. f i l l e d w i t h crude oil and d i s p l a c e d w i t h carbon d i o x i d e a t a h i g h p r e s s u r e . Cdifornia 90007 I.MISCIBLE GAS DISPLACEMENT 267 OIL RECOVERY BY CARBON DIOXIDE THE RESULTS OF SCALED PHYSICAL MODELS AND FIELD PILOTS TODD M. However. The r e s u l t s of t h i s s t u d y have r a t h e r u n e q u i v o c a l l y i n d i c a t e d t h a t t h e h i g h r a t i o s of i n j e c t e d c a r b o n d i o x i d e t o r e c o v e r e d c r u d e o i l s h o u l d i n f a c t be expected. are t o b e e x p e c t e d or a r e due t o some d e f e c t s in t h e a p p l i c a t i o n of t h e p r o c e s s i n t h e f i e l d . . T h e e f f i c i e n c y o f t h e p r o c e s s is t h w a r t e d by t h e high m o b i l i t y of t h e carbon d i o x i d e which l e a d s t o v i s c o u s f i n g e r i n g and i t s low d e n s i t y . a n d t h e n t h e d i s p l a c e m e n t of t h e s w o l l e n . INTRODUCTION Much of t h e o p t i m i s m c o n c e r n i n g t h e p o t e n t t a l of carbon d i o x i d e i n r e c o v e r i n g r e s i d u a l c r u d e o i l h a s been based on s l i m t u b e experiments. The e x p e r i m e n t s have a l s o shown t h a t a v e r y h e i g h t e n e d e f f i c i e n c y of t h e p r o c e s s c a n b e a c h i e v e d by u s i n g a s l u g of c a r b o n d t o x t d e f o l l o w e d by water. Los Angeles. is as e f f e c t i v e as c a r b o n d i o x i d e i n t h e s e model s t u d i e s . however t h e r e s u l t s of f i e l d p i l o t o p e r a t i o n s have i n d i c a t e d t h a t very l a r g e q u a n t i t i e s of carbon d i o x i d e are r e q u i r e d t o r e c o v e r t h e r e s i d u a l oil. which changes b u t l i t t l e between e n t r a n c e and o u t l e t b e c a u s e of t h e h i g h p e r m e a b i l i t y of t h e system. Universiq of Southern clllifornia. which a l s o d i s p l a y s a h i g h s o l u b i l i t y i n o r g a n i c compounds. ABSTRACT C o n s i d e r a b l e e f f o r t s h a v e b e e n e x p e n d e d in t h e p a s t d e c a d e on t h e p o t e n t i a l u s e of c a r b o n d i o x i d e f o r t h e r e c o v e r y of r e s i d u a l c r u d e oil. which l e a d s t o g r a v i t y s e g r e g a t i o n . m o b i l e o i l by a g a s d r i v e ( i f carbon d i o x i d e is i n j e c t e d c o n t i n u o u s l y ) or by water i f a s l u g of carbon d i o x i d e is f o l l o w e d b y t h e l a t t e r . o b s e r v e d i n f i e l d d e m o n s t r a t i o n p r o j e c t s .

1: THE EFFECT OF PRESSURE ON THE 5 ' CRUDE I N A SLIM TUBE DISPLACEMENT OF A 4 Warner'' s t u d i e d t h e n u m e r i c a l s i m u l a t i o n of c a r b o n d i o x i d e d i s p l a c e m e n t of r e s i d u a l c r u d e o i l w i t h a model t h a t depended on t h e mixing of t h e carbon d i o x i d e w i t h t h e o i l f o r i t s m o b i l i z a t i o n and t r a n s p o r t . shows t h e e x i s t e n c e of a m i s c i b i l i t y gap13 in carbon dioxidec r u d e o i l s y s t e m s .268 5. The s e c o n d p h a s e is t h e m i x t u r e of t h e l i g h t c o m p o n e n t s of t h e crude and t h e e x c e s s carbon d i o x i d e in t h e system.6. As t h e c a r b o n d i o x i d e c o n t e n t of t h e s y s t e m is inc r e a s e d . E a r l i e r work o n v i s c o u s f i n g e r i n g of c o u r s e showed t h a t s t a b l e p i s t o n . \ 400 800 PRCSSURC. The j u x t a p o s i t i o n of t h e p h y s i c a l p r o p e r t i e s of c a r b o n d i o x i d e w i t h t h e s l i m t u b e r e c o v e r y a s a f u n c t i o n of p r e s s u r e in F i g u r e 1 c e r t a i n l y s u g g e s t s t h a t t h e i n c r e a s i n g r e c o v e r y is due t o g r a d u a l l y decreasi n g a d v e r s e m o b i l i t y and g r a v i t y r a t i o s . There a p p e a r e d t o b e a n i m p l i c i t s u g g e s t i o n t h a t good recovery would be a c h i e v e d as l o n g as t h e d i s p l a c e m e n t p r e s s u r e was e q u a l t o o r g r e a t e r t h a n t h i s v a l u e . The p h a s e b e h a v i o r a s d e s c r i b e d in t h e c u r r e n t l i t e r a t u r e .7 E a r l i e r i n v e s t i g a t o r s have def i n e d 1 * 2 s 3 * 4 . No o i l bank was developed i n t h e s e s i m u l a t i o n s and t h e p a r a m e t e r t h a t e x e r c i s e d t h e c h i e f c o n t r o l o v e r t h e p r o c e s s w a s t h e g r a v i t y s e g r e g a t i o n of t h e i n j e c t e d carbon d i o x i d e i n t h e w a t e r . Warner's n u m e r i c a l r e s u l t s would have s t i l l been p o o r e r had h e i n c l u d e d t h e v e r y f i n e g r i d t h a t C l a r i d g e l 2 h a d e a r l i e r s h o w n was n e c e s s a r y f o r v i s c o u s f i n g e r i n g t o b e p r o p e r l y e x h i b i t e d in a n u m e r i c a l model. 1200 PSI 1600 Fig. t h e s o l u b i l i t y is a b o u t 60 m o l p e r c e n t c a r b o n d i o x i d e .f i l l e d r e s e r v o i r . and i t c a n b e r e a d i l y shown t h a t t h e d i s p e r s i o n c o e f f i c i e n t s a c h i e v e d in s l i m t u b e s a r e v e r y d i f f e r e n t from t h e d i s p e r s i o n c o e f f i c i e n t s o b t a i n e d in r e a l r e s e r v o i r s 9 . F o r v i r t u a l l y a l l c r u d e o i l s t h a t h a v e b e e n s t u d i e d a n d r e p o r t e d i n t h e l i t e r a t u r e . t h e s o l u t i o n of carbon d i o x i d e in t h e c r u d e f r a c t i o n a t e s i n t o two o r more phases.g. t h e p r e d i c t a b i l i t y of a "minimum m i s c i b i l i t y p r e s s u r e " . 'OOt A . et. even d e s p i t e a n a d v e r s e m o b i l i t y r a t i o i f t h e d i a m e t e r of t h e f l o w s y s t e m approaches t h e t h i c k n e s s of t h e f i n g e r s t h a t can be g e n e r a t e d in t h e s y s t e m u n d e r s t u d y l o . The l a t t e r phase remains .and t h e n d e b a t e d . G a r d n e r . R e f e r e n c e 8..8 h o w e v e r s h o w e d t h a t t h e h i g h r e c o v e r i e s o b t a i n e d i n s l i m t u b e s is r e l a t e d t o t h e a t t a i n m e n t of a l o w p h y s i c a l d i s p e r s i o n c o e f f i c i e n t (D/vL). t h a t p r e s s u r e a t which 95% of t h e o i l c o n t a i n e d in t h e s l i m t u b e is r e c o v e r e d b e f o r e carbon d i o x i d e breakthrough.al.l i k e d i s p l a c e m e n t of o n e f l u i d by a n o t h e r is e f f e c t e d w h e t h e r t h e f l u i d s a r e m i s c i b l e o r n o t . One is a s o l u t i o n of carbon d i o x i d e i n t h e heavy components of t h e c r u d e . e.

a s much as 60% 0% t h e o i l c o n t a i n e d i n a p i l o t p a t t e r n may have been r e c o v e r e d a t a r a t i o of 27.1875 The r e s e r v o i r chosen f o r t h i s p a r t i c u l a r set of e x p e r i m e n t s . w a s one of r e l a t i v e l y low p e r m e a b i l i t y .i n j e c t i o n of produced carbon d i o x i d e . a p i l o t o p e r a t i o n i n t h e Meakin s a n d i n d i c a t e s t h e r a t i o of recovered o i l t o i n j e c t e d carbon dioxide w i l l be 28. a n d t h e l a s t a v a i l a b l e f i u r e s f o r t h e Two F r e d s f i e l d i n West Texas i n d i c a t e a r a t i o of 18 MSCF/Bblq6. M i s s i s s i p p i . A t L i t t l e Creek. Even i n t h e a b s e n c e of v i s c o u s f i n g e r i n g . MDS.2 3. then t h e economic v i a b i l i t y of carbon dioxide i n j e c t i o n p r o j e c t s f o r t h e r e c o v e r y of r e s i d u a l o i l must b e re-examined. INJECTION PATTERN SPACING. INJECTOR T O T PRODUCER.. 28. RESULTS OF THE EXPERIMENTAL STUDIES The e x p e r i m e n t a l work was c a r r i e d o u t i n p h y s i c a l l y s c a l e d models of a d i r e c t l i n e d r i v e p a t t e r n . FT. A t L i c k C r e e k . T h i s was done t o minimize t h e e f f e c t s of g r a v i t y s e g r e g a t i o n and v l s c o u s f i n g e r i n g and t h u s s e c u r e a r e l a t i v e l y f a v o r a b l e performance. i t can be s e e n . RESERVOIR THICKNESS.6 MSCF/Bb116.269 a highly mobile f l u i d and can be expected t o f i n g e r through t h e r e s e r v o i r j u s t a s w o u l d any low v i s c o s i t y f l u i d . The S a c r o c t e r t i a r y i l o t r e c o v e r e d 3% o f t h e r e s i d u a l o i l a t a CO /OIL r a t i o of 36 MSCF/BbllP. A r k a n s a s . PARAMETER PERMEABILITY. F . MODEL VALUE PROTOTYPE VALUE 20. 111. and a c c o u n t i n g f o r i n t e r e s t c h a r g e s due t o t h e d e l a y b e t w e e n i n j e c t i n g c a r b o n d i o x i d e and r e c o v e r i n g t h e c r u d e 01. I f t h e s e r a t i o s are p r o j e c t e d t o b e v a l i d f o r f u l l s c a l e o p e r a t i o n . Even r e . a f t e r r e q u i r e d p u r i f i c a t i o n and d r y i n g .) equipment i s d e s c r i b e d elsewhere9. These s t u d i e s were c a r r i e d o u t i n a n a t t e m p t t o d e t e r m i n e whether s u c h h i g h c a r b o n d i o x i d e / o i l r a t i o s a r e t o b e e x p e c t e d o r a r e due t o s p e c i f i c p e c u l i a r i t i e s of t h e p i l o t and d e m o n s t r a t i o n tests from which they have emenated. The f i e l d p e r f o r m a n c e o f t e r t i a r y p i l o t o p e r a t i o n s have i n d i c a t e d t h a t i n d e e d carbon d i o x i d e can m o b i l i z e and d i s p l a c e r e s i d u a l crude o i l . TABLE 1. see T a b l e 1. T h e d e t a i l s of t h e s c a l i n g p r o c e d u r e s t h a t h a v e b e e n u s e d a n d t h e c o n s t r u c t i o n of t h e r e q u i r e d h i g h p r e s s u r e ( t o 5000 psi. . t h e c i t e d r a t i o s of c a r b o n d i o x i d e i n j e c t e d t o produced o i l would add o v e r $20 t o t h e c o s t of r e c o v e r i n g a b a r r e l of o i l by t h e i n j e c t i o n of carbon dioxide.000 LINE DRIVE 3.4 MSCF/Bbl. LINE DRIVF 462.08 0. high t e m p e r a t u r e ( t o 250'F. a Buckley L e v e r e t t a n a l y s i s 1 4 of t h e i n j e c t i o n of s o l v e n t i n t o a w a t e r e d o u t r e s e r v o i r shows t h a t e a r l y breakthrough of a mobile s o l v e n t s h o u l d be expected.).

21. P11400 PSI.2 70 A. CUMULATIVE RECOVERY OF OIL AND WATER FOR FXG. o f a p o r e v o l u m e f o l l o w i n g b r e a k t h r o u g h . n e o u s l y w i t h t h a t of c a r b o n d i o x i d e .c r i t i c a l c a r b o n d i o x i d e a t 1400 p s i a n d 73OF. The P r o d u c t i o n H i s t o r y of R e s i d u a l O i l R eco v er y by C ar b o n D i o x i d e Figures 2 through 4 present a t y p i c a l production h i s t o r y f o r the d i s p l a c e m e n t of r e s i d u a l c r u d e o i l by s u b . 2 The p r o d u c t i o n r a t e of oil r e a c h e s a maximum v a l u e w i t h i n 02 t o 03 . . 2 PRODUCTtON HISTORY FOR A TYPICAL C 0 2 DIS LACEMENT . i t i s t h e o n l y m o b i l e p h a s e i n t h e r e s e r v o i r f o l l o w i n g a water f l o o d . INJECTED (PV1 F i g . Carbon d i o x i d e a p p e a r s a t t h e p r o d u c i n g end of t h e s y s t e m a f t e r t h e i n j e c t i o n of a b o u t 02 of a p o r e v o lu m e. I t is r e a d i l y s e e n t h a t w a t e r a l o n e is p r o d u c e d a t f i r s t .I. 45°A. COa INJECTED IPV) F ig. . b u t t h e r a t i o o f oil t o c a r b o n d i o x i d e in t h e e f f l u e n t c o n t i n u o u s l y d e c r e a s e s a f t e r i t s f i r s t ap p ear an ce. a n d c r u d e oil p r o d u c t i o n is i n i t i a t e d s i m u l t a .CRUDE. T=73 g CO. S0=0.P.3 .

e x p r e s s e d i n terms of MSCF/Bbl. . 4. I t is i n f o r m a t i v e t o n o t e t h a t t h e a v e r a g e m o l a r c o n c e n t r a t i o n of c a r b o n d i o x i d e in t h e o i l p h a s e w i t h i n t h e m o d e l . and t h i s i s t r u e even which a c o m p l e t e l y m i s c i b l e hydrocarbon (dodecane) i s s u b s t i t u t e d f o r t h e crude o i l . i s s h o w n i n F i g u r e 6 f o r v a r i o u s i n i t i a l s a t u r a t i o n s of c r u d e o i l . T h e e f f i c i e n c y o f t h e r e c o v e r y p r o c e s s . The i m p o r t a n c e of t h i s o b s e r v a t i o n i n connectiom w i t h t h e r e l a t i o n s h i p of phase b e h a v i o r t o recovery w i l l be discussed later. F i g u r e 5. Fig. r e a c h e s a v a l u e of w e l l o v e r 0. I t i s s e e n t h a t t h e C 0 2 / 0 1 1 r a t i o s a r e i n d e e d in t h e r a n g e of 20 t o 30 MSCF/Bbl when r e s i d u a l s a t u r a t i o n s of less t h a n 30% are b e i n g recovered. 5 . c a l c u l a t e d by m a t e r i a l b a l a n c e .6 by t h e time carbon d i o x i d e b r e a k s t h r o u g h in t h e e f f l u e n t . 2 The recovery of t h e r e s i d u a l o i l is never c o m p l e t e even w i t h t h e i n j e c t i o n of s e v e r a l p o r e volumes of c a r b o n d i o x i d e . AVERAGE MOLAR CONCENTRATION OF CARBON D I O X I D E IN OIL PHASE FOR EXPERIMENT DESCRIBED BY FIGURE 2. PRODUCTION RATE HISTORY FOR FIG.271 Fig.

The s o l u b i l i t y of c a r b o n d i o x i d e i n t h e p r i n c i p a l c r u d e oil u s e d i n t h i s s t u d y is shown i n F i g u r e s 7 and 8. From c h e m i c a l t h e r m o d y n a m i c s i t i s known t h a t f o r t h e c a s e w h e r e t h e r e i s no n e t volume c h a n g e a t t e n d a n t upon mixing. E f f e c t of T e m p e r a t u r e a n d P r e s s u r e The t e m p e r a t u r e a n d p r e s s u r e a f f e c t t h e d i s p l a c e m e n t of o i l by c a r b o n d i o x i d e i n two i m p o r t a n t a s p e c t s . 7. The s e c o n d i s t h e e f f e c t of t e m p e r a t u r e a n d p r e s s u r e on t h e s o l u b i l i t y of t h e c a r b o n d i o x i d e i n t h e c r u d e o i l . SCF/BbL . SOLUBILITY OF CARBON D I O X I D E I N 450 CRUDE. P=1400 P S I . T h i s i n d e e d ' w a s f o u n d t o b e t h e case f o r t h e s o l u b i l i t y of l i q u i d c a r b o n d i o x i d e i n h y d r o ~ a r b o n s l ~ . - B. The f i r s t is r e l a t e d t o t h e f a c t t h a t t h e s e parameters a f f e c t t h e p h y s i c a l p r o p e r t i e s . 6. which. THE EFFICIENCY OF OIL RECOVERY BY CARBON D I O X I D E AS A FUNCTION OF INITIAL OIL SATURATION. T 73OF.The s o l u b i l i t y of s u p e r c r i t i c a l carbon d i o x i d e i n hydrocarbons i n c r e a s e s w i t h i n c r e a s i n 8 p r e s s u r e and d e c r e a s e s w i t h i n c r e a s i n g t e m p e r a t u r e . Pool 1 A\ I Fig. i n t u r n .272 I F i g . s o l u b i l i t y of one l i q u i d i n a n o t h e r is n o t a p p r e c i a b l y a f f e c t e d by p r e s s u r e . v i s c o s i t y and density. a n d i s i n c r e a s e d by t e m p e r a t u r e . a f f e c t f l u i d f l o w i n c l u d i n g t h e d i s p l a c e m e n t of o n e f l u i d by a n o t h e r .

b u t a f u r t h e r p r e s s u r e i n c r e a s e above t h e s a t u r a t i o n v a l u e h a s o n l y a minor e f f e c t on recovery. t o 2650 p s i . and o v e r t h e p r e s s u r e r a n g e o f 650 p s i . t o 1. The d e n s i t y of t h e c a r b o n d i o x i d e i n c r e a s e s f r o m 0.273 If P a W i t h t h i s b a c k g r o u n d i n h a n d . b u t i n c r e a s e s f r o m 63% t o o n l y 71% when t h e p r e s s u r e is i n c r e a s e d a n o t h e r 1000 p a l t o 2000 p s i .78.P.I. A s i m p l e f o r c e b a l a n c e t h e n shows t h a t t h e same d e g r e e of g r a v i t y s e g r e g a t i o n t h a t o c c u r s a t a p r e s s u r e o f 2150 p s i c a n b e a c h i e v e d . a n d 0.. It i s a p p a r e n t t h a t t h e e f f e c t of p r e s s u r e a t a subc r i t i c a l t e m p e r a t u r e i s v e r y a i g n i f i c a n t i f t h e p r e s s u r e is l e s s t h a n t h e s a t u r a t i o n v a l u e ( a b o u t 900 p s i a t 75OF. and f i n a l l y t o 2650 p s i . Both c r u d e o i l a n d a m i s c i b l e h y d r o c a r b o n (dodecane) were u s e d a t r e s i d u a l a n d h i g h i n i t i a l s a t u r a t i o n s .70 t o 0.).000 p s i . 0. a t t e m p e r a t u r e s t o 130°F. t h e r e c o v e r y i n c r e a s e s f r o m a meager 20% t o o v e r 6 0 % u p o n i n c r e a s i n g t h e p r e s s u r e f r o m 6 5 0 p s i .74 g/CC.11 t o 0. s o m e t w e n t y r u n s were c o n d u c t e d t o e v a l u a t e t h e e f f e c t of t e m p e r a t u r e a n d p r e s s u r e on t h e d i s p l a c e m e n t of c r u d e o i l by c a r b o n d i o x i d e a s r e v e a l e d i n t h i s p h y s i c a l l y s c a l e d m o d e l s t u d y . c r u d e o i l a t s u b . The d r a m a t i c i n c r e a s e i n t h e r e c o v e r y as t h e p r e s s u r e is i n c r e a s e d f r o m 6 5 0 p s i t o 1000 p s i a t 75OF.30. p a r a l l e l s a m a r k e d i n c r e a s e i n t h e d e n s i t y of c a r b o n d e n s i t y f r o m 0.c r i t i c a l t e m p e r a t u r e s is shown i n F i g u r e 9 f o r b o t h r e s i d u a l a n d h i g h i n i t i a l saturations. The e f f e c t of p r e s s u r e on t h e r e c o v e r y of t h e 45OA.22 t o 0. For t h i s h i g h i n i t i a l o i l s a t u r a t i o n .76 a s t h e p r e s s u r e i n c r e a s e s f r o m 1150 t o 2150 psi. This translates t o a density d i f f e r e n c e b e t w e e n d i s p l a c i n g f l u i d a n d d i s p l a c e d f l u i d ( w a t e r ) o f 0.24 g/cc.

10.1 PnrvreW) ZOO0 02 .9.18 01 . H e n c e .1 0 800 0 1870 01 .274 a t 1 1 5 0 p s i o n l y by i n c r e s i n g t h e r a t e by a f a c t o r of t h r e e . THE EFFECT OF PRESSURE ON RECOVERY OF CRUDE OIL A 130°F* T .c r i t i c a l v a l u e s a l t h o u g h t h e r e c o v e r y l e v e l s a r e s o m e w h a t less t h a n a t t h e l o w e r t e m p e r a t u r e .8 - S 21BO 0. t h e small d i f f e r e n c e i n r e c o v e r y i s p r o b a b l y d u e t o t h e c o m b i n e d e f f e c t of s m a l l d i f f e r e n c e s i n b o t h s o l u b i l i t y and g r a v i t y segregation.1 lo00 01 . INJECTED WVI Fig.c r i t i c a l t e m p e r a t u r e t h e e f f e c t o f p r e s s u r e on t h e r e c o v e r y of r e s i d u a l c r u d e o i l i s s i m i l a r t o i t s e f f e c t a t s u b . Between 2150 p s i and 1650 p s i . F i g u r e 10.8 CO. t h e p o o r e r e f f i c i e n c y a t 1150 p s i as compared t o t h a t a t 2150 p s i is p r o b a b l y due p r i m a r i l y t o a g r e a t e r d e g r e e o f g r a v i t y s e g r e g a t i o n a t t h e l o w e r pressure.8 1 1180 01 . leiend 0 A lo00 02 . THE EFFECT OF PRESSURE ON RECOVERY OF CRUDE O I L BY CARBON D I O X I D E AT 73'F* A t a s u p e r . COa INJECTED (PV) F i g .

. THE EFFECT OF PRESSURE ON THE DISPLACEMENT OF DODECANE BY CARBON D I O X I D E . r e s p e c t i v e l y . 11.82. --I U COa INJECTED IPV) Fig. THE COMPENSATING EFFETS OF TEMPERATURE AND PRESSURE ON THE DISPLACEMENT OF RESIDUAL CRUDE O I L To g a i n s t i l l f u r t h e r i n s i g h t i n t o t h e d i s p l a c e m e n t of crude o i l by carbon d i o x i d e .275 To f u r t h e r s t u d y t h e r o l e of d e n s i t y on t h e d i s p l a c e m e n t of carbon d i o x i d e . F i g u r e 12. a n d 125OF. i t is t h e r a t e of c h a n g e of s o l u b i l i t y w i t h p r e s s u r e t h a t a f f e c t s t h e r a t e of change of recovery w i t h p r e s s u r e whether t h e d i s p l a c i n g f l u i d is r i s c i b l e or not. The r e s u l t s of two r u n s . e v e n t h o u g h t h e t e m p e r a t u r e s a r e below and above t h e c r i t i c a l t e m p e r a t u r e . t h e e f f e c t of p r e s s u r e on t h e d i s p l a c e m e n t of a c o m p l e t e l y m i s c i b l e hydrocarbon was i n v e s t i g a t e d as a f u n c t i o n of p r e s s u r e . t h e d e n s i t y of c a r b o n d i o x i d e i s 0. It i s a p p a r e n t t h a t t h e e f f e c t of p r e s s u r e is t h e same f o r t h e d i s p l a c e m e n t of d o d e c a n e a s i t i s f o r t h a t o f c r u d e o i l . --I I COi INJECTED (PVI Fig. one a t each of t h e f o r e g o i n g sets of p a r a m e t e r s . i s s h o w n i n F i g u r e 11 a n d i t i s r e a d i l y s e e n t h a t t h e r e s u l t s of t h e t w o r u n s can b e superimposed on each o t h e r . a set of e x p e r i m e n t s were performed o v e r a range of t e m p e r a t u r e s and p r e s s u r e s where t h e d e n s i t y c o u l d b e m a i n t a i n e d r e l a t i v e l y c o n s t a n t by m a n i p u l a t i n g t h e s e two p a r m e t e r s . a s i t i s a t 2700 p s i . 1 2 . A t 1 4 0 0 p s i . a n d 75’F.

The E f f e c t of I n i t i a l Oil S a t u r a t i o n . 80 COa INJECTION IPV) Fig. see Fig. INJECTED IPV) 1. e f f e c t is s e n s e d o n l y i n t h e l a t e r l i f e of t h e f l o o d . Some e x p e r i m e n t s were c o n d u c t e d a t s t i l l l o w e r r a t e . t h e i n d i c a t e d r a n g e r e s u l t s i n a s l i g h t improvement i n recovery. The o i l r e c o v e r y is o n l y 50% of t h e r e s i d u a l s a t u r a t i o n b u t c l i m b s t o 80% of t h e i n i t i a l o i l s a t u r a t i o n when t h e l a t t e r is 80%.63 I I 29. The i n i t i a l o i l s a t u r a t i o n h a s a very d i r e c t e f f e c t on b o t h t h e f r a c t i o n a l o i l r e c o v e r y . b u t g r a v i t y s e g r e g a t i o n dominated t h e s y s t e m and t h e c r u d e o i l recovery decreased rapidly.8 2D Fig. THE EFFECT OF INITIAL OIL SATURATION ON OIL RECOVERY .8 1. 13. F i g .2 0. 1 4 . %(PVl 1 1 .4 f o o t p e r day.4 CO.6 0. THE EFFECT OF VELOCITY (OR INJECTION RATE) ON THE RECOVERY OF CRUDE O I L BY CARBON DIOXIDE D.6 1.05 t o 0. I 23 - 0 0.2 1. 1 3 s h o w s t h a t i n c r e a s i n g t h e v e l o c i t y o v e r t h e However.0 1. E f f e c t of I n j e c t i o n Rate E x p e r i m e n t s were c o n d u c t e d a t p r o t o t y p e v e l o c i t i e s v a r y i n g f r o m 0.276 C.4 0.77 1 . 14. a n d t h e r e s u l t i n g c a r b o n d i o x i d e / o i l r a t i o .

. . t h e v i s c o s i t y r e d u c t i o n and.which is n e c e s s a r y b e f o r e t h e c a r b o n d i o x i d e c a n c o n t a c t a n d d i s s o l v e i n t h e c r u d e o i l ) .I. is much h i g h e r t h a n would be p r e d i c t e d from t h e r e l a t i v e p e r m e a b i l i t y r e l a t i o n s h i p f o r t h e o i l .I. The E f f e c t of O i l ComDosition CRUDE BY CO2 F o u r d i f f e r e n t " o i l s " were u s e d t o s t u d y t h e e f f e c t of o i l c o m p o s i t i o n : d o d e c a n e . h e x a d e c a n e w h i c h d i s p l a y s a m i s c i b i l i t y g a p . T h e r e is i n f a c t v i r t u a l l y no d i f f e r e n c e i n t h e r e c o v e r y of t h e o i l p h a s e .29. E. c r u d e .P. The o v e r a l l recovery is h i g h e s t when carbon d i o x i d e is completely m i s c i b l e w i t h t h e o i l phase. The C 0 2 / 0 I L r a t i o is o n l y a b o u t 7 MSCF/Bbl f o r a n I n i t i a l s a t u r a t i o n of 0.277 The d i f f e r e n c e I n t h e e f f i c i e n c y of t h e d i s p l a c e m e n t is d i r e c t l y r e l a t e d t o t h e f a c t t h a t when a low o i l s a t u r a t i o n is d i s p l a c e d . t h i s l e a d s t o t h e c o n c l u s i o n t h a t t h e m o b i l i t y of t h e o i l It IS obviously i n which carbon d i o x i d e h a s d i s s o l v e d h a s been increased. viz. as l o n g as i t s v i s c o s i t y is less t h a n 6 c e n t i p o i s e s . a n d a s o l u t i o n o f a 14O A.P. The s w e l l i n g of t h e 45' c r u d e oil u s e d i n t h i s s t u d y is shown i n F i g u r e 1 5 i t is s u b s t a n t i a l . o n l y a s l i g h t l y lower 0 1 1 r e c o v e r y is a c h i e v e d when h e x a d e c a n e is s u b s t i t u t e d f o r t h e dodecane. t h e s w e l l i n g of t h e c r u d e o i l phase which c a u s e s t h i s i n c r e a s e i n m o b i l i t y . The r e s u l t s a r e shown i n F i g u r e 16.5 p o r e v o l u m e of c a r b o n d i o x i d e . c r u d e i n t h e 45O c r u d e . dodecane. w h i c h is c o m p l e t e l y m i s c i b l e . m i s c i b l e o r not. e s p e c i a l l y a f t e r carbon d i o x i d e breakthrough. The f l o w of o i l . most i m p o r t a n t l y . much of t h e i n j e c t e d c a r b o n d i o x i d e is b e i n g u s e d m e r e l y t o d i s p l a c e t h e water i n t h e r e s e r v o i r (. b u t i n c r e a s e s to a v a l u e of 15 MSCF/Bbl when t h e i n i t i a l s a t u r a t i o n is d r o p p e d t o 0. Fig.I.w a t e r system. THE SWELLING FACTOR OF A 45°A. perhaps.77.P. However. 15. d u r i n g t h e i n j e c t i o n of t h e f i r s t 0. t h e 4S0A. The s l i g h t d i f f e r e n c e s i n r e c o v e r y t h a t d e v e l o p u p o n t h e i n j e c t i o n of more c a r b o n d i o x i d e are perhaps b e s t understood by r e f e r r i n g t o t h e experiment u s i n g a m i x t u r e of 14O and 45O crude o i l s . b o t h o i l a n d water a r e p r o d u c e d t h r o u g h o u t t h e run. When t h e i n i t i a l o i l s a t u r a t i o n is a b o v e t h e r e s i d u a l v a l u e .

l. . t h e r e s u l t s a r e n o t s i g n i f i c a n t l y d i f f e r e n t f r q m t h o s e t h a t have been d e s c r i b e d f o r t h e dead crude.0 . Exacerbated v i s c o u s f i n g e r i n g .d /: &: /* . 1.H.4- c- FO+"-----" 11**"1 / / / d ' 2 . t h a t most of t h e o i l t h a t is r e c o v e r e d h a s t h e same g r a v i t y a s t h a t of t h e m i x t u r e .4 . of 06 t o 0. b 16 . 17.4 0 A -50 . Although a good p a r t of t h e methane a p p e a r s t o be s t r i p p e d from t h e c r u d e as t h e carbon d i o x i d e d i s s o l v e s i n i t ./o O.> U u 0 2 -3e5 -34( Y) a .I1 .0 . A d d i t i o n a l work h a s shown t h a t when a l i v e c r u d e oil i s u s e d . O n l y a t t h e t a i l e n d of t h e r e c o v e r y is t h e r e some e v i d e n c e of t h e l i g h t e r f r a c t i o n b e i n g p r e f e r e n t i a l l y r e c o v e r e d . ' 5 3 i t s r e c o v e r y by c o n t i n u o u s carbon d i o x i d e i n j e c t i o n was n o t i c e a b l y l e s s t h a n t h e r e c o v e r y w i t h t h e o t h e r o i l s w h i c h h a d a v i s c o s i t y of l e s s t h a n 6 centipoises.I . ' f.O .'/ .ll.6 .. I t is i n f o r m a t i v e t o n o t e .7. .4 18 2. l o w e r s o l u b i l i t y and s w e l l i n g probably a l l c o n t r i b u t e t o t h e l o w e r recovery f o r t h e more v i s c o u s c r u d e . .../f .PI . t h e m o b i l e p h a s e c o n t a i n i n g a l a r g e f r a c t i o n of c a r b o n d i o x i d e a n d a s m a l l f r a c t i o n o f t h e l i g h t e n d s of t h e c r u d e o i l i s b e i n g p r e f e r e n t i a l l y produced. d d 0 0 . # *r. 12 0 .f' cJJ-e-J-~Mo /o-'/ 0 / -42. i t a l s o a p p e a r s t h a t f r a c t i o n a t i o n o c c u r s a t a somewhat l o w e r m o l a r concent r a t i o n of c a r b o n d i o x i d e a n d a more v o l a t i l e oil is p r o d u c e d s o m e w h a t e a r l i e r . Fig. / / 0 m - 46 12- /2' . W N / // / B 5. However t h e o v e r a l l r e c o v e r y d o e s n o t seem t o b e s u b s t a n t i a l l y a f f e c t e d by t h e p r e s e n c e of moderate amounts of methane i n t h e c r u d e o i l .278 T h i s m i x t u r e had a v i s c o s i t y of 20 c e n t i p o i s e s and a g r a v i t y of . B e c a u s e o f t h e f r a c t i o n a t i o n of t h e s y s t e m i n t o t w o or m o r e phases when t h e c a r b o n d i o x i d e c o n t e n t of t h e s y s t e m exceeds a mol f r a c t i o n .2 . C. f.I1 A h .

4 0. t h a t t h e comparis o n s t h a t w i l l b e made on t h e e f f i c i e n c y of t h e v a r i o u s s l u g s w i l l be f o r a ----. PV Fig. a n d 73OF.1 0. 130°F* .22 p o r e volume f o r a t o t a l i n j e c t i o n of 1. 18. when t h e s i z e o f t h e s l u g i s i n c r e a s e d a b o v e 0. t h e recovery does n o t i n c r e a s e a n y f u r t h e r .11 t o 0. a n d t h e r e s u l t s a r e p r e s e n t e d i n F i g u r e s 18 a n d 19. 19 THE EFFECT OF SLUG SIZE ON THE RECOVERY OF RESIDUAL CRUDE O I L 18 PSI. a n d t h e r e s i d u a l o i l s a t u r a t i o n w a s c o n s i s t e n t l y brought down t o 0.v.22 p o r e v o l u m e .. 0. The 4 5 O c r u d e w a s u s e d i n a l l t h e s e e x p e r i m e n t s .2 p o r e volumes.2 P. . a s l o n g a s t h e t o t a l f l u i d i n j e c t e d is l i m i t e d t o 1.279 F.6 0. carbon d i o x i d e o r carbon d i o x i d e and l i m i t e d volume water.of t o t a l f l u i d i n j e c t e d .1 L 3 o a t FLUID INJLCTED.7 FLUID INJECTED. 73OF.0 t o 1. THE EFFECT OF SLUG SIZE ON THE RECOVERY OF RESIDUAL CRUDE O I L 1400 P S I . t h e oil r e c o v e r y i n c r e a s e s l i n e a r l y w i t h a n i n c r e s e i n s l u g s i z e f r o m 0. However. I t is i m p o r t a n t t o n o t e i n t h e f o l l o w i n g d i s c u s s i o n . A s a m a t t e r of f a c t . The E f f e c t of S l u g S i z e The e f f e c t of u s i n g s l u g s of c a r b o n d i o x i d e on t h e recovery of r e s i d u a l c r u d e oil was s t u d i e d . PV as a e a 9 11) Fig. b e f o r e i n i t i a t i n g t h e test. F o r o p e r a t i n g c o n d i t i o n s of 1000 p s i .v. t h e r e c o v e r y a c t u a l l y d e c r e a s e s as t h e s l u g s i z e is i n c r e a s e d above a v a l u e of 0.2 a3 0. a n d t h e t o t a l f l u i d i n j e c t e d i s k e p t c o n s t a n t a t a b o u t one p o r e volume.21 p.22.

t h e r e b y d e f e a t i n g t h e e n t i r e p r o c e s s . which is r e l a t i v e l y i n e f f i c i e n t i n d i s p l a c i n g t h e s w o l l e n c r u d e o i l .8 0. w i l l be much more e f f i c i e n t l y d i s p l a c e d by a r e l a t i v e l y v i s c o u s f l u i d . On t h e o t h e r hand. The e f f i c i e n c y of t h e u l t i m a t e d i s p l a c e m e n t c a n be i n c r e a s e d s l i g h t l y i f t h e v i s c o s i t y of t h e c h a s e w a t e r i s i n c r e a s e d by t h e a d d i t i o n of a g l y c o l o r a polymer. v i z . any f u r t h e r i n j e c t i o n of carbon d i o x i d e r e s u l t s i n t h e development of a (dense) gas d r i v e . t h e optimum s l u g s i z e showed no c o n s i s t e n t change.26 p o r e volume. 7 3 O t o 130°F. a s e a r c h .3 (L6 FLUID INJECTED. 4 0 L I I I I 01 . a n d f r o m 1 0 0 0 p s i t o 1800 psi. n i t r o g e n is i n j e c t e d . PV 0. f o l l o w i n g t h e i n j e c t i o n of a n optimum s l u g of c a r b o n d i o x i d e .4 I (LEI 1 I 1 I 0.. 20. 0. Experiments proved t h a t n i t r o u s o x i d e performed i n t h e p h y s i c a l models i n a v i r t u a l l y i d e n t i c a l manner t o carbon dioxide. N i t r o u s Oxide r n o r d e r t o g a i n f u r t h e r c o r r o b o r a t i o n f o r t h e h y p o t h e s i s t h a t i t is t h e s w e l l i n g of t h e r e s i d u a l c r u d e o i l is t h e key f a c t o r i n t h e recovery of t h e l a t t e r by t h e i n j e c t i o n of carbon d i o x i d e .9 I Fig.2 0.. w a s made f o r o t h e r s u b s t a n c e s t h a t w o u l d d i s s o l v e t o t h e same e x t e n t a n d s w e l l t h e c r u d e o i l e q u i v a l e n t l y : n i t r o u s oxide h a s been d e s c r i b e d t o b e v i r t u a l l y e q u i v a l e n t t o c a r b o n d i o x i d e i n many p h y s i c a l p r o p e r t i e s 1 8 . t h e n t h e s w o l l e n crude o i l phase. ( I t is a f a r more e x p e n s i v e substance. t h e n t h e r e s u l t i n g r e c o v e r y o f o i l is markedly reduced. rendered mobile by t h e i n c r e a s e i n i t s p o r e volume s a t u r a t i o n . The i n c r e a s e d e f f i c i e n c y of s l u g s o f c a r b o n d i o x i d e i n r e c o v e r i n g r e s i d u a l crude o i l is w e l l i l l u s t r a t e d by t h e r e s u l t s of t h i s work which are p l o t t e d i n F i g u r e 20.20 t o 0. THE EFFICIENCY OF SLUGS OF CARBON DIOXIDE I N RECOVERING RESIDUAL CRUDE O I L G.2 80 Over t h e range of t e m p e r a t u r e s and p r e s s u r e s i n v e s t i g a t e d i n t h e s t u d y of t h e s l u g s of c a r b o n d i o x i d e . I f . The n i t r o g e n i s a n i n e f f i c i e n t d i s p l a c i n g f l u i d . water. Again. see F i g u r e 21. I t i s h y p o t h e s i z e d t h a t t h e o p t i m u m s l u g s i z e i s t h a t v o l u m e of c a r b o n d i o x i d e w h i c h c a n b e i n j e c t e d i n t o t h e s y s t e m w i t h o u t e s t a b l i s h i n g a f r e e and c o n t i n u o u s s a t u r a t i o n throughout t h e e n t i r e model. i f carbon d i o x i d e i n j e c t i o n is h a l t e d b e f o r e a f r e e gas phase s a t u r a t i o n i s e s t a b l i s h e d throughout t h e model.) .7 0. i t m u s t be n o t e d t h a t t h e r e s e r v o l r p r o t o t y p e m o d e l l e d i n t h i s work i s o n e w h i c h s h o u l d show up c a r b o n d i o x i d e a t i t s very best. Once such a mobile g a s s a t u r a t i o n is e s t a b l i s h e d . moreover It s t r i p s some of t h e d i s s o l v e d c a r b o n d i o x i d e f r om s o l u t i o n I n t h e c r u d e o i l . i t ranged from 0.

THE INEFFICIENCY OF NITROGEN AS A CHASE FLUID AFTER A C02 SLUG IV./-- I. I I I I I Nitrogen S. mobile water.F S T oll = 2 3 e . viz. A l t h o u g h v a l u e s a p p r o a c h i n g 5 MSCF/B h a v e b e e n a c h i e v e d i n t h e s e model s t u d i e s .mhn * 150. Although t h e r e s i d u a l s a t u r a t i o n of t h e o i l phase ( a s o l u t i o n of carbon d i o x i d e i n o i l ) c a n b e l o w e r e d by c o n t i n u i n g t h e f l o w of c a r b o n d i o x i d e . i t is c a u t i o n e d t h a t t h e model used was a v e r y f a v o r a b l e one. and t h e d e c r e a s e i n t h e v i s c o u s t o g r a v i t y f o r c e s encountered I n r a d i a l f l o w away from t h e w e l l b o r e s w i l l encourage g r a v i t y s e g r e g a t i o n .~ ~ i h * 335 SCF/BbI i - s $ 0 =24- - A%=OOm . riuq size * 0 3 PV Pmssun 2500 psi 48- U W -c Y U s P - Tmp.o c c l u d i n g . Even minor h e t e r o g e n e i t y i n a f i e l d o p e r a t i o n w i l l encourage channeling. and 3) t h e i m m i s c i b l e d i s p l a c e m e n t of t h e mobile s o l u t i o n of carbon d i o x i d e i n o i l by t h e c o n t i n u i n g f l o w of carbon d i o x i d e or water. 21. Continuous i n j e c t i o n of c a r b o n d i o x i d e w i l l r e c o v e r a s i g n i f i c a n t f r a c t i o n of a w a t e r f l o o d r e s i d u a l o i l s a t u r a t i o n . * 37 PV Water Sa s 28 Pv d - 56- 0 CO. t h e .281 72 - I I 64- - -- I .------- I ____ ____----. l o w p e r m e a b i l i t y . r e s u l t i n g i n some c o n t i n u i n g e v a p o r a t i o n o f c r u d e o i l f r a c t i o n s . 2) t h e s o l u t i o n of carbon d i o x i d e i n t h e crude o i l and i t s subsequent s w e l l i n g t h a t develops o i l phase m o b i l i t y .. and c o n s i s t e n t w i t h t h e p r i n c i p l e s of f l u i d f l o w and phase behavior.. b u t t h e r e s u l t i n g carbon d i o x i d e / o i l r a t i o s w i l l b e above 20 MSCF/B. uniform and l i n e a r . The performance of t h e d i s p l a c e m e n t e x p e r i m e n t s l e a d s t o t h e conclusion t h a t t h e mechanism by which carbon d i o x i d e d i s p l a c e s r e s i d u a l crude o i l is comprised of t h r e e s e q u e n t i a l s t e p s : 1) t h e i m m i s c i b l e d i s p l a c e m e n t of t h e o i l . The u s e of s l u g s of carbon d i o x i d e f o l l o w e d by water w i l l e f f e c t i v e l y reduce t h e r e s u l t i n g carbon d i o x i d e / o i l r a t i o w i t h o u t s e r i o u s l y a f f e c t i n g t h e amount of o i l t h a t can be r e c o v e r e d by t h e i n j e c t i o n of a t o t a l of about o n e p o r e v o l u m e o f f l u i d . and may b e as high as 30.= . CONCLUSIONS P h y s i c a l l y s c a l e d m o d e l s t u d i e s of t h e d i s p l a c e m e n t a n d r e c o v e r y of c r u d e O i l by c a r b o n d i o x i d e y i e l d r e s u l t s w h i c h a r e c o n s i s t e n t w i t h t h e r e s u l t s of f i e l d d e m o n s t r t i o n and p i l o t p r o j e c t s . AaSOlO I It AS0 * 0060 00 0 1 07 03 04 05 06 07 08 09 I0 I2 Fig.

V. (1976) 7.W..A.J.. a n d e n d o w m e n t f u n d s a t t h e U n i v e r s i t y o f Southern C a l i f o r n i a .C. M. TRANS AIME (1954) 201. O f f e r i n g a .M. L. I.. E l A r a b i . J. " S t u d y o f C o n d i t i o n s o f P e t r o l e u m . JPT. " D e t e r m i n a t i o n a n d P r e d i c t i o n o f C O P Minimum M i s c i b i l i t y P r e s s u r e " . Jr. "Carbon D i o x i d e f o r S o l v e n t F l o o d i n g f o r I n c r e a s e d O i l Recovery".282 r e s u l t i n g i n c r e m e n t a l c a r b o n d i o x i d e / p r o d u c e d o i l r a t i o s w i l l b e v e r y high. N a t i o n a l P e t r o l e u m C o u n c i l . N e f t . S t a l k u p .. (1979). ( 1 9 8 0 ) . 4 6 t h A n n u a l Meeting of SPE of AIME (1971) 4. " D i s p l a c e m e n t o f O i l F r o m P o r o u s . which d i s s o l v e s i n a n d swells c r u d e o i l s s i m i l a r l y . Holm. D. R. a n d P a t e l .. ( 1 9 7 8 ) .W. The more p r a c t i c a l l i m i t t o t h e r e c o v e r y i s r e a c h e d when t h e r e s i d u a l s a t u r a t i o n of t h e low v i s c o s i t y o i l p h a s e t o t h e s u b s e q u e n t g a s o r water d r i v e i s approached. 59-61 6. ACKNOWLEDGEMENTS The work on t h i s p r o j e c t w a s s u p p o r t e d by t h e U n i t e d S t a t e s Department o f E n e r g y .. a n d v a n d e r P o e l . H. J. Media by M i s c i b l e L i q u i d s " . M... R. U n i v e r s i t y of S o u t h e r n C a l i f o r n i a . REFERENCES 1. JPT. 3. G a r y E n e r g y Co. PbD. The s o . 5 4 t h Annual M e e t i n g of SPE of AIME. a n d O r t l o f f . 5 D u n n y s h k i n . A. 310-317 11.I.D.. "The E f f e c t of P h a s e B e h a v i o r on CO2 Flood D i s p l a c e m e n t E f f i c i e n c y " .D. W a r n e r . W. R a t h m e l l . i s as e f f e c t i v e as c a r b o n d i o x i d e i n r e c o v e r i n g c r u d e o i l . Holm. 1417-1438. M i s c i b i l i t y w i t h Carbon Dioxide". S l i m t u b e e x p e r i m e n t s s i n c e t h e y do n o t c o r r e c t l y model t h e d i s p e r s i o n c o e f f i c i e n t s a n d t h e r e l a t i o n s b e t w e e n g r a v i t y a n d v i s c o u s f o r c e s do n o t provide adequate i n s i g h t i n t o a r e s e r v o i r recovery process. P.I. Y e l l i g .. L a s Vegas. N i t r o u s o x i d e . "A L a b o r a t o r y I n v e s t i g a t i o n o f M i s c i b l e D i s p l a c e m e n t b y C02".F. SPE 8367.. 8.. "Mechanism of O i l D i s p l a c e m e n t by Carbon Dioxide". Law. D i s s e r t a t i o n .. Washington.c a l l e d minimum m i s c i b i l i t y p r e s s u r e a s i n t e r p r e t e d f r o m s u c h e x p e r i m e n t s i s a c t u a l l y t h e p r e s s u r e above w h i c h no s i g n i f i c a n t i n c r e a s e i n r e c o v e r y w i l l b e achieved. 38891 2.. 10. Orr.C. (1979) 9. a n d N a m o i t . TRANS AIME (1959) 216. The s u b s t i t u t i o n of n i t r o g e n f o r water as a c h a s e f l u i d i n j u r e s t h e r e c o v e r y b e c a u s e t h e g a s i s n o t as good a d i s p l a c i n g a g e n t f o r t h e s w o l l e n c r u d e .. R... P.. The complex p h a s e b e h a v i o r of c a r b o n d i o x i d e w i t h c r u d e o i l a p p e a r s t o c o n t r i b u t e l i t t l e t o t h e r e c o v e r y p r o c e s s . and J o s e n d a h l . "An E v a l u a t i o n o f W i s c i b l e C02 F l o o d i n g i n W a t e r f l o o d e d S a n d s t o n e R e s e r v o i r s " . a n d M e t c a l f e . 160-168.. C. JPT ( 1 9 7 4 ) . TRANS AIME (1959) 216. D. '!Laboratory I n v e s t i g a t i o n of t h e WaterD r i v e n Carbon D i o x i d e P r o c e s s f o r O i l Recovery". 225-231 3.. SPE 3 4 8 3 . Beeson.S.. J u n e 1981. t h e e f f e c t of t h e f r a c t i o n a t i o n of t h e c r u d e i n t h e p r e s e n c e o f c a r b o n d i o x i d e r e s u l t s i n s o m e s l i g h t a d d i t i o n a l r e c o v e r y a t t h e t a i l e n d of t h e f l o o d . The r e c o v e r y mechanism is s t i l l e f f e c t i v e a t l o w e r p r e s s u r e s . v. a n d H a s s i n g e r . C. Khoz. 1339-1347 - - . J. G a r d n e r . F. "Enhanced O i l Recovery An A n a l y s i s of t h e P o t e n t i a l f o r Enhanced O i l Recovery f r o m Known F i e l d s i n t h e U n i t e d S t a t e s 1976 t o 2000.

C o m b e r i a t i . E.S. K a n e .Department of Energy. a n d Z a m m e r i l l i . T u l s a . SACROC U n i t . W. S e e a l s o .H. P a p e r N4. D.. A. 1940.M.. 352-61 13. D i s p l a c e m e n t of R e s d i u a l Oil by M i s c i b l e CO2 i n L i n e a r Geometry".. New York (1950). SPE 8896. p. . T.. " T a r g e t R e s e r v o i r s f o r C O P M i s c i b l e F l o o d i n g " .S. " P h y s i c a l l y S c a l e d M o d e l s S i m u l a t i n g t h e .. G l a s s t o n e . J. K a m a t h .. Van Nostrand. S.I. Washington. 713. G e r r a r d . K.. SPE I m p r o v e d O i l F i e l d Recovery Symposium. D o s c h e r . G r u y F e d e r a l . R. SPEJ ( 1 9 7 2 ) . T e x t Book of P h y s i c a l C h e m i s t r y .E l e c t r o l y t e s . H i l d e b r a n d . A. D. Plenum P r e s s .. N e w Y o r k ( 1 9 7 6 ) . 5 0 t h Annual C a l i f o r n i a R e g i o n a l M e e t i n g of SPE of AIME (1980) 1 5 . A G r a p h i c A n a l y s i s .L. R e i n h o l d .283 12.C. (1980) 17. . T u l s a 1978 16. U. " P e r f o r m a n c e Review o f a L a r g e S c a l e C a r b o n Dioxide-WAG P r o j e c t . 14.Department of Energy Symposium. New York. a n d G h a r i b .K e l l y S n i d e r F i e l d . " D i s c u s s i o n of t h e Use of C a p i l l a r y T u b e N e t w o r k s i n R e s e r v o i r Peformance S t u d i e s " . SPE 7 0 9 1 . S o l u b i l i t y of Gases a n d L i q u i d s .R..L. S. Inc. "The R o l e of R e s e r v o i r T e m p e r a t u r e i n Carbon D i o x i d e F l o o d i n g " . C l a r i d g e . Oklahoma 1979 18. p r e s e n t e d a t t h e U. a n d S c o t t ..V. T h e S o l u b i l i t y of N o n . J.

This Page Intentionally Left Blank .

and be b e t t e r a b l e t o e v a l u a t e l a b o r a t o r y results i n l i g h t of t h e i r f i e l d experience. General o b j e c t i v e s f o r each type of test are l i s t e d . Some of t h e f a c t o r s that have t o be s t u d i e d are t h e reservoir geology. In t h i s paper.MISCIBLE GAS DISPLACEMENT 285 LABORATORY TESTING PROCEDURES FOR MISCIBLE FLOODS S. The t o p i c s t r e a t e d include s i n g l e . I r v i n e . These w i l l i n c l u d e t h e measurement of t h e phase behavior and d i s placenent d a t a of r e s e r v o i r crude o i l . An a d d i t i o n a l a i m of t h e paper is t o give r e s e r v o i r and production e n g i n e e r s i n s i g h t i n t o t h o s e procedures. C a l i f . The o b j e c t i v e of t h i s paper is t o provide a state-of-the-art review and c r i t i q u e f o r r e s e a r c h e r s i n the f i e l d . apd economics. g e n e r a l s c r e e n i n g criteria are presented f o r the s e l e c t i o n of s u i t a b l e candidate r e s e r v o i r s f o r m i s c i b l e flooding.t h e . U. G. The process is complex and involves many parameters t h a t have t o be optimized so t h a t a flood can l e a d t o a t e c h n i c a l and economic success. McCAFFERY* Petroleum Recovery Institute. G.A.Alberta. o i l solvent displacement characteristics. as w e l l as r e s e r v o i r engineering a s p e c t s such as s o l v e n t production and o i l i n j e c t i o n s t r a t e g i e s . SAYEGH and F.o f . recommended p r a c t i c e s are o u t l i n e d . .s o l v e n t systems. waterflood performance.a r t review and c r i t i q u e of l a b o r a t o r y t e s t i n g procedures f o r m i s c i b l e f l o o d i n g f o r r e s e a r c h e r s irr t h e f i e l d . l a b o r a t o r y t e s t i n g procedures f o r m i s c i b l e flooding w i l l be d i s c u s s e d .s o l v e n t phase behavior. o i l and o i l .S . Chlgury. IXTRODUCTIOY One of t h e p r i n c i p a l enhanced recovery methods c u r r e n t l y under consideration f o r l i g h t o i l r e s e r v o i r s is miscible f l o o d i n g w i t h carbon dioxide and/or hydrocarbon s o l v e n t s . An a d d i t i o n a l aim of t h e paper is t o g i v e r e s e r v o i r and production engineers i n s i g h t i n t o l a b o r a t o r y t e s t i n g procedures SO that they may a p p r e c i a t e t h e i r p o t e n t i a l s and l i m i t a t i o n s and t h u s be b e t t e r a b l e t o e v a l u a t e l a b o r a t o r y results i n l i g h t of t h e i r f i e l d experience. expected performance under b o t h water and solvent f l o o d i n g . and involve slim-tube and c o r e displacement t e s t s . 80 t h a t they may a p p r e c i a t e t h e i r p o t e n t i a l s and l i m i t a t i o n s . Gnuah T2L 2A6 ABSTRACT The o b j e c t i v e of t h i s paper is t o provide a s t a t e . and many examples from t h e l i t e r a t u r e are referenced. For o t h e r * P r e s e n t address: Occidental Research Cozporation. and how such d a t a may be used in e v a l u a t i n g t h e s u i t a b i l i t y of a solvent flood f o r a p a r t i c u l a r a p p l i c a t i o n . I n a d d i t i o n .and multiple-contact phase behavior and p h y s i c a l p r o p e r t i e s measurements. 92713.

I n g e n e r a l .achieves m i s c i b i l i t y w i t h . b u t o f t e n i n the form of a s l u g t y p i c a l l y about 20-30% o f t h e hydrocarbon pore volume (HCPV). When s u b c r i t i c a l s o l v e n t s a r e used a t p r e s s u r e s above t h e i r bubble p o i n t . c o m n l y c a l l e d t h e water-alternating-gas (WAG) process. and low o i l recovery.286 ze. PROCESS DESCRIPTION AND GENERAL SCREENING CRITERIA I n a m i s c i b l e f l o o d t h e s o l v e n t c o n t a c t s t h e o i l and a mixing zone is formed.o i l viscosity reduction . I n t h e mixing zone.is readily available t h e o i l a t r e s e r v o i r conditions .good h o r i z o n t a l c o n t i n u i t y .c o n t a i n s no f r e e g a s s a t u r a t i o n .h o r i z o n t a l p e r m e a b i l i t y . I t i s r a r e l y t e c h n i c a l l y or economically f e a s i b l e t o i n j e c t a s o l v e n t that i s d i r e c t l y m i s c i b l e w i t h t h e o i l . t h e p r o c e s s is one of l i q u i d l i q u il e x t r a c t ion1 ’ 1 5 . t h e r e is a gradual change in composition from o i l t o s o l v e n t . w i t h o u t a n i n t e r f a c e . poor sweep e f f i c i e n c y . t h e s o l v e n t i s u s u a l l y n o t i n j e c t e d c o n t i n u o u s l y . The s l u g is t h e n followed by a chase f l u i d . m i s c i b i l i t y is g e n e r a l l y achieved through what are known a s t h e m l t i p l e . the reader is r s f e r r e d t o t h e works by Holm’.high microscopic displacement .:irvs of t h e E i s c i b l e f l o o d i n g process and i t s f i e l d a p p l i c a t i o n s . The s l u g may be i n j e c t e d i n small p o r t i o n s a l t e r n a t i n g w i t h water. w a t e r may be co-injected w i t h t h e s o l v e n t . Burnett and DamC have reviewed s c r e e n i n g tests f o r a v a r i e t y of enhanced o i l recover:* p r o c e s s e s .3. I n s t e a d . a good c a n d i d a t e r e s e r v o i r f o r h o r i z o n t a l m i s c i b l e f l o o d i n g should have t h e f o l l o v i n g characteristics: .contains undersaturated o i l . These latter i n j e c t i o n modes h e l p c o n t r o l t h e h i g h m o b i l i t y of t h e s o l v e n t . A l t e r n a t i v e l y .t h i n pay zone.t o .o i l e x t r a c t i o n by s o l v e n t . Two such mechanisms can occur when gaseous or s u p e r c r i t i c a l s o l v e n t s a r e used: a condensation mechanism and a v a p o r i z a t i o n mechanism. Stalkup2.low v e r t i c a l . u s u a l l y water o r l e a n gas. This c a u s e s t h e s o l v e n t t o o v e r r i d e t h e o i l and f i n g e r through i t . up t o 5 m . For economic r e a s o n s . t o d r i v e i t through t h e r e s e r v o i r towards t h e production w e l l s .c o n t a i n s no mobile water ?ins s o l v e n t should be chosen such that i t : ratio .o i l swelling . and M~ngan‘’~.c o n t a c t m i s c i b i l i t y (MCM) mechanisms7-13.l o r i n t e r f a c i a l tension .r e l a t i v e l y homogeneous .blowdown recovery efficiency The s o l v e n t s used (C02 and hydrocarbons) are g e n e r a l l y less dense and v i s c o u s t h a n the o i l s .is cheap . The high o i l recovery i n m i s c i b l e f l o o d s i s a t t r i b u t e d t o t h e following factors: . Dosher e t a1. These are a d v e r s e f a c t o r s i n h o r i z o n t a l f l o o d s and lead t o e a r l y s o l v e n t breakthrough.Zot f r a c t u r e d .

and volumetric p r o p o r t i o n s of c o e x i s t i n g phases as f u n c t i o n s of p r e s s u r e . t h e phase behavior s t u d i e s i n v o l v e t h e f o l l o w i n g measurements: .densities . and of D.287 For v e r t i c a l downward displacements. This p e r m i t s unhindered v i s u a l o b s e r v a t i o n o f t h e e n t i r e cont e n t s of the c e l l . The c e l l i s placed i n a t h e n m s t a t e d oven f o r temperature c o n t r o l . Samples of t h e s e phases may a l s o be withdrawn f o r d e n s i t y and v i s c o s i t y measurements.m u l t i p l e phase . of Connor and Pope25. Robinson have t h e i r sampling l i n e s d i r e c t l y connected t o gas chromatographs f o r analysis. 17-26 A connuon t y p e o f a p p a r a t u s c o n s i s t s o f a windowed c e l l whose volume may be manipulated by means of a p i s t o n o r mercury from a p o s i t i v e displacement pump.s o l v e n t system determine t h e mechanism by which m i s c i b i l i t y is achieved f i n e . o r by c i r c u l a t i n g t h e f l u i d s . Once e q u i l i b r i u m has been reached.26 has two interconnected This g i v e s a greater f l e x i b i l i t y of o p e r a t i o n and p e r m i t s t h e measurement of v i s c o s i t y w i t h o u t using a s e p a r a t e viscometer.16 Apparatuses used i n connection w i t h C02 and hydrocarbon systems were described by o t h e r . The a p p a r a t u s described by Orr e t a l .to . Mixing is u s u a l l y done w i t h a magnetically-coupled stirrer. cells. and f o r compositional a n a l y s e s .the . The a p p a r a t u s e s of Orr e t a l .to . 2 4 d i f f e r s from t h e o t h e r s i n t h a t i t resembles a continuously s t i r r e d t a n k r e a c t o r . i n c l u d i n g both l i q u i d and s o l i d phases Phase Behavior Measurement Equipment High p r e s s u r e phase e q u i l i b r i u m experimental techniques f o r a v a r i e t y of a p p l i c a t i o n s have r e c e n t l y been reviewed by Eubank e t a1. The a p p a r a t u s d e s c r i b e d by LeeP3 and Sayegh e t a1.solubilities . Constant composition expansions may a l s o be c a r r i e d o u t t o determine bubble and dew p o i n t s . t h e requirements a r e somewhat l e s s constraining: .viscosities formation. v i s u a l o b s e r v a t i o n s of t h e c o e x i s t i n g phases may be c a r r i e d o u t .t u n e t h e phase behavior packages i n compositional s i r m l a t o r s I n g e n e r a l . Robinson (personal communication) a t t h e U n i v e r s i t y of A l b e r t a has t h e c e l l constructed e n t i r e l y from sapphire. The d e s i r e d components of t h e mixture are loaded i n t o t h e cell and t h e n mixed. by rocking t h e c e l l . r e s e a r c h e r s . .to c h a r a c t e r i z e t h e o i l . 2 4 .the r e s e r v o i r should not c o n t a i n p e r m e a b i l i t y b a r r i e r s t o v e r t i c a l flow displacement should be c a r r i e d o u t a t a s u i t a b l e r a t e such that t h e flood is g r a v i t y s t a b l e PHASE BEHAVIOR MEASUREMENTS Phase behavior measurements are c a r r i e d o u t f o r s e v e r a l purposes: . D.o i l swelling .

The w e l l should be produced a t a slow r a t e d u r i n g sampling t o minimize p r e s s u r e drawdown e f f e c t s and t h e r e s u l t a n t phase changes. Thus. ASTM s t a n d a r d procedures and are p r e f e r a b l y c a r r i e d o u t w i t h bottomh o l e samples. d i s t i l l a t i o n curve.laltene p r e c i p i t a t i o n could leqd t o a r e d u c t i o n i n r e s e r v o i r permeability. Nore a c c u r a t e p r e d i c t i o n s of t h e displacement p a t h may be made u s i n g a q u a t e r n a r y diagram. d i f f e r e n t i a l l i b e r a t i o n . T e s t i n g of o i l p r o p e r t i e s should b e p e r i o d i c a l l y repeated d u r i n g t h e production l i f e t i m e o f a r e s e r v o i r . ~ ”Data f o r hydrocarbon f l o o d s o f r e s e r v o i r crudes were presented by ~ ~ s e v e r a l author^^'^'^^'^^. s u f f i c i e n t sampling time should be allowed t o ensure t h a t t h e sample bomb i s f i l l e d w i t h f r e s h oil. These tests are g e n e r a l l y c a r r i e d o u t a t r e s e r v o i r temperature using. I n g e n e r a l . t h e s l u g could break down. d e n s i t y . and c o n s t a n t composition expansion charac t e r i s tics. Also. while asp. however. f o r example. ~ ’ ~ ~ ’ ~ ~ ’ ~ ~ T n i s provides information on t h e phase b e t a v i o r of t h e C02-011 s y s t e m i n t h e v a r i o u s l o c a t i o n s of t h e r e s e r v o i r where t h e p r e s s u r e may vary. . Large volumes of r e s e r v o i r f l u i d s a r e necessary t o c a r r y o u t a complete l a b o r a t o r y study of a m i s c i b l e flood. The phase diagrams of C02crude o i l s stems a r e o f t e n presented i n t h e form o f t e r n a r y phase d i a g r a n s . g a s . ’ A second type o f test is t h e c o n s t a n t c o n p o s i t i o n e ~ p a n s i o n . molecular weight.ll p12’14p18’23’24’26p30-38 This r e f l e c t s t h e growing i n t e r e s t i n u s i n g C02 as a m i s c i b l e f l o o d i n g a g e n t . a t c o c i i t i o n s where m l t i p l e l i q u i d phases appear. Standingl’l and Henry e t a1. This is e s p e c i a l l y Important where t h e r e s e r v o i r p r e s s u r e f a l l s below t h e o r i g i n a l bubble p o i n t o f t h e o i l . compressib i l i t y . % It should.o i l r a t i o . bubble p o i n t .28 presented d e s c r i p t i o n s of bottomhole sampling procedures. 25 Phase Behavior Tests t o C h a r a c t e r i z e t h e Crude Oil-Solvent System The g e n e r a l phase behavior of hydrocarbon f l u i d s have been w e l l re~ i e w e d . 9 p 1 2 y Y 4 y 2 4Such a r e p r e s e n t a t i o n provides a convenient form f o r t h e v i s u a l i z a t i o n of t h e com o s i t i o n a l p a t h d u r i n g a c o n s t a n t temperature and p r e s s u r e d i s lacementl 11g7 and f o r determining t h e mechanism of a c h i e v i n g m i s c i b i l i t y . i t i s unreasonable t o u s e bottomhole samples f o r a l l t h e s e tests. formation volume f a c t o r . t h e sampling w e l l should be s e l e c t e d so that i t is r e p r e s e n t a t i v e o f t h e average r e s e r v o i r c o n d i t i o n s . b e remembered t h a t t h e t e r n a r y r e p r e s e n t a t i o n is n o t thermodynamically r i g o r o u s and hence should n o t be i n t e r p r e t e d l i t e r a l l y .288 Phase Behavior T e s t s t o C h a r a c t e r i z e t h e Crude O i l Typical t e s t s f o r t h e c h a r a c t e r i z a t i o n of t h e crude o i l involve t h e measurement of i t s composition. v i s c o s i t y . c o n s t a n t volume d e p l e t i o n . w h i l e most of t h e r e c e n t l y published s t u d i e s have d e a l t with t h e phase behavior o f C02-011 systems. For example. The following d i s c u s s i o n w i l l c o n c e n t r a t e on C02-reservoir crude o i l systems s i n c e t h e s e are o f most i n t e r e s t t o t h e i n d u s t r y . then recombine them t o n a t c h t h e p r o p e r t i e s o f t h e bottomhole sample. The normal procedure i s t o t a k e l a r g e samples of s e p a r a t o r o i l and gas. and be c a r r i e d o u t on samples from t h e d i f f e r e n t producing zones o r horizons o f a pool t o determine i f t h e r e are any v a r i a t i o n s i n o i l p r o p e r t i e s .

I n g e n e r a l . 39 The procedure d e s c r i b e d by O r r et a1. t r u e e q u i l i b r i u m may n o t be a t t a i n e d and t h e p r e d i c t i o n w i l l be o p t i m i s t i c . The f i r s t t y p e o f t e s t is t h e g e n e r a t i o n o f a Benham p l o t by a stagewise approximation o f t h e continuous m u l t i p l e . It i s important t o c a r r y o u t b o t h t y p e s o f tests i n a l a b o r a t o r y s t u d y s i n c e each one p r o v i d e s d i f f e r e n t i n f o r m a t i o n n e c e s s a r y f o r t h e e v a l u a t i o n o f a f i e l d a p p l i c a t i o n . based on changes i n phase volume. LABORATORY DISPLACEMENT TESTS Laboratory displacement tests p r o v i d e i m p o r t a n t i n f o r m a t i o n on t h e behavior of r e s e r v o i r f l u i d / s o l v e n t systems under dynamic displacement c o n d i t i o n s . T h i s is probably a r e a s o n a b l e assumption i n many cases s i n c e r e s e r v o i r flow rates a r e q u i t e low. These tests a r e o f two t y p e s : slim-tube and c o r e d i s p l a c e m e n t s . As such. f i n g e r i n g .289 The d e n s i t y .o i l systems. and v i s c o s i t y of t h e C02-saturated 0 i 1 1 8 ’ 2 6 ’ 3 1 a r e u s u a l l y measured i n g a r a l l e l w i t h t h e phase-envelope measurements d e s c r i b e d above. Phase Behavior T e s t s t o Determine t h e Mechanism o f M u l t i p l e Contact M i s c i b i l i t y The tests mentioned p r e v i o u s l y are a l l s t a t i c . a c e r t a i n p r o p o r t i o n o f o i l and s o l v e n t are mixed i n a PVT c e l l and allowed t o reach e q u i l i b r i u m . more s o l v e n t g a s d i s s o l v e s i n t o t h e o i l c a u s i n g i t t o swell and t h u s t o reduce i t s d e n s i t y and v i s c o s i t y . The e n t i r e p r o c e s s i s r e p e a t e d u n t i l o n l y one phase appears i n t h e c e l l . Each t y p e of t e s t w i l l now be d i s c u s s e d i n f u r t h e r d e t a i l . t h e rate o f s o l v e n t i n j e c t i o n i n t o t h e c e l l would have t o be c a r e f u l l y s e l e c t e d t o o b t a i n meaningful r e s u l t s . R e s e n r o i r parameters such as t h e m o b i l i t i e s o f t h e h a s e s and flow rates should be taken i n t o account t o determine a realistic ratio. a v a p o r i z a t i o n p r o c e s s is involved. These tests are c a r r i e d o u t i n a c o n t r o l l e d manner i n a PVT c e l l . b u t i f s e v e r e c h a n n e l l i n g . On t h e o t h e r hand. as t h e p r e s s u r e i n c r e a s e s . The tests d e s c r i b e d i n t h i s s e c t i o n are designed t o s i m u l a t e t h e dynamic. o r g r a v i t y s e g r e g a t i o n o c c u r i n t h e r e s e r v o i r . i t i s i m p l i c i t l y assumed t h a t t h e o i l and s o l v e n t i n t h e r e s e r v o i r have enough time t o reach e q u i l i b r i u m . t h e l i q u i d phase i s purged i f . s w e l l i n g f a c t o r . t h e d e n s i t y and v i s c o s i t y curves could s t a r t i n c r e a s i n g because the e f f e c t o f p r e s s u r e on t h e f l u i d p r o p e r t i e s predominates o v e r t h e e f f e c t o f s o l v e n t d i s s o l u t i o n . Carbon d i o x i d e i s g e n e r a l l y more e f f e c t i v e i n t h i s r e g a r d t h a n hydrocarbon s o l v e n t g a s e s . j 6 A t v e r y h i g h p r e s s u r e s . .24 i s a v a r i a t i o n o f t h e above method i n t h a t t h e n u l t i p l e c o n t a c t s are c a r r i e d o u t c o n t i n u o u s l y . m u l t i p l e c o n t a c t p r o c e s s o c c u r r i n g i n a r e s e r v o i r between t h e i n j e c t e d s o l v e n t and the r e s e r v o i r crude o i l . Connor and Pope2 r e c e n t l y p r e s e n t e d such d a t a f o r h y d r o c a r b o n . and a f r e s h b a t c h o f o i l is i n t r o duced i n t o t h e c e l l . Another problem a s s o c i a t e d w i t h d e s i g n i n g t h i s t y p e o f batchwise experiment is t h e c h o i c e o f v o l u m e t r i c r a t i o s o f gas-to-liquid c o n t a c t e d i n each s t e p . a t which p o i n t MCM h a s been a t t a i n e d .’ 2’ 9’ 24’2 5 ’ 39 I n t h i s procedure. I n such a n e x p e r i ment. which o n l y approximates t h e continuous c o n t a c t s i n a r e s e r v o i r . The p r o p o r t i o n s and p r o p e r t i e s o f t h e r e s u l t a n t vapor and l i q u i d a r e t h e n measured. t h e vapor phase is t h e n purged and a f r e s h b a t c h o f s o l v e n t is i n t r o d u c e d i n t o t h e c e l l . I f a condensation p r o c e s s o c c u r s . s i n g l e . t h u s t h e p r o c e s s p a r a m e t e r s a r e w e l l d e f i n e d .c o n t a c t tests.c o n t a c t process. ’ The drawback o f t h i s method is t h a t i t is a s t a g e w i s e p r o c e s s .

OOIP) . t h e c r i t e r i a f o r m i s c i b i l i t y b e i n g achieved i n a carbon d i o x i d e f l o o d a r e : . breakthrough l a t e s o l v e n t b r e a k t h r o u g h ( a t around 0. A l a b o r a t o r y i n v e s t i g a t i o n i n v o l v i n g a series o f r u n s ' c o u l d be done w i t h e i t h e r o r b o t h o f t h e f o l l o w i n g objectives : . l y i n g between t h e wre r e a l i s t i c c o r e f l o o d s and t h e more i d e a l i s t i c m u l t i p l e .solvent screening p r e s s u r e ( E ) etermination M'd Orr et al. a s t h e f l o o d i n g a g e n t . w h i l e t h e second f l o o d would i n v o l v e f i r s t . or later) a s m o t h t r a n s i t i o n from o i l t o s o l v e n t i n t h e mixing zone w i t h o u t t h e a m e a r a n c e of a n i n t e r f a c e h i g h u l t i m a t e recovery ( g r e a t e r t h a n 95% o f t h e o r i g i n a l o i l . composition and g a s . A s l i m tube t est is c a r r i e d o u t p r i m a r i l y t o determine i f a s o l v e n t a c h i e v e s m i s c i b i l i t y w i t h a n o i l a t a c e r t a i n temperature and p r e s s u r e . f o r example. number o f p h a s e s . The s l i m t u b e is normally c o n s t r u c t e d from h o r i z o n t a l l y c o i l e d s t a i n l e s s steel t u b i n g . as the d i s p l a c i n g a g e n t .24' 31'40'41 y 4 2 .T h i s i s due i n p a r t to t h e uniformi t y o f t h e packing and t h e dampening e f f e c t o f t h e t u b e ' s w a l l s . The f i r s t f l o o d could be conducted under d e f i n i t e l y immiscible c o n d i t i o n s u s i n g n i t r o g e n .i n .e a r l y breakthrough . For f u r t h e r d i s c u s s i o n s . a s w e l l a s o i l recovery and p r e s s u r e drop a c r o s s t h e coil--each as a f u n c t i o n o f t h e volume o f solvent injected.c o n t a c t m i s c i b l e c o n d i t i o n s u s i n g benzene. The b a s i c assumption i n slim-tube tests i s that t h e displacement i s p i s t o n l i k e and t h a t l i t t l e o r no f i n g e r i n g o c c u r s . The t u b e i s 10-20 m l o n g .p l a c e . The d a t a o b t a i n e d from t h e test i n c l u d e e f f l u e n t c o l o r . t h e y may be thought of a s b e i n g a n i n t e r mediate approximation t o r e s e r v o i r f l o o d s . The e f f l u e n t from t h e slim-tube p a s s e s through a s i g h t g l a s s f o r visual o b s e r v a t i o n .Slim-Tube Displacement Tests Slim-tube displacement tests are l a b o r a t o r y tests that are c a r r i e d o u t i n a n i d e a l i z e d porous medium. u n i f o r m i t y of bead s i z e and packing). I t would be i n s t r u c t i v e t o c a r r y o u t two i n i t i a l d i s p l a c e m e n t s t o c h a r a c t e r i z e t h e p a r t i c u l a r s l i m tube b e i n g used.. a n i m i s c i b l e displacement is c h a r a c t e r i z e d by: . As such. f o r example.r i c h and s o l v e n t . This a l s o depends t o some e x t e n t on t h e c h a r a c t e r i s t i c s o f t h e s l i m t u b e i t s e l f ( t u b e d i a m e t e r . t h e n flooded w i t h C O P . The c o i l is f i r s t s a t u r a t e d w i t h o i l .r i c h p h a s e s i n t h e mixing zone .o i l r a t i o .8 pore volumes o f s o l v e n t i n j e c t e d .t h e appearance of a methane bank p r i o r t o s o l v e n t breakthrough .24 have made a summary o f slim-tube displacement a p p a r a t u s e s used by v a r i o u s i n v e s t i g a t o r s .t h e o b s e r v a t i o n o f a n i n t e r f a c e between t h e o i l . about 5 m u i n t e r n a l d i a m e t e r . and packed w i t h f i n e g l a s s beads o r s a n d s t o a p o r o s i t y o f about 30% and t o a p e r m e a b i l i t y of 3-15 urn2. Produced l i q u i d and gas phases are metered s e p a r a t e l y .low u l t i m a t e r e c o v e r y A l l o f t h e above-noted symptoms o f a n immiscible displacement should appear i f t h e p r e s s u r e is w e l l below t h e MEip. . i s sampled f o r a n a l y s i s . . On t h e o t h e r hand.minimum m i s c i b i l i t y . Accordingly. t h e r e a d e r is r e f e r r e d t o o t h e r p u b l i s h e d works. and is t h e n f l a s h e d t o a t m s p h e r i c p r e s s u r e through a b a c k p r e s s u r e r e g u l a t o r .no appearance o f - - a methane bank p r i o r t o .c o n t a c t PVT c e l l tests.

w h i l e Jacobson43 s t u d i e d t h e c o n t r i b u t i o n o f a c i d g a s e s t o m i s c i b i l i t y . . t h e r e s u l t s of slim-tube experiments may be used t o c a l i b r a t e compositional s i m u l a t o r s . m u l t i p l e . o i l .dynamic o i l .recovery mechanisms9’1 9 .c o n t a c t experiments12’19’25 although.water. hence displacement tests on r e s e r v o i r rocks must follow. a s l i m tube l e n g t h between 10-20 1 is recommended. r e l a t i v e p e r m e a b i l i t y .e f f e c t of t h e following f a c t o r s on displacement e f f i c i e n c y o r o i l recovery: . This is supported by t h e lower number of c o n t a c t s (about 1 0 ) r e q u i r e d i n PVT c e l l . Core Displacement Tests Following slim-tube displacement t e s t s t o confirm t h e establishment of m i s c i b i l i t y w i t h t h e o i l f o r a given s o l v e n t a t a p p r o p r i a t e r e s e r v o i r c o n d i t i o n s of temperature and p r e s s u r e . I n g e n e r a l .c o n t a c t m i s c i b i l i t y i s achieved f a i r l y e a r l y i n t h e l i f e of t h e displacement ( w i t h i n t h e f i r s t two m e t e r s ) . as mentioned p r e v i o u s l y . I n a d d i t i o n t o studying dynamic m i s c i b i l i t y c o n d i t i o n s . The following s e c t i o n d e a l s w i t h c o r e displacement tests i n a n attempt t o provide more d e t a i l e d i n s i g h t i n t o t h e displacement behavior as i t may occur i n t h e r e s e r v o i r i n r e g i o n s contacted by t h e s o l v e n t . w h i l e n i t r o g e n and methane i n c r e a s e i t .291 A v a r i e t y of s l i m t u b e l e n g t h s have been used by v a r i o u s researcher^.r o c k i n t e r a c t i o n s w i t h gas and b r i n e . slim-tube displacement tests are an extremely u s e f u l t o o l f o r studying t h e m i s c i b i l i t y r e l a t i o n s h i p between o i l and s o l v e n t systems under c o n t r o l l e d dynamic c o n d i t i o n s .5 5 m) were required t o develop m i s c i b i l i t y when carbon d i o x i d e was i n the l i q u i d form.s o l v e n t phase behavior 5 8 . For t h i s reason. - Benham e t a1. and gas r e l a t i v e p e r r n e a b i l i t i e ~ ~ ” ~ ~ . Such tests c a n be used t o e v a l u a t e a v a r i e t y of displacement phenomena t h a t have b e a r i n g o n t h e m i s c i b l e f l o o d i n g process. such experiments are open t o i n t e r p r e t a t i o n .d i f f u s i o n and d i s p e r s i o n c o e f f i c i e n t s .m i s c i b l e and compositional s i m u l a t o r tuning48’ 6 4 . Other r e s e a r c h e r s 1 O S 3 l ’40’41 ’42’44 have i n v e s t i g a t e d t h e e f f e c t of t h e d i f f e r e n t p r o c e s s v a r i a b l e s on t h e carbon d i o x i d e MMP. The r a t e a t which slim-tube d i s placements are r u n a f f e c t s t h e s t a b i l i t y o f t h e displacement f r o n t and t h e time allowed f o r c o n t a c t between t h e o i l and s o l v e n t . Y e l l i g 1 5 concluded t h a t l o n g e r l e n g t h s (2.chromatographic s e p a r a t i o n of components1 ’48 . w e t t a b i l i t y . which provides f o r good d e f i n i t i o n o f t h e minimum m i s c i b i l i t y p r e s s u r e . The u s e o f r e l a t i v e l y low rates a l s o minimizes t h e p r e s s u r e drop a c r o s s t h e slim tube. o t h e r w i s e a high o i l recovery would not be o b t a i n e d . and dead-end pore volume^^^-^^ . On t h e o t h e r hand. 39’45’46 Wang and L ~ c h en ~ e~ t i g a t e d t h e r e l a t i v e e f f i c i e n c y of d i f f e r e n t wateri v s a l t e r n a t i n g . Thus. These i n c l u d e . displacement rates are b e s t kept a t less t h a n 1 0 m/day.^^ I t would appear t h a t m u l t i p l e . I n summary. t h e ME’ i n c r e a s e s w i t h d e c r e a s i n g o i l g r a v i t y and i t s C5 t o C30 c o n t e n t . and pore geometry) have n o t been t a k e n i n t o account.g a s c y c l e s and concluded t h a t t h e t o t a l o i l recovery w a s i n s i g n i f i c a n t l y a f f e c t e d by t h e i n j e c t i o n sequence provided t h a t t h e t o t a l amount of carbon d i o x i d e i n j e c t e d remained t h e same.8 have presented c o r r e l a t i o n s f o r t h e minimum enrichment of d r y gas (by LPG) r e q u i r e d t o a c h i e v e m i s c i b i l i t y . c o r e f l o o d i n g measurements are g e n e r a l l y recommended. and w i t h i n c r e a s i n g temperature and molecular weight of t h e o i l C5+ f r a c t i o n . Hydrogen s u l f i d e and LPG i n t h e carbon d i o x i d e decrease t h e I W . Caution must be e x e r c i s e d when t r a n s p o s i n g t h e r e s u l t s of such s t u d i e s t o r e s e r v o i r systems s i n c e t h e e f f e c t s of t h e r e s e r v o i r rock p r o p e r t i e s (homogeneity.

and Eatycky e t a1.p e r m e a b i l i t y c h a r a c t e r i s t i c s Combining p l u g s from d i f f e r e n t f a c i e s and w i t h w i d e l y v a r y i n g p r o p e r t i e s makes t h e i n t e r p r e t a t i o n of t h e d i s p l a c e m e n t r e s u l t s d i f f i c u l t and o f q u e s t i o n a b l e v a l u e a s i n p u t d a t a f o r s i m u l a t o r p r e d i c t i o n s of f i e l d performance. more l i k e l y . shown t h a t t h e h e t e r o g e n e i t y e x h i b i t e d by i n d i v i d u a l c o r e segments d e c r e a s e s when t h e segments are b u t t e d t o g e t h e r t o form a l o n g e r c o r e assembly. and l a m i n a t i o n s would tend t o g i v e r e s u l t s that e x a g g e r a t e t h e e f f e c t s o f t h e h e t e r ~ g e n e i t i e s . and t h e n f l o o d i n g w i t h c r u d e o i l u n t i l t h e water s a t u r a t i o n approaches t h e connate water s a t u r a t i o n . rsoocl v e n t type’ 4’60 k . i n j e c t i o n water. h o r i z o n t a l p l u g s have t o be d r i l l e d o u t of t h e f u l l d i a m e t e r c o r e . These p l u g s a r e t y p i c a l l y 2-3 cm i n d i a m e t e r and 6-10 cm long. v e r t i c a l c o r e s may be used f o r e v a l u a t i n g v e r t i c a l f l o o d s w h i l e . s a t u r a t i n g w i t h r e s e r v o i r lsrine. water s a t u r a t i o n (secondary o r t e r t i a r y f l o o d i n g mode)59’60’63rC5 . i t is l e f t t o a g e f o r s e v e r a l d a y s f o r t h e purpose of . f o r h o r i z o n t a l f l o o d s . A t y p i c a l t e s t procedure u t i l i z e d w i t h c l e a n e d .67 have. phase behavior ( m u l t i l e l i q u i d and s o l i d phases)32 . Non-preserved c o r e needs t o be cleaned thoroughly by e x t r a c t i o n or displacement w i t h s o l v e n t s such a s toluene-methano166. mounted d r y i n a c o r e h o l d e r . Cores w i t h l a r g e h e t e r o g e n e i t i e s s u c h as f r a c t u t e s .to mounting them i n a tiraxial c o r e h o l d e r . blowdown . The c o r e s a v a i l a b l e f o r t e s t i n g may b e i n t h e preserved state o r . are i n a n aged c o n d i t i o n . t h e n methods u t i l i z i n g gas flow a n d / o r e v a p o r a t i o n c a n be Following placement of crude o i l i n t h e c o r e . I f t h i s procedure cannot p r o v i d e a s u f f i c i e n t l y low i n i t i a l water s a t u r a t i o n . and t h e n have i t s w e t t a b i l i t y and i n i t i a : o i l s a t u r a t i o n r e . vugs. displacement c piroens ur ar et e” l 5’’6 ’ s . The c o r e is connected t o r e s e r v o i r o i l and b r i n e .e s t a b l i s h e d by c o n t a c t w i t h t h e r e s e r v o i r f l u i d s . F u l l d i a m e t e r . ~ ~ u d i e s St of Rosman and Simon66. however. and s o l v e n t c o n t a i n e r s . p a r t i c u l a r l y i f t h e development o f m u l t i p l e c o n t a c t m i s c i b i l i t y i s i n v o l v e d . t h e c o r e f a c e s c a n be machined s q u a r e on a l a t h e . It i s recommended t h a t t h e plugs b e chosen s u c h that t h e y come from t h e same f a c i e s i n t h e r e s e r v o i r . solve C02-foam ’63 A c o r e d i s p l a c e m e n t a p p a r a t u s c o n s i s t s of a c o r e h o l d e r i n which t h e c o r e is placed under a c o n f i n i n g p r e s s u r e . . c o r e i s flooded a t r e s e r v o i r t e m p e r a t u r e and p r e s s u r e w i t h t h e s e f l u i d s i n t h e p r o p e r sequence. and t h e r e is t h e o p t i o n o f p l a c i n g f i l t e r paper between t h e c o r e f a c e s p r i o r . t h e c o r e s can be used d i r e c t l y i n t h e displacement experiments. I f p r e s e r v e d . The s e l e c t i o n of r e s e r v o i r c o r e s f o r t h e s e tests is a n important procedure which r e q u i r e s a n u n d e r s t a n d i n g o f t h e geology o f t h e e n t i r e r e s e r v o i r . low i n t e r f a c i a l tension66 . s o l v e n t i n and t iC02-polymern ti ns jl eucgt isoinz)eb 2WAG. The c o r e s should be sampled from t h e pay zone of i n t e r e s t and chosen t o p r o p e r l y r e p r e s e n t t h e main r o c k t y p e s o c c u r r i n g i n t h e r e s e r v o i r . solvent inje t ’ . Although o u t c r o p c o r e s may a l s o be used f o r c e r t a i n m e c h a n i s t i c s t u d i e s . f l o o d i n g mode jec on. and t h e f l u i d p r o d u c t i o n and p r e s s u r e d r o p s are monitored. To a c h i e v e good c a p i l l a r y c o n t a c t between t h e c o r e s . and t h a t t h e y have similar and r e p r e s e n t a t i v e p o r o s i t y . About 20 p l u g s should be b u t t e d t o g e t h e r i n a c o r e h o l d e r t o g i v e a s u f f i c i e n t l y l o n g assembly f o r t h e d i s p l a c e m e n t t e s t . The. V i s u a l o b s e r v a t i o n o f t h e c o r e ’ s e f f l u e n t s c a n b e made through a s i g h t g l a s s .292 . It is recommended that c o r e from t h e a c t u a l r e s e r v o i r be used i n t h e d i s placement tests. water-solvent c( cooi nnjtei ncut ioouns. non-preserved c o r e i n v o l v e s e v a c u a t i n g .

t h e c o r e is water~ flooded w i t h i n j e c t i o n water down t o r e s i d u a l o i l s a t u r a t i o n . F i n a l l y . T. The disadvantage of u s i n g aged c o r e i s t h a t one is seldom s u r e of t h e adequacy o f t h e measures t a k e n t o r e s t o r e t h e c o r e t o its o r i g i n a l state. Tech.. 3. "Status of C02 and Hydrocarbon Miscible O i l Recovery Methods". s o l v e n t .W.s o l v e n t system. STALKUP. e t a l .S. (January 1976) 76. Laboratory t e s t s have been c a t e g o r i z e d i n t o s t a t i c and dynamic measurements. A prime r e a s o n f o r a t t e m p t i n g t o r e s t o r e t h e r e s e r v o i r w e t t i n g c o n d i t i o n i n t h e c o r e relates t o t h e r e p o r t e d t r a p p i n g o r s h i e l d i n g o f o i l by mobile water i n water-wet It is g e n e r a l l y b e l i e v e d that mixed o r i n t e r m e d i a t e l y w e t systems provide optimum t e r t i a r y recovery e f f i c i e n c i e s w i t h s o l v e n t floods. t y p i n g t h e manuscript. and t h e i r m i x t u r e s under c o n t r o l l e d c o n d i t i o n s .M. "Carbon Dioxide Miscible Flooding.g. S t a t i c phase behavior tests e n a b l e t h e measurement o f t h e p r o p e r t i e s of t h e o i l . Once t h e p r e l i m i n a r y s e l e c t i o n has been made. HOLM. of Energy P u b l i c a t i o n No.. I f t h e s o l vent f l o o d is t o be a secondary one. and d i f f e r e n t t y p e s of tests that may be c a r r i e d o u t under each category have been l i s t e d .t e r a g i n g . The w a t e r . and t o B. A L i t e r a t u r e Search t o June 30. P e t . Tech.I. ACKNOIJLEDQ4ENI'S The a u t h o r s wish t o e x p r e s s t h e i r thanks t o P.e s t a b l i s h i n g t h e o r i g i n a l ~ e t t a b i l i t y ~ A f. P a s t . The d i s t i n c t advantages of u s i n g non-preserved c o r e are its ease of handling d u r i n g t h e d r i l l i n g of p l u g s . L. DOE/BETC/5785-1 (1980). U. Moore f o r . J . P e t . F i n a l l y . t h e waterflood s t e p is then n a t u r a l l y omitted .293 r e . l a b o r a t o r y tests can be c a r r i e d o u t t o reduce t h e t e c h n i c a l and economic u n c e r t a i n t i e s a s s o c i a t e d with f i e l d tests. - . F. Sigmund f o r c o n s u l t a t i o n s . The l a b o r a t o r y t e s t s should be supplemented w i t h g e o l o g i c a l ( r e s e r v o i r d e s c r i p t i o n ) and computer s i m u l a t i o n studies'. (August 1978) 1102. RECAPITULATION The f i r s t s t e p i n t h e implementation of a f i e l d .s c a l e m i s c i b l e flood is t h e s e l e c t i o n o f s u i t a b l e c a n d i d a t e r e s e r v o i r s and s o l v e n t s . DOSCHER. These should be augmented by o t h e r l i m i t a t i o n s and/or i n c e n t i v e s (e. J. 2. Dept. A set of t e c h n i c a l s c r e e n i n g criteria has been provided t o a i d i n t h e s e l e c t i o n . P r e s e n t and Outlook f o r t h e Future". t h e core i s s o l v e n t flooded. and t h e a b i l i t y t o examine t h e cores and measure t h e i r p r o p e r t i e s (such as a i r p e r m e a b i l i t y and p o r o s i t y ) p r i o r t o t h e f l o o d tests. REFERENCES 1. c o r e displacement tests h e l p determine t h e e f f e c t of t h e process c o n d i t i o n s and rock p r o p e r t i e s o n t h e displacement e f f i c i e n c y i n t h e swept zone of the reservoir. Slim tube tests determine t h e dynamic m i s c i b i l i t y c h a r a c t e r i s t i c s of t h e o i l .o i l r e l a t i v e p e r m e a b i l i t y may be c a l c u l a t e d from t h e p r e s s u r e drop and production h i s t o r y o f t h e waterflood. "Carbon Dioxide f o r t h e Recovery of Crude O i l .. economic) s p e c i f i c t o each locale. 1979 F i n a l Report".

*'Miscible Flood Displacement P r e d i c t i o n o f M i s c i b i l i t y " . "A L a b o r a t o r y I n v e s t i FATIWELL. and HOLSTE. Trans. Texas. 'The E f f e c t o f Phase Behavior on Cog Flood Displacement E f f i c i e n c y " . SOC. 9. "The E f f e c t of Phase E q u i l i b r i a on t h e C 0 2 Displacement Mechanism".S. o f P e t . (1961). Nevada (September 23-26. o f P e t . - 10. p r e s e n t e d a t t h e 5 4 t h Annual F a l l T e c h n i c a l Conference and E x h i b i t i o n o f t h e SOC.UK.. SOC. o f P e t .A. P r o c e e d i n g s o f 2nd I n t e r n a t i o n a l Conference on Phase E q u i l i b r i a and F l u i d P r o p e r t i e s i n t h e Chemical I n d u s t r y . paper SPE 8367.. 675. g a t i o n o f M i s c i b l e Displacement by Carbon Dioxide". o f P e t . .C. F r a n k f u r t / Main. of CIM.. - 6. of AIME.B. paper SPE 3483. prepared f o r 4 6 t h Annual F a l l Meeting o f t h e SOC. " E f f e c t o f O i l Composition o n Miscible-Type Displacement by Carbon Dioxide". T u l s a . o f P e t . 17.A. SPE/DOE p a p e r no. Eng. Eng. P e t . MNGAN. Las Vegas. Oklahoma (Apr. 5. p r e s e n t e d a t t h e 1981 SPE/DOE Second J o i n t Symp.I.J.C. o f CIM. 1971). HUANG.L. . o f P e t . Oklahoma ( A p r i l 22-24. 1979).R. o f CIM paper no. paper SPE 4735. o f C I M paper no. P. 1981).. "Volumetric and Phase Behavior o f O i l F i e l d Hydrocarbon Systems". METCALFE.W. AIME (1960) 219. o f P e t . p r e s e n t e d a t t h e 1981 Permian B a s i n O i l and Gas Recovery Symposium o f t h e SOC.Fundamentals". paper SPE/DOE 9785. 5-8. HALL. 1981).E.M. T u l s a . P. J. 'M. o f A I I E . "Carbon D i o x i d e Flooding . DOWDEN. and PATEL..Ch. J. F. HOLM. Tulsa.. and HASSINGER. L. GARDNER. paper SPE 9710. Eng. 12. C a l g a r y . Eng. (August 1979). "Phase R e l a t i o n s of M i s c i b l e Displacement i n O i l Recovery". 80-31-04.. o f AIME. F. 229. 242. paper SPE 8814. and DANN. J. 64. D a l l a s . SOC. 11. C. p r e s e n t e d a t t h e F i r s t J o i n t SPE/DOE Symp. 1980). "Microscopic I n v e s t i g a t i o n o f Cog Flooding Process". and BRAUN. ORR. .. V. C a l g a r y .J. 1980). "Carbon Dioxide Displacement of a West Texas R e s e r v o i r O i l " . STALKUP. P r o j e c t s " . Dechema. 16.C 13. and T A m J.. A l b e r t a (May 25-28. Oklahoma ( A p r i l 5-8.E.. o n Enhanced O i l Recovery of t h e SOC. p r e s e n t e d a t Improved O i l Recovery Symposium o f t h e SOC. L. Eng. N... Oklahoma ( A p r i l 20-23. o f AIME. P. HUTCHINSON. Midland. 1980).B.H.T.. P e t . "Screening T e s t s f o r Enhanced O i l Recovery BURNETT. W. W. R C . Eng. INNGAN. EIJB. 7. J. 9788. A. "Carbon Dioxide Flooding A p p l i c a t i o n s " . Eng..D.1.294 4. New O r l e a n s .. 14. o f AIME. STANDING. on Enhanced O i l Recovery. p r e s e n t e d a t 31st Annual T e c h n i c a l Meeting o f t h e P e t .. West B e r l i n (Ilareh 17-21.S. and YARBOROUGH.W.T. R. L o u i s i a n a (October 3-6. SOC. p r e s e n t e d a t 3 1 s t Annual T e c h n i c a l Meeting o f t h e P e t . Eng. P e t . . D..1981). WANG. 1 9 7 4 ) . W .F. "A Review o f Experimental Techniques f o r Vapor-Liquid E q u i l i b r i a a t ltigh P r e s s u r e " . SOC... SOC. 15.. and KUNZMAN.. and JOSENDAL. 81-31-22. G. N. T u l s a .H. J R . Texas (March 12-13.. K. R. J. 1981). "The Displacement o f R e s i d u a l O i l by Carbon Dioxide". 7 (l). E. p r e s e n t e d a t t h e 1981 SPE/DOE Second J o i n t Symposium o n Enhanced O i l Recovery o f t h e SOC. M. 8. A l b e r t a (May 25-28.W. YELLIG. BENHAM. A. E i g h t h P r i n t i n g (1977)..

" M i s c i b i l i t y R e l a t i o n s h i p s i n t h e Displacement of Oil by RUTHERFORD. J. (October 1967).K... Eng.D. J. J . S e r .F.* HENRY.-Tech. 1980).L.. L. SOC..U. Eng. o f C I M . 30.D "The P r o p e r t i e s of Petroleum Fluids". E. of CIM. "Phasengleichw i c h t e und k r i t i s c h e Erscheinungen i n b i n a r e n Mischsystemen b i s 1500 b a r C02 m i t n-Octan. and YARBOROUGH. (December 1962) 340. W. D. Morgantown. P e t . and POPE. Symp. SOC.. A l b e r t a (May 3-6. p r e s e n t e d a t 32nd Annual T e c h n i c a l Meeting o f t h e P e t . "Laboratory Experiments t o E v a l u a t e F i e l d P r o s p e c t s f o r C02 Flooding". "Development of Experimental Phase E q u i l i b r i a f o r M i s c i b l e Hydrocarbon Enhanced O i l Recovery". and FUSSELL. and JOSEXDAL. and VOGEL. paper SPE/WE 9786. T u l s a . L. Chemie-1ng. 27 . J . o f P e t . Eng.D. . SHELTON.I " V i s c o s i t i e s o f Carbon Dioxide-Sulfur Dioxide Mixtures and t h e E f f e c t o f Methane on S u l f u r Dioxide-Oil M i s c i b i l i t y " .. (September 1977) 1171.. 24. West V i r g i n i a (November 5-7. W . 22.L.. 1. S. Petroleum P u b l i s h i n g Company. A l b e r t a (March 1979). P e t ..295 18. and DUNLOP. R. Calgary. P e t . and LWi.E. J. Soc. . G. SILVA... C. . J. of AIME E a s t e r n Regional Meeting.. Tech. P e t . HORVATH. "A New System f o r Obtaining Vapor and L i q u i d Sample Analyses t o F a c i l i t a t e t h e Study of Multicomponent Mixtures o f Elevated P r e s s u r e s " . Tulsa. B. and YARBOROUGH. J. 1981).. SOC.G. (August 1963) 873. SCHNEIDER. LIEN. .T. " U t i l i z a t i o n of Composition O b s e r v a t i o n Wells i n a West Texas C02 P i l o t Flood". . "PVT Measurements o n Petroleum Reservoir F l u i d s and t h e i r Uses".W P e t .R. n-Tridecan und n-Hexan".L. 19. E. "Mechanisms of O i l Displacement by Carbon Dioxide". M. H.. L.M. Calgary. Ind. HOLM. (1967) 2 ( l l ) . Eng. P e t . Calgary.. 23. Tech.. J R .. Z. J R . J . D. (February 1978) 20. P e t . A. 20.W. SMITH. SIMON. (December 1974) 1427. G.L. P e t . . Trans. and HLAVACEK. JACOBY. NAJMAN. . MENZIE. "Phase Equilibrium and Fluid P r o p e r t i e s o f Pembina Cardium Stock-Tank Oil-Methane-Carbon Dioxide-Sulfur Dioxide Mixtures".E.... 31. "Phase Behavior P r o p e r t i e s o f C02R e s e r v o i r O i l Systems". 33. 1247. V.. and NIELSEM. and FRANCK. Oklahoma ( A p r i l 5-8..R. A. Oklahoma (1973). J . FEATHER. SOC. Tech. 11. 28. 225. 81-32-41.. Eng. R. P e t . J . n-Undecan. SAYEGH. Tech. L.W. R.. 29. "Physical P r o p e r t i e s o f Carbonated O i l s " . paper no. "A Study o f t h e V a p o r i z a t i o n o f Crude O i l by Carbon Dioxide Repressuring". 25. o f CIM. 649. Prog. A l b e r t a (May 3-6. (November 1963). o f CIM. CONNOR. 48. L. p r e s e n t e d a t t h e 1981 SPE/DOE Second J o i n t Symp. p r e s e n t e d a t 32nd Annual Meeting of t h e P e t . h g . ORR. Petroleum Recovery I n s t i t u t e . ALWANI.. F. on Enhanced O i l Recovery o f t h e Soc. . A P E (1959) 216. Soc.. Chem. 1981). L. D. paper SPE 9534 p r e s e n t e d a t 1980 Soc.. and PELLETIER. paper 81-32-14. ROSMAN. 32. Chem. McCAIN. R.J. o f P e t . (1967) 63 (811.L. "Carbon Dioxide S o l v e n t Flooding f o r I n c r e a s e d O i l Recovery". 26.H. L i g h t Hydrocarbons". WELKER. 1981). . HOLM. Research l b t e RN-9. LEE. "Multiple Phase Behavior i n Porous Media During COP o f Rich-Gas Flooding". E.A. 21. J.. YARBOROUGH.

on Enhanced O i l Recovery. G. Research Report RR-40.X. " E f f e c t i v e n e s s of Carbon Dioxide Displacement Under M i s c i b l e and I m n i s c i b l e Conditions". P.. ''Determination and P r e d i c t i o n of COP Minimum M i s c i b i l i t y P r e s s u r e s " . A l b e r t a (August 1 9 7 8 ) . Oklahoma ( A p r i l 20-23.. R. . Houston. "Carbon Dioxide Dynamic M i s c i b i l i t y w i t h L i g h t R e s e r v o i r O i l s " . YELLIG. WILLIAMS.. C a l g a r y .M. R.. L.I. Bucharest . LEE. of AIME. of P e t . Research Note RN-8...S... '?leasurement and C o r r e l a t i o n o f CO2 M i s c i b i l i t y P r e s s u r e s " . " E f f e c t o f I m p u r i t i e s o n Minimum M i s c i b i l i t y P r e s s u r e s and IETCALFE. K. 36. ORR. Oklahoma ( A p r i l 20-23. 46. J. 45. "Use of t h e Peng-Robinson Equation of S t a t e t o P r e d i c t Hydrocarbon Phase Behavior and M i s c i b i l i t y f o r F l u i d Displacement".L. C. . Tech. p r e s e n t e d a t t h e 5 5 t h Annual F a l l T e c h n i c a l Conference and E x h i b i t i o n o f t h e SPE o f AIME. 42. paper SPE 9230. p r e s e n t e d a t t h e 1 9 8 1 SPEIDOE Second J o i n t Symposium o n Enhanced O i l Recovery o f t h e SOC. Texas (September 21-24. p r e s e n t e d a t t h e F i r s t J o i n t SPEIDOE Symposium on Enhanced O i l Recovery. E.. p a p e r SPE 8813.L. R. . J. J . and WLLIN. J. paper SPE 8817.A. CRONQUIST. 1980).S.P. J.S. 1981). L. YARBOROUGH. J. p r e s e n t e d a t t h e World Petroleum Congress. Eng. T u l s a . p r e s e n t e d a t t h e 53rd Annual F a l l T e c h n i c a l Conference and E x h i b i t i o n of t h e SOC.. 44. C a l g a r y .. W. 147.A.. ' h b o r a t o r y C02 Floods and T h e i r Computer Simulation". A l b e r t a (September 1978). LEE. TUREK.p r e s e n t e d a t t h e F i r s t J o i n t SPEIDOE Symp. p a p e r SPE 9790. F. of A m . D.. 'Phase Behavior and Displacement S t u d i e s of Carbon Dioxide-Hydrocarbon Systems". E. Texas (September 21-24. A Z I Z . p a p e r SPE 7477. 43. ZANA. o f P e t . R. 1980). and ROBINSON.M. and HUMPHRYS.I. 37. 1980).. C. LEE. R. NGHIEM. C a l g a r y .D. P e t . 'Thase Behavior o f C02 and Crude O i l i n Low Temperature Reservoirs". Dallas. . Petroleum Recovery I n s t i t u t e . 40.. paper SPE 9231. 38. p r e s e n t e d a t 4 t h Annual D E Enhanced O i l Recovery Symposium. p r e s e n t e d a t t h e 5 5 t h Annual F a l l Technical Conference and E x h i b i t i o n o f t h e SOC. Eng. Oklahoma ( A p r i l 5-8. 35.S... S e p a r a t i o n Model f o r t h e Study o f M u l t i p l e C o n t a c t M i s c i b i l i t y i n Rich Gas Drives". Can.296 34. of P e t . T u l s a . LEE. "Acid Gases and Their C o n t r i b u t i o n t o M i s c i b i l i t y " . Research Report RR-37..I.E. T u l s a .S Minirmm Enrichment L e v e l s f o r CO2 and Rich Gas Displacement". T u l s a . . H. 1972) 56.. P e t .. A l b e r t a (March 1 9 7 9 ) .D.. R. and LIEN.N. and METCALFE. Rumania (September 1979). and SIGMUND. Eng.... o f AIME. 1978). 1 9 8 0 ) . "Phase E q u i l i b r i a i n Carbon Dioxide-Multicomponent Hydrocarbon Systems: Experimental Data and a n Improved P r e d i c t i o n Technique". SOC. JACOBSON. Eng. P. Petroleum Recovery I n s t i t u t e . JR. Texas (October 1-3. J . I.L. METCALFE. W. 41. J R .A. 39. " M i s c i b i l i t y of Carbon Dioxide-Sulfur Dioxide-Hydrocarbon Systems". and I H . and SHELTON. A. (April-June.F. J. FUSSELL. (June 1 9 7 3 ) . Dallas. 1978). C. "A M u l t i c e l l E q u i l i b r i u m METCALFE. Petroleum Recovery I n s t i t u t e . J. O Oklahoma (August 29-31.M... JOHNSON. E R4 SIGMUND. .

Flow Following S o l v e n t I n j e c t i o n i n Carbonate Rocks". W.F. 'The E f f e c t o f Microscopic Core H e t e r o g e n e i t y o n M i s c i b l e Flood R e s i d u a l Oil S a t u r a t i o n " . presented a t t h e 53rd Annual f a l l T e c h n i c a l Conference and e x h i b i t i o n of t h e Sac.L. R. and LOCKE. SOC.N. . Soc. 61. and OWENS. (March 1964) 73. and BAKER. B.. (February 1974) 91.. R. - . and SCHNEIDER. J. P e t . 69. 54. A.N. P e t . "Displacement o f O i l by S o l v e n t a t High Water S a t u r a t i o n " . T. J. " F a c t o r s A f f e c t i n g M i s c i b l e Flooding Dispersion C o e f f i c i e n t s " . 1978). Eng. paper SPE 8812. P e t . Dallas. J.. '%ixing E q u a t i o n s i n S h o r t Laboratory Cores". R. STALKUP. J. P e t . 53. SOC. SOC. F. . P e t . U. HENRY.. G. BLACKWELL. IJANG.. and JOHNSTON.P. and LOCKE. (August 1980) 278. Houston. o f P e t . SOC. P e t . - - "Laboratory Study o f Microscopic D i s p e r s i o n Phenomena". VAN DER POEL. "A Laboratory Study o f t h e E f f e c t s o f C02 I n j e c t i o n Sequence o n T e r t i a r y O i l Recovery". Eng. 1980). COATS. paper SPE 7474.C. Oklahoma ( A p r i l 20-23. Calgary. P e t . Eng. "A Technique f o r t h e Laboratory lleasurement o f Carbon Dioxide NATRINS. (1962) L. HELLER. T u l s a . SOC. J. J . L.P. p r e s e n t e d a t t h e F i r s t J o i n t SPE/DOE Symposium o n Enhanced O i l Recovery. P e t . Texas (October 1-3. 48.. 57. "The E f f e c t s of Water I n j e c t i o n on M i s c i b l e Flooding Methods Using Hydrocarbons and Carbon Dioxide". SPENCE. C. 55.C. "Compositional Model S t u d i e s CO2 O i l Displacement Mechanisms". R. of C I M . SOC. W.I. J . 50. of CIM paper no. " R e l a t i v e P e r m e a b i l i t y S t u d i e s o f Gas-Water SCHNEIDER.L. 59. O. 49. SOC. M. P e t . P e t . (February 1981). Eng. 51. Dept. Porous Media".. .297 47.W. (June 1975) 217. W.J. A l b e r t a (May 25-28. C.D.S.E. Eng. J. Eng.J SOC. Texas (September 21-24.. Eng. p r e s e n t e d a t 31st Annual T e c h n i c a l Meeting. (February 1976) 23. K. SOC. and YELLIG. Unit Displacement E f f i c i e n c y i n R e s e r v o i r Rock". BRIGHAM. BAKER. 52. (June 1977) 219. of P e t .. C. o f AINE. 62. J .D. Eng. (1962) 1 77.H. TABOR.S. " M u l t i p l e Phase Generation During C02 Flooding". W..F. 1980). Eng. and WATKINS. p r e s e n t e d a t t h e 5 5 t h Annual F a l l T e c h n i c a l Conference and E x h i b i t i o n of t h e SOC. L. 80-31-06. SOC. " E f f e c t s o f D i s p e r s i o n and Dead-End Pore Volume i n Miscible Flooding". . 1980). R. "Dead-End Pore Volume and D i s p e r s i o n i n Porous Fledia".P. Eng.W. and SMITH. 60.E. Eng. "Effect o f Lateral D i f f u s i v i t y on M i s c i b l e Displacement i n H o r i z o n t a l Reservoirs".. J.W. 58. Eng..E. (1962) L 93. paper SPE 9229. LEACH. DOE/MC/10689-3 (1980). J. . P e t . (December 1970) 337... P e t . of Energy P u b l i c a t i o n no... F. o f A M E . J R . "A Review o f D i f f u s i o n and D i s p e r s i o n i n PERKINS. J . J . Eng. "Mobility C o n t r o l f o r COP Floods A L i t e r a t u r e Survey".D. P e t . YELLIG. 89.. 3. SOC. 56. SOC.K. and METCALFE. F. SHELTON.

and SIMON. JACKSON. l. ... . 51-32-16. Tech. J. p r e s e n t e d a t t h e 1981 SPE/DOE Second J o i n t Symposium o n Enhanced O i l Recovery of t h e SOC...H. and ZANA.J.. and DLAClrJJELL. 1981).H.. (1981) 20 ( ) 104.L i t t l e Escambia Creek F i e l d Unit". J.. SHIRER. J. "Field I n j e c t i v i t y Experiences With M i s c i b l e Recovery P r o j e c t s Using A l t e r n a t e Rich-Gas and Water I n j e c t i o n " . p r e s e n t e d a t t h e 52nd Annual F a l l T e c h n i c a l Conference and E x h i b i t i o n o f t h e Soc.. R. " P h y s i c a l Model. "The Dependence of Water GRIST.L.T. J. ROSMAN. P e t . G. J. West Wilmar R i c h Gas P i l o t " . Calgary. p r e s e n t e d a t t h e Middle E a s t O i l T e c h n i c a l Conference of t h e Soc. T u l s a . Tech.M.. Tech.. J . 66. ROSMAN. . P e r m e a b i l i t y on Core C l e a n i n g Methods i n t h e Case of Some Sandstone Samples". 68. of P e t . "Planning a T e r t i a r y O i l Recovery P r o j e c t for J a y . WANSLEEBEN... p a p e r SPE 6723. "On t h e N e c e s s i t y o f R e s p e c t i n g R e s e r v o i r C o n d i t i o n s i n L a b o r a t o r y Displacement S t u d i e s " . p r e s e n t e d a t t h e 32nd Annual T e c h n i c a l Meeting of t h e P e t . "Flow H e t e r o g e n e i t y i n R e s e r v o i r Rocks". M . o f P e t .J. JR. Tech.L.. BATYCKY. E n g . (April-June 1975) 48.. P. 1977). J. 69.. E. Oklahoma ( A p r i l 5-8. C a n P e t . L. "Experimental S t u d i e s o f Low I n Displacement by C02 I n j e c t i o n " . (September 1977) 1051. Eng. (December 1976) 1427. of P e t . 70. HARVEY. 1979). E. P e t . Colorado (October 9-12. paper SPE/DOE 9805. T. 1981). SHELTOM. paper SPE 7785. D. and BESSERER. Manama. Can. and SIGMUND. and NEUSTADTER. Soc. G. of CIM. P e t . Bahrain (March 25-29..A. Denver. E. LANGLEY.D. C. L.L. 67. 64. a n d KELM. Eng. LONGERON. of C M p a p e r no.O.. CIIRISTIAN.M. A l b e r t a (May 3-6. I . Soc.E. D . R. RANDALL. M I R K I N . A... 65.. P e t . J.. M. CUIEC.. KIMBLE.298 63.P. A. C. "Miscible and Immiscible Displacement S t u d i e s on Carbonate R e s e r v o i r Cores". J. and PACSIXSZKY. gf AIME.

The computer model was used t o simulate p a r t i a l l y m i s c i b l e carbonated n a t u r a l gas f l o o d i n g i n t h e western area o f t h e Budafa o i l f i e l d . Among t h e o t h e r p o s s i b i l i t i e s t h e e f f e c t o f carbon d i o x i d e was a l s o studied. and t h e import of crude o i l imposes a considerable economic burden on a c o u n t r y developing i t s i n d u s t r y . Based upon t h e o r e t i c a l c o n s i d e r a t i o n s a t e c h n o l o g i c a l scheme has been developed t o increase t h e sweep e f f i c i e n c y .and multi-phase systems. &VA FARKAS. PVT s t u d i e s have proved t h a t gas c o n t a i n i n g 81 mole % carbon d i o x i d e can be used f o r EOR purposes. INTRODUCTION The o i l resources o f Hungarian r e s e r v o i r s cover o n l y a small p a r t o f t h e country’s demand. JANOS TOROK Hungarian Hydrocarbon Institute VAL~R BALINT General Contractingand Designing Office for the Oil Industry. based on l a b o r a t o r y PVT and displacement studies. TIBOR PAAL. and . three-phase mathematical model developed t o simulate carbon d i o x i d e f l o o d i n g i s s u i t a b l e f o r t r e a t i n g s i n g l e . The production h i s t o r y match and p r e d i c t i o n agreed w e l l w i t h t h e f i e l d data. Apart from t h e n e e d .t o search f o r new o i l f i e l d s . it became e v i d e n t as long ago as t h e f i f t i e s t h a t it was important t o consider secondary and l a t e r t h e t e r t i a r y recovery methods. t h u s t h e generation and disappaerance o f phases can be followed by t h e model w i t h o u t d i f f i c u l t i e s . GABOR ACS. f i e l d experiment on depleted r e s e r v o i r . A ten-component. The s t u d i e s covered t h e v o l u m e t r i c and phase behaviour o f carbonated n a t u r a l gas f l o o d i n g under f i e l d c o n d i t i o n s and t h e r e s u l t s proved t h a t such f l o o d i n g was e f f i c i e n t even i f t h e gas i s n o t pure carbon dioxide. “Olajterv” ZOLTAN B I R ~ lhznsdanubian Oil and Gas Roduction Company ABSTRACT A systematic program o f carbonated n a t u r a l gas f l o o d i n g has been c a r r i e d o u t i n Hungary. followed by composit i o n a l mathematical s i m u l a t i o n and. The d i f f e r e n c e equations handle t h e systems w i t h d i f f e r e n t number o f phases I n a u n i f o r m way.MISCIBLE GAS DISPLACEMENT 299 COMPLEX STUDY OF COZ FJBODING IN HUNGARY SANDOR DOLESCHALL.

PVT AND PHASE BEHAVIOUR MEASUREMENTS F e a s i b i l i t y s t u d i e s o f t h e a p p l i c a t i o n p o s s i b i l i t i e s o f carbon d i o x i d e and carbonated n a t u r a l gases s t a r t e d i n 1955 w i t h a s e r i e s o f PVT measurements. I t i s i n t e r e s t i n g t h a t these phenomena occur i n t h e presence o f carbonated n a t u r a l gases. and o n l y t h e change o f t h e given p r o p e r t i e s was c o r r e l a t e d w i t h t h e d i s s o l v e d carbon d i o x i d e cont e n t . - The aim o f these s t u d i e s i s t o determine exact K values according t o t h e need o f compositional mathematical s i m u l a t i o n . t h e p r e d i c t i o n o f v i s c o s i t i e s o f saturated and undersaturated crudes under d i f f e r e n t c o n d i t i o n s i s p o s s i b l e w i t h t h e use o f o n l y one measured v i s c o s i t y value. s e t s o f curves have been developed t o p r e d i c t t h e s o l u b i l i t y . too. The existence o f t h e mentioned multiphase systems had t o be considered i n planning vapour--liquid e q u i l i b r i u m studies. I n t h i s way simple. even i f they a r e r e l a t i v e l y r i c h I n l i g h t hydrocarbon f r a c t i o n . E q u i l i b r i u m r a t i o s had been d e t e r mined f o r c h a r a c t e r i s t i c r e s e r v o i r oil--carbonated n a t u r a l gas as w e l l as r e s e r v o i r oil--water--carbonated n a t u r a l gas systems and a method f o r e s t i mation was developed. P a r t i t i o n o f l i g h t and intermediate hydrocarbons between t h e r e s e r v o i r o i l and t h e second l i q u i d phase has been proven i n agreement w i t h o t h e r experience. Based upon t h e r e s u l t s o f more d e t a i l e d PVT measurement. t h a t no complete m i s c i b i l i t y occurs even i f .300 very soon most a t t e n t i o n focused on t h e questions o f C02 f l o o d i n g because i n Hungary t h e occurrence o f n a t u r a l carbon d i o x i d e i n h i g h carbon d i o x i d e content n a t u r a l gases i s more o f t e n found and t o a g r e a t e r e x t e n t than t h e world average. s w e l l i n g and v i s c o s i t y o f monophase r e s e r v o i r oil--carbonated n a t u r a l gas systems. 1 . T h i s r e g i o n depends upon t h e t o t a l composition o f t h e system and t h e phenomenon i s connected w i t h t h e r e s t r i c t e d s o l u b i l i t y o f carbon d i o x i d e i n r e s e r v o i r o i l s . As a r e s u l t o f a d d i t i o n a l measurements and comparison of experimental w i t h computed data u s i n g d i f f e r e n t equations o f s t a t e i t i s concluded t h a t f u r t h e r improvements a r e necessary both f o r t h e development o f generalized K f u n c t i o n s and equations o f s t a t e t o g e t h e r w i t h t h e improvement o f interaction coefficients. I n t h e case o f c e r t a i n Hungarian crude o i l s r e v e r s i b l e p r e c i p i t a t i o n o f semi-solid p a r t i c l e s has a l s o been observed b u t mostly under such circumstances which cannot be r e a l i z e d i n a c t u a l r e s e r v o i r s . Repeated measurements i n a windowed PVT c e l l revealed t h e presence o f a carbon d i o x i d e r i c h second l i q u i d phase which e x i s t s w i t h i n a d e f i n i t e pressure-temperature range above a c e r t a i n gas--oil r a t i o . . mainly i f t h e carbon d i o x i d e c o n t e n t o f t h e d i s s o l v e d gas i s above 60 mole 5 . I n t h e course o f t h e thorough examination o f t h e PVT data "unusual" behaviour was observed. easy t o use equations w i t h good accuracy have been developed. The very f i r s t PVT s t u d i e s proved t h a t carbonated n a t u r a l gas a l t e r s t h e v i s c o s i t i e s and v o l u m e t r i c p r o p e r t i e s o f crudes w i t h very d i f f e r e n t d e n s i t i e s i n a favourable way compared w i t h t h e e f f e c t o f lean o r wet n a t u r a l gases under t h e same c o n d i t i o n s . t h e d i s s o l v e d gas i s pure carbon d i o x i d e . bearing i n mind t h e a c t u a l r e s e r v o i r conditions. For example. I t has been proved t h a t i n t h e case o f Hungarian crude o i l s . Some r e s u l t s o f C02 f l o o d i n g i n Hungary can be found i n Ref. The a c t u a l PVT p r o p e r t i e s o f t h e o r i g i n a l gas saturated o i l were chosen as a reference s t a t e t o e l i m i n a t e t h e p o s s i b l e l a r g e e r r o r s coming from t h e unknown parameters o f such very complex systems.

and gas-phases under r e s e r v o i r c o n d i t i o n s w i t h a r a t i o corresponding t o t h e a c t u a l s a t u r a t i o n a t a given. The pressure was increased t o t h e o r i g i n a l r e s e r v o i r pressure d i r e c t l y by t h e i n j e c t i o n gas. - A PVT model was used t o f o l l o w t h e change o f t h e v o l u m e t r i c and phase behaviour and t h e e q u i l i b r i u m composition o f phases i n t h e course o f f l o o d i n g . which depends upon t h e pressure. The l i n e a r model t o examine t h e e f f i c i e n c y o f C02 used f o r t h e measurements was 1 m long and 25 mm i n diameter. t h e r o l e o f t h e l i g h t hydrocarbon f r a c t i o n present and supported t h e conclusion p r e v i o u s l y drawn on t h e b a s i s o f PVT s t u d i e s o f mono-and two-phase systems. Experimental data on s o l u b i l i t y . I t has a l s o been found t h a t d e s p i t e t h e d i l u t i o n o f t h e s l u g by hydrocarbon gas i n t h e pores.30 1 Judging by t h e r e s u l t s o f o t h e r studies. T h i s c r i t i c a l value. T h i s technique i s s u i t a b l e f o r s t u d y i n g p r o d u c t i o n h i s t o r i e s . I t has been pointed o u t t h a t because o f t h e i n t e r a c t i o n o f carbonated water and r e s e r v o i r rocks c e r t a i n c l a y m i n e r a l s c o n t r a c t and t h i s may improve t h e e f f i c i e n c y o f t h e process i n p r a c t i c e . r e l a t i v e l y s i g n i f i c a n t v a p o r i z a t i o n o f t h e o i l takes place i f t h e carbon d i o x i d e c o n t e n t o f t h e f r e e gas phase i s above a c e r t a i n c r i t i c a l concentration. I n t e r p r e t a t i o n o f t h e r e s u l t s revealed t h e Importance o f t h e dynamic pressure-increase process being a p p l i e d f o r carbonated n a t u r a l gas f l o o d i n g . T h i s model contained water-. t h e vapour phase g r a d u a l l y being changed by t h e i n j e c t i o n gas a t each i n t e r mediate pressure u n t i l e q u i l i b r i u m composition was approached. depleted f i e l d . LABORATORY DISPLACEMENT STUDIES F o l l o w i n g encouraging PVT r e s u l t s dynamical l a b o r a t o r y s t u d i e s were c a r r i e d o u t displacement processes. o i l . v a r i o u s forms o f C02 . temperature and t h e c h a r a c t e r i s t i c s o f t h e o i l . i t has been concluded t h a t it i s n o t p o s s i b l e i n p r a c t i c e t o replace a l l t h e f r e e and d i s s o l v e d gas by a carbon d i o x i d e s l u g o f reasonable s i z e . These observations confirmed i n d i r e c t l y t h e idea about t h e probable formation o f a m i s c i b l e f r o n t i n t h e r e s e r v o i r under dynamic c o n d i t i o n s d u r i n g carbonated n a t u r a l gas f l o o d i n g . I n another s e t o f experiments t h e f i n a l pressure was reached step by step.s a t u r a t e d w i t h carbonated n a t u r a l gases having d i f f e r e n t composition. Volumetric and phase behaviour as well as water c o n t e n t and hydrate forming c o n d i t i o n s o f carbonated n a t u r a l gases i n Hungary were a l s o s t u d i e d and t h e r e s u l t i n g data used t o formulate generalized r e l a t i o n s h i p s . These experiments were repeated f o r d i f f e r e n t f i e l d c o n d i t i o n s u s i n g i n j e c t i o n gases w i t h d i f f e r e n t carbon d i o x i d e content. Taking i n t o c o n s i d e r a t i o n t h e composition o f t h e g r a d u a l l y displaced gas. even i n t h e presence o f calcium carbonate and r e s e r v o i r rocks c o n t a i n i n g c l a y minerals t o g e t h e r w i t h vapour--liquid e q u i l i b r i u m r a t i o s supplied f u r t h e r i n f o r m a t i o n e n a b l i n g a b e t t e r understanding o f t h e mechanism o f carbonated n a t u r a l gas f l o o d i n g . as well as f l o o d i n g and t h e a c t u a l mechanism o f t h e process. t h e i n t e r f a c i a l t e n s i o n decreases w i t h i n c r e a s i n g carbon d i o x i d e c o n t e n t i n gas--oil--water systems. can a l s o be exceeded by using carbonated n a t u r a l gases f o r t h e i n j e c t i o n . As t o t h e v o l u m e t r i c p r o p e r t i e s and v i s c o s i t i e s o f t h e e q u i l i b r i u m l i q u i d phases no s u b s t a n t i a l d i f f e r e n c e c o u l d be found on comparing t h e e f f e c t o f a carbon d i o x i d e s l u g and a l a r g e r volume o f carbonated n a t u r a l gas w i t h higher carbon d i o x i d e content. s w e l l i n g and v i s c o s i t y o f t y p i c a l r e s e r v o i r waters . Nonconsolidated r e s e r v o i r sandstone cores and r e s e r v o i r f l u i d were used f o r these displacement t e s t s .

i . To reach t h e i n j e c t e d e q u i l i b r i u m c o n c e n t r a t i o n o f t h e carbonated water i n t h e e f f l u e n t 4-8 p o r e volumes o f s a t u r a t e d water were necessary. Carbonated water s a t u r a t e d a t r e s e r v o i r p r e s s u r e and t e m p e r a t u r e was i n j e c t e d i n t o t h e p r e v i o u s l y water f l o o d e d c o r e . Gaseous carbon d i o x i d e was i n j e c t e d i n t o t h e model when s t u d y i n g t e r t i a r y r e c o v e r y methods f o r d e p l e t e d r e s e r v o i r s . and f a v o u r a b l e e f f e c t s o f f r e e gas s a t u r a t i o n were observed. I f t h e w e t t a b i l i t y o f r e s e r v o i r r o c k changes f r o m water-wet t o o i l . b u t b e t t e r r e s u l t s i n most 2 cases i f t h e p r o p e r d i s p l a c e m e n t t e c h n o l o g y i s used. -50-60 $.w e t t h e f a v o u r a b l e e f f e c t s o f f r e e gas s a t u r a t i o n and t h e l a t e r w a t e r f l o o d i n g a r e reduced. g i v e n o t worse. I f carbon d i o x i d e i s i n j e c t e d i n t o t h e d e p l e t e d o i l r e s e r v o i r i t i n t e r a c t s w i t h t h e r e s e r v o i r f l u i d and component mass t r a n s f e r s t a r t s among t h e phases. The a d d i t i o n a l o i l r e c o v e r y was always r e l a t e d t o t h e r e s i d u a l o i l s a t u r a t i o n o f t r a d i t i o n a l water f l o o d i n g .302 As a f i r s t s t e p t h e e f f e c t o f carbonated w a t e r was examined.p h a s e a possible increase i n the o i l . t o o . t h e r e b y changing t h e w e t t a b i l i t y p r o p e r t i e s o f t h e system and l e a d i n g t o t h e r o c k becoming more o i l . t h e a d d i t i o n a l o i l r e c o v e r y was 12-16 % o f t h e o r i g i n a l o i l i n p l a c e f o r systems h a v i n g a h i g h i n i t i a l gas s a t u r a t i o n and 8-12 % f o r t h e case o f h i g h i n i t i a l w a t e r s a t u r a t i o n .. . Using t h i s . A f t e r t h e p r e s s u r e b u i l d . However even i n t h i s case.t h e d e p l e t e d r e s e r v o i r has a h i g h gas s a t u r a t i o n . The a d d i t i o n a l o i l was 5-7 % o f t h e o r i g i n a l o i l an p l a c e . t o o .w e t character counteracts these favourable e f f e c t . mentioned above were performed w i t h p r a c t i c a l l y p u r e carbon d i o x i d e . 3-5 p o r e volumes o f carbonated w a t e r were used t o obtain t h i s result. The a d d i t i o n a l o i l reached 10 % o f 0 .2 PV. o v e r s a t u r a t e d water was used i n t h e n e x t s e r i e s o f experiments. A l t h o u g h t h e carbon d i o x i d e c o n t e n t o f t h e o i l phase decreases t h e v i s c o s i t y o f c r u d e r i c h i n h i g h m o l e c u l a r components and s w e l l s t h e o i l . Two d i f f e r e n t i n i t i a l s a t u r a t i o n c o n d i t i o n s were used as average r e s e r v o i r c o n d i t i o n s f o r m o d e l l i n g p r o d u c t i o n his t o r i e s : .r e s e r v o i r water--carbon d i o x i d e . Carbon d i o x i d e appeared i n t h e e f f l u e n t a f t e r i n j e c t i n g one p o r e volume o f s a t u r a t e d w a t e r . The r e s u l t s showed t h a t t h e use o f c a r b o r a t e d n a t u r a l gases h a v i n g a CO c o n t e n t above 80 mole $. p . The parameters i n f l u e n c i n g t h e e f f e c t i v e n e s s o f C02 f l o o d i n g must be i n d i v i d u a l l y determined f o r each p r o j e c t . Consequently. A l l o f t h e dynamic displacement t e s t s . The p r o b a b l e o p t i m a l s l u g s i z e was about 0. I n extreme cases some o f t h e components w i t h i n t e r f a c i a l a c t i v e c h a r a c t e r i s t i c s may adsorb on t h e r o c k s u r f a c e .r e s e r v o i r r o c k systems. . T e s t s were conducted u s i n g carbonated n a t u r a l gases. on t h e q u a n t i t y o f i n j e c t e d C02 ( 2 ) .. Because o f t h e i n j e c t i o n o f a l a r g e volume o f w a t e r and t h e modest a d d i t i o n a l o i l recovery. T h i s change of w e t t a b i l i t y depends upon many f a c t o r s among o t h e r s . Complex f l o w c o n d i t i o n s and physico-chemical processes e x i s t i n r e s e r v o i r o i l . The a d d i t i o n a l o i l r e c o v e r y a s a f u n c t i o n o f s l u g s i z e was s t u d i e d .t h e d e p l e t e d r e s e r v o i r has a low gas s a t u r a t i o n and h i g h w a t e r s a t u r a t i o n . -25-35 %.w e t . t h i s method i s uneconomic. To i n c r e a s e t h e amount o f i n j e c t e d carbon d i o x i d e . The p r e s s u r e was increased t o t h e o r i g i n a l r e s e r v o i r p r e s s u r e by i n j e c t i n g carbon d i o x i d e gas.u p . As a r e s u l t o f t h i s process t h e o i l phase w i l l be r i c h e r i n components h a v i n g h i g h e r m o l e c u l a r w e i g h t s . t h e a d d i t i o n a l o i l was produced w i t h a r a t h e r h i g h water c u t . d i f f e r e n t s i z e s o f C02 s l u g s were i n j e c t e d and f o l l o w e d by r e s e r v o i r w a t e r f l o o d i n g .

COMPUTER MODEL A three-phase. carbon d i x i d e 7+ and water. as well. The governing d i f f e r e n t i a l equations o f t h e compositional model w r i t t e n i n a usual form a r e as f o l l o w s : = div [5 2 j Aj kjfj C. The components a r e : seven hydrocarbon components /C.I0 water j = gas.t h e d i s s o l v e d gas i n t h e water phase c o n s i s t s o f carbon d i o x i d e o n l y . b u t t h e l i t t l e i n f l u e n c e o f t h i s on t h e phase e q u i l i b r i u m made i t reasonable t o n e g l e c t it. o i l .igrad C j + 9i i = l. C6.. The b a s i s o f t h e c a l c u l a t i o n s i s t h e assumption t h a t local thermodynamic e q u i l i b r i u m e x i s t s d u r i n g displacement. P a r t I . C3. Besides three-phase regions. t h e s o l u t i o n o f t h e carbon d i o x i d e component i n t h e water phase cannot be neglected. t h e formation f l u i d o f Budafa o i l f i e l d was considered as a ten-component system.w e t . I n t h i s way. two-. C2. (grad p j Jti + $9 grad z) + @zSjSjDj. C /..Z. As t h e water phase e x i s t s everywhere i n t h e formation.303 R e l a t i v e p e r m e a b i l i t y curves f o r saturated carbonated water systems were a l s o measured. and d u r i n g t h e water i n j e c t i o n a g r e a t amount of carbon d i o x i d e i s t o be transported by water. C4. ten-component mathematical model has been developed t o study carbon d i o x i d e displacement experiments and t o p r e d i c t performances (3. The three-phase e q u i l i b r i u m was i n t e r p r e t e d as t h e simultaneous existence o f two two-phase e q u i l i b r i u m s . Decreasing o i l and i n c r e a s i n g water p e r m e a b i l i t i e s c o u l d be observed i n c e r t a i n s a t u r a t i o n ranges. depending upon t h e CO content o f t h e gas used./ .t h e gas and o i l phases do n o t c o n t a i n a water component. The increase 2 i n r e s i d u a l o i l s a t u r a t i o n was a l s o observed w i t h i n c r e a s i n g C02 content. /When checking t h e c a l c u l a t i o n s t h e d i s s o l v e d gas i n t h e water phase contained methane. moreover one-phase regions occur d u r i n g t h e processes. To s i m p l i f y t h e e q u i l i b r i u m c a l c u l a t i o n s t h e f o l l o w i n g assumptions were made: . C5. The c h a r a c t e r o f r e l a t i v e p e r m e a b i l i t y curves j u s t i f i e d t h e e f f e c t mentioned above. . nitrogen. thus a method has been developed t h a t a l l o w s one t o e a s i l y c a l c u l a t e a change i n t h e number o f phases. t h e r e l a t i o n s h i p s c o r r e l a t e d w i t h l a b o r a t o r y PVT and e q u i l i b r i u m measurements can d i r e c t l y be employed. ) . I n accordance w i t h t h e l a b o r a t o r y measurements. .. Under some circumstances t h e porous medium became more o i l . The bases o f comparison were t h e r e l a t i v e curves o f hydrocarbon gas saturated s y s terns..

I n o r d e r t o overcome t h e energy reduction. I n accordance w i t h these c o r r e l a t i o n s t h e formation volume f a c t o r i s c a l c u l a t e d as a f u n c t i o n o f t h e d i s s o l v e d g a s / f l u i d r a t i o f o r both f l u i d phases. edge water d r i v e worked w h i l e t h e formation pressure decreased t o an average level o f 2900 kP3. Thus t h e q u a n t i t y o f t h e dissolved gas has t o be known. The e f f e c t i v e formation t h i c k n e s s v a r i e s from 1-2 m t 30 m. The sandstone formations o c c u r r i n g a t an average depth o f 850 m have a temp e r a t u r e o f 64 OC. convergence problems o f i t e r a t i v e techniques occur. and a numerical procedure combining t h e Newton-method and t h e method o f halving. 139 m i l l i o n m hydrocarbon gas was i n j e c t e d i n t o t h e r e s e r v o i r from 1942 t o 1958. T h i s accumulation was unfavourable from t h e p o i n t o f view o f t e r t i a r y recovery because t h e o i l zones o f t h e two lower l a y e r s were s i t u a t e d under t h e gas caps o f t h e two upper I ayers.817. t h e c a l c u l a t i o n o f three-phase e q u i l i b r i u m make i t necessary t o s o l v e a coupled system o f two n o n l i n e a r a l gebraic equations. During t h e primary and secondary displacements t h e s o l u t i o n gas d r i v e . The i n i t i a l pressure level j u d g i n g by t h e h y d r o s t a t i c c o n d i t i o n a t t h e beginning o f p r o d u c t i o n was 9800 kPa. Occasionally. which was considered as an 3 abandon pressure. F o l l o w i n g t h e r a i d increase i n t h e ! ? number o f wells. The area i s a s e c t i o n o f t h e Lower-Pannonian /Lower-Pliocene/ Budafa r e s e r v o i r which c o n s i s t s o f f o u r separable sequences o f s t r a t a o f t h e same hydrodynamic system. The energy o f the formation decreased because o f t h e h i g h p r o d u c t i o n l e v e l and r e s t r i c t e d egge water d r i v e . . t h e d i s s o l v e d g a s / o i I r a t i o can be determined from t h e composition o f t h e o i l phase by normal f l a s h c a l c u l a t i o n . i t s average d e n s i t y a t 20 C being 0. mainly when t h e number o f phases changes. The producgd crude i s o f an i n ermed a t e . t o a s l i g h t extent.10' kg/mf. FIELD EXPERIMENT A f t e r some p i l o t t e s t s t h e f i r s t large-scale process was s t a r t e d i n t h e western area o f t h e Budafa o i l f i e l d i n 1972. Very good matches could be achieved by modifying t h e molecular weight o f t h e C7+ component by 5 $. As f o r t h e dissolved gas/water r a t i o . The average p o r o s i t y i s 21 $. I n order t o check t h e PVT and e q u i l i b r i u m c a l c u l a t i o n s l a b o r a t o r y pressure-build-up measurements were simulated by a one-volume element model. however. o n s a r e used t o d e t e r mine t h e mole f r a c t i o n s o f t h e phases. When determining d e n s i t i e s o f t h e f l u i d phases t h e labor a t o r y c o r r e l a t i o n s a r e applied. i t was assumed t h a t water i n i t s normal s t a t e i s f r e e o f gas.p a r a f f i n character.6 $. ensure f a s t convergence i n every case.800 m /year i n t941 which was t h e peak p r o d u c t i o n o f t h i s area. Because t h e composition o f t h e phases i s known. f l a s h c a l c u l a t i . The average recovery e f f i c i e n c y was 22. A t o t a l o f 1 m i l l i o n m o i l and 600 m i l l i o n m gas was produced. crude p r o d u c t i o n amounted t o 89.304 When c a l c u l a t i n g t h e phase e q u i l i b r i u m . The r e s e r v o i r o i I was i n i t i a l l y saturated. The d e n s i t y o f t h e gas phase i s c a l c u l a t e d using t h e Redlich-Kwong equation o f s t a t e . t h e average h o r i z o n t a l 9 p e r m e a b i l i t y 0. t h e two upper sequences o r i g i n a l l y had an e x t e n s i v e gas cap. Production was begun i n J u l y o f 1937.1 pm . The system was transformed i n t o one n o n l i n e a r a l g e b r a i c equation. The formations a r e heterogeneous v e r t i c a l l y and h o r i z o n t a l l y . t h e energy o f gas caps and.

305 I .

and t h e system c o u l d be c h a r a c t e r i z e d by an i r r e g u l a r m u l t i . an average v a l u e . The changed c o n d i t i o n s can be seen i n F i g . 71 p r o d u c t i o n and 9 o b s e r v a t i o n w e l l s . The carbon d i o x i d e c o n t e n t o f t h e produced gas remained above 65 mole % d u r i n g t h e w a t e r i n j e c t i o n . t h e g a s / o i I r a i o decreased f r o m t h e p r e v i o u s y e a r s ’ l e v e l o f 5000 m /m t o a b o u t 600 m3/$. t h e Budafa-West m u l t i . t h e water i n j e c t i o n r a t e f 5 0 0 m /day. T h i s disadvantageous e f f e c t was caused by t h e h i g h gas s a t u r a t i o n d a t i n g back t o t h e p r i m a r y and secondary r e c o v e r y .900 kPa. which made i t e v i d e n t t h a t i n j e c t i o n o f a d d i t i o n a l c rbonated n a t u r a l gas was n o t necessary. The w e l l p a t t e r n used i s shown i n F i g .072 m i l l i o n m w a t e r had been produced and t h e average r e c o v e r y e f f i c i e n c y had been 27.7 % 0 . t h e a d d i t i o n a l o i l from Section I I . I n t h e f i r s t phase o f t h e t e r t i a r y recovery.9 T h i s amount o f a d d i t i o n a l o i l i s . q u i t e considerable i n t h a t t h e e a r l i e r value was ‘12. The f i n a l amount o f a d d i t i o n a l . HISTORY MATCH AND PREDICTION The f i e l d e x p e r i m e n t was analysed by s i m u l a t i o n o f performance h i s t o r y (3. G R response t o w a t e r f l o o d i n g was observed f r o m t h e end o f 1974 when t h e O c h a r a c t e r o f p r o d u c t i o n changed remarkably. t h e e x i s t i n g w e l l s i n t h e a r e a were t a k e n i n t o account.). r e l a t i v e l y homogeneous s e c t i o n s h a v i n g low w a t e r s a t u r a t i o n .7 I The p r o d u c t i o n o f a d d i t i o n a l o i l proved t o be economi c a l l y worth w h i l e . and t h e e f f e c t i v e n e s s was poor. Gas and l i q u i d f l o w always o c c u r r e d s i m u l t a n e o u s l y i n t h e l a y e r s . When d e s i g n i n g t h e technology. U n t f I 1 s t January 1981. carbonated n a t u r a l gas was i n j e c t e d i n t o t h e f o r m a t i o n d u r i n g which c o n t r o l l e d p r o d u c t i o n was r e a l i z e d . i . The carbonated n a t u r a l gas used was produced f r o m a h i g h p r e s s u r e r e s e r v o i r d i s c o v e r e d i n t h e a c t u a l area o f Budafa. 1. The carbon d i o x i d e appeared i n t h e p r o d u c t i o n w e l l s 1-2 months a f t e r t h e b e g i n n i n g o f i n j e c t i o n .l a y e r s f o r m a t i o n under t h e gas caps. w a t e r i n j e c t i o n was s t a r t e d a t t h e g a s . I t should be mentioned t h a t t h e g r e a t e r p a r t o f t 9 e i n j e c t e d gas was used t o f i l l up t h e gas caps. o i l expected i s 5.013 m i l l i o n m w a t e r had been i n j e c t e d i n t o t h e f o r m a t i o n . t h u s an a r e a l model was used and t h e e f f e c t s o f c a p i l l a r i t y and g r a v i t a t i o n were n e g l e c t e d .s p o t p a t t e r n .was i n j e c t e d i n t o t h e low p r e s s u r e o i l r e s e r v o i r by means o f n a t u r a l energy. a b o u t 1-2 % f o r t h e m u l t i . 2. For example. 3 %. however. 2/ show t h a t t h e GOR amounted t o a v e r y h i g h l e v e l /3000-5000 m /m / d u r i n g t h e i n j e c t i o n . 694 m i l l i o n m carbonated n a t u r a l gas and 3. T e r t i a r y r e c o v e r y by carbonated n a t u r a l gas was r e a l i z e d by u s i n g 41 i n j e c t i o n .h a v i n g a carbon d i o x i d e c o n t e n t o f 81 m l e % and l i g h t hydrocarbons . and t h e whole a r e a was w a t e r f l o o d e d f r o m t h e summer o f 1975. T h i s gas . Along w i t h i n c r e a s i n g o i l p r o d u c t i o 3 r y t e . Because of t h e complex p e t r o g r a p h i c and h e t e r ogeneous s a t u r a t i o n c o n d i t i o n s .l a y e r r e s e r v o i r . A t t h a t t i m e t h e averag? p r e s s u r e o f t h e r e s e r v o i r was 10. 173.306 A t t h e b e g i n n i n g o f t e r t i a r y r e c o v e r y t h e o i l zones o f t h e a r e a had a h i g h gas s a t u r a t i o n .5 %. t h u s t e r t i a r y r e c o v e r y r e s u l t e d i n a d d i t i o n a l o i l o f 3. P a r t I I. The d i s p l a c e m e n t i s s t i l l c o n t i n u i n g . laminated and n e a r l y h o r i z o n t a l . The method has proved t o be s u c c e s s f u l f o r one-layer. p . heterogeneous. Data r e l a t i n g t o i n j e c t i o n and p r o d u c t f o n 3 r a t e s / F i g . By January 1981. The r e s e r v o i r i s t h i n .000 m o i l and 3 1.o i l c o n t a c t o f t h e two upper l a y e r s i n t h e autumn o f 1974. I n o r d e r t o d i m i n i s h t h e h i g h GOR v a l u e o f t h e p r o duced f l u i d . No o i l bank f o r m a t i o n c o u l d be observed i n any o f t h e p r o d u c t i o n w e l l s .

L ---n I FIGURE 2 Grp Butlofa-West Unit performcnce history 307 .

w i d e t e r t i a r y displacement by carbon d i o x i d e c a r r i e d o u t i n Hungary. W had t o assume i n t h e m o d e l l i n g . R e l a t i v e p e r m e a b i l i t y c u r v e s f o r three-phase carbonated systems were n o t a v a i l a b l e . 4 . and parameters o f t h e c u r v e s were determined by h i s t o r y matching. Based upon l a b o r a t o r y measurements and p u b l i s h e d d a t a a s i m p l e f o r m o f p a r a m e t r i c r e l a t i v e p e r m e a b i l i t y c u r v e s were c o n s t r u c t e d .s p o t element. 3 y e a r s o f water i n j e c t i o n / were used. / The s e c t i o n i s shown i s F i g . we c o u l d n o t d e s c r i b e a l l t h e i n j e c t i o n and p r o d u c t i o n w e l l s o f t h e s e c t i o n .5 y e a r s /2. For t h i s reason an e a s i l y separable. The f i g u r e shows p r e d i c t i o n s u n t i l 1983 t o g e t h e r w i t h t h e a c t u a l p r o d u c t i o n parameters o f t h e l a s t t h r e e y e a r s . To some e x t e n t we had t o m o d i f y t h e i n j e c t e d gas t o g e t t h e good p r e s s u r e h i s t o r y match. r e l a t i v e l y homogeneous s e c t i o n s h a v i n g low w a t e r s a t u r a t i o n . t h e t h i c k n e s s was d i v i d e d i n t o a good and a poor p e r m e a b i l i t y l a y e r . t h e r e f o r e on t h e bases o f a v e r a g i n g t h e d i s t a n c e s o f t h e i n j e c t i o n and p r o d u c t i o n w e l l s o f t h e s e c t i o n an e i g h t h o f a f i v e . 3 .2 % f o r t h i s s e c t i o n . The r e s u l t s o f t h e h i s t o r y match can be seen i n F i g .5 y e a r s o f gas i n j e c t i o n . The h e t e r o g e n e i t y was t a k e n i n t o a c c o u n t i n a s i m p l e way. P r e s s u r e and p r o d u c t i o n d a t a o f 5. T h i s f a c t was proved by t h e c a l c u l a t i o n s . / I t i s t o be noted t h a t d e t a i l e d d a t a on f o r m a t i o n parameters were a l s o i n a c c e s s i b l e . CONCLUSIONS A f t e r t h o r o u g h and e x t e n s i v e s t u d i e s economic f i e l d . t h e model was a p p l i e d t o t h e o t h e r 5 s e c t i o n s o f t h e area. . V o l u m e t r i c and phase behaviour o f t h e three-phase system can be modelled w i t h good a c c u r a r c y u s i n g t h e l a b o r a t o r y c o r r e l a t i o n s . The CO d i s p l a c e m e n t has proved t o be s u c c e s s f u l f o r 2 one-layer. Because o f computer r e s t r i c t i o n /an ICT 1905 computer w i t h a memory o f 32 Kwords was used/. and a c a l c u l a t e d gas l o s s o f o n l y 6 % r e s u l t e d . /Primary and secondary displacement I r e s u l t e d i n 45. t h i s i s n o t t h e case. t h e s e m o d i f i c a t i o n s were e q u a l i z e d f r o m t h e v i e w p o i n t o f t h e whole area. A n a l y s i s o f h o r i z o n t a l p e r m e a b i l i t y d i s t r i b u t i o n i n t o v e r t i c a l d i r e c t i o n u s i n g c o n t i n u o u s c o r e samples o f t h e r p s e r v o i r examined showed t h a t 20 % o f t h e p e r m e a b i l i t y d a t a were above 0. The r e s u l t s o f h i s t o r y matching a r e summarized i n F i g . and t h e average r o c k parameters and i n i t i a l s a t u r a t i o n r e f e r r i n g t o t h e b e g i n n i n g o f t h e t e r t a r y r e c o v e r y were used. / The s e c t i o n was c o n s i d e r e d t o be o f c o n s t a n t t h i c k n e s s . The f l o o d i n g process i s v e r y s t r o n g l y i n f l u e n c e d by t h e presence o f h i g h p e r m e a b i l i t y zones. However. The i n j e c t i o n and t h e p r o d u c t i o n d a t a o f t h e model were c a l c u l a t e d from t h e c u m u l a t i v e d a t a o f t h e s e c t i o n u s i n g t h e p o r e volume r a t i o o f t h e s e c t i o n and t h o s e o f t h e e i g h t h o f t h e f i v e . A f t e r h a v i n g good r e s u l t s on t h e h i s t o r y match f o r S e c t i o n I I . as i s t o be expected. W wished t o make use o f a l l t h e measured e data.s p o t element was cons t r u c t e d . I t seemed t h a t no parameter group can be chosen t o s i m u l a t e e a r l y breakthrough o f carbon d i o x i d e .308 c o n s t i t u t e s a c o m p l i c a t e d system. ‘ as Section 1 1 . 15-25 s i m u l a t i o n s were used t o reach t h e f i n a l r e s u l t s f o r each case. The computed average p r e s s u r e s d i f f e r e d f r o m t h e measured ones by o n l y a b o u t 5 %.31 pm which d i f f e r e d remarkably f r o m t h e average v a l u e . The c a l c u l a t i o n s were performed i n 1978.l a y e r s e c t i o n o f t h e r e s e r v o i r was examined b e l o n g i n g t o t h a t a r e a where t h e h i g h e s t amount o f o i l o r i g i n a t e d from. h o r i z o n t a l . o u r i n t e n t i o n was t o o b t a i n an o v e r a l l p i c t u r e o f t h e process. e t h a t no f l o w b o u n d a r i e s e x i s t e d between t h e s e c t i o n s though. one.

Comparison of measured and cornouted data Budofo-West 309 unit Sert. .FIGURE 3.on II.

/ r01...mom\ FIGURE L. Comparison of measured a n d computed data Budafa-West unit .b c / / ' / .r .

... Ko'olaj 6 s Folzga'z. S.. T. Z. B a ' l i n t . i n t h i s case. B i r 6 . Co. I. Nedra Publ. 1979 3. Paa'l. Zabrodin. Farkas c ... 1977 2. D o l e s c h a l l . : "HBromfa'zisu.r e n d s z e r e k por6zus kozegben V a l 6 a'ramoltata'sakor~-/"Changes o f w e t t a b i I i t y c o n d i t i o n s and f l o w c h a r a c t e r i s t i c s f o r f l o w i n g c a r b o n d i o x i d e s a t u r a t e d f l u i d system i n porous media"/. Feb.. kompozici6s modell 6s alkalmaza'sa a Budafa-nyugat t e l e p sz6n-dioxidos muvel6s6nek l e ira'sa'ra" /"A three-phase. 1981 . G. c o m p o s i t i o n a l model and i t s a p p l i c a t i o n f o r d e s c r i b i n g CO d i s p l a c e m e n t o f t h e Budafa-West r e s e r v o i r " / . D o I e s c h a I I . Ba'n. S. KColaj 6s Foldga'z. Torok. P a r t I I. P a r t I. J. Ko'olaj 6 s Foldga'z. t o be a s u b s t a n t i a l f a c t o r i n a c h i e v i n g economic a d d i t i o n a l o i l p r o d u c t i o n t h e r e b y overcoming t h e e f f e c t s o f unfavourable r e s e r v o i r conditions. V. Moscow.311 U t i l i z a t i o n o f t h e l o c a l p o t e n t i a l proved.: " P r i m e n e n i j e u g l e k i s l o v o gaza v d o b i c h e n e f t i " / " A p p l i c a t i o n o f carbon d i o x i d e i n o i l p r o d u c t i o n " / . i981. Jan. NOMENCLATURE C mass c o n c e n t r a t i o n di f f usi v i t y gravitational acceleration permeability tensor r e l a t i v e permeability pressure mass s i n k p e r u n i t volume p e r u n i t t i m e sa t u r a t i on time depth viscosity density porosity D 9 K - k P 9 S t Subscripts i j r e f e r s t o i t h component r e f e r s t o j t h phase REFERENCES 1.: "A n e d v e s i t 6 s i a ' l l a p o t 6s a z a'ramla'si jellemzo'k va'ltoza'sa CO d a l t e l i t e t t f l u i d u m . B a ' l i n t . Nov. Acs. P. V. A...

This Page Intentionally Left Blank .

MISCIBLE GAS DISPLACEMENT 313 AN ITERATIVE METHOD FOR PHASE EQUILIBRIA CALCULATIONS WITH PARTICULAR APPLICATION TO MULTICOMPONENTMISCIBLE SYSTEMS NIKOS VAROTSIS. The paper discusses briefly the specific use of the model in an experimental phase behaviour study for UK oil-C02 systems and as an integral part of a compositional reservoir simulator. injection of CO requires a knowledge of the vapour-liquid equilibria especially 2 . ADRIAN C. GEORGE STEWART Petroleum Engineering Department. modified Soave-Redlich-Kwong and . The use of CO to improve oil recovery is not a new idea since C02 has been investigated $or miscible displacement. The co extracts hydrocarbons from the oil phase and 2 at the same time co2 is absorbed into the liquid phase up to the moment that The study and prediction of oil recovery involving miscibility is achieved. Heriot. iteration method used requires a minimum number of variables for which simultaneous iteration is required and an algorithm based on the Broyden's modification of the full Newton step gives consistent phase properties and rapid convergence even near the very sensitive for a miscible displacement critical point area. The model has been tested against published data including simple binaries. TODD. for immiscible displacement of reservoir oil. The current industry interest in co flooding is mainly concentrated on the mass transfer effect that takes place begween the injected CO2 phase and the reservoir oil inside the reservoir.Watt University ABSTRACT equation of state based method is used to establish phase behaviour and properties for mixtures of injection gases and reservoir fluids with specific application to multicomponent miscible systems including CO2' An The modified Soave-Redlich-Kwong or the Peng-Robinson or a version of the The Redlich-Kwong equation of state can be selected to be used in the model. A method is needed. at the very sensitive critical point. for producing well stimulation and for carbonated water flooding. Such a model will be described which using any of the Peng-Robinson. Good agreement between the predicted and the experimental values has been found together with a minimum number of iterations required to solve each problem. first to calculate the saturation conditions for the mixtures of the injected gases and reservoir oils from which a prediction of the miscible pressure can be made and second to carry out the isothermal flash calculations for different pressures so that the phase behaviour of the system can be studied in detail. INTRODUCTION One of today's more promising oil recovery techniques is miscible C02 flooding. ternaries and multicomponent mixtures of reservoir oil and C02 injection gases.

extrapolation technique is used to improve the initial estimates for the consequative calculations of the saturation pressure of a reservoir oil-CO2 mixture across the phase envelope and up to the crLtica1 point. SCHEMATIC TERNARY DlAGRAM . An MISCIBILITY MECHANISMS - DIFFERENT MODELLING APPROACHES Two of the most important and promising gas injection enhanced oil recovery practices are C02 flooding and lean gas injection. The major mechanisms to improve the oil recovery in a carbon dioxide flooding are vaporization and condensation. Although the model has been specifically applied to CO -oil systems is obviously applicable to 2 any injected gas or flowing system. This representation although not quantitative demonstrates how F I G U R ~ 1. The extraction of hydrocarbons by CO and its condensation into 2 the reservoir fluid results finally in an one phase miscible fluid. The development of miscibility can be visualised conceptually with a ternary diagram (Figure 1 ) .314 a version of the Redlich-Kwong equations of state can give predictions of the vapour-liquid equilibria of multicomponent mixtures and especially good and rapid convergence in the critical point region where most of the methods according to the literature fail to converge. Mass transfer takes place between the C02 rich phase and the oil rich phase and the initially immiscible phases gradually become miscible as they are enriched in intermediate and even heavy hydrocarbons and C02 respectively.

critical point and it is its relative position in respect of the point that represents the reservoir fluid composition that defines whether under certain conditions the mixture of the injected qas and the reservoir fluid can obtain miscibility (Figure 2). The same remarks apply more or less for a lean gas injection flooding where the vaporization of the light hydrocarbons from the reservoir fluid to the gas phase controls the whole process.2high solubility of CO in water which reduces the water density and therefore the overriding of the ca2-water mixture and the acidic effect on the rock which increases the permeability of the reservoir. There are also some minor mechanisms to improve the enhanced oil recovery by injection of CO These are: oil swelling. reduction of oil viscosity. The phase envelope of the mixture at different conditions is required to determine the minimum miscibility pressure and the equilibrium lines (tie-lines) in order to study in detail the distribution of the different components in the two-phases. The theoretical studv of a miscible displacement experiment or of a miscible reservoir flooding requires accurate and reliable phase behaviour data. increase in oil density.315 important it is to be able to predict the critical point of a mixture for a The miscibility path passes throush the multiple contact miscible process. The requirement for the generation of a miscible displacement is that the reservoir fluid composition must lie either to the right of the extension of the tangent to the phase boundary curve at the critical point or above the critical point in the single phase region. MISCIBILITY CONDITIONS . CRITICAL POINT A TWO PHASE REGION FIGURE 2. .

542261 - 2 026992W -.45724 RT b = 0. Due to inconsistent phase properties near the critical point and the requirement for comprehensive experimental data for each oil composition of the latter.m = 0. the key area for any miscible displacement. B R T r2 = bP RT (for pure components) . The system of equations consists of: (i) An overall material balance equation L + V = 1 (ii) Component material balance equations Lxi + vy = zi i = l.3 16 Either an equation of state based method is used to establish phase behaviour and properties or equations are used which have been obtained by curve fitting experimentally derived data. Most of the current published equation of state based methods appear to suffer from requiring a great number of iterations or do not converge at all in the critical point area.n Three different equations of state can be used to provide values for the These are: compressibility factor of the vapour and liquid phase.0778 pC R2Tc2 pC .37464 + 1. (1) The Peng-Robinson equation of state p=-RT v-b v(v+b) T a( I + b(v-b) (P-Rl or in terms of the compressibility factor: Z3 where: - (1-B)Z 2 + 2 (A-3B -2B)Z - 2 3 (AB-B -B ) = 0 a(T) = 0. the equation of state based method is now widely preferred. i = l n c y i = l i = l (iv) Thermodynamic phase equilibria equations fiL = fiv i = l.n i (iii) Restrictive equations on the phase compositions n c x i = l .A = J il) +mT (: . PHASE EQUILIBRIA MODEL FOR A MISCIBLE OIL RECOVERY PROJECT The technique being presented here for calculating vapour-liquid equilibria using an equation of state includes a system of non-linear equations and an iterative sequence to solve the equations.

0794 0.07259 Ci nC4 nCIO 0.0867 2.46512 .4198 0.) m = 0.5 = S RTC a = R A Tc -b t pc & C ~ h = bP s A = 2 2.0734 RA COATS 0.4154 0. 10 RB 0.0759 0. 2 R T lml.551711 - 0.4265 0.'.0794 0. RB calculated by our model are compared against those obtained bg Coats and Fussell for a ternary mixture of C . B = - aP R T bP RT (pure components).0862 0.nB are supposed to be functions of temperature and of the nature of each component.A = pC B = bP RT - (for pure components) (3) Modified Redlich-Kwong Equation of State RT p=--v-b aT-4 v(v+~) (M-R-K) or in terms of the compressibility factor where: 2 2.070452 A ' FUSS % FUSS 0.nC at 160 F (344.317 (2) The modified Soave-Redlich-Kwong equation of state: p - RT v-b v(v+b) (M-S-R-K) or in terms of the compressibility factor Z -Z 3 2 2 +(A-B-B )Z-AB = 0 where: a(T) = 0.0859 0.086173 0.419367 0.42727 R2T pC J -ap 2 .nC4 . RA.3K). In Table I values of the parameters RA.15613W 2 RT b = 0. +(-.42617 0.451875 % COATS 0.48508 + 1.4251 0.4638 0. The values of these parameters are calculated from generalised correlations applicable over a wide range of temperature.

297 0. 3. experimental ones are given in Table 11.exp{b 1 iv (Zv-l) 1 4 i S n fiv - B i' (2 -B ){I+ " ) v v zV .661 : 0.174 0. 972.7 psia 975. . K-Val.1 ) ) - (2 -B ) L L iL L B~ i ' {1+ -} zL 1<-lSn For the vapour phase: - Px .exp{b fiL (2 .5 (1-ki j ) b = n C xibi i=l . = ai 0.00 3 -174 0. EXPER.2969 0086 .008 Satur.013 0.1 psia 1 0 psia 00 For multicomponent mixtures the following mixing rules proposed by Soave are used: n a = i=l n j=1 C C x x ai.086 the K-values and the saturation pressure estimated using the Modified Redli' Kwonq Equation of State compared to the values predicted by Coats and to th. aij .173 nC 4 nC 10 3.297 0. Press.318 For the same mixture and for composition (mole fraction) CH4 n-Butane n-Decane : : : 0.253 0. Kij = interaction parameter The fugacity coefficients of component i in a mixture are calculated using the following equations For the liquid phase: Px.5 a 0. COATS K-Val OUR MODEL K-Val.

L and V. = AL(2aiL-biL)/BL. A . Wi = Av (2a -biv)/Bv iv The equilibrium ratios K are defined as: i L Ki = Yi/xi = (fi /xiP)/(fiV/YiP) = $iL/$iv DESCRIPTION OF THE PROGRAMME The programme is written in Fortran Iv language and is implemented as a conversational time-share package. correction step is accomplished by dividing the unknown variables into two groups. The user is guided through the data input and calculation options by a question and answer sequence at the visual display unit . compressibilities and densities Composition of each phase by mole fractions x i. critical temperatures and pressures and accentric factors.3 19 where: U. There are four modes of calculations. The first group contains iteration (independent) variables which are The second grouv contains dependent variables the unknowns to be corrected. (i) Isothermal flash calculation (ii) Bubble or dew-point calculation (iii) K-values prediction (iv) Binary coefficient optimization The programme storesphysical properties for the pure components as molecular weights. yi and k-values for each component Saturation pressure or temperature Liquid and vapour phase enthalpies and liquid yield Mass ratio of vapour to feed and volume ratio of vapour to feed Homogeneous mixture density Binary interaction coefficient (if requested) SOLUTION TECHNIQUE Iteration Method minimum variable iteration method is used to reduce the size of the Correction The size reduction of the step by eliminating as many unknowns as possible. The results printed out by the programme comprise the following items: Liquid and vapour phase mole fractions. and there is an equal number of equations to define them.

1 c i n must be less than the error tolerance.i n For the isothermal flash calculations the corresponding saturation conditions are used as initial estimates. Correction Step The correction step used is a modified full Newton step and it requires the calculation of an approximation to the Jacobian obtained by numerical differentation of the function f(G) for the first iteration and the Broyden’s updating technique to improve the matrix for the rest number of iterations.} P i T i ri . For a saturation calculation these n iteration variables are n-1 compositions plus the saturation pressure or temperature. x Error Equations The Euclidean norm of the residuals of the thermodynamic phase equilibria equation Mi - fiL - fiv.37(1+Wi)(1. .320 The iteration sequence is a four step process: - Select the iteration variables and assume values for these variables Use the defining equations to calculate the dependent variables Use the error equations to calculate the error Use a correction step to update the variables - Initial Estimates of the Iteration Variables For the prediction of the saturation conditions a first estimation of the unknown phase composition has to be taken using the component K-values calculated by the empirical equation )/r 1 K = exp {5. For an isothermal flash calculation are n-1 compositions plus the vapour of liquid phase fraction tv or L). o The array contains n elements (the number of components present in the systein).. 1 <. This technique avoids analytical differentation of very complex functions and requires only one numerical differentation of the function f(x) per calculation The value of the iteration variable x at the K + 1 iteration is given by: k+l G -k k = x + A ( x ) where: Btk) is the Broyden’s approximation t the Jacobian.

the extrapolated from the previous calculation initial estimates of the vapour phase CO2 compositions. . Table IV gives the Euclidean norms for the same synthetic mixture with and without extrapolation. Table I11 indicates the total number of iterations required for a complete bubble-point curve calculation.70% Oil No. are. iterations for bubble-point 30% COP . The step for each successive calculation changes as the physical properties of the two phases are approaching each other. (At 18 different compositions of an eleven cmponents synthetic mixture ranging from 0% CO up to 81% CO which corresponds to the 2 2 critical composition). As the overall composition approaches the critical one. the step is being reduced to a minimum because in this region of the phase diagram very small changes in composition cause very significant and radical changes in the phase properties. 2 TABLE I11 - NUMBER OF ITERATIONS FOR A SYNTHETIC MIXTURE No Extrapolation Total No.18% oil Extrapolation 100 46 5 2 5 1 Using the extrapolation technique and the step by step approach to the critical region the final calculation forthe critical Doint itself usually requires only one or two iterations. than 60% the rumber of iterations required to achieve convergence. Table V demonstrates how close to the actual value. Either the estimated phase compositions and saturation conditions of the former step are used as initial estimates of the iteration variables in the next step or these values are extrapolated using a combination of quadratic and linear extrapolation to the new composition The second approach improved the method drastically by reducing more step. The extrapolated values are also compared with those that would be the initial estimates if the extrapolation technique has not been applied. Iterations Bubble-point curve (18 points) No. It also presents the number of iterations required for bubble point calculation for two different compositions of the synthetic oil-CO mixture. Two different approaches have been tried to carry out this series of calculations. iterations for critical point 82% C02 .32 1 EXTRAPOLATION TECHNIQUE TO IMPROVE INITIAL ESTIMATES There is an option available in the programme to calculate the whole phase envelope of a certain mixture of reservoir oil and injection gas starting from the saturation conditions of the reservoir fluid and ending at the critical point of the mixture where usually the injection gas composition is relatively high and the two fluids are becoming first contact miscible.

5329 0.000397 0.0000033 Euclidean norms for bubble point 30% COP 0.5329 0. ternaries.4176 0.00697 0.6356 Actual CO mole 2 fraction next step 30%C02-4O%CO2 40%C02-50%C02 50%C02-6O%CO 0. These tests include binaries.6393 0.322 There is also an option incorporated in the model to plot the pressure-composition data in an X . Iteration No Extrapolation 0.7241 APPLICATIONS - DISCUSSION OF RESULTS In order to test the accuracy of our computer model various calculations have been performed for hydrocarbon/CO mixtures for which the phase behaviour data 2 has been published. synthetic oils and mixtures of injection gases and reservoir oils.EXTRAPOLATED INITIAL ESTIMATES I Synthetic oil CO mixture calculation steps Extrap.000502 0.4704 Extrapolation 0.Y diagram. this method can be applied only when the composition step is very small. .04629 0. TABLE IV - EUCLIDEAN NORMS FOR A SYNTHETIC MIXTURE No. An attempt was made to calculate the matrix used for the iteration sequence only once in the beginning of each series of calculations and then to update it continuously all across the saturation curve avoiding the recalculation of the It has been found that approximation to the Jacobian at each composition step. initial estimates CO mole 2 fraction 0. smethirgthat seems to be time consuming and not practical at all.0000257 TABLE V .5358 0.6356 0. initial estimates CO mole 2 fraction 0.7298 Non extrap.

04-0. S . bubble point curve using the Peng-Robinson equation of state is plotted in Figure 7 . The M-R-K equation of state seems to fit perfectly well the dew-point curve and the critical point. The maximum deviation between the predicted and the experimental points is 3. Synthetic oil C02 system The composition of This mixture has been studied by Metcalfe and Yarborough. The phase envelope was The fitting of the predicted calculated at 100°F and is illustrated in Figure 3 . The maximum observed deviation between the predicted and the experimental points is 3.323 Isobutane - CO.105 was used in the modified Redlich-Kwong equation of state as the iteraction parameter for the mixture. system The phase behaviour of this mixture has been measured by Besserer and Robinson and theoretical predictions reportedbypeng and-inson. N-Butane-Decane-C02 system The experimental data for this ternary mixture has been published by Metcalfe and Yarborough. Figure 5 the synthetic oil and the phase envelope were calculated at 15OoF. have to be obtained in order to establish a complete view of the phase behaviour of the reservoir fluids. .6%. coefficient of 0. . The experimental results will be a test for our compositional simulator and A valid and reliable theoretical especially for the assumptions we used. 0 three pseudo cdmponents and the bubble point curve was calculated at 255 F using the M-R-K equation of state (Figure 7). and Figure 6 present the experimentally obtained pressure-composition data and the predicted data derived using the M-R-K and the M-S-R-K equations of state. Once again in the near the critical point region the fitting of the predicted curve is almost perfect. Reservoir oil-C02 mixture The C7+ cut has been divided into This oil has been studied by Simon et al. oils and reservoir conditions.5% and occurs at the critical composition. A binary phase boundaries to the experimental data is almost perfect. The theoretical model described above. The next stage will be the generation of experimental phase behaviour data for CO2-oils for a whole range of North Sea crudes using a rig for multiple contact equilibrium experiments which has now already started to operate. The phase envelopes were calculated for two different pressures 1700 psia and 1500 psia at 160°F (Figure 4).05. is a part of a research programme on dynamic contact CO miscible studies. model can decrease dramatically the cost of a miscible flooding because only a few experimental results. The various calculations that have 2 already been performed are for U . Rangely field oil - injection gases 1 & 2 The calculated The experimental data has been published by Graue et al. The interaction coefficient used for CO 2 hydrocarbons is 0 1 and for methane -C6+ is 0. which are very expensive and time consuming.

70 .4a .go I . 40.80.20.50. POINTS 0 EXPER.80 .".30.70.0a FIGURE 3.m . .30 .ISO-BUTANE MIXTURE TEMPERATURE=IBBF R-K EOS 50-50 '02- FIGURE 4.20 .60.32 4 1.10. I.OO . PHASE " @@ MOLE FRACTION CARBON DIOXIDE 50-50 ENVELOPES FOR THE CO~-NC~-C~O TERNARY PRESSURE COZ - Nc4 1500. POINTS .I 0 . 1700 PSIA TEMPERATURE 160O~ .OO .PHASE ENVELOPE FOR C02.90Q 6 ) Q c + PRED.so . X .

10I .S-R-K EOS .325 MOLE FRACTION. PHASE ENVELOPE FOR'C02-SYNTHETIC OIL MIXTURE TEMPERATURE=lSBF M0D.CARBON D I O X I D E FIGURE 5 .OO MOLE FRACTION CARBON D I O X I D E FIGURE 6.40- 2 a a 1.501.20I . PHASE ENVELOPE FOR CO2-SYNTHETIC OIL MIXTURE TEMPERATURE=I50F R-K EOS ( 0 a &! L 11 1.30- I .

90 .50.50 2.50 .00w I: ) P I .30 .00.OO GAS.I J .50r I J K l $ 2 w 2.50 6) GAS-2 < I E 3.40 .20 .00 3.60 . BUBBLE POINT CURVE FOR RANGELY FIELD OIL-CM MIXTURES TEMPERATURE=tGBF P-R EOS 2 :2. BUBBLE POINT CURVE FOR CU2-RESERVOIR OIL MIXTURE TEMPERATURE=255F R-K EOS . I0 .001 .326 0 EXPER.OO MOLE FRACTION CARBON D I O X I D E F I G W 7.OO .50- .OO MOLE FRACTION CARBON DIOXIDE FIGURE 8. 1.70 . POINTS 4.

Modified RedlichKwong equation of state has been developed applicable to the severe reservoir conditions encountered in miscible gas flooding enhanced oil recovery schemes. Various tests indicated that the MSRK equation appears in some cases to give closer approximation to the experimental data than the PR equation. iL fiv K i L P pc r ' R T TC Tr V V X i Mole fraction of component i in liquid phase Mole fraction of component i in vapour phase Global mole fraction of component i Compressibility factor of vapour phase Accentric factor Yi i ' z W L .321 CONCLUSIONS robust computer programme for isothermal flash. Soave-Redlich-Kwong. psia Equilibrium ratio y. b = Temperature. bubble and dew point calculations using one of the Peng-Robinson. The treatment of the pseudo components in the oil mixtures is under more investigation. pressure and composition dependent parameters Fugacity of component i in liquid phase. found situations where the MSRK or PR equations of state converge to unrealistic solutions. however./x i i Mole fraction of liquid phase Pressure Critical pressure Reduced pressure P/P Gas constant Temperature Critical temperature Reduced temperature T/T Vapour phase mole fraction Molar volume f. A case has not yet been found where the proposed scheme does not converge although we have. NOMENCLATURE a. A The demonstrated examples show very good and rapid convergence at any point across the phase boundaries of the CO /hydrocarbon mixtures and particularly the 2 sensitive critical point region. psia Fugacity of component i in vapour phase. The MRK equation appears to fit better the dew point curve than the bubble point one.

363 "An Equation of State Compositional Model".. No. "A Compositional Phase Equilibrium Model Applied to Pressure Drop Prediction in North Sea Oil Wells". L. SPEJ.. 10. "Study of a Possible C02 Flood in the Rangely Field. 20 8. Ph. 475 2 SIMON...D Thesis. p.328 REFERENCES 1. 3.D.. "Phase Behaviour Properties of CO2 Systems". Vol. "An Iterative Sequence for Phase Equilibria Calculations Incorporating the Redlich=Kwong Equation of State". SPE 7060. 7. Vol. & Eng. 1973. 1979 FUSSELL.M.. 3. PATEL. P. Eng. ROSMAN.S. No. Colorado". .. Eng. R..Y. . G. . D. SPE 8285. DAUBERT. R...W... "Equilibrium Phase Properties of i-Butane-CO2 System". 1949. 1978 METCALFE. ZANA. L .. SPE 8367. HeriotWatt University. 1976 Equation of State".. F.Reservoir Oil 11. YANOSIK. Chem. 1978 BESSERER. GRAUE. "Phase Equilibria in Hydrocarbon Systems N-Butane CO System". ZANA. SPE 7061. REAMER. ROBINSON. L. Chem.X. 1978 12. LACEY. American Chemical Society Journal. AZIZ. 6. J. "A Modified Soave Equation of State for Phase Equilibrium Calculations". YARBOROUGH. 9. "Effect of Phase Equilibria on the C02 Displacement Mechanism. October 2. 4. K. 1980 COATS. Data. 1. 3. J . E. June 1978 PENG. D. M. p.. SPEJ... D. SAGE. 1980.H.J. p. K. Ind.B.D.. 5. D. "The Effect of Phase Behaviour on CO Flood Displacement Efficiency".S. 1978. T. D. McGLASHAN. Feb. Fundam. GRABOSKI. 298 OLDS. Chem.. "A New %-Constant Ind. 1979 2 NGHIEM. ROBINSON. "A Robust Iterative Method for Flash Calculations using the Soave-Redlich-Kwong or the Peng-Robinson Equation of State". SPEJ. ORR. p. J. GARDNW.

An important problem in equilibrium calculations is to avoid false trivial solutions where the vapor and the liquid phase are identical. An accelerated and stabilized successive substitution method is presented. Newton type methods are however dependent on good initial estimates. making the method convergent also in the single phase region. partly because of the demand for accurate and consistent phase predictions encountered in connection with enhanced oil recovery techniques like gas miscible flooding. INTRODUCTION During recent years. The present work is part of a research project concerned with the development of numerical simulation models for enhanced oil recovery processes. The general conclusion is that the accelerated successive substitution method is faster and much more stable than the Newton type methods considered. They both perform well on hydrocarbon mixtures. This aspect has been discussed by Maddox and Erbar 111. Maps of an error norm which measures the fugacity deviations. Both the Soave-Redlich-Kwong (SFX) equation /1/ and the Peng-Robinson (PR) equation /2/ have been extensively used. Their method was extended to three. They also presented a method to detect single-phase states. It was . In a recent paper Nghiem and Aziz /4/ presented an algorithm using Powell's method which is a combination of a Newton method and the steepest decent method. or even no convergence. are presented to illustrate how the different solution techniques perform. Also Mott / 6 / has presented a two-phase algorithm based on Powell's method. The approach is based on an equation of state. Fussel and Yanosik /3/ introduced Newton type iteration methods. A procedure for disappearing phases is included. To overcome the convergence problem. the PR equation being slightly better in predicting liquid densities. JAN WAR JENSEN Continen tal Shelf Institu te ABSTRACT The present paper addresses the problem of phase equilibrium calculations in the critical point region.and four-phase systems by Mehra 151.'close to saturation pressures. The basic solution procedure for flash calculations is the successive substitution method (SSM).M I S C I B L E GAS DISPLACEMENT 329 PHASE EQUILIBRIUM CALCULATIONS IN THE NEAR-CRITICAL REGION RASMUS RISNES Norsk Agip AIS VILGEIR DALEN. and both the Soave-Redlich-Kwong and the Peng-Robinson equations are considered. The accelerated successive substitution method has been compared with Newton type methods like Powell's method. and since then the trend has been towards such refined numerical solution techniques. hydrocarbon phase equilibrium calculations based on cubic equations of state has received considerable attention. It has however been reported to show poor convergence.

P.. The acceleration routine employs an Aitken type formula for correcting K-values. composition and the type of phase considered. = f. L. the fugacities can be calculated from the equation of state. = y. = (1) L. pressure. The resulting computer program is called COPEC. EQUILIBRIUM CONDITIONS If we consider N moles of mixture or feed of composition z . Formulas for the Soave-Redlich-Kwong (SRK) and the Peng-Robinson (PR) equations of state are given in Table l . the same equation is used both for the liquid and the vapor phase. V. BASIC SUCCESSIVE SUBSTITUTION METHOD The successive substitution method is based on the concept of equilibrium constants K defined by: i K. yi: n is the number of components. The fugacity will depend on temperature. which separate into L moles of liquid of composition x. The method is designed to converge also in the single-phase region. (T. type) 1 1 With cubic equations of state. + vy. f. xi. and V moles of vaior of composition y. and both the SRK and the PR equations are included. A cubic equation may give 3 solutions in volume. As the compositions are given in mole fractions we have the following constraints: zz. xi. but the basic solution method is an accelerated and stabilized successive substitution method (ASSM). The remaining n equations needed are provided by the thermodynamic criterion stating that the fugacities in the liquid and the vapor phase must be equal: These 2n+2 equations define the two-phase equilibrium problem. Several solution options are available including Powell's method. (2) and one of the restricting equations ( 3 ) constitute a system of n+2 equations in the 2n+2 unknowns. = Nz. (l). and it contains a bring-back procedure that brings the solution back to the two phase region if it reaches the single-phase region too soon./x. In an equation of state approach. The program is based on an equation of state approach.= zy. = 1 (3) Eqs. we have an overall material balanceland a component balance equation for eich component: L + V = N Lx. The distinction between liquid and vapor phase is then made by choosing the smallest volume for the liquid phase and the greatest volume for the vapor phase. 1 1 1 (5) .330 directed towards the development of a thermodynamic simulator capable of predicting the phase behaviour of mixtures of hydrocarbon reservoir fluids and possible injection gases.

.45724 L Y ~= ~ ’ 1 -U wom5 = 1 m + m(l -)*T ’: + m(1 .07780 R ~ pC R ~ T a(T) = 0..and B = .26992~ The cubic equation for the compressibility factor Z = Pv/RT is: The cubic equation for the compressibility factor Z = Pv/RT is: z3 . Soave-Redlich-Kwong: p = RT a Peng-Robinson: p=RT v-b b v-b v(v+b) RT .0. = (1 .42747 pC * T ~ ~ U a(T) = 0. the fugacity coefficient of component h is given by: In Vh = bk (Z-1) In Wk bk (Z-1) A =a i ’ - ln(Z-B) - .54226~ . j where tiij are binary interaction coefficients .0 . = 0 aP where A = . 1J .AB = o Z3 -(l-B)Z 2+(A-3B 2 -2B)Z-(AB-B2-B3) bP RT where A = . i j J 1J 8.v(v+b)a + b(v-b) RT PC b = 0.37464 + 1..Tro”) = 0.574~.ln(Z-B) - B ( a aik .and B = bP RT R ~ T ~ With mixing rules as given below.08664 2 pC = 0.480 + i.. the fugacity coefficient of compoponent k is given by: aP R ~ T ~ With mixing rules as given below.a) B bk ln(l+Z) The mixing rules employed for both equations are: b = 1 xi bi i a = I I xi x 8.5 0 5 ai .331 Table 1 Suormary description of the Soave-Redlich-Kwong and Peng-Robinson equations of state. 1 7 6 ~ ~ 2 = 0..) 1J 0.z2 + (A-B-B2 )z .ti.

Introducing the fugacity coefficients JI. and the'cycle . When the value of V is determined. If we consider one mole of feed. the system of flash equations (11.-values can be obtained. V. If values for the equilibrium chstanti are assumed. eliminate L in Eq.epeated.332 where y. is the fugacity ratio fiL/f. (10) we see. and y. We then calculate the non-existing phase as if the system were at the saturation pressure: . the compositions can be calculated in a straightforward manner. there will only be one root of interest. are mole fractions in equilibrium. During the iteration process when the fugacities are not yet equal.. When the root of Eq. is based on the fugacity rakios. and the fugacity equations ( 4 ) are replaced by the Ki-equations (5). 1 ' the fugacities can be written In equilibrium the fugacities are equal and hence the equilibrium constants are given by Ki = JIiL1JIiV (7) This is an important relation as it allows the defin tion of K-values also outside the two-phase region.that V is determined as the root in the equation P(V) = 0 (12) This equation is readily solved by Newton's method. (2). (12) gets outside the interval [0.1]. ( 3 ) and ( 5 ) can most conveniently be solved by introducing the g V func() tion following Nghiem and Aziz 141.1) 2 < E When the equilibrium constants are given. the resulting set of equations can easily be solved for the unknowns L.) = 1 () ~ l+(Ki-l)V From Eq. and x. The criterion for acceptance of a solutio. and then sum up all the equations we obtain 1Xi + m y i - Xi) = 1 (10) The g V () function is defined by (Ki-l)zi (11) g V = I(Yi -'x. the improved K-value estimates are obtained by where j is the iteration number and R. this indicates a single-phase state.". (2). N=l. With these compositions improved K. As the g V function () always has a negative slope. To comply with other solution methods the following error norm is used p = I(Ri . x.

The estimates should assure that the calculations start in the two-phase region in order to avoid false single-phase solutions. the fugacity coefficients are calculated. 1 1 If VtO.333 K. and the fugacity ratios will converge to a common constant value. This limiting value will be different from unity except if the system is at the saturation pressure. The often quoted empirical formula from Wilson /8/ 1 K. then V is set equal to 0. 3. It may however break down along the lower dew point curve. The evaporation must be stopped before we run out of any component in the liquid phase. From the resulting compositions.3727 (l+ui) (1 ' r i - . but if possible. the compositions are corrected only to the extent that the fugacity ratios deviate from the average value. . An alternative to the empirical formula can be based on the equation of state. A common factor in the K-values has no effect on the composition of the nonexisting phase. The basic idea is the following. xi = zi. 2. yi = Z Ki xi The normalization is necessary as this is not automatically assured outside the two-phase region. There are 3 conditions these estimates should meet: 1. = . The estimates should have sufficient spread to avoid false solutions where all K-values become equal to one. The estimates should be as close as possible in order to obtain a rapid solution.3. The evaporation rate for each component is assumed to be proportional to the fugacity of that component. These fugacity based K-values work very well in the near critical region. There. K-VALUE ESTIMATION The set of initial K-values is the starting point for the iteration procedure. The reason is probably that in addition to start with K-values consistent with the equation of state. The system will converge to a definite composition. . and from these we obtain the K-value estimates. and the fugacities are calculate We then assume a gas phase to be formed by evaporation from this liquid. The mixture or feed is assumed to be 1 quid at the temperature and pressure given. may be needed in order to assure that the assumed liquid behaves liquid-like.x. The method also works well along the bubble point curve and in most of the two phase region.11 Tri normally meets these requirements well.exp [5. we start in the middle of the two-phase region with both L and V equal to one half. the proportionality constant being the same for all components. and when the K-values are corrected according to the fugacity ratios. a restriction of the liquid 2-factor to say ZL<0. the evaporation should proceed until only half of the liquid remains.

the convergency may be very slow. it must always be tested that it leads to an improved solution in the way that it brings the fugacity ratios closer to unity. + log : R + . A simplified approach is to assume a constant value for the exponent to the fugacity ratio. also in conjunction with proper acceleration.. The equilibrium constants can be regarded as long products. an exponent of 2 may normally safely be employed. log Ki = log Ki + l o g : R (1 + k + k2 + . = KJ 1 ' 1 R . THE BRING-BACK PROCEDURE When the system is in a single-phase state. + l o g : + log : R R + .. This can bi written Taking the logarithm we obtain 1 log Ki = log KY + log R. We routinely apply this exponent after 10 iterations.. The process can now be accelerated by replacing the remaining part of the series by a geometric series where k is the ratio between the terms . ) = log KJ + 1 ~ log : R 1-k (18) The quotient k is calculated as the ratio between the last two consecutive terms (omitting the subscript i) and the resulting accelerating step is K. If we wait until the system is well on the right track. the situation is characterized by a monotone and steady approach towards the solution. Between each accelerated step there must be a single step in order to determine the quotient k in the exponent in Eq. When this acceleration step is used. it must be rejected and replaced by a single step. causing alternate signs in the series above. starting with the initial estimate Ko and then multiplied by the fugacity ratios R. If not. ( 2 0 ) . .334 THE ACCELERATION PROCEDURE When the system is close to the critical point. As a common factor in the set of 1) equilibrium constants has no influence on the composition of the non-existing . But after say 20 to 50 iterations. (l-k) 1 where l/(l-k) is exponent to the fugacity ratio. which approaches unity as the number of iterations increases. (17) In the first part of an equilibrium calculation the fugacity ratios may change from values smaller than one to values greater than one. the composition of the non-existing phase is calculated by formula (13) or ( 4 ... A method similar to Aitken's accelerating formula can then be used to speed up the convergence rate.

for all but the last component. and the multiplication factor y is determined from I(yKi - 1 ) zi = 0 (21) which gives a new set of equilibrium constants = yK. Secondly. /f. Physically this corresponds to creating a nucleus gas bubble and see if it will grow or disappear. V and yf for all but the last component. V equals zero. the K-factors are reduced bringing the system kacttv to the single-phase state. or in other words. 5 . If we consider a single-phase liquid. niL forla11 components. This corresponds to having correction factors (f. . When the correction factors (fi /f are mostly less than one. As V equals one. the equilibrium problem is stated as a system of n nonlinear equations in n independent variables by and the solution is accepted when the residual norm p of Eq. the robustness is increased by combining the Newton method with the more stable steepest descent method.335 phase. niV for all components. This algorithm has been used by several other workers during recent times 1 4 . this gives the possibility to adjust the K-values to make the gV() function zero. A gas bubble will grow spontaneously if the fugacity is lower in the gas phase than in the liquid phase. First. but differs from that method in two important ways. we may reason in a similar way. Powell's algorithm is based on the classical Newton method. (9) gets below a selected tolerance as before. Using Powell's method.-)zi=o Y Ki NEWTON-TYPE METHODS As a supplement to the acceleration procedure described above. If the single phase is gas. we have also implemented the Newton-type method described by Powell / 9 / . 6 / and is claimed to be a robust and efficient tool for the problem at hand. = 1 Ki Z Ki zi (22) on which we apply the normal correction factors (fiL/fiV). Independent variables are selected from the following options: =-iteration: LY-iteration: m-iteration: NV-iteration: L and x . ) mostly greater than one. inversion of matrices at each iteration step is avoided by using an approximate matrix-updating scheme directly on the inverse of the Jacobian. The K-values will then 1L be further increased and kge system will return to the two phase state. to keep the non-existing phase at the edge of the two-phase region while we test for its existence. the multiplication factor is in this case determined from Z(1 . If. however. the bubble will disappear spontaneously. the fugacities are higher in the gas phase than in the liquid phase. In summary.

and all results shown here are obtained for a temperature of 311 K (0') 10F. All cases are run in double precision on a VAX-11/780 computer. Following Nghiem and Aziz 141.5.336 n. and some tests of this approach has been made utilizing the generalpurpose minimization routines E04JAF and EO4FDF contained within the NAG library /lo/.VJ-ll < E E Default valugs of E ~ E. Mixtures of Isobutane and Carbon Dioxide Tests with this binary system has been made with the PR version. and niv are liquid and vapor fractions per component.01 and 10 . the strategy adopted here is to start with successive substitutions and then switch to Powell's method if the convergence is slow. i. niL = Lxi and nii = vyi. and some applications on relatively simple fluid systems are discussed in the following.e. ~ cV and . 6 1 . -CALCULATED 0 I I--\\ MEASURED 'A ! 0 ' 0 0 " m I I I 40 %a 1 '80 I Y) MOLE PERCENT CO2 Figure 1 Phase envelope for binary mixtures of isobutane and carbon dioxide at 311 K (0') 10F. Relevant component properties are summarized in Appendix.~ . . ~ E. V of Eq. 0. 1 and is in good agreement with previous calculations and experimental data 1 2 . APPLICATIONS The above algorithms have been implemented into a computer program COPEC. A switch after i iterations reauires all of the following conditions to be fulfilled: E~ < pj < E~ and and --&> pJ-l E~ 0 < VJ < 1 1 VJ . respectively. The flash equilibrium problem may also be formulated as a pure minimization problem. 0. The phase behaviour of this system for a wide range of CO contents is depicted in 2 Fig. ( are set to D .

1 and 89. we encountered no serious problems obtaining saturation points as indicated in Fig. Figure 2 Same system in the nearcritical region (calculated points are indicated bydots). 3 . and they demonstrate thus the ability to perform flash calculations very close to a critical point. 2. With the refined successive substitution method presented here. Details of the volumetric behaviour in this near-critical region are shown in Fig. More detailed results in this region are given in Figs. 100 80 60 ---T MOLE PERCENT &' / 40 20 0 m 970 980 geo loo0 1010 1020 PRESSURE (PSIA) Figure 3 Volumetric behaviour of I-C4 . solutions are obtained easily even with pure successive substitutions. 2 and 3 . A critical C02 content between 89.2% is predicted. . These points are obtained by repeated flash calculations and not by direct saturation point calculations. whereas the region indicated by a circle may pose more difficult problems. 1 .C02 at 311 K (0') 10F in the nearcritical region (calculated points are indicated by dots). I 1 I I I 1 8 5 8 4 8 5 8 6 8 7 8 8 8 9 MOLE PERCENT CO.337 In most parts of the two-phase region of Fig.

f= l1016 ss.CO mixture (89 mole percent C02) at 311 K (0') 10F for different sofution alternatives. trivial solution. . Such solutions are characterized by all K-values approaching unity and may appear when the feed composition yields one and only one root in the cubic equation for the compressibility factor.96 6. These are plots of the fugacity residual norm p as function of the independent variables corresponding to a W-iteration and depict in detail the performance of the different solution alternatives.f o 1 6 ss. 4. and several observations can be made. it can be seen that a relatively high tolerance level on the8fugacity residuals ( E in E q .98 69 .€=10-8 Powell.338 In Fig. Contour values refer to the logarithm of p (log p ) . The failure of Powell's method is investigated further in Figs. the Powell method converges to a false or. First.97 PRESSURE (MPal 6. The problem of avoiding such trivial solutions is probably the biggest problem encountered with flash calculations near to a critical point. The vanishing norm for a I-C vapor mole fraction of 0. If such a condition is suspected.977 MPa. more precisely. the sensitivity of the solution with respect to the tolerance level should always be investigated.94 6.16 80 c t P Y 0 K 70 W I P P 2 9 I . the predicted flash behaviour for a C02 content of 89 mole percent is considered in detail.95 6. E = 10. A more serious matter is the observation that for pressures above 6.0. 5 and 6. 90 t o A 0 0 ss.60 50 40 30 6.9 Figure 4 Flash behaviour of a I-C4 . E = 10 9) also gives a smooth curve. ( ) tends to widen the two-phase region. but is obviously a too high tolerance this close to a critical point. = 4.11 correspondsl?o the trivial solutions of K.

the accelerated successive substitution method proceeds in rather large steps towards a true solution whereas Powell's method rather quickly finds a trivial solution. It should be understood that the small-scale contour features in the lower left-hand parts of Fig. and Fig. 6 are artifacts originating from that map (and all the other maps shown here) being contoured from a 101 by 101 network of points.reach a norm of 10 whereas the accelerated version reaches a norm of 10 in just 31 iterations. a maximum norm of only 10 separates the true solution from the trivial ones. '' Fig.980 MPa. After 15 iterations the residual norm is approximately 10 . 5 clearly shows that is not an appropriate tolerance level in the present case. while after 300 iterations in the latter case. VAPOR MOLE F R A C T I O N Figure 5 Residual norm plot for a 11% I-C4 T = 311 K). that it takes 172 pure successive substitutions. Moreover. The fi ure alsogshows what happens when Powell's method is employed at a norm of lo' and 10 . a trivial solution is rapidly found. 5 shows the solution space at a relatively large scale and depicts the performance of the first 15 pure successive substitutions. It should be noted gqso. K-value estimates both by formula (15) and the fugacity approach are included. A large region in the-fglution space satisfies this tolerance. 6 depicts what happens if acceleration or Powell's method is started at this point. but in this case this makes very little difference when some 15 iterations hays been performed. - 89% C 0 2 mixture ( = 6.339 Fig. P . As may be seen. The fugacity approach starts off somewhat better. the solution is stuck at what appears to be something like a saddle point. In the former case.

981 MPa corresponds to a single-phase liquid. Powell's method does not give a true solution at all at 6 . and it might be an idea to employ the K-values as independent variables in Newton-type methods as well. 4. If anything. A further illustration of the acceleration process is given in Fig. It would be interesting to see the performance in such a space. AS we have seen. 7 shows similar fugacity residual norm plots for a decreasing pressure. The others show how the delineation between true two-phase solutions and trivial solutions is gradually improved as the pressure drops off from the bubble-point. P = 6. namely P = 6 . At 6 . The independent variables selected for the norm plots are probably not the natural ones for successive substitution. successive substitution must be considered as a process taking place in a space spanned by the K-values. and the iteration performance is plotted for two pressures in Fig. but at approximately twice the no. 9 4 MPa a proper solution is obtained. of iterations required with acceleration. T = 311 K . 8.340 VAPOR MOLE FRACTION Figure 6 Residual norm plot for a 11% I-C4 89% CO mixture showing the 2 behaviour of different solution alternatives (P = 6. 9 8 MPa. It is seen that the acceleration process improves the successive substitution method dramatically. ) - Fig. In addition. 9 4 and 6.98 MPa. Again the 89% C02 system is considered. a Powell iteration generally takes more computer time than a successive substitution iteration.980 MPa. .

51 P > z- 0.10 a f zV z V U 0 9 w 0’ I 0.980 MPa Figure 7 Residual norm plots for a 11% I-Cq - P . 0.61 \ VAPOR M O L E F R A C T I O N i \ I I a p f V U z V 0 a l L w W I $ d 5 0.01 0.11 P - 6 9 7 9 MPa 0.11 P = 6. 9 8 1 MPa 89% COP mixture at 311 K.6 .11 P = 6.10 P - 6.940 MPa VAPOR MOLE F R A C T I O N 0.970 MPa 0. .01 VAPOR M O L E F R A C T I O N VAPOR MOLE F R A C T I O N \ a .341 0.

__ __ __ 10-n p 6. 10-5 P K z 10-1( : : 5 10-l! I I I I . . . The solution escapes the two-phase region after 3 iterations. The figure also shows the favourable effect of the multiplication factor y defined by Eq. and the no.3% CO mixture is considered and the pressure step in question is from 6.6% m a . The solution is much more quickly returned to the two-phase region. (23). corresponding to a decrease in liquid mole fraction from 9. . .98 - mh 1 1 I l l l . . (14) defining a hypothetical liquid phase.01 to 0. i x I I . A 89. 9. l . the iteration is continued and brings the solution back into the twophase region before the tolerance level is met.342 0 I I OSS X SSwithur.94 MPa p 8. OF ITERATIONS loo 0 Figure 8 Iteration performance for a 11% I-CL - 89% CO mixture at 311 K. but with Eq. 50 NO. of iterations required to achieve convergence is significantly reduced. 2 Another aspect of the refined successive substitution method presented here is illustrated in Fig. It depicts the iteration performance for a step in a series of flash calculations where resulting K-values from one point is used as initial estimates for the next.46%.82 to 6.

This mixture has a l s o been s t u d i e d by Gundersen i l l / . A Ternary Mixture F i g s . 10.08 B B 10- 10- 10- I < 10 \ Y LO2 \ \ \ \ 10- 1 m . P The p r e d i c t e d f l a s h behaviour a t some temperatures i n t h e v i c i n i t y of t h e c r i e t i c a l temperature i s shown i n Fig. 0 NO. The SRK equation of s t a t e and t h e a c c e l e r a t e d successive s u b s t i t u t i o n method i s used throughout this example.f a c t o r s . OF ITERATIONS Figure 9 E f f e c t of bring-back procedure. W had no d i f f i c u l t i e s obtaining convergence even more c l o s e t o t h e c r i t i c a l p o i n t then i n d i c a t e d i n t h e f i g u r e .12 10- 1. he was a b l e t o perform f l a s h c a l c u l a t i o n s up t o some 0.02 M a from the c r i t i c a l p o i n t . 40% propane and 20% n-butane. 10 and 11 show some r e s u l t s obtained with a t e r n a r y mixture of 40% ethane. 7 . Using a stepping procedure towards t h e c r i t i c a l region and a s p e c i a l treatment of 2 .343 0.10 10- 1. .

344 4. and that the tolerance has to be gradually reduced to get accurate results as a critical point is approached. to set a background. The effect on the isotherms is to create a discontinuity from two-phase to single-phase. and the corresponding effect on the P-T phase diagram is to cut a top off the two-phase region as shown in Fig.1 5.6 and 366. In Fig. the calcula1) gets less than O281.2 PRESSURE (MR) Figure 10 Flash behaviour of the ternary test mixture at different temperatures. However. 8 The algorithms considered in this paper have been extensively tested also on systems consisting of a larger numbers of components. 8 1121 will be presented here. 11. the flash behaviour obtained at several temperatures is ploited in Fig. 5 ' F . 12. and the critical temperature is estimated to approximately 5. In Fig.8 6. consistency near a critical point is probably more important than to pursue solutions very close to this critical point. However. The previous example clearly shows that such irregularities are to be expected.8 4. 10. all isotherms between 365. and this criterion tions are terminated when 1 z. and some results obtained for a 6-component synthetic oil mixture commonly referred to as Yarborough mixture no. Specifically. - .(K. with a "normal" tolerance level of some of the isotherms were found to become irregular as the saturation pressure was approached. These results will concentrate on the solution performance. Yarborough Mixture No. In most practical applications. Fair agreement with experimental results is obtained at 200 F. 10 an attempt is made to define a critical point vicinity where a phase separation is not insisted on and the solution is interpreted as a single-phase "critical" mixture.0 6. is felt to be well adapted t'h ott tolerance level of 10 Nghiem and Aziz 141 indicated a similar approach.7 K experience this discontinuity (points beyond this discontinuity are not plotted).

27) (0.48) (0. 8 .16) (0.23) (0.35) (0.42) (2. The isotherm of 75'F is sufficiently close to the critical point to yield relatively hard flash equilibrium problems.3740 0.67) (1.31) (0.45) (0.3953 0.3878 Pure SSM with overshoot (0.345 I I ! I I / I / I I TEMPERATURE I K I Figure 11 Phase diagram of the ternary test mixture.16) (0. and when converging.Resulting liquid mole fractions are included in Table 2 .Not converged within 300 iterations NC NC . and results of some testing of different solution alternatives for this temperature is given in Tables 2-4. Yarborough mixture no.48) NC .45) (2.3188 0.37) (0.34) (0.41) (2. Liquid mole fraction 0.98) SSM 19 32 47 66 147 282 SSM with acceleration 13 20 31 39 28 34 41 (0.3987 0. Seven pressures are considered.71) 13 20 28 37 224 293 (0.41) (0. of iterations (and CPU-time) for different versions of SSM. 514 the alternatives has been used in the yield essentially the same results.3537 0.23) (0. A tolerance of 10 present context. Table 2 Pressure (psia) 2000 2500 2750 2875 3000 3050 3075 No.2621 0.

With the acceleration procedure. but not so well more close to the saturation point.0306 0. however.0330 0. of iterations for the higher pressure values. First a simple iteration is done in order to determine k and the exponent of Eq. (20). and thereafter an accelerated step is made in accordance with Eq. The acceleration procedure proceeds in pairs of steps. (20).0244 0 PRESSURE IPSIA) Figure 12 Volumetric behaviour of Yarborough mixture no. With overhoot the fugacity ratios are raised to the power of 2 in Eq.346 Table 2 compares different version of the successive substitution method. that the n Mole fractions c1 c2 c3 N-C5 c7 C10 0. This may be seen to function well for the lower pressure values.0457 0. A important detail is. (8) after a fixed no.8097 0. The pure version yield a prohibitively high no. .0566 0. of iterations (10 in the present case). 8 at different temperatures. and the testing step included in the acceleration procedure should therefore be emphasized. The reason why the overshoot feature in some instances fails is probably that it is too uncritically employed. a maximum of 41 iterations is used for all the pressure values considered.

A value of 10 been found t o be s u i t a b l e i n most cases and y i e l d s t y p i c a l l y some 15-30 i t e r a t i o n s before a c c e l e r a t i o n i s attempted.16) (0.45) (0.21) (0.764 -2. If it i s too high. and ill u s t r a t e s t h a t some caution should be used when s e t t i n g t h i s parameter.-koo has many i t e r a t i o n s a r e made before a c c e l e r a t i o n i s attempted.21) (0.41) (0.59) (0.58) (0.37) (0. 8 Residual norm a t s t a r t I Pressure (psis) 2000 2500 2750 2875 3000 3050 3075 SSM with acceleration 13 20 31 39 28 34 41 (0.67) (0.i t e r a t i o n s .860 ( r e j e c t e d ) 27.59) (0. 22 24 26 28 30 32 Fugacity r a t i o exponent 17.16) (0.52) (0.69) + + + + + 28 26 27 30 39 + + + + + 17 18 19 26 30 . I f it i s too low.125 I n essence w e have applied t h e same c r i t e r i o n f o r s t a r t of a c c e l e r a t i o n a s f o r switch t o Powell's method. too many a c c e l e r a t i o n s t e p s a r e r e j e c t e d . I n t h i s case j u s t t h e simple s t e p i s ' t a k e n and i s followed by a new p a i r of s t e p s .940 ( r e j e c t e d ) 47. Yarborough mixture no.70) (0.481 SSM + Powell W-iteration 13 20 23 32 17 21 23 (0.45) (0.544 ( r e j e c t e d ) 27. Here. Comparisons with Powell's method a r e made i n Table 4 . Table 3 No. The a c c e l e r a t i o n performance recorded a t 3050 p s i a i l l u s t r a t e s t h i s point: I t e r a t i o n No. see Eq. b u t both a l t e r n a t i v e s a r e somewhat slower than t h e a c c e l e r a t e d successive s u b s t i t u t i o n method.16) (0. namely &u.347 a c c e l e r a t e d s t e p i s r e j e c t e d i f t h e f u g a c i t y r e s i d u a l norm f a i l s t o be decreased by t h i s s t e p .83) SSM + Powell NL-iteration 13 20 23 32 17 21 23 (0.21) (0. Table 3 compares d i f f e r e n t a l t e r n a t i v e s f o r t h e most important parameter i n t h i s c r i t e r i o n .72) (0. NL-iterations a r e somewhat more e f f i c i e n t than W .34) (0.007 0. ( 2 5 ) . of i t e r a t i o n s a s f u n c t i o n of s t a r t of a c c e l e r a t i o n .

K. it is also faster than Newton type methods. but acceleration steps are never taken unless they lead to improved solutions. Ri T V V i ii X Zi Y E P " i W . = y. Newton-type minimization routines /lo/. However. a more direct expression for Gibbs free energy is much better than a fugacity residual norm. CONCLUSIONS An accelerated and stabilized successive substitution method (ASSM) has been formulated for flash calculations of multi-component systems and has been especially designed for applications in the near-critical region. it can easily be adapted to other equations of state.348 We also did some tests with general-purpose. NOMENCLATURE Equation of state coefficients Equation of state coefficients Liquid and vapor phase fugacities Gibbs free energy Acceleration parameter Equilibrium constants. Generally. The acceleration procedure is based on an Aitken type formula for correcting the K-values. Working with the object function only. In the examples presented. L1 n N P R. Liquid moles or liquid mile flacgion No. but one observation is worth mentioning. As special saturation pressure calculations are not needed to delineate the two-phase region. of moles Pressure Gas constant and fugacity ratios Temperature Molar volume Vapor moles or vapor mole fractions Mole fraction of component i in liquid Mole fraction of component i in vapor Mole fraction of component i in system Compressibility factor K-value multiplication factor Tolerance Fugacity residual norm Fugacity coefficients Acentric factor k K.and will therefore detect single-phase solutions automatically. The method presented is based on the Soave-Redlich-Kwong and the Peng-Robinson equation of state. the ASSM method has been shown to be a highly stable and efficient method. Both the routines E04JAF and E04FDF were found to be less efficient than the other solution alternatives considered here./x. the greatest advantage of the ASSM method is its stability close to saturation pressures. The method is made convergent also in the case of a disappearing phase. Compared with Powell's method and other Newton type methods. the method is well adapted for incorporation in compositional simulators. of components Total no.

(ed. 59-64. and YANOSIK. Eng. pp. The project is fully financed by Norsk Agip A/S. J. Eng. MEHRA. 3. Sci.B. in RABINOWITZ. G. Chem. 1969. M. REFERENCES 1. Chem.: "Equilibrium Constants from a Modified Redlich-Kwong Equation of State". Eng. SPE Paper 8285 presented at the 54th Annual Fall Meeting of SPE of AIME. pp.349 Subscripts C i. FUSSEL. K. and ROBINSON. 8.: "Development and Evaluation of a Method for Calculating the Phase Behaviour of Multi-Component Hydrocarbon Mixtures Using an Equation of State". POWELL. WILSON. Dorchester (1980). J. AEE Winfrith Report 1331. 1 (1976).J. Vol. Dallas (1980).D. Fundam. (June 1978).K. Iteration no. 18. paper no. 173-182. Ind. MADDOX.-Y. et el. Vol.. SOAVE. 2. 1197-1203. and ERBAR. No. PENG. London (1970). SPE Paper 9232 presented at the 55th Annual Fall Technical Conference and Exhibition of SPE of A W E . Numerical Algorithms Group Ltd. pp. R. 27 (1972). D. 15C presented at the AIChE 65th National Meeting. (Feb.: "A New Two-Constant Equation of State". 2.: "A FORTRAN Subroutine for Solving Systems of Non-Linear Algebraic Equations".. NAG Library Manuals. Las Vegas (1979). 6.L. j j L r V = = = = = = Critical Component no.: "Equilibrium Calculations by Equations of State".: "Computation of Multiphase Equilibrium for CoIppositional Simulation". (as superscript) Liquid phase Reduced Vapor phase ACKNOWLEDGEMENT This research is part of a joint project between Norsk Agip A/S and the Continental Shelf Institute (IKU). L.): "Numerical Methods for NonLinear Algebraic Equations". R. D. . 10. 5. NGHIEM. D.: "An Iterative Sequence for PhaseEquilibrium Calculations Incorporating the Redlich-Kwong Equation of State". 9. Sac. 7. 198l).: "A Modified Redlich-Kwong Equation of State. Cleveland. R. P.: "A Robust Iterative Method for Flash Calculations Using the Soave-Redlich-Kwong or the Peng-Robinson Equation of State". May 4-7..N. and AZIZ. J.. 15.H. Application to General Physical Data Calculations". Oxford (1978). Ohio. The authors wish to thank Norsk Agip A/S for permission to publish this paper. pp.D. Vol. 74-78. Oil and Gas Journal. 4. Pet. Gordon and Breach Science Publishers. G.X.E. MOTT.

Table 5 Comp. pp.Jr.4885 7.096 . 17 (1972). 13.070 44.65 Acentric factor 0.c3 . "The Properties of Petroleum Fluids". J.350 11.0986 0. Eng.82 408.58 305. 129-133. GUNDERSEN.COMPONENT DATA The critical properties used in the computer program COPEC are taken from McCain /13/.2539 0. I.370 2.387 4. Eng . For the tertiary test system.18 469.21 190.251 3.606 4.N-C4 .: 12.001 0.882 4. W.26 617. the PR equation with a binary interaction coefficient of 0..205 142. Gulf Publ. "Vapor-Liquid Equilibrium Data for Multicomponent Mixtures Containing Hydrocarbon and Nan-Hydrocarbon Components". YARBOROUGH.D.799 3.: "Numerical Aspects of the Implementation of Cubic Equations of State in Flash Calculation Routines". c02 c1 c2 c3 144 N-C4 N-C5 c7 c10 Hole weight 44.: Tulsa (1973).2250 0.012 For the six-component Yarborough mixture the PR equation is used with all binary interaction parameters equal to zero. T.124 72.3498 0. .124 58.010 16. McCAIN.151 100.650 3.42 369.N-C4 : : : 0.13 has been used.286 Component properties Pc (ma) Tc (K) 304. Chem. For the binary test system considered in this paper.0104 0.043 30.2010 0. to appear in Comp 6r Chem. APPENDIX .009 0.737 2. and those used in the example calculations are given in Table 5. Data.14 425. the SRK equation has been used with binary interaction coefficients as follows: C2 c2 c3 . Vol.65 540.1848 0.097 58.1524 0. Comp.

together w i t h $he a n a l y t i c a l methods used t o assess changes i n core c h a r a c t e r i s t i c s . f i e l d r e s u l t s t o date have been s u f f i c i e n t to i d e n t i f y several major problems and opportunities with t h e carbon dioxide technique. While many have reported t h i s to be depositfon of high molecular weight materials upon mixing of crude i n situ pluggiag tests have n o t proved the dioxide 2' of t h i s type of p r e c i p i t a t i o n Observed reductions i n . F i e l d experience. only limited empirical d a t a is currently available.o f reduced i n j e c t i v i t y experienced i n sane on i n j e c t i n g carbon dioxide. The paper presents r e s u l t s from i n i t i a l tests on four d i f f e r e n t carbonate containing core materials. ANDREW TWEEDIE Department of Petroleum Engineering. thus. Heriot-Watt University Both formation damage and stimulation e f f e c t s have been experienced during "miscible" carbon dioxide f i e l d and laboratory tests i n the USA. ADRIAN C.MISCIBLE GAS DISPLACEMENT 351 THE EFFECT OF SIMULATED COZ FLOODING ON THE PERMEABILITY OF RESERVOIR ROCKS GRAHAM D. Nor has any established theory f o r the formation damage been i d e n t i f i e d . and i n p a r t i c u l a r t h e carbonated w a t e r carbonate mineral reaction in sandstones during a C02 enhanced recovery process. Although encouraging. has only recently begun to accumulate. although it seems l i k e l y t h a t i n some instances formation damage may be caused by formation f i n e s . - - (1) General Successful laboratory i n v e s t i g a t i o n s of miscible. . A l l the p r o j e c t s reported havebegunsince 1972 (mostly i n t h e United S t a t e s ) .w a t e r mixtures through l i n e a r rock cores are described. One of the reservoirs due to the and carbon occurrence problems is t h a t . flooding have been w e l l documented i n the literature. however. carbon dioxide. released by d i s s o l u t i o n and subsequently migrating i n t o pore throats. This paper describes a laboratory investigation i n t o the e f f e c t s of rock f l u i d i n t e r a c t i o n under s i m u l a t a r e s e r v o i r conditions. While the stimulation e f f e c t s have been a t t r i b u t e d to d i s s o l u t i o n of t h e reservoir rock by carbon dioxide enriched flood water no work has been done t o i d e n t i f y and quantify t h i s phenomenon. ROSS. TODD and J. The design and operation of experimental equipment f o r flowing CO .

Since the cementing material i n sandstone is located between sand grains adjacent to flaw channels. combinations of calcium and magnesium (dolomite) and iron ( s i d e r i t e ) . although varying. w i l l form carbonic acid. but usually minor amounts of d e t r i t a l carbonate grains may also be present. These minerals have a l o w solubility in pure water a t atmospheric conditions. I n view of the lack of data and uncertainty in the published r e s u l t s relating to carbonate dissolution on carbon dioxide flooding. a and movement of particulate matter i n t o the throats of i n t e r s t i t i a l pores. a research programme has been i n i t i a t e d to study the phenanenon. but become increasingly soluble w i t h increasing water carbonation (or C02 concentration) and pressure. H Thus. mixing with either injection water or connate water. carbonates are cormnonly found as pore f i l l i n g and replacement cements consolidating the sand grains. together w i t h the results of experiments undertaken to establish the mechanism(s) of carbonate dissolution i n porous media. a relatively s m a l l change i n the pore framework due to carbonate dissolution may significantly a f f e c t the total permeability. Although no work has been carried out in the 0 to 100 C m f r a t u r e range a t pressures above 100 bars. . The carbonates most conrmonly found i n reservoir rocks are those of calcium ( c a l c i t e ) . the following equation representing the chemical reaction for calcium carbonate: Similar chemical reactions take place w i t h the other carbonates. Upon injection. . indications from other studies are that c a l c i t e solubility: (1) increases w i t h increasing temperature a t constant t o t a l pressure and COP concentration. increases in i n j e s t i v i t y have also been experienced i n the course of carbon dioxide f i e l d tests These were in t u r n attributed to dissolution of carbonate minerals i n carbon dioxide enriched floodwater (carbonated water).352 injectivity can probably therefore be attributed t o other mechanisms. Conversely. One characteristic of carbonic acid is t h a t a t very low carbon dioxide p a r t i a l pressure. The carbonate form is converted to that of the soluble bicarbonate. (2) Carbonate Dissolution i n Reservoir Rock Many producing formations contain carbonates i n sme form. This paperpresentsthe first phase of the study. causing increased permeability. the developpent and operation of apparatus for flowing C02-water mixtures through linear rock cores. carbon dioxide. I n sandstones. and to determine how formation permeability characteristics are likely to be altered during a carbon dioxide flood. carbonates constitute the bulk of the formation rock. The solubility trends 6f calcium carbonate in carbonated water as a function of pressure and temperature are gresented i n Figure 1. one of which m y be the disintegration of carbonate cements i n the reservoir rock. The objectives of the programme are to evaluate the dissolution effects of carbonated water on formation carbonates. carbonated water w i l l retain its acid nature w i t h very l i t t l e C02 i n solution. I n the case of limestone and dolomite reservoirs. the p is reduced considerably.

. A process flow scheme of the core flooding apparatus is presented i n Figure 2. 250[ 10 2( 3 2 CO* PRESSURE (bars) Figure 1 Solubility of c a l c i t e i n carbonated water Carbonated water. whether the reaction effects a reduction i n rm permeability i n the reservoir by releasing particles which then migrate and plug flow channels. formed upon injection of carbon dioxide i n t o a w e l l . i s capable of operation i n moderately corrosive liquid environments under controlled conditionerof temperature. This dissolution effect w i l l be more pronounced i n the vicinity of the wellbore since the carbonated water solution w i l l approach t o t a l bicarbonate saturation as the water moves away f o the well. and 2 increases up t o a maximum a t five weight per cent COP concentration before falling again a t higher CO concentrations a t constant temperature and tota9 pressure. high temperature penneameter was designed and constructed t o permit an examination of carbonated water dissolution effects. or an increase i n permeability.353 increases with t o t a l pressure a t constant temperature and CO concentratSon. However. shown i n Plate 1 . w i l l react with the carbonate minerals i n the rock and transport the dissolved products through the reservoir. The apparatus. is not apparent from tests undertaken t o date. pressure and flow rate. EXPERIMENTAL A high pressure.

5 a long) n to be f i t t e d i n the c e l l . I t c o n s i s t s of a thick-walled s t a i n l e s s steel outer cylinder with removable l i d . f i t t e d i n t e r n a l l y with a sleeve core holding assembly.354 Plate 1 A Front view of experimental apparatus d e t a i l e d description of the major e q u i p e n t components follows: (a) Core Holder: The core holder c e l l was designed f o r high pressure core flooding i n corrosive l i q u i d environments. The sleeved core is secured between t h e c e l l l i d / i n l e t end p l a t e and the o u t l e t end p l a t e by three t i e rods. Both end p l a t e s a r e scored with l i n e s r a d i a t i n g from the c e n t r e and a l s o with concentric c i r c l e s about the centre. The o u t l e t end p l a t e can be p r e c i s e l y adjusted on the t i e rods t o enable s h o r t cores (down to 1. the core e f f l u e n t and a thermocouple probe. These l i n e s allow even f l u i d and pressure d i s t r i b u t i o n across the ends of a core. The c y l i n d r i c a l s h e l l has four entry p o r t s o r taps. The end p l a t e s serve a s d i s t r i b u t o r and receptor respectively f o r the f l u i d flowing through the core. one i n the side-wall f o r t h e core sleeve confining pressure and the o t h e r s i n the l i d s one each f o r the core i n f l u e n t . The c e l l l i d is secured t o .

1 Brine Preparation and vacuum System Viscometer Core Holder P W k==i Figure 2 ul ul Experimental F l o w Apparatus .

000 p s i . I t serves as a mixing v e s s e l during carbonated water preparation and as a f l u i d separator i n which pressure and volume changes between the d r i v e f l u i d (hydraulic o i l ) and t h e core flooding f l u i d (brine o r carbonated brine) are tzansmitted through t h e f l e x i b l e rubber membrane. Consequently an "in l i n e " c a p i l l a r y tube viscometer was incorporated i n t h e flow apparatus to enable l i q u i d v i s c o s i t y measurements t o be made under test conditions. From the c a p i l l a r y tube dimensions and measurement of the pressure drop across t h e tube a t known constant flow rate. i s controlled manually by a remotely controlled a d j u s t e r f r a a t h e v a r i a b l e speed motor. The d r i v e f o r t h e mechanism is supplied by a Kopp v a r i a b l e speed motor. (c) Transfer B a r r i e r : The t r a n s f e r barrier unit is a f l u i d pressure t r a n s f e r device. by o f f e r i n g r e s i s t a n c e to the j e r k s caused by the rocking mechanism. I n t e n s i f i c a t i o n is obtained by a l a r g e area reciprocating p i s t o n pushing a s m a l l C02 compression p i s t o n with a r a t i o of 100 to 1 between t h e p i s t o n areas. comprising an open-ended rubber bladder o r membrane enclosed i n a 5 l i t r e capacity c y l i n d r i c a l steel pressure vessel. (b) Viscosity Measurement System: Required d a t a on carbonated brine v i s c o s i t y are n o t reported i n the l i t e r a t u r e . supply t h e core sleeve confining pressure. connected through a d r i v e arm and couplings to a steel c r a d l e holder bolted to the t r a n s f e r b a r r i e r . A compressed a i r driven hydraulic pump d r i v e s t h e l a r g e area piston.are (1) a 20 QD length of 0 2 mm precision bore s t a i n l e s s steel . The s p i r a l s help to maintain the i n t e g r i t y of various connections. The general arrangement of t h e v i s c m e t e r i s shown diagrauanatically i n Figure 2. and the rocking rate from 15 to 90 cycles per minute. The main elements . (f) Displacement System: The flow rate was determined i n a l l cases by employing an Eldex Precision Pump i n conjunction with t h e back pressure regulator. The f l u i d l i n e s to and from the t r a n s f i r barrier are s p i r a l l e d around the axis of rocking. t h e required v i s c o s i t i e s can be calculated from the HagenP o i s e u i l l e Equation. The d r i v e arm length is fixed to give a rocking angle of 30 degrees. The Eldex p o s i t i v e displacement punp .356 t h e base by twenty high t e n s i l e b o l t s and sealed by Water from a hydraulic pump is used to an O-ring. a rocking mechanism was attached t o t h e t r a n s f e r b a r r i e r . The core holder has been t e s t e d and c e r t i f i e d f o r use up to a maximum working pressure of 6. tube (secured by epoxy r e s i n i n s i d e a length of support tubing) and (2) a d i f f e r e n t i a l pressure transducer. (e) Transfer Barrier Rocking Mechanism: To enable e f f i c i e n t and rapid preparation of equilibrium s o l u t i o n s of carbon dioxide i n w a t e r . (d) I n t e n s i f i e d C02 Supply: Carbon dioxide pressures g r e a t e r than cylinder pressure (830 p s i ) are obtained using a gas booster.

connected immediately upstream of the back pressure regulator to monitor system back pressure. 21/V models: .5 cc per minute. to maintain the core and fluids entering the core a t the desired temperature level. - (2) 0 to 10 bar gauge. which s e t s the length of stroke. connected i n the l i n e to the vacuum pump used during i n i t i a l vessel and pipework evacuation. The drive o i l was drawn fran a perspex reservoir by the Eldex pump and delivered a t constant volume to the base of the transfer barrier.357 delivers a steady flow t h a t can be varied from 0 to 4. (3) Heat to the transfer barrier and core holder is supplied electricakly by close f i t t i n g mesh elements and controlled i n each case w i t h i n t 1 C ) by a thermostat/thermomuple controller.. (1) 0 t o 5000 p s i absolute pressure transducer. A non-corrosive fluid (hydraulic o i l ) was used as a drive fluid t o displace the core flood liquid from the membrane i n the transfer barrier. (3) 0 t o 600 bar gauge. The pump is capable of delivery pressures i n excess of 5000 psi. the flow apparatus is equipped with four pressure gauges and two pressure transducers. (9) Pressure Measurement System: As shown i n Figure 2. I t is linked to a s t r i p chart recorder to provide a continuous record of the pressure d i f f e r e n t i a l data. connected across the core holder and viscosity measurement capillary tube. connected i n the compressed to a i r supply l i n e to the gas booster monitor the a i r pressure to the gas booster - and hence the level of gas intensification. The gauges are a s follows: (1) 0 t o -1. (4) 0 to 400 bar precision gauge with a stainless s t e e l measuring element. - (h) Temperature Control System: The temperature control System consists of three independent sub-systems: (1) (2) to heat khe contents of the transfer barrier. connected to the core to monitor sleeve confining pressure l i n e - core sleeve pressure. - Both pressure transducers a r e S E Labs. (2) 0 50 p s i d i f f e r e n t i a l pressure transducer.0 bar vacuum gauge. The flow r a t e within this range is adjusted by a micrometer screw on the pump. connected t o the transfer barrier to measure the system "upstream" pressure. It is e l e c t r i c a l l y connected to an Analogic d i g i t a l u n i t for visual observation. Xnmlatfon for the vessels is provided by 4 cm thick layers of rock wool encased in aluminised glass . and to maintain the viscometer temperature.

The separator is a sealed perspex cylinder w i t h a capacity of 300 ccs. (3) Porous Media To enable study of the carbonated water-carbonate mineral reaction i n sand- stone without interference from other effects such a s clay or mica alteration.358 cloth jackets. was also partly based on i t s high carbonate content.e. the f l u i d l i n e immediately upstream of the core holder is coiled tightly around the core c e l l lid. reduced to atmosuheric uressure on discharse from the back pressure regulator. to gain a more complete understanding of the reaction e f f e c t of carbonated water on carbonate mineral i n reservoir rock. A back pressure above the carbonation pressure is maintained throughout the t e s t t o ensure t h a t only liquid phase exists a t a l l p i n t s i n the flow system. two other sandstones and a limestone were tested i n this i n i t i a l study. porosity. a relatively pure quartz-carbonate sandstone. (2) Experimental Procedure I n i t i a l testing consists of flowing base water (i. and the core is divided i n t o a series of segments. To ensure that a l l fluid entering the core is a t t e s t temperature. These materials are l i s t e d and described i n Table 1. It has an i n l e t for the core effluent near the top and o u t l e t s i n the l i d and base for the separated gas and liquid respectively. Liquid from the base of the separator flows via a five-way selector valve t o sealed glass collection vessels for measurement and analysis. enters the gas/liquid separator. The meter is a precision device provided with a 150 mm d i a l of 100 divisions and a s i x d i g i t revolution counter form of t o t a l i s e r . is injected into the core a t constant r a t e by hydraulic o i l displacement. pore size distribution and overall dissolution e f f e c t i n each of the core segments is then assessed. fran a quarry i n the Yorkshire Jurassic. The permeability of the core is measured as a function of time. Subsequently brine. Following a core flood experiment a s e r i e s of analyses are performed on the core and effluent liquid. both before and a f t e r core flooding were: . was chosen for the i n i t i a l stage of the study. The permeability. brine or d i s t i l l e d water) through the core t o establish the i n i t i a l or reference (stabilised) permeability. The analytical procedures used i n the description of these materials. The volume of carbon dioxide produced is measured by a wet-type volumetric meter connected directly to the gas o u t l e t fran the separator. A s e r i e s of chromel-alumel thermocouples are used to monitor temperature throughout the flow system. However. it was necessary to choose material w i t h a relatively simple mineralogical composition. These are linked via a selector unit to a d i g i t a l thermometer for visual display and recording. carbonated to the desired level inside the rubber membrane of the mixing vessel. The selection of this particular sandstone. a calcareous g r i t . (i) Effluent Collection and Measurement System: Core effluent. The capillary tube viscometer is enclosed i n a water bath where it i s maintained a t the desired temperature by hot water circulation. The effluent liquid is analysed for content of calcium and magnesium by EDTA t i t r a t i o n . Thus. and a l l core effluent is collected for chemical analysis. The carbonated water and core temperatures are carefully controlled to represent o i l reservoir conditions.

The dolomite is evenly distributed. occurring as rho& shaped crystals and crystalline masses in the voids between sand grains 90% Permeability 20OmD Quartz Dolomite 10% Porosity 10% Felspar and Clay less than 1% Rotliegende Sandstone Calcitic Sandstone Composed of subrounded to ro*mded quartz grains with patchy calcite pore fill and clay Quartz 95% Calcite 2.TABLE I Summarised descriptions of core materials Formation Rock Type Description Mineral Content Physical Properties Yorkshire Jurassic Calcarenaceous Composed of subrounded detrital quartz grains Sandstone and detrital carbonate debris cemented by micritic calcite Quartz 80% Ferroan Calcite 20% Permeability 10OmD Porosity 16% Fife Carboniferous Dololdtic Sandstone Composed of angular to subrounded quartz grains partially cemented by secondary dolanite.5% Felspar and Clay 2% Permeability 3 0 W Porosity 15% Oxfordshire Jurassic Oolitic Limestone Composed of ooliths and shell fragments cemented by micritic calcite Calcite Quartz 98% 2% Permeability 6OmD Porosity 15% W u l W .

3 60
(1)

thin section petrographic analyses, scanning electron microscope analyses, porosity, pore size distribution and permeability measurements.

(2) differential dye staining for carbonate identification
(3)
(4)

For each series of experiments a number of 2 5 cm diameter X 7.5 cm length . cylindrical cores were drilled and trimmed from the same block of rock, so that variation in the properties would be kept to a minimum. As a precaution against collapse on dissolution, the cores were coated on the cylindrical surface with epoxy resin.

RESULTS
(1) The initial series of experiments were carried out on the Yorkshire Jurassic calcitic sandstone. First tests with distilled water and brine (no carbonation) were aimed at establishing a stabilised or reference permeability, prior to any carbonated water flood. The results of two such tests, R7 and R9, are presented in Figure 3. Significant increases in permeability were obtained in the tests, with little apparent levelling off in the rate of permeability increase and attainment of a reference value, upon injection of up to 500 pore volumes. Chemical analysis of the core effluents for calcium showed the permeability increases to be attributable t the dissolution of calcite cement in the flood liquids. o
As

a comparison with the base liquid experiments, a series of tests with carbonated brine were then undertaken. As shown in Figure 4 , much greater permeability increases were obtained, although again there was little indication of any fall in the rate of permeability increase. The results of the core effluent analysis as compared with those from a brine flood (R9) are presented in Figure 5 . As expected, the calcium concentrations in the effluent samples from the carbonated brine tests were far higher than in the brine test, although there was a significant difference between the results for carbonated brine tests R20 and R21. Examination of the flooded cores showed this was because in R20 a thin band was preferentially dissolved, whereas in R21 a more uniform dissolution took place (Plate 2). Presumably in R20, as the flood progressed, the main flow was through the thin permeability "streak", resulting in lower total dissolution than in R21.

To gain an understanding of the variation in local permeability, the cores from the various tests were retrieved after flooding and cut into three 25 cm long segments. The permeability of each segment was then measurod . and plotted as a function of axial position in the core. The plots for R9 and R2O are presented in Figure 6 The profiles obtained show the . permeability at the inlet end of the cores was increased considerably more than that at the outlet end. Also, the fact that the profiles have approximately the same shape, infers that the location of each profile is simply determined by the level of carbonation of the brine. Since constant flow rate was used in the experiments, this result implies that a zone of increasing permeability, which can be considered as a front, was moving through the cores. The velocity of the permeability front migration is in turn a function of the liquid flow rate through the core and carbonation level of the brine.

361

2

M0

R7 Calcitic Sandstone
3

/

R9 Calcitic Sandstone 3% Brine, 2Ooc, 1000 psi

1

100

200

300

400

500

PORE VOLUMES INJECTED Figure 3 Permeability changes during runs 7 and 9

22

19
16

13
10
7
4

1

i 00

200

300

PORE VOLUMES OF CARBONATED WATER INJECTED Figure 4 Permeability changes during runs 20 and 21,

362

0.12

-

R20

-

thin band preferentially dissolved

0.08

R21

- uniform - brine
w

dissolution

0.04

-

W

R9

flood
A

W

w

Q
1800

600

1200

CORE FLOOD VOLUME (cc)

Figure 5

Comparison of effluent calcium concentration profiles for runs 9, 20 and 21

. ) .

.

Plate 2

Comparison of Yorkshire Jurassic sandstone cores before (left) and after (right, run 2 1 ) a carbonated water flood

363

600 ,

400

.

200

- - - INITIAL PERMEABILITY - - - - - I
I
2.5
AXIAL

I
5.0

1
7.5

DISTANCE ALONG CORE (all)

Figure 6

Permeability profile for runs 9 and 20

The porosity and pore size distribution of each of the 2.5 a long segments n from all the above tests were measured and compared to initial whole core values. Generally it was found that although large increases in the permeability had occurred, the porosity had changed little. This result illustrates that the main mechanism for the increase in permeability is probably not the uniform dissolution of carbonate cement, but rather the removal of constrictions in the larger pores. This is confirmed by the mercury porosimeter pore size distribution results, which show that it was primarily the diameters of the larger pores which were increased during the tests.
(2) To further, and more realistically, test the permeability front migration phenomenon, a series of tests were initiated on material with a much lower carbonate concentration than the Yorkshire Jurassic sandstone. Difficulty in acquiring calcite cemented sandstone led to a dolomitic material from the Fife Carboniferous being used at this stage. It was hoped that it would be possible to dissolve out all the dolomite cement from this sandstone and thus eventually achieve constant permeability. However, the low reaction rate of dolomite in carbonated water, compared to that of calcite, effectively ruled out this possibility.

The permeability profiles of two tests, R22 and R23, on the dolomitic sandstone are presented in Figure 7. The very slow reaction rate of dolomite

under ambient temperature conditions meant virtually no dissolution effects were observed in R22, while in R23, although a fairly significant permeability increase was obtained, chemical analysis of the core effluent showed that only a small proportion of the dolomite cement was leached out.
(3) Some tests were then carried out on a calcitic Rotliegende Sandstone from a Southern North Sea gas field, but a series of core collapses, caused by weakening on wetting, meant abandonning the use of this material and continuing the search for other sources of calcitic sandstone.

364

600

-.

R23 Dolomitic Sandstone

400

200

"

-

R 2 2 Dolomitic Sandstone 1000 p s i Carbonation, ZOOC
v

-

0

200 400 PORE V L M S O CARBONATED WATER INJECTED GU E F Figure 7 Permeability changes during runs 22 and 23

600

R26 O o l i t i c Limestone
1500 p s i Carbonation, 80°C

10

20

30

40

50

Pore Volumes of Carbonated Water Injected Figure 8 Permeability change during R26

( 4 ) A test w a s run on an o o l i t i c limestone from t h e Oxfordshire Jurassic, the permeability p r o f i l e of which is presented in Figure 8. A very rapid increase i n permeability was obtained, with the d i f f e r e n t i a l pressure across t h e core f a l l i n g t o almost zero a t maximum flow r a t e , a f t e r i n j e c t i o n of only 50 pore volumes. Examination of t h e flooded core shaved t h i s was because two 1.5 mm diameter "wormholes" of roughly c i r c u l a r cross section had formed over t h e length of t h e core.

365
Pore s i z e d i s t r i b u t i o n a n a l y s i s of the limestone i n d i c a t e d an extremely wide pore diameter d i s t r i b u t i o n and, as expected, it w a s the s e l e c t i v e enlargement of t h e l a r g e pores a t t h e upper extreme of t h e d i s t r i b u t i o n that c o n t r i b u t e d s i g n i f i c a n t l y to t h e i n c r e a s e i n p e r m e a b i l i t y i n this test.

CONCLUSIONS
(1) The high p r e s s u r e , high temperature carbonated water permeameter c o n s t r u c t e d to i n v e s t i g a t e carbonate d i s s o l u t i o n e f f e c t s on carbon dioxide flooding i s providing new i n s i g h t i n t o the v a r i a b l e s that c o n t r o l t h e d i s s o l u t i o n process.

( 2 ) Only i n c r e a s e s i n c o r e p e r m e a b i l i t y from d i s s o l u t i o n of carbonate minerals were experienced. N o evidence f o r f i n e s migration o r p a r t i c l e plugging w a s obtained i n the experiments.

(3) The d i s s o l u t i o n of carbonate mineral from c o r e s produces a change i n local p e r m e a b i l i t y which t r a v e l s as a f r o n t through the c y l i n d r i c a l core.
(4) The dramatic i n c r e a s e i n p e r m e a b i l i t y of a core d u r i n g a carbonated w a t e r f l o o d i s probably due t o removal of c o n s t r i c t i o n s and s e l e c t i v e d i s s o l u t i o n of t h e l a r g e r pores.

REFERENCES

1.

NEWTON, L. E. and McCLAY, R. A.; "Corrosion and Operation Problems, C02 P r o j e c t , SACROC Unit", Paper SPE 6391, presented a t t h e SPE-AIME Permian Basin O i l and G a s Recovery Conference, Midland, TX, March 10-11, 1977
"North Cross U n i t C02 Flood WNTIOUS, S. B. and T A , M. J.; H M Review of Flood Performance and Numerical Simulation Model", Paper SPE 6390, presented a t the SPE-AIME Permian Basin O i l and G a s Recovery Conference, Midland, TX, March 10-11, 1977 HANSEN, P. W ; . "A CO T e r t i a r y Recovery P i l o t , L i t t l e Creek F i e l d , Mississippi'!, Paper S8E presented a t the SPE-AIME 52nd Annual F a l l Technical Conference and Exhibition, Denver, O c t . 9-12, 1977

2.

-

3 .

6747,

4.

DOSCHER, T. M. and KUUSKRAA, V. A.; "Carbon Dioxide f o r Enhanced Recovery o f Crude O i l " , paper presented a t t h e European Symposium on Enhanced O i l Recovery, Edinburgh, J u l y 5-7, 1978
"SACROC Carbon Dioxide I n j e c t i o n A Progress R e p o r t " , paper presented a t t h e API Production Department Annual Meeting, Los Angeles, A p r i l , 1976 CAMERON, J. T.;

5 .

-

6 .

MILLER, J. P.; "A P o r t i o n of t h e System CaC03-C0 0, with Geological Implications", Am. Jour. Sci. , (March 7953) 250,

161-203 7 .
"The S o l u b i l i t y o f Calcite i n Carbon Dioxide SOlUtiOnS", ELLIS, A. J.; Am. Jour. Sci., (May 1959) 354-365

257,

366
8 .

SEGNIT, E R, HOLLAND, H. D and BISCARDI, C J.; . . . . "The Solubility of Calcite in Aqueiaus Solutions", Geochim. et Cosmochim. Act., (1962) 26, 1301-1331 "The System CaO-C02-H20 in the Two Phase Region Calcite SHARP, W E; . . and Aqueous Solution", PhD Thesis, Univ. of California, 1964 "The System Ca0-C02-H20 in the Two S W , W. E and KENNEDY, G C.; . . Phase Region Calcite and Aqueous Solution", Jour. of Geol. (1965) 73, 391-403

9 .
10.

-

NUMERICAL METHODS

367

COMPUTER MODELLING OF EOR PROCESSES
KHALID AZIZ

Computer Modelling Group, 3512-33Street, N. W., Calgary,Alberta T2L 2A6, Canada

ABSTRACT This paper presents a rather personal view of recent developments, current problems and future prospects for the computer simulation of enhanced oil recovery schemes. While substantial progress has been made over the past twenty years o r so, some problems of significant practical importance remain unresolved. INTRODUCTION This paper is neither a comprehensive review of past work on reservoir simulation also referred to as reservoir modelling nor a complete catalogue of current activities in this field. Instead it presents the author's view of (a) the status of simulation technology, and (b) current and future problems. The paper is intended primarily for individuals interested in using models rather than those who are engaged in the development of models.

-

-

The contents of the paper are heavily influenced by work conducted by the author's students at the University of Calgary and his colleagues at the Computer Modelling Group (CMG). Important work underway at other institutions may not be mentioned here primarily because of the lack of up-to-date information available to the author. CMG is, however, a vehicle for cooperative research in reservoir simulation among universities, research organizations, government agencies and industry. Currently 34 such organizations are members of CMG and these organizations have a rather direct and significant influence on its work. Hopefully, because of this type of interaction, problems being investigated by CMG reflect current industry.needs. Modelling is an iterative process consisting of the following major stepsl:
1.
2.

3.
5.

4 .
6. 7. 8.

Describe Reservoir Describe Recovery Mechanism Write Mathematical Model Develop Numerical Model Develop Computer Model (Program) Validate Model Match History Predict Future Performance

Often during steps 6, 7 and 8 it becomes necessary to go back to steps 1 , 2, 3 or 4 and alter some of the assumptions made earlier. Assumptions are necessary at various stages to (a) allow simulation of processes where recovery mechanisms are

368
n o t f u l l y understood, (b) make t h e problem tractable, and ( c ) reduce c o s t o f simulation. Obviously t h e need f o r t h e assumptions is c o n s t a n t l y changing with improved understanding o f t h e p h y s i c a l and chemical a s p e c t s o f t h e recovery processes, development of new numerical techniques, and hardware innovations. S t e p s 6 and 7 d e a l i n g with t h e v a l i d a t i o n and use o f models w i l l n o t be considered i n t h i s paper. CLASSIFICATION OF MODELS
A large v a r i e t y o f models are i n c u r r e n t u s e and t h e number is c o n s t a n t l y i n creasing. New models are developed t o ( a ) s i m u l a t e new processes, ( b ) s i m u l a t e behaviour of r e s e r v o i r s with s p e c i a l c h a r a c t e r i s t i c s , (c) reduce c o s t , (d) i m -

prove accuracy, (el have access t o a s u i t a b l e model under a c c e p t a b l e c o n d i t i o n s , or ( f ) understand r e s e r v o i r simulation. Table 1 p r o v i d e s a c l a s s i f i c a t i o n based on Recovery Mechanisms, Reservoir/Well C h a r a c t e r i s t i c s , Numerical Approximations, FluidIRock P r o p e r t i e s , S o l u t i o n Techniques and Computer Type.
Table 1

C l a s s i f i c a t i o n o f Models
1.

RECOVERY MECHANISMS
1.1

WATERFLOOD OR PRIMARY DEPLETION 1.1.1 Three component, t h r e e phase 1.1.2 Two component, two phase
GAS OR SOLVENT INJECTION 1.2.1 Multicomponent, s i n g l e phase 1.2.2 Multicomponent, multiphase

1.2

1.3

CHDfICAL FLOOD 1.3.1 Four component, two or t h r e e phase (polymer) 1.3.2 Multicomponent, multiphase ( s u r f a c t a n t , c a u s t i c )

1.4

THERMAL MODELS 1.4.1 Three component steam 1.4.2 Compositional steam 1.4.3 Steam w i t h a d d i t k v e s 1.4.4 I n s i t u combustion

2.

RESERVOIR/WELL CHARACTERISTICS
2.1 2.2

RESERVOIR W L COUPLING EL FRACTURES 2.2.1 Static fracture 2.2.2 Dynamic f r a c t u r e 2.2.3 Uniformly d i s t r i b u t e d f r a c t u r e s
CONSOLIDATION OF RESERVOIR ROCK 2.3.1 Sand flow 2.3.2 Ground subsidance

2.3

3.

NUMERICAL APPROXIMATIONS
3.1

PRIMARY VARIABLES 3.1.1 Selection of variables 3.1.2 Selection of equations 3.1.3 Alignment o f v a r i a b l e s and e q u a t i o n s

369
Table 1 (Cont'd)

3.2

LINEARIZATION

3.2.1 3.2.2 3.3 3.3.1 3.3.2 3.3.3 3.3.4 3.3.5 3.4 3.4.1 3.4.2 3.4.3 3.4.4 3.4.5 3.4.6 3.5 3.5.1 3.5.2 3.5.3 3.5.4 3.5.5 3.5.6 3.5.7 3.5.8 4.

Newton's method Other methods Fully i m p l i c i t Sequential I m p l i c i t Pressure E x p l i c i t Saturation (IMF'ES) Dynamic I m p l i c i t Band reducing techniques Single point upstream Two point upstream Harmonic average Centralized upstream Other interblock mobility c a l c u l a t i o n methods Nine-point schemes Standard f i n i t e - d i f f e r e n c e s Higher order f i n i t e - d i f f e r e n c e s Variational Semi-analytical Location of g r i d point i n a block Curvilinear g r i d Local g r i d refinement Moving g r i d

DECOUPLING

INTERBLOCK F O LW

TRUNCATION ERROR

FLUID/ROCK PROPERTIES 4.1
RELATIVE PERMEABILITY CALCULATION 4.1.1 Three phase model

4.1.2 4.1.3 4.1.4 4.2

Temperature effect model Composition e f f e c t model Hysteresis model

FLUID PROPERTIES 4.2.1 Empirical c o r r e l a t i o n s 4.2.2. Equation of s t a t e

5.

SOLUTION TECHNIQUES 5.1 5.2 5.3
ORDERING O EQUATIONS F

GAUSSIAN ELIMINATION
ITERATIVE METHODS

6.

COMPUTER TYPE 6.1 6.2 6.3 STANDARD
VECTOR PROCESSORS

INTERACTIVE DATA INPUT AND ANALYSIS O RESULTS F

3 70

This classification provides a suitable framework for comment on the status of some aspects of the technology. RECOVERY MECHANISMS The simplest models that can be used for primarily depletion and water or hydrocarbon gas injection studies are referred to as black-oil or models. Themodels of this type have been in use for over twenty years and are based on the assumption that the reservoir fluids can be assumed to consist of only three pseudocomponents oil, water and gas at standard conditions. This rather gross assumption works well for systems that remain far from the critical or the retrograde region during the recovery process and where the injected fluids consist of the same components as in the in situ fluids. Even in this relatively simple case different models can yield different results for the same problem2. Most of these differences may be attributed to the numerical aspects to be discussed later.

-

Compositional models allow for the representation of oil and gas by a mixture Of several m o m pseudo-components each. They can handle complex phase behaviour associated with, for example, the injection of C02. Chemical flood models are even more complicated compositional models with capabilities to handle important rock/fluid and fluid/fluid reactions. Each component or pseudo-component yields one conservation or mass balance equation to be solved for each grid point. Hence as the number of components increases, the number of equations to be solved increases in direct proportion. Thermal models can vary in complexity from the simple three component steam model to the complex in situ combustion model. In addition to the conservation of mass we must also add the conservation of energy to our system of equations to be solved. The problems in defining the recovery mechanism from the point-of-view of the modeller usually relate to the lack of experimental information for the selection of pseudo-components, to predict their physical and chemical properties, and to validate the assumptions of the mathematical model. Contrary to the belief held by some, reservoir simulation does not reduce or eliminate the need for experiments it allows us to get the most out of laboratory and field experiments we can afford to run.

-

Examples of the phenomena that can not be handled in a satisfactory fashion at this time are (a) formation and flow of emulsions, and (b) flow of more than two liquid phases. RESERVOIR/WELL CHARACTERISTICS The intimate interaction between the reservoir and the flow in the wellbore (tubing or annulus) of both injection and production wells must be recognized for realistic simulation. While it is relatively easy to do single phase well flow calculations to any desired accuracy, the same is not true when two or three phases exist in the wellbore. The transient nature of the flow causes further complications. Modellers often underestimate the importance of the wellbore/ reservoir coupling and overestimate the reliability of correlations for performing wellbore calculations. Errors of the order of 20% are possible even when "best" available methods are utilized. Since there are no clear schemes for the determination of what wellbore flow calculation method may be the best in each situation, even higher than 20% errors are possible. Simulation of the initiation, extension and closing of a fracture requires the coupling of rock and fluid mechanics. This important field has only recently started receiving attention. Much work is required before this technology can be used to improve the design of massive-hydraulic fractures now being conducted in tight formations4,5. These models are also required to predict fracture orienta-

371
tion and size in unconsolidated oil sands, where fracturing is used to provide initial communication between the injectors and the producers. Except for single well studies in cylindrical coordinates, economic constraints demand that blocks containing wells be orders of magnitude larger than the Size of the well. The problem then is to relate the calculated conditions at the grid point in a block to the well that may be located anywhere in that block. Analytical solutions based on single phase flow theory are used to relate the well pressure to the block pressure6. Detailed simulation of the zone near the well through the use of small blocks and cylindrical coordinates is necessary when saturation and/or thermal effects become important. The information generated from such a local study of the well vicinity is used in the form of pseudofunctions for the simulation of the reservoir. A better solution of this problem would be to have the capability to do local grid refinement without placing small blocks where they are not needed. The multi-grid approach may offer a solution to this problem7. Recently it has been possible to generalize the well treatment to handle vertical fractures that go through a number of grid pointsa. For single phase flow, where it is possible to compare numerical and analytical solutions, the agreement is excellent. For multiphase flow, in addition to the problems encountered for wells, we also have the unresolved problem of multiphase flow in the fracture. Reservoir rocks that are naturally fractured behave in a significantly different fashion from conventional reservoirs. They may be simulated through the concept of double-porous-media with separate equations for each system and appropriate transfer terms for interaction between the systemsg. Some of the problems with the practical use of this concept are (a) determination of the value of the transfer terms, and (b) experimental verification, particularly for multiphase flow. Some reservoirs are either unconsolidated or only partially consolidated. The flow of sand in such systems alters rock properties. In shallow reservoirs removal of fluids (and/or solids) may also cause ground subsidance. Little is known about these two mechanisms and their simulationlo. NUMERICAL APPROXIMATIONS The mathematical model of flow in a conventional reservoir consists of one partial differential equation for each pseudo-component. Furthermore, for thermal processes an additional partial differential equation for temperature is obtained from the conservation of energy. In addition several constraints and algebraic relations must also be satisfied. The equations of the mathematical model may be manipulated to obtain a set that is more amenable to numerical treatment. At this stage a set of primary variables, equal in number to the partial differential equations to be solved, is selected. Sometimes the selection is postponed until after the application of some technique to translate the partial differential equations to difference equations. For example in black oil simulation, the oil phase pressure, and gas and water saturations form a suitable set of primary variables. The selection of primary variables and the alignment of these variables with appropriate equations can have a substantial effect on the eventual performance of the model. Numerical difficulties can result due to the appearance and disappearance of a phase during simulation. This happens, for example, in thermal and in variable bubble point black oil problem simulation. This problem may be circumvented by variable substitution or b using a technique that does not allow the phase to disappear completely6,ll 9 1z~13. The same result is obtained by both techniques, however program complexity and computer time can differ substantially. With the use of variable substitution it is possible to solve for one less equation and

Hence f o r a t y p i c a l problem. coupled algebraic e q u a t i o n s of t h e form: F(X) = 0 (1) number Of where X is t h e v e c t o r o f unknowns ( = number o f primary v a r i a b l e s blocks) f o r a time s t e p .d i f f e r e n c e approximation f o r second o r d e r space d e r i v a t i v e s a s s o c i a t e d w i t h t r a n s p o r t terms and two-point backward d i f f e r e n c e approximation f o r first o r d e r d e r i v a t i v e s a s s o c i a t e d w i t h t h e accumulation terms. Most r e s e r v o i r s i m u l a t o r s u t i l i z e three-point f i n i t e . s p e c i a l care is necessary f o r a smooth t r a n s i t i o n from one set o f v a r i a b l e s t o a n o t h e r s e t l 2 . depending on the problem. The Jacobian matrix is s p a r s e w i t h t h e form shown i n F i g u r e 1. However. t h e f r a c t i o n o f t o t a l computer time t h a t is spent on ( 2 ) also i n c r e a s e s . Each non-zero e n t r y i n t h e m a t r i x is a NEQxNEQ block element where NE4 is the number of primary v a r i a b l e s . A s t h e number o f blocks i n c r e a s e s .372 t h u s save computer time. Such a set o f non-linear e q u a t i o n s can o n l y be solved by some i t e r a t i v e technique. equation ( 2 ) must be solved many hundreds of times. Recent research t o reduce t h i s e f f o r t w i l l be discussed i n the following s e c t i o n .fi) = -F(Xv) (2) where (v) is the l e v e l of i t e r a t i o n and A is t h e Jacobian w i t h elements a f i / a x j . 7 xx xxx xxx xx X x x x x x X x X X X X X X xx x xxx x xxx x xx X xx X x xxx x xxx x xx X X X X X X X X X X X X X X X X - X xx x xxx x xxx x xx X xx x X x xxx x x xxx x x xx X X xx x xxx x xxx x xx X - Figure 1 S t r u c t u r e of Matrix A f o r a 4 x 3 ~ 2 Grid (Each X r e p r e s e n t s a NEQxNEQ block matrix) Each time s t e p u s u a l l y r e q u i r e s 2 t o 5 Newton i t e r a t i o n s f o r t h e s o l u t i o n of ( 1 ) . The f u l l y i m p l i c i t method has u n l i m i t e d . Application o f Newton’s method y i e l d s : A(fi+1 . The end r e s u l t of such a n e x e r c i s e is a set o f non-linear. These elements can be evaluated either numerically or a n a l y t i c a l l y .

KO et al. The rules for the computation of relative permeability are not well defined.= k kr8)values at the block boundaries. f fWIBB = 4 fWuu . in some cases.fWu + 4 fwd and harmonic total mobility (HTM): . increase the number of iterations necessary for convergence over the time step or fail to converge. This results in a limitation on stabilityl. The behaviour of these schemes is different for compressible systems and when the saturation change is not monotonic. They found that the centralized upstream for . Many different methods have been investigated with a view to reducing the grid orientation and truncation errorl5. worked best. but Newton's method may not converge. Although the nine-point formulation works. or converge to an unreal solution if the initial guess (previous time step) is too far from the solution. The reliability of such methods is questionable for difficult problems. For some situations ourvilinear grid can be used t o reduce both grid orientation and truncation error. This can be accomplished through a nine-point (as opposed to five-point ) scheme for two-dimensional problemsl3. . The value of absolute permeability at the boundary is computed as the harmonic average of the two adjacent blocks. Other variations of Newton's method like the semi-implicit or linearized implicit also work well for some problems.373 stability. l6 and a twenty-seven (as opposed to seven-point) scheme for threedimensional problems. A problem of convergence to unreal solutions. . However their tests were for incompressible water flood problems. its use at this time is prohibitively expensive.15 expressed transmissibilities for the two phase pressure and saturation equations as and AT (CUF): (5) (3) (4) fw AT respectively. The size of the matrix equation to be solved can be reduced by suitable approximations that partially or fully decouple the equations (SEQUENTIAL METHOD) and reduce the number of implicit equations to one (IMPES). the difference equations. M exceeded 2 . Grid orientation is a major unresolved problem that raises some serious questions about the credibility of simulation for highly unfavourable mobility ratios. However this approach is also not suitable for general applications. Non-linearities associated with the production/ injection terms can have a significant influence on the stability and time truncation error of the modell4. Even this method failed when the shock mobility ratio. Another approach to reduce the grid orientation effect is to allow flow in directions that are both parallel and diagonal to the grid. or the matrix. The decoupling can take place at the Jacobian level of the partial differential equations. The flow into and out of a block depends upon the permeability (kp. has been eliminated through the addition of a "penalty source" term to the inert gas equationl2. In the I W E S method the pressure is solved for implicitly while the saturations are treated explicitly. Abou-Kasseml3 has observed significant reduction in grid orientation with the nine-point scheme for a steam displacement problem where five-point shows substantial effect of the orientation of the grid. which arises in the simulation of steam displacement with a non-condensable gas. The most comon approach is to use the relative permeability of the upstream block. Approximations of this type do.

and ( b ) combustion f r o n t . t h e e f f e c t of space t r u n c a t i o n can mask t h e t r u e phenomena i n processes where block s i z e is t o o l a r g e t o d e f i n e e v e n t s i n t h e r e s e r v o i r . I n v e s t i g a t i o n s are a l s o underway on methods of s e l e c t i n g an optimum number of components t h a t can be used t o r e p r e s e n t t h e r e s e r v o i r and i n j e c t e d f l u i d s l g . R e l a t i v e l y simple equations of state when properly tuned and used o f f e r a powerful means of computing f l u i d p r o p e r t i e s i n an a c c u r a t e and c o n s i s t e n t fashionl7. and on the e f f e c t of temperature and i n t e r f a c i a l t e n s i o n on r e l a t i v e permeability. t h e modeller must ensure t h a t the scheme w i l l converge i n d i f f i c u l t s i t u a t i o n s with r e l a t i v e l y few i t e r a t i o n s l 7 . An example of t h i s is t h e k i n e t i c s of low tem- p e r a t u r e oxidation f o r i n s i t u combustion processes. FLUID/ROCK PROPERTIES R e a l i s t i c p r e d i c t i o n of f l u i d and rock p r o p e r t i e s f o r t h e changing c o n d i t i o n s during simulation is of c r u c i a l importance i n r e s e r v o i r simulation. and r i s t h e r e s i d u a l vector.374 Space and time t r u n c a t i o n e r r o r s can be maintained a t t o l e r a b l e l e v e l s f o r convent i o n a l simulation. The work required f o r t h e d i r e c t s o l u t i o n of a system l i k e t h i s is given by . Models are o f t e n used t o p r e d i c t t h r e e phase r e l a t i v e permeability from two phase data. and t h e effects o f temperature. 1 8 . i n t e r f a c i a l tension and h y s t e r e s i s phenomenon. ( b ) using higher o r d e r methods. Examples o f t h i s s i t u a t i o n are ( a ) m i s c i b l e o r chemical s l u g s . including CMG appear t o be making s i g n i f i c a n t advances i n t h i s area. o r by some i t e r a t i v e method which involves t h e repeated s o l u t i o n of s e v e r a l sets o f s m a l l e r matrix equations by Gaussian elimination. t h e r e s u l t s can be e s p e c i a l l y s e n s i t i v e t o t h e s i z e o f t h e blocks i n t h e zone where m i s c i b i l i t y i s being e s t a b l i s h e d . These equations can be imbeded within a compositional model. More data than what is c u r r e n t l y a v a i l a b l e are required t o v a l i d a t e and r e f i n e these models. i t is t h e r e s p o n s i b i l i t y of t h e modeller t o make t h e l i m i t a t i o n s of t h e r e s u l t s c l e a r t o t h e u s e r of t h e information derived from t h e simulation. and c a p i l l a r y pressure. and are u s u a l l y a v a i l a b l e . x*l is a vector r e p r e s e n t i n g change i n t h e primary v a r i a b l e s from v t o * l i t e r a t i o n . A s t h e processes become more complex t h e data requirements i n c r e a s e while the a v a i l a b i l i t y of data decrease. SOLUTION OF MATRIX EQUATIONS The heart o f a r e s e r v o i r simulator is a program f o r t h e s o l u t i o n o f a l a r g e set of l i n e a r equations t h a t may be expressed as where A is a s p a r s e matrix with a well defined s t r u c t u r e . There is a l s o concern t h a t . I n s i t u a t i o n s of t h i s type. Several groups. The accuracy of simulation could be improved by ( a ) a d a p t i v e g r i d refinement. Since compvtations with t h e equation of s t a t e a r e i t e r a t i v e . One exception t o t h i s is data on t h r e e phase r e l a t i v e permea b i l i t y . None of t h e s e approaches have f u l l y succeeded so far. t h e assumption of thermodynamic equilibrium between phases i n a block may not be valid. I n a m u l t i p l e c o n t a c t miscible d r i v e process. Most of t h e p r o p e r t i e s required f o r t h e simulation of primary d e p l e t i o n o r water flooding i n crude o i l r e s e r v o i r s can e a s i l y be measured i n t h e l a b o r a t o r y . However. Equations of t h i s type may be solved d i r e c t l y by Gaussian e l i m i n a t i o n . t h i s a s p e c t of t h e problem is n o t t o t a l l y i n t h e c o n t r o l of t h e simulator developer. a t least f o r some processes l i k e t h e i n j e c t i o n of CO2 i n heavy o i l . However. Often l a c k o f good experimental data and the need f o r answers w i t h i n t i g h t time c o n s t r a i n t s f o r c e s one t o make assumptions t h a t may o r may n o t be j u s t i f i e d . o r ( c ) using a n o t h e r ( p o s s i b l y a n a l y t i c a l ) model w i t h i n t h e block t o provide t h e necessary d e t a i l .

In difficult problems it is necessary to treat the coupling between the well and the reservoir in a fully implicit fashion. In general iterative methods that work are cheaper than direct elimination for larger problems. One such method. One disadvantage of this approach is that as efficiency on one machine increases. known as COMBINATIVE. To minimize work I is chosen to be the direction with the largest number of grid points. If the well goes through more than one layer or block. the program becomes less and less portable. and K are the number of grid points in the three directions. The cross-over point depends upon the methods and the problem. (b) solve the pressure equation by Gaussian elimination with D4 ordering and obtain initial estimate of pressure. and (e) apply ORTHOMIN20 acceleration. Program structure and solution algorithms are being developed for this class of computers. additional terms are introduced. The current compilers can only go partways in achieving high efficiency with such processors. J . CONCLUSIONS The need for robust. has worked even for extremely difficult thermal problems20. These are (a) the development of pipeline and parallel processors. (C) use this pressure estimate to form new residuals for (7). Their intelligent use can play a key role in optimizing oil recovery. economical. Some rather powerful iterative methods have been developed recently. (d) do an LU factorization of the whole set and obtain an initial estimate of the remaining variables and an extra contribution to the initial pressure estimate obtained in step (a). This method became more economical than the direct elimination if (J*K*NEQ + NEQ-1) 2 50 (9) The combinative method involves the following steps: (a) decouple pressure equation by neglecting appropriate terms in (7).5 for D4 orderingl. and (b) the development of display and interactive techniques. . Unless properly handled. Interactive preparation of data and graphical display of results can make it much easier to run simulators and analyze results.375 WD f I(NEQ J K ) 3 (8) where I. This is particularly true of the new or infrequent users of a complex model. The coefficient f=l for standard ordering may be reduced to between . Other iterative methods based on the multi-grid21 approach now being developed show even greater promise. Work required for iterative methods is difficult to predict since the number of iterations required depends on the problem. there are two other developments that are beginning to have a profound influence on reservoir simulation. Simulators are an essential tool for understanding and predicting reservoir performance.l9 and . Iterative methods also require less storage than direct methods. The pipeline and parallel processors can perform a large number of operations (up to 5 x lo8 floating point operations per second) very quickly provided the software is designed to take full advantage of the hardware. Another problem with iterative methods is their reliability in difficult situations. work required to solve the equations can increase substantially22. Within the next few years this is expected to become the normal procedure for conducting simulation studies. COMPUTER HARDWARE In addition to the computers becoming faster with larger and larger memory. realistic and easy to use simulators is increasing as the oil recovery mechanisms being applied become more and more complex. This procedure is repeated until convergence is achieved.

The work at the University of Calgary has been supported over the past sixteen years by the National Science and Engineering Research Council (previously National Research Council). Significant new developments in numerical techniques. Number of equations per grid block ( = number of primary variables) Residual vector Vector of change in primary variables over an iteration Vector of primary variables (unknowns) NE4 r X X Pa 9 viscosity of phase . experimental studies must be continued to provide data and correlations for the prediction of fluid and rock properties.K k kr-9. NOMENCLATURE A Jacobian matrix I. but much more needs to be done and will be done over the next few years. Grid nodes along the three directions absolute permeability relative permeability of phase 9. The author is indebted to these organizations for financial support and to his students and colleagues for the generation of ideas and for their implementation in practical simulations. SubscriDts BB d U Block boundary 1 point downstream of block face in question 1 point upstream of block face in question uu 2 points upstream of block face in question . Additional support is provided by Associate'Members of CMG through the membership fees.J. Model validation with carefully conducted experiments is also essential. provide partial funding for the work of CMG through the AlbertaKanada Energy Resources Research Fund. process understanding and hardware have taken place over the last few years. ACKNOWLEDGEMENTS The Department of Energy and Natural Resources of the Province of Alberta and the Department of Energy. Mines and Resources of the Government of Canada.376 Along with the development of new numerical techniques.

11. Ph. 6. J. CIM 81-32-50.. Thesis. AU.H. P.. "Analysis of Vertical Two Phase Flow Calculations: Crude Oil . Calgary. for Thermal Recovery Processes". .. "Elimination of the Constraint .. 15 - . Compositional. J.S.D. and AZIZ.138. ABOU-KASSEM.02. L. K. A.Gas Flow in Well Tubing".. 9.. 5. CIM 81-32-23. and SETTARI. G. University of Calgary (1980). K. Soc. Thesis. 7. tion and Associated Ground Subsidence under Non-Isothermal Two-Phase Flow Conditions". and AZIZ..K. T and Farouq Ali..A. 12. (January 1980). SETTARI. presented at the 32nd Annual Technical Meeting of the Petroleum Society of CIM. D. Eng. A. 37-58. Equation and Modelling of Problems with a Non-Condensable Gas in Steam Simulation". (1980) 3.. CMG. K. B M . "Multi-Level Adaptive Solution to Boundary Value Problems". "Comparison of Solutions to a Three-Dimensional Black Oil Reservoir Simulation Problem". Y3tatic Fracture Model". 3. . (February 1981). GREGORY. 333-390.. Petrol. AZIZ. Applied Science Publishers. M. 86-92. and CHEN. K. W. J. A. R B . Dallas. CMG. 10. J. Math. Petrol.. presented at the 55th Annual Fall Technical Conference and Exhibition of the SPE of AIME. NGHIEM. "Numerical Modelling of Reservoir Compac. FOGARASI. Three-Phase Coning Model". 6. .. W. "Investigation of Grid Orientation in a Two-Dimensional. 8. W..RB. Pet. (February 1979) 9. 1. RUBIN. A. 1. J. Calgary. SPE 9302. London (1979).. presented at the 32nd Annual Technical Meeting of the Petroleum Society of CIM. CULHAM. S.C. FORSYTH.E. llTreatmentof Nonlinearities and Production Allocation in a Fully Implicit.H. Tech. K. Ertekin. 13. R. 487-500. Alberta (May 1981). S. A.A. Comp. BRANDT.01. Three-Phase Steam Model". 2 . University of Calgary (1981). Jr. Houston. 13-25. presented at the SIAM Fall Meeting. B and VINSOME. Cdn. K. Tech.L. (January 1981) 3 1. B and VINSOME. "A Numerical Simulation Model .Sc. KO. (December 1980) 20. 14. M.M..Re. BUCHANAN. Texas (November 1980). Soc. presented at the 32nd Annual Technical Meeting of the Petroleum Society of CIM.. Reservoir Simulation". (April 1977) 2. GESHELIN. WADE. CROOKSTON.M..S. Eng. and VINSOME.. Calgary. Pet.. llModellingInfinite-Conductivity Vertical Fractures Using Source or Sink Terms". RUBIN. ODEH. A. FONG. Alberta (May 1981). Alberta (May 1981)..377 Superscript V Iteration level REFERENCES 1. "Simulation of Hydraulic Fracturing Processes". "Petroleum Reservoir Simulationf1. CIM 81-32-35. BEHIE. "Techniques for Fully Implicit . 4. llStimulatingthe Triassic Carbonates in the Foothills Gas Trend of Northeast British Columbia". Texas (September 1980). "A Critical Comparison of FiniteDifference Interblock Mobility Approximations in Numerical Reservoir Simulation".

"Block I t e r a t i v e Methods f o r F u l l y I m p l i c i t Res e r v o i r Simulation".M. and AU. Texas (September 1980). LEE.D i f f e r e n c e Scheme f o r Eliminating t h e Grid O r i e n t a t i o n E f f e c t i n Numerical Reservoir Simulation". Las Vegas.. SIAM J. "A Robust I t e r a t i v e Method f o r Flash C a l c u l a t i o n s Using t h e Soave-Redlich-Kwong o r t h e Peng-Robinson Equation o f Statell. Nevada (September 1979). GEORGE. W.. HEIDEMA".A. J r .A. A. Nevada (September 20. Nevada (September 1979). L. Wxperimental and F l u i d P r o p e r t i e s Required f o r Simulation o f Therpresented a t t h e 54th Annual F a l l Technical Cont h e SPE o f AIME. Dallas.H. CHEN. SPE 8248.. and A Z I Z . Las Vegas.K. 21.. SPE 9303. CMG. V a l c u l a t i o n of Multiphase Equilibrium f o r Compositional Simulationf1. G. f e r e n c e and Exhibition of 1979).H. Texas (September 1980). Las Vegas. S. R. . and CULHAM. . Dallas. BEHIE. 17.01 ( J u l y 1981).. presented a t t h e 54th Annual F a l l Technical Conference and Exhibition of t h e SPE o f AIME. presented a t t h e 54th Annual F a l l Technical Conference and Exhibi t i o n of t h e SPE o f AIME. P. SPE 9232. R. 2. BEHIE. presented a t t h e 55th Annual F a l l Technical Conference and Exhibition of t h e SPE of AIME.A. 19. Numer. K. W. JACOBY. (June 1974) 11. 18. MEHRA.E. R.. N G H I D l . A. G.T. T h e o r e t i c a l S t u d i e s on t h e mal Processest1. and FORSYTH. K. KO. K. "On Block Elimination f o r Sparse Linear Systems". and V I N S O M E . 585-603. SPE 8293... and A Z I Z .A. Anal. S. 22. presented a t t h e 5 5 t h Annual F a l l Technical Conference and Exhibition of t h e SPE o f AIME. **A Weighted Nine-Point F i n i t e . SPE 8285.Rl7..C. "Multi-Grid S o l u t i o n o f t h e Pressure Equation i n Reservoir Simulationf1.378 16.

water and steam) heat flow in the reservoir a heat & conduction in the surrounding formations. This formulation has been mentioned in the literature6 but not tested otherwise than in isothermal black-oil model.three-phaseflow (oil. weakly or highly implicit. and is nevertheless highly stable. owing to the fact that the water and gas flow equation and the energy equation are solved simultaneously with implicit treatment of the gas transmissibility. The efforts have been concentrated on methods of solution. INTRODUCTION TWIST (Tool When Injecting Steam) is a three-dimensional steamflood model. The various reservoir heterogeneities and temperature dependent parameters (including relative permeabilities) are considered. A semi-implicit method is used for time formulation. Vertical heat losses to overlying and underlying strata and wellbore heat losses between the surface and the reservoir are taken into account. The oil phase equstion is decoupled with a scheme of the type p-T-S /S This formulation enables this thermal simulator to be very efficient $n a k m s of coinputing time and stability . Mass conservation and energy equations are solved simultaneously to improve stability. Rueil Malmaison. We use a formulation which requires significantly less computing time per grid block-time step than the implicit scheme.NUMERICAL METHODS 379 THREE-DIMENSIONAL NUMERICAL SIMULATION OF STEAM INJECTION P. The literature on the simulation of steamflooding is extensive 1-7. steam) and heat flow in the which describes. The equations are solved sequentklly or simultaneously and the formulation is explicit. Wellbore heat losses between the surface and the reservoir are taken into account. France ABSTRACT k three-dimensional thermal model has been developed for simulating both cyclic steam injection and steam drive. LEMONNIER Institut FranGais du Pktrole. Numerical results are presented showing e steam stimulation history of five cycles and the influence of steam quality. . initial reservoir pressure and steam injection rate on steamflood performance in a five-spot pattern. Distillation effects are approximated through the decreased residual oil saturation when steam is present. water reservoir. The numerical model !IWIST describes three-phase flow (oil.

ed shoving a steam stimlation history of five cycles and the influence of steam quality a i reservoir pressure on oil rd recovery in steam drive. The four unknowns are oil pressure. We have three additional equations for obtaining oil saturation. s +s +s o w s = 1 Po .Pw = PCW P. Numerical results are present. ve Our experience includes steam stimulation and steam drive for pilots o f various pattern shapes and various oil viscosities. saturated steam. steam and water saturations. . ( 3 ) conservatioo of energy and ( 4 ) equilibrium constraints. The equilibrium constraints are expressed by one of the three equations ( 4 ) for the following cases : no steam. MODEL DESCRIPTION Simulator equations The model consists of four equations expressing (1) conservation of mass for water and steam. ( 2 ) conservation of mass for oil phase. ( 6 ) and ( 7 ) capillary pressures.380 i ' encountered no difficulties in simulating field c a s e s rith TWIST. . superheated steam. temperature.Po = Pcg The phase velocity 7: is defined as The condensation term is eliminated by summing the water aad steam mass conservation equations 3 . gas and water pressures : (5) saturations constraint.

The equation is approximated by the standard finite -difference approximation.Shut in. The injection pressure and the steam quality are s ecified at surface or reservoir conditions. We assume negligible effects of heat conduction in the horizontal directions 3 . homogeneous o r heterogeneous by layers or by cells.Bottom-hole pressure 2 . At. . 6 7 . .381 Additional assumptions 1 2 . The heat conduction equation for the surrounding rock is not solved simultaneously with the reservoir equations. The model is not compositional. We use the method o f Rameyl3. SatterP4 for wellbore heat losses computations 20. wellhead pressure.Temperature dependency of the physical and thermal parameters is accounted for.vityare takeo into account.Water or steam injection rate . .Liquid production rate . Well model The wells have the following specifications : 1 .Numerical simulations include steam drive and steam stimulation.The model can operate in one. Reservoir rock and fluids areampressible.Any variation in steam pressure with depth is negligible. 8 . Distillation effects are approximated through decreased residual oil saturation in the presence of steam. Wellbore heat losses in the injection wells are calculated. temperature and quality. Heat loss to overburden and underburden Heat loss by conduction to the overlying and underlying strata is calculated from the numerical solution of the heat conduction equation.Three-phase relative permeabilit'es at each temperakure value are i calculated using Stone's method .Reservoir can be anisotropic. - . The basic assumptions o f the method are as follows: Steam is injected at a constant rate. 3 4 - Reservoir dip and gra. two or three dimensions with Cartesian or'radial grid.time step n+l the equation is solved using the reservoir boundary temperature at the previous time step. 5 .Oil production rate 3 4 5 Each well can operate successively in injection and production modes (for huff and puff process for example).

Sn. T+) n' bX = = f (p". Tn) + df df b~ b p +-6. fluid densities and f l u i d enthalpies. temperature or saturations contours or profiles. Various print. just as pressure. mobilities and enthalpies in the flow terms. the same assumption is made for the. . Residual oil and irreductible water saturations are represented as linear functions of temperature. WOR. Array maps can be selected and printed with any orientation.%pproximcttions9 are used for time discretization. Oil and water enthalpies are expressed as polynomial functions of temperature. Semi-implicit .) and well behaviour versus tine (presviscosity. The influence of temperature on the relative permeability curves is described by shifting the curves wit. Graphic plotting of input data (oil ..relatire permeability end points.5 b S df +z 6T (9) x"" . S.difference form with upstream densities. Explicit dating is used f o r water and oil relative permeabilities end semi-implicit formulation is used for steam relative permeability. . relative permeabilities.e. sure. Water and steam viscosities are expressed as functions o f temperature.) are available. Steam enthalpy am2 all fluid densities are treated as functions of pressure and temperature. The rock specific heat and thermal conductivity are represented as linear functions of temperature.hout changing the curves shape5. Capillary terms are expressed explicitly in saturation. .f with X = p.. Arrays are dimensioned automatically with appropriate values at the beginning of each run.ingoptions are allowed for input data and output rpsults. time level n) dating is considered for fluid viscosities. where sn+l. All input cards are checked for validity and for inconsistencies. T. SOLUTION METHOD Discretization The three equations ( 1 ) ( 2 ) ( 3 ) are discretized into finite .. The accumulation terms are written with implicit dating. formulation is then linearized as follows : The resulting f (pn+'. Explicit (i. oil recovery. Interblock transmissibilities o r flow terms are treated as follows. All errors encountered are listed at the end of the dataprocessing.382 Fluid and rock properties The steam and water properties are expressed through correlations fron the steam tables. on account of the weak sensitivity to implicit versus explicit dating encountered in numerical simulations. User facilities User facilities have been developed in the simulator. Oil \-iscosity is entered into the model as tabular function o f temperature with exponential interpolation between adjacent entries in the table.

13 -. The definition of 6 X can vary from one grid cell to an other and from one time step to an other according to the equilibrium condition in the grid cell at time step n+1. The unknown bX is equal t o b T o r 6Ss. 6T. = A . eliminating 6T 6X = 6 T and equation (4a) is used 6 X = 6 S S and equation (4b) is used for 6 X =6T and equation (4c) If superheated steam is present at time step n+l.3 F21 R . If steam is present at time step n+l. 8 S s are reduced to two equations in the two uknowns bp. The equations ( 4 ) (13) (14) involving the three unknowns 6 p . The first three equations can he represented by The explicit treatment o f oil and water relative permeabilities allows the water saturation unknown 6 s to be eliminated in equations (10) and (12) by means of equation ( 1 1 ) . for eliminating 6 S s . Resolution Procedure The procedure of solution of the equations for the time step n+l is as follows : 1 Solve the two equations ( 1 3 ) (14)for bp and 6 X using Gaussian direct solution with I) 4 ordering 1 1 . A semi-implicit formulation for production terms similar to that described by Spivak and Coatsto is used for increasing computational stability. l4 X A A24 2j F t i l = Fil - A i4 A24 j = 1. A. made with the equilibrium conditions at time step n - . 6Ss andbs. If no steam is present at time step n + l . . i = 1 et 3 .383 With these approximations the four equations ( 1 ) ( 2 ) ( 3 ) ( 4 ) are expressed in terms of the four unknowns bp. Explicit dating of saturation-dependent production terms can give saturation oscillations in grid cells near the wellbore. is used f o r eliminating 6 S s .. The"e1imination of 6 s W results in a system of two coupled equations ( 1 3 ) ( 1 4 ) : A I 13 I . The substitution of the unknowns 6s or bT is . 6X by use of the equilibrium constraints ( 4 ) . = R . bT. --xR2 A24 A 14 ..

5.25 . The run were made in a two-dimensional radial configuration (r. z) with a 12 x 6 grid. from 01 to 0.2 0.16 ft) . . choose an other equilibrium condition and solve again the two equations for 6 p and 6X. Wellbore heat losses computed by the model give at the end of injection a quality of 90% at the bottom of the well. 12 .384 2 Check the validit. 3 - - Solve equation ( 1 1 ) for 6 sW using the values dp. If not.DATA FOR W L L STIMULATION PROBLEM 24 m (78.3 I’ 41 . The steam quality was 1 0 at surface conditions and the injection temperature was 264OC (5070F).4 1C-4 psi-1\ 26OC (790F) 2800 kPa (406 psia) 0. The top of the reservoir is located at a depth o f 228 m (748 ft). II 180 II 1 9 55 . The data are given in Table 1. Numerical simu’ lation studies of a steam drive pilot12 were made with a two-disensional grid.9 01 . TABU 1 .74 ft) 200 m (656.43 at 26OC and 0 1 at 22OOC . Two example cases are presented to illustrate the use of the model in the case of well stimulation and steam drive.y of the solution for each grid cell. from 0 8 to 2 p 2 (800 to 2.tion The first application of the model consisted in simulating the production history of a well submitted to five successive steam stimulation cycles.7 Well stimulation problem 8924 mPa.000 nd) 10 Zone thickness Exterior radius Porosity Horizontal permeability Anisotropy kh/kv Irreductible water saturation Residual oil saturation to water Residual oil saturation to steam Rock compressibility Initial temperature Initial pressure Initial water saturation TABLE 2 Temperature 200’2 ( 680F) 5OoC (1220F) 1 oooc (2120F) 150OC (3020F) 2OOOC (3920F) 26OOC ( 5000F) 0. Other values are given in Table 2. Oil viscosity at initial temperature of 26OC (790F) is 4270 mPa.s (cP) 545 I ’ 36 ’’ 8.OIL PRASE VISCOSITT Steam drive problem 5159 n f a S (CP) i‘.10-5 Ha-’ (3. 6T and 6 s s .s (cP). Simulations of the whole seven-spot pattern are now pursued with a three dimen) sional grid (12 x 9 X 5 . 2 5 It . 26 . Well stimula. We specified a steam injection rate of 1 0 0 metric tomes per day (629 B/D 0% cold water equivalent) into the six layers. EXAX’LES OF API‘LICATIOXS k e used the model for simulating a number of field cases.2 .

A new cycle is initiated when the oil rate has decreased to 5m3/da. According to this criterion the durazions of the successive production cycles have been 152. DAYS 600 t 0 Fig.Water-cut versus time for five stimulation cycles . 130 and 133 days. 1 1. - Oil production rate for five successive steam stimulation cycles 0. 9 9 . 2 . DAYS Fig. ) .3 1 6 I T I M E .63 0 - a I g 0. One cycle involves 25 injection days. 136. The well bottom-hole pressure evolution is shown in Fig.4- ? f o ! 0 I I 200 400 T I M E . 1 and Fig. \ ) P I- S 0. The variations of oil rate and water cut with time for the successive cycles are shown in Fig. 2.385 We simulated five cycles. 5 soaking days and a producing period with a totai fluid production rate of 20m3/ day (126 BID).8 - Cycle 1 .y (31 BID).

L higher depletion and a better performance for the first cycle and to a relatively poor per%ormance forthe second cycle (Fig.eam ratio may be essentially attributed to the relatively low porosity of the reservoir and to the high viscosity of the oil. .386 during the 800 days of history.Well bottom-hole pressure versus time for five stimulation cycles An other simulation performed with predefined values for the length of the five successive production phases leads to the following values of the .32 after 120 producing days of the . DAYS 600 I 8 0 Fig.64. 1).500 tomes (78.23 and 0.19. 0. 0. The history includes 125 days of injection.9 third cycle.24 after 90 producing days of the second cycle.1271113 (51.18. 0. The total oil recovery was 3. 3 . 0.616 STB cold water equivalent) and a total water production o f 8 . 0 2 after 150 producing days o f the fourth cycle and 0.346 STB) for a total steam injection of 12. The total oil recovery was the same.113 STB) The values of the oil/steam ratio for each of the five cycles are the Dollowing: 0.31.27 after 180 production days of the last cycle. The fair values of the oil/st. oil/steam ratio: 0 4 after 90 producing days o f the first cycle. Cycle 5 END OF INJECTION 60 START OF PRODUCTION a 0 240 0 u 5 v) t 3 t 20 0 I 1 200 400 T I M E . Cycle 1 Cycle 2 Cycle 3 Cycle 4 . 0. .570m? (24. 25 days o f soaking and 650 days of production. The criterion chosen for the end of the production phase leads to a greater length.

4 ft) and Az = 4m (13. One eighth of a five-spot pattern was represented by a 6 X 3 X 5 grid (Fig. 5 for two temperature . The water-oil relative permeability curves are shorn in Fig.387 Steam-drive Many studies have been devoted to the evaluation of the performance of steam flooding technique .l 5-18. 6 . 4 .14m (46.Water-oil relative permeclbility curves Fig. The relative permeabilities were temperatwe dependent.Gas-oil relative permeability curves .4) with A x =Ay = 14. However the influence of some operating parame-cershas not yet been clarified. 5 . The effects of steam quality and reservoir pressure on steamflood performance were investigated in a five-spot pattern with the simulator.12 ft) Table 3 summarizes the data for this problem and Table 2 shows the oil viscosity versus temperature.Simulation grid for PRODUCER one--eighth of five spot pattern Fig. 3 I U 1 2 3 4 5 6 INJECTOR Fig.

8 the oil recovery for a given heat input is little sen% sitive to steam quality when quality is above 6 . 500 kPa (72. Fig.1 J/e O C ( 0 .ft -OF) ( 3 2 Btu/ft (32 Btu/ft OC . 0 .5 - OC O C (37 Btu/cu. The high increase of injection pressure results from the formation of a high viscosity oil bank downstream from the condensation front. 5 Btu/lb OF) 0. .35 J/cm3 3 2 . 5 J/cm p2 (1000 md) 1 (6.1 25°C (77OF) - 0. 1 Area (5-spot) Reservoir thickness Porosity Horizontal permeability Vertical permeability Formation compressibility Specific heat of formation Specific heat of overburden and underburden Thermal conductivity of formation Thermal conductivity of overburden and underburden Oil compressibility Thermal expansion coefficient of oil Specific heat of oil Stock-tank oil density Irreductible water saturation Residual oil saturation to water Residual cil saturation to steam Initial temperature Initial water saturation Initial pressure Injection rate for full pattern (WE) Production bottomhole pressure 2.5 psia) and 4000 lla i’ (580 psia) 50 m3/day (314 B/D) 300 kEa (43. Injection pressure starts to decrease when the oil bank becomes mobile.388 TABLE 3 - DATA FOR STEAM-DRIVE PROBIXM 2 10.000 m (2. 9 shows earlier steam breakthrough when steam quality increases and initial pressure decreases.day - OF) OF) kPa-’ ( 4 . 4 at 25°C and 0. Well injectivity and productivity indices as calculated according to Peaceman l 9 were multiplied b two.6 OF-l) 2. As shown in Fig. 7 shows the effect of steam quality on injection pressure. The heat input is equal to the cumulative enthalpy o f steam at sand face referred to initial reservoir temperature.35 2 . The gas-oil relative permeability We studied the effects of varying bottom-hole steam quality from 0 to 1 for t w o values o f initial reservoir pressure. Specified injection rate for both set of cases was 50 tonnes/day (314 B/D cold water equivalent) for the full pattern.25 at 25°C and 0 4 at 175OC . 5 p2 (2500 md) I. steam is injected only into the two bottom layers.95 g/cm3 (60 lb/cu. Fig.5 psia) and 3800 kPa ( 5 5 1 psia) values 09 25OC (77OF) and 2OO0C (3920F).ft -OF) - 2. curves are shown in Fig.8 psi.5 acres) 20 m (65.4 500 kPa (72. ft) 0. for both Fa cases respectively.4 6.5 psia) and 3800 k ’ ( 5 5 1 psia).day . 4 10-6 psiA’) OC-’ (3.5 psia) and 4000 kPa (580 psia) respectively.) OC ( 3 5 Btu/cu.3 W/m 2 .2 at 17joC 0. y Production wells produce from the five layers at deliverability against a bottom-hole pressure of 300 kPa (43.6 ft) 0. 6. 3 Wim 6.

9 .Effect of steam quality and initial reservoir pressure on oil recovery after 30 TJ heat input.2 0.I y 30a s i= 2 152 - P R E S S U R E = 500 k P o -O-0 Fig. 4 INITIAL ----.389 60 - 2 w -r 45- 2 4 Lu & 3 v.Effect of steam quality and initial reservoir pressure on steam breakthrough. Fig. 7 ----------- - 1 I 2 3 T I M E . .YEARS .6 STEAM OUALITY 0.Effect of steam quality and initial reservoir pressure on injection pressure 0 0. 8 .4 0.8 1 S T E A M OUALITY Fig.

Low heat.5% of heat input. after 6 years of injection. It appears that the oil/steam ratio is improved when steam quality increases and initial reservoir pressure decreases. 13 shows the effect o f steam quality and initial reservoir pressure on the energy yield.8 0 0. Hence we introduce an adimensional parameter. 12 versus steam quality and initial reservoir pressure. f o r comparing the performances of the tests. 1 3 .Effect of steam quality and initial reservoir pressure on vertical heat loss after 6 years Fig. 1 1 ) .6 1 STEAM O U A L l T r Fig.Effect of steam quality and initial reservoir pressure on heat produced The cumulative oil/steam ratio is plotted in Fig. I a 6o - 2 -- - I N l T l P L PRESSURE 500 k P o ( 72 p i 1 0 I 4000kPo~580p8is) 50- z 0 02 . The oil/steam ratio does not take into account the variation of heat input due to the variation of steam quality. The value of EY is equal to one when the energy content of the steam at sand face is equal to the calorific value of the produced oil (calorific valae of oil = 38 GJ/m3). instead of 29% in the low-pressure case considered by Gomaa. less heat losses are achieved for the lower initial pressure when quality is above 3@.2 0. 10 show that the curves are also reservoir pressure dependent. depending on initial reservoir pressure and on the duration of steam injection.emperature and faster heating of the reservoir.4 0.. 10. For these two values the vertical heat l o s s curves of Gomaal7 obtained with an initial pressure of 414 kPa give the same Talue of 31%. . 0. The heat injection rate for this case is 460 kJD-m3 reservoir and t. The results in Fig.4 0. the energy yield EY. Fig.390 Vertical heat losses to overlying and underlying strata are relatively independent of time after steam breakthrough.he reservoir thickness 50 m.6 0.6 STEAM OUALlTY 0. due to lower steam t. 11 . 10 . The energy yield is defined as the ratio between the calorific value of the cumulative oil produced and the heat input previously defined. For a steam quality of 6% and an initiai reservoir firessure of 500 kPa (72 psia) the vertical heat loss is 31% of hea. and the heat loss is 38.t input. loss to overburden strata and early breakthrough result in a high amount of heat produced by the wells when operating at low initial reservoir pressure (Fig. For a pressure of 4000 kPa (580 psia) the heat injection rate is 505 kJ/D-m3 res. An optimum steam quality value can be determined in Fig. Heat losses after 6 Sears of steam injection are presented in Fig.

The dispersion o f the data is less than in the case of Kern-River since the simulations were carried out on a uique pattern whereas the correlation f o r Kern-River had been obtained from the field results o f several pilot tests.08 00 simulating a 9-year steam drive (582 time steps) with 4 0 kPa initial pressure .39 1 T n e selisitilTitY t o steam qualitj.Fig. An optimum steam in.Effect of steam quality and initial reservoir pressure on cumulative oil/steam ratio after 6 years Fig. 13 - Effect of steam quality and initial reservoir pressure on energy yield. 12 .is much stronger at l o w pressure than 2 t high pressure. 14. This optimum steam rate corresponds to the value of 50 m3/day (314 B/D ) used in the sensitivity study for the full fivespot pattern. -5 - Model running time The formulation described above is very efficient in terms of computing time. The curve in Fig. The same reason may explain the shift o f the optimum sream quality towards lower values when t.5 B/D -A-ft for Kern-River16). The computation time f o r . This is relared t o t h e higher amount of heat transported by the produced fluids in the case of lokT pressure tests (Fig. 11).ime increases. The simulation of the steam drive pilot12 with a three-dimensional grid 9 X 12 x 5 (405 active cells) requires 0.0018 seconds per grid block-time step.5 psia) initial reservoir pressure. We made a similar study with the simulator in the case of 6% bottomhole steam quality and 500 kPa (72. The case presented here of a three-dimensional steam-drive with a 6 X 3 x 5 grid ( 6 0 active cells) requires 0 0 0 seconds per grid block-time step.8 1 Fig. after the project had reached a peak oil-production rate. 0 02 0:4 0:6 STEAM OUALlTY 0. was determined at Kern River from field results16.5 10 m3/day/m3 o f reservoir volume (1. A three-dimensional run with a 25 x 8 x 4 grid (800 cells) requires 0. The optimum volumetric injection rate. 14 has the same shape as the curve developed for Kern-River. As 2 matter of fact the heat produced after 6 years of injectiori is about tvice the value obtained after 3 years.002 seconds per grid block-time step on the CDC 7600 computer.94 B/D A-ft) was found (1. jection rate of 2. as a result optimizes the rate o f t e instantaneous oil/steam ratio.

CONCLUSIONS 1 The semi-implicit formulation of the. It appears that these parameters affect the heat loss to the surrounding formations. The model may be used for simulating a wide variety of thermal problems. the heat transported by the produced fluids and the performance of the process.5 f o r a 12 X 12 X 4 grid (576 cells).392 BOTTOM-HOLE STEAM CUALITY = 6041~ 0 0.M 3 OFRESERVOIR VCUJME Fig. .25 05 .25 PRODUCTION 1 6 4 d / L J A Y . Partial financial support for the realfzation of the simulator used in this study was provided by Soci6t6 Nationale Elf-Aquitaine (Production). The ratio between computing times on a CDC 7600 and a vector computer CUP 1 has been 5.0 Oj5 1. 14 - Optimum injection rate and 6% downhole steam quality has been 28 seconds on the CDC 7630 computer. Better performances and higher sensitivity to steam quality are observed at lower pressures. - 3 The influence of steam quality and initial reservoir pressure on steamflood performance has been investigated.60 1. BURGER of Institut Francais du PBtrole f o r his helpful and constructive discussions. J. He also expresses his appreciation to Mr. 4 The analysis of the steamflood performances obtained for various steam injection rates at givenquality and pressure indicates the existence of an optimum injection rate. - 2 Five successive steam injection cycles in a low porosity reservoir have been simulated to evaluate the decline of the oil/steam ratio from cycle to cycle.solution method used for the three-phase three-dimensional model TWIST enables the simulator to be very efficient in terms of computing time and stability. - - ACKNOWLEDGEMENTS The author wishes to thank Institut Franqais du PCtrole for permission to publish this paper.G.

: "Simulation of Steamflooding with Distillation and Solution Gas". P. J. E G : "Numerical Model for Thermal Processes".. .M. . J (Oct. 4 . measured vertically downward (m) specific weight (kPa/m) time difference operator. J. W. and Rubin. J. presented at SPE 54th Annual Fall Meeting. .Weinstein. 1973). B E : "Three-Dimensional Simulation of S'. Eng.393 NOMEXCLATURE = = = = = = = = qL S t T TS = = = = = = U = = = = = = = enthalpy (J/g) absolute permeability (m2) relative permeability thermal conductivity (W/m OC) pressure (kPaj capillary pressure (pa) mass injection or production rate (Ton/day) enthalpy production rate (J/day) heat loss rate (J/day) saturation time temperature ("C) temperature of saturated steam (OC) internal energy (J/g) phase velocity depth. Three-Phase Model of the Two-Dimensional Steamflood Process". SOC. Pet. 1970) 405-417. N.g. Chieh. Eng. Farouq Ali. 1976) 235-247. no 4 .. H. (Feb. H. .. Sept. Woods. 1974) 573-592. 6~ = Xn+l viscosity (Pa..Coats. SOC. Las Vegas. 23-26. "Estimation of Three-Phase Relative Permeability and Residual Oil Data". B. Chu..L. Can. Eag... SOC. 2 .C. K. (Oct. J. George. Compositional Thermal Simulator for Steam Injection Precesses" . e. Vinsome.. J.. W.G.. SOC. Calgary. J. Lin.s) porosity density (g/cm3 ) - .W. . June 7-11 1936. S. 3 . Grabowski. A. Pet. Two-Dimensional. (Oct. (Dec. (Dec. Eng.H.: "Numerical.Paper 7613 presented at 1 5 6 7 - - 8 - 27th Annual Technical Meeting of the Petroleum Society of CIM..X" Subscripts and superscripts = steam = time level = oil = rock = steam = water RGFERENCES . Pet. Behie.: "A Three-Phase. Eng.K. Wheeler. K. Pet. Stone. 1978) 369-383.Ferrer. R. J. Tech.eamflooding". 1977) 65-78.. SOC. Coats.A. J. K B : "A Highly Implicit Steamflood Model". Pet..H.D. Petr. 1979.C.Shutler. V 1 2 . Marcum.: "A fully Implicit General Purpose Finite-Difference Thermal Model for In-SLtu Combustion and Steam?. paper SPE 8396. Coats.

1965) 845-851. Pet. . : "Direct Methods in Reservoir Simulation". Tech. 1975) 765-776.. : "Steam Drive Pilot in a Fractured Carbonated Reservoir Lacq SupCrieur Field. J. J. : "Wellbore Heat Transmission". Peaceman. G. D. 1978) 183-194.. (August. Pet.W. H. Sahuquet. Ehrlich. J. SOC. Eng.394 9 10 11 - Lolen. (July. 1 5 . J (June 1972) 253-266. 18 . Dallas.J. J.H. Ferrier. SOC. Crookston. Tech. Paris. E. K.. Texas.H. H. : "Kumerical Simulation of Coning Using 1mpli. R. Pet. C.B. Eng. (Sep. To be published by Editions Technip. . Sourieau. J. (June.E.Gomaa. Pet. :"Thermal Methods of Oil Recovery: Chapter 4 .B. J. R. Eng.. : "Performance of Steam Displacement in the Kern River Field".Spivak. : I 1 Interpretation of Yell-Block Pressures in Numerical Reservoir Simulationt9. 1970) 257-267. 1980.J. 1980) 325-332. Pet.S.Chu. D. 1980. Pet. (June 1974) 295-308. presented at SPE 55th Annual Fall Meeting. (June. Trimble.J. J.Nolan. (Feb. (Apri1. Applicability of S t e m 13 :I1 19 20 - flooding for Carbonate Reservoirs*!.. J. Oklahoma. Eng. Tech. :"Numerical Simulation of Steam Displacement Field Performance Applications". P.E.Bursell. SOC. Pet. "paper SPE 9453. .Satter. Coats. J. Pet. SOC.M. presented at the First Joint SPE/DOE Symposium of Enhanced Oil Recovery. Pithan. d . A..G. J. :"Heat Losses During Flow of Steam Down a Wellbore". Burger. 17 . K. 21-24. Berry. Tulsa. Tech. . Tech.S.W. Coats. JR. :"Correlations for Predicting Oil Recovery by Steamflood".cit Production Terms".1962) 12 427-435. Sept. 14 . : "Tests o f the Stability an2 Time-Step Sensitivity of Semi-Implicit Reservoir Simulation Techniques". J... Ramey. Pet. April 20-23. A.C. 16 .Price. E.paper SPE 8821. 1975) 997-1004. C.

I. particularly if the relative permeability curves are highly non-linear. this gain in efficiency is realised only if the special problems associated with long time steps can be overcome. reservoir engineers are freed from the need to consider the internal working of their simulator. The solution of the linearised equations using iterative techniques becomes more difficult as the time step is increased. Of these problems. R. Hawell. The convergence of the non-linear equations which are solved at each time step can also present difficulties for long time steps. for many problems. Experience with large 3D problems arising from North Sea applications has been most encouraging. A new 'nested factorisation' algorithm for solution of these equations is described. In our experiencecl). coning). as well as the most robust alternative. It is shown that. The wide applicability of fully-implicit methods also reduces the need for special purpose simulators designed for particular applications (e. It is often thought that fully-implicit simulators are less efficient in their use of computer time than IMPES and semi-implicit alternatives. Finally. as the strong stability of the method allows the simulator to take much longer time steps than would otherwise be possible. by using a new form of time averaged relative permeability. The application of this technique also reduces the non-linearities in the mass conservation equations which are solved at each time step.g. However. APPLEYARD. Indeed. M. bubble point crossing. a fully-implicit simulator is the most efficient. high permeability contrasts. POLLARD Atomic Energy Research Establishment. England ABSTRACT This paper addresses some problems which arise when a fully-implicit black oil simulator is allowed to take large time steps. etc. this need not be the case. the most obvious is the increased numerical dispersion arising from time truncation errors (as distinct from space truncation errors) resulting in additional smearing of flood fronts. methods I s motivated principally by the much greater stability and robustness of fully-implicit methods when applied to problems involving strong gravity segregation. INTRODUCTION The use of fully-implicit numerical methods in reservoir simulators is becoming This shift away from IMPES and semi-implicit increasingly wide~pread('*~'~). The new method is shown to be more efficient than existing techniques on a set of 2D test problems. and can concentrate on more important issues. it is possible to reduce time truncation errors to a very low level. coning. CHESHIRE and R. As a direct result of this improved robustness. Oxfordshire.NUMERICAL METHODS 39 5 SPECIAL TECHNIQUES FOR FULLY-IMPLICIT SIMULATORS J. . K.

............ T h i s p a p e r a d d r e s s e s each o f t h e s e problems i n t u r n ........l i n e a r r e l a t i v e p e r m e a b i l i t y terms... Elements o f Y r e p r e s e n t t h e mass of a phase i n a c e l l ..... A t (3) t: F(t+At) .. and p r e s e n t comparisons w i t h o t h e r widely used methods. (2) method t h e r i g h t . and e l e m e n t s o f F t h e sum o f flows from n e i g h b o u r i n g c e l l s and w e l l s ... F i r s t l y ..... w e show t h a t i t is p o s s i b l e t o reduce t i m e t r u n c a t i o n errors s i g n i f i c a n t l y u s i n g a new t e c h n i q u e f o r computing t i m e averaged flows. i n p r a c t i c e .i m p l i c i t flow i n t e g r a l can b e w r i t t e n a s F(t')dt'= X(t+At)kr(t+At)At ...396 s o l u t i o n o f t h e l i n e a r e q u a t i o n s u s i n g i t e r a t i v e t e c h n i q u e s becomes more d i f f i c u l t as t h e t i m e s t e p i s i n c r e a s e d .....i m p l i c i t approximated by I t+At F<t')dt' t ...... T h i s s u g g e s t s t h a t i t may b e p o s s i b l e t o reduce t i m e t r u n c a t i o n e r r o r s s i g n i f i c a n t l y by performing a c a r e f u l t i m e i n t e g r a t i o n i n which s p e c i a l a t t e n t i o n i s g i v e n t o t h e n o n ........h a n d s i d e of e q u a t i o n ( 2 ) is and t h e t i m e d i s c r e t i s e d e q u i v a l e n t o f t h e d i f f e r e n t i a l e q u a t i o n (1) i s -= rAt FCt')dt' = F(t+At) .......... w e i n t r o d u c e a new and h i g h l y e f f i c i e n t t e c h n i q u e f o r i t e r a t i v e s o l u t i o n o f t h e l i n e a r e q u a t i o n s ..l i n e a r convergence problems.. I n t e g r a t i n g (1) o v e r a t i m e s t e p A t g i v e s AM = I n the s t a n d a r d f u l l y ... The a p p l i c a t i o n o f t h i s t e c h n i q u e a l s o reduces t h e s e v e r i t y o f n o n ...... i t i s o f t e n n e c e s s a r y t o l i m i t t h e t i m e s t e p i n o r d e r t o p r e v e n t t h e growth o f t i m e t r u n c a t i o n errors. E = F(t+At) - F(t) ................. (4) Equation (4) is s t r o n g l y s t a b l e which m a k e s i t p o s s i b l e t o a c h i e v e h i g h computing e f f i c i e n c y by t a k i n g l a r g e t i m e s t e p s ... F i n a l l y .. t h e main t i m e t r u n c a t i o n errors o c c u r i n t h o s e c e l l s where s a t u r a t i o n s are changing r a p i d l y ............ (5) Using e q u a t i o n ( 5 ) i t i s p o s s i b l e t o p l o t g r i d maps o f t h e t i m e t r u n c a t i o n e r r o r and.............. T I W TRUNCATION ERRORS The s p a c e d i s c r e t i s e d f i n i t e d i f f e r e n c e e q u a t i o n s governing t h e flow o f o i l . ... However... water and g a s c a n be summarised i n t h e form -= dM dt F ..... (1) where M and F are v e c t o r s ........ An estimate o f t h e s e errors can b e o b t a i n e d by comparing flow and w e l l terms a t t h e b e g i n n i n g and end o f each t i m e s t e p .. as one might e x p e c t ..... Time T r u n c a t i o n C o r r e c t i o n The s t a n d a r d f u l l y . (6) ......

..... if t h e s a t u r a t i o n change is small........ t h e i n t e g r g l s i n equation (10) can b e performed a n a l y t i c a l l y ..0(6-f(s)) ds dt ........ r a t h e r than focusing a t t e n t i o n a t one or two p o i n t values....-ds dt 0 ... (7) R e l a t i v e p e r m e a b i l i t i e s a r e f u n c t i o n s of s a t u r a t i o n .... If t h e r e l a t i v e permeability curves a r e approximated by piecewise l i n e a r f u n c t i o n s .... I t may b e shown t h a t t h e t i m e averaged r e l a t i v e permeability is c l o s e t o t h e time-centred value.. (9) as t h e s a t u r a t i o n approaches i t s l i m i t i n g value... (10) Equation (10) forms t h e b a s i s of a p r a c t i c a l technique for computing average r e l a t i v e p e r m e a b i l i t i e s during l a r g e t i m e s t e p s . Equation ( 8 ) can now be used t o transform t h e t i m e i n t e g r a t i o n i n ( 7 ) t o an equivalent s a t u r a t i o n i n t e g r a l giving krds ................ &(kr(t+At) + k r ( t ) ) .E At a z ax .. k. AM A (11) where is obtained by r e p l a c i n g r e l a t i v e p e r m e a b i l i t i e s by t h e i r t i m e averaged values..397 where kr(t+At) is t h e value of t h e r e l a t i v e permeability a t t h e end of t h e t i m e s t e p ....... a s t h e s a t u r a t i o n change increased... and necessary t o transform t h e t i m e i n t e g r a t i o n i n ( 7 ) t o an i n t e g r a l . A l l o t h e r terms a r e evaluated a t t h e end of t h e t i m e s t e p ....... The coupled non-linear equations (11) are solved for p r e s s u r e and s a t u r a t i o n i n each g r i d block by Newtonian i t e r a t i o n : - [---Z]AZ=E..=-L t A t+At kr(t')dt' . 2 The d i s c r e t i s e d equivalent of (1) may now be w r i t t e n a s _t ... The c o n s t a n t .. (12) ............ A approximate transformation can be obtained n (a) t h e flow is l o c a l l y incompressible (b) c a p i l l a r y forces a r e negligible (c) t h e flow o u t of a c e l l depends only on the average so t h a t it is t h e r e f o r e equivalent saturation by assuming t h a t saturation i n the c e l l _ ........ i t is p o s s i b l e t o take l a r g e t i m e s t e p s without l o s i n g accuracy........... (8) where 0 is a c o n s t a n t ..... To o b t a i n a more a c c u r a t e approximation w e wish t o r e p l a c e t h e r e l a t i v e permeability by i t s t i m e averaged value kr ...... is set t o 1 f o r an invading phase and zero for a d i s p l a c e d phase t o ensure t h a t ..... 6 .. k r ( t + A t ) ............ Because t h e c a l c u l a t i o n takes d e t a i l e d account of t h e shape of t h e r e l a t i v e permeability curve... and f(s) is t h e f r a c t i o n a l flow curve f o r t h e phase under consideration... but approaches t h e i m p l i c i t value................P ....... so t h a t both ir and can be computed e x a c t l y a t modest o v e r a l l c o s t .

are - I n some c a s e s . b u t o u r concern h e r e i s s o l e l y w i t h t i m e t r u n c a t i o n e r r o r s . F o r example.. which makes t h e s o l u t i o n much more d i f f i c u l t t o f i n d . T h i s t e c h n i q u e h e l p s t o reduce numerical d i s p e r s i o n . b u t i t a l s o i n t r o d u c e s a d i s c o n t i n u i t y i n t o t h e f u l l y . The convergence rate i s shown . A s a r e s u l t t h e r e i s no d i s c o n t i n u i t y i n t h e t i m e averaged e q u a t i o n (11) Numerical Examples o f T i m e T r u n c a t i o n C o r r e c t i o n The e f f e c t o f t h e T i m e T r u n c a t i o n C o r r e c t i o n (TTC) is i l l u s t r a t e d u s i n g l D . T h i s is because Er i s e v a l u a t e d as a n i n t e g r a l o v e r t h e t i m e s t e p .i m p l i c i t method e v e n t u a l l y converges t o the TTC r e s u l t as t h e t i m e s t e p is r e f i n e d .692 .700 32 . A l l t h e r e s u l t s show c o n s i d e r a b l e numerical d i s p e r s i o n due t o s p a c e t r u n c a t i o n errors. t h i s i t e r a t i o n converges v e r y slowly. 2D and 3D examples.. The a n a l y t i c Buckley-Leverett r e s u l t i s also shown f o r comparison.698 512 .. and as such v a r i e s c o n t i n u o u s l y i n t i m e . The f i r s t i s a s t a n d a r d Buckley-Leverett problem w i t h e q u a l o i l and w a t e r viscosities. The J a c o b i a n m a t r i x .698 256 -695 .398 where t h e s o l u t i o n v a r i a b l e s ( p r e s s u r e and s a t u r a t i o n changes) are r e p r e s e n t e d and r e s i d u a l v e c t o r .687 . (1* . The e f f e c t of t h e t i m e t r u n c a t i o n e r r o r c o r r e c t i o n is similar t o t h a t observed w i t h s i n g l e p o i n t upstream w e i g h t i n g .. F i n a l l y .697 . x.. ax ax AM - F..698 F i g u r e 1 a l s o shows r e s u l t s u s i n g t w o p o i n t upstream ~ e i g h t i n g ' ~ ) . A series of runs showed t h a t t h e t i m e t r u n c a t i o n error i s h a l v e d as t h e number of t i m e s t e p s i s doubled and t h a t t h e s t a n d a r d f u l l y . by x. These problems are much reduced i f t i m e averaged r e l a t i v e p e r m e a b i l i t i e s are used..2].677 .i m p l i c i t two p o i n t upstream w e i g h t i n g r e s u l t i s v i r t u a l l y i d e n t i c a l t o t h e a n a l y t i c result Jz .me km kro s2 + (I-s) .698 64 .in Table 1 f o r c e l l number 12. w e n o t e t h a t i f t h e r e l a t i v e p e r m e a b i l i t y o f water is 1 set t o z e r o below t h e Buckley-Leverett s a t u r a t i o n (--) t h e n t h e f u l l y . TABLE 1 THE WATER SATURATION I N G R I D BLOCK 12. R e s u l t s are d i s p l a y e d f o r one p o i n t upstream w e i g h t i n g u s i n g 32 t i m e s t e p s b o t h w i t h and w i t h o u t t h e t i m e t r u n c a t i o n error c o r r e c t i o n .. e n g i n e e r s f r e q u e n t l y u s e r e l a t i v e p e r m e a b i l i t y c u r v e s f o r t h e i n j e c t e d phase which a r e s e t t o z e r o below t h e Buckley-Leverett s a t u r a t i o n ... The f r a c t i o n a l flow o f water used i s kl-w + f = = 2 .i m p l i c i t e q u a t i o n ( 4 ) . even i f k r ( t + A t ) does n o t . when s t u d y i n g t h e e f f e c t s o f water i n j e c t i o n . RESULTS OBTAINED USING SINGLE POINT UPSTREAM WEIGHTING Number o f t i m e s t e p s Standard r e s u l t TTC r e s u l t 16 -659 ... (13) F i g u r e 1 shows water s a t u r a t i o n d i s t r i b u t i o n s a f t e r twelve c e l l p o r e volumes have been i n j e c t e d .698 128 . At At e v a l u a t e d a t t h e c u r r e n t b e s t estimate of t h e s o l u t i o n .

The second layer is inactive and the reservoir is therefore in two sections which communicate only through the wells. The results indicate that the time averaging technique virtually eliminates time truncation errors.... 1 PT Upstream ITTCI 1 PT UDstream -. Both wells are completed in layers 1. 3-5 and 7-8.c 0 L ... The results Case 3 is a 3 dimensional gas/oil problem described by Odeh(5). Instead.. Time stepping in PORES is not controlled by maximum permitted saturation and pressure changes as in most other simulators. shown in figure 3 again illustrate that time truncation errors are dramatically reduced using time averaged relative permeabilities. Case 2 is a 38 x 8 cross section with a water injector in column 1 and an oil producer in column 38.0 5 ) (c) PORES default TTE controls with time averaged relative permeahilities.-.399 h c 0 ..Analytic Solution c 3 0 l n J1 2 P T Upstream ITTCI 2 P T Upstream .l..3) (b) PORES default TTE controls/lOO ( z ASmax = 0 .. the estimate of time truncation error (equation 5) is converted to a local material balance error by scaling with cell pore volumes and formation volume factors. a fully-implicit black oil simulator described in reference 1. SATURATION PROFILES FOR O N E DIMENSIONAL TEST OF TIME TRUNCATION CORRECTIONS Problems 2 and 3 were run on PORES. Cell Number FIG. The time stepping algorithm attempts to keep the root mean square of this local material balance error within specified bounds. . Further details of this problem are given in reference 1. Thus time stepping is tied directly to the best available estimate of time truncation error. Figure 2 shows the water cut as a function of time for Case 2 using (a) PORES default TTE controls (z ASmax = 0.

6- 0.7 - 0..0.” .Standard TTE Contrds 1100 .. W A T E R CUT AGAINST T I M E FOR TWO D I M E N S I O N A L I X S E C T I O N 1 TEST OF T I M E TRUNCATION CORRECTION . G O R AGAINST T I M E F O R T H R E E D I M E N S I O N A L T E S T OF T I M E T R U N C A T I O N C O R R E C T I O N F I G ....2- ...Standard TTE Controls 1100C TTE Controls + T T I h Y S .TTE Controls t T T C I Standard TTE Contrds I Years F I G 3 .3- 0.5- P L 0.... Standard TTE Controls I I : .-.Standard --... I .. 2 .....4- a 0. ---..Standard..

a s t h e mass accumulation t e r m i n t h e Jacobian matrix (equation (12)) which makes it d i a g o n a l l y dominant (and t h e r e f o r e non-singular) i s i n v e r s e l y As a r e s u l t .. Nested F a c t o r i s a t i o n A l l i t e r a t i v e methods f o r t h e s o l u t i o n of t h e l i n e a r equations Ar = b ... (15) I n t h i s paper. i t i s necessary t o s o l v e t h e l i n e a r i s e d equations (121.53 2.. A t . t o o b t a i n an updated e s t i m a t e of t h e s o l u t i o n t o t h e non-linear equations (11).. This observation i s p a r t i c u l a r l y t r u e of f u l l y ....401 LARGE TIME STEPS AND THE SOLUTION OF THE LINEAR EQUATIONS A t each Newton i t e r a t i o n .. w e p r e s e n t a way of approximating such a matrix by n e s t e d f a c t o r i s a t i o n . A = d + 1 + U1 + E2 + U2 + E3 + U3 ....... I n t h e following s e c t i o n w e d e s c r i b e t h e method used i n PORES....46 4.. Five p o i n t f i n i t e d i f f e r e n c e methods g i v e rise t o t h e n e s t e d block t r i d i a g o n a l s t r u c t u r e shown i n Fig.. i t i s not e a s i l y adapted t o d e a l with g e n e r a l s p a r s e matrices.i m p l i c i t s i m u l a t i o n s with long t i m e s t e p s .... i n g e n e r a l . For small problems. for a f a i r l y e v e n t f u l period i n a t y p i c a l simulation...i m p l i c i t codes e f f i c i e n t . t h e s e equations may be solved by Gaussian e l i m i n a t i o n .. The r a t e of convergence of t h e i t e r a t i o n depends p r i m a r i l y on how w e l l B approximates t o A.. t h e l i n e a r equations a r e proportional t o the t i m e s t e p length.. Because t h e algorithm e x p l o i t s t h e s t r u c t u r e of t h e matrix t o t h e f u l l . depend on t h e s t r u c t u r e of A...31 3.. 4...35 55 64 116 Because t h e l i n e a r equations a r e more d i f f i c u l t t o s o l v e with l a r g e t i m e s t e p s i t is important t o devise powerful i t e r a t i v e methods t o make f u l l y . This e f f e c t i s i l l u s t r a t e d i n Table 2 . The e f f i c i e n c y and robustness of t h e r e s u l t i n g procedure depend c r i t i c a l l y on t h e e f f e c t i v e n e s s of t h e s o l u t i o n method adopted. some form of i t e r a t i v e s o l u t i o n method must be used. a s s t o r a g e and computing t i m e i n c r e a s e very r a p i d l y with t h e problem s i z e . t o t h e c o e f f i c i e n t matrix A such t h a t B-'8 i s e a s i l y c a l c u l a t e d f o r any v e c t o r 8.... . The b e s t choice of B w i l l . Howe v e r t h i s i s not p r a c t i c a b l e f o r l a r g e 3 dimensional s t u d i e s . (14) depend on t h e e x i s t e n c e of an approximation.. I n such c a s e s ... more d i f f i c u l t t o s o l v e i f long time s t e p s a r e taken. Time S t e p s for Tot a1 Simulation Period 12 18 23 48 T o t a l Number of Newton I t e r a t i o n s 67 T o t a l Number of Linear I t e r a t i o n s 366 237 226 273 Linear I t e r a t i o n s Per Newton Iteration 5.... which shows how t h e number of l i n e a r i t e r a t i o n s r e q u i r e d f o r each Newton i t e r a t i o n i n c r e a s e s with t h e t i m e s t e p s i z e . B .

... (18) 8 = (y + E1)y -1 ( y + U1) ................ .................. (16) ..colsum(E 2 8 -1u 2 + E........u1 Bands connect Cells in X Direction 1 2 ........ (17) a = (8 + E2)6-1(8 + uz) ..402 I. .colsum(X) is the diagonal matrix formed by summing t h e elements of X i n COlUmaS... u 3 Bonds connect Cells in Z Direction FIG.L... u2 Bands connect Cells in Y Direction 1 3 . (19) Y d .......Ely -1 u1 ..........................THE STRUCTURE OF THE COEFFICIENT MATRIX FOR A F I V E POINT F I N I T E DIFFERENCE SIMULATION The n e s t e d f a c t o r i s a t i o n approximation for a 3 system may be summarised a s D B = (a + E3)a-l(a + u 31 ......a-lu3) 3 Here..........

. it would be more appropriate to use it as preconditioning for a symmetric conjugate gradient algorithm as described by Meijerink and Van der Vrt'...... + It will be noted from equation ( 2 0 ) that y has been set in such a way that the column sum of the error matrix... E3 = u3 = 0......... .. At each stage.... the equations are solved a line at a time using = B( -B ' - E2p) ........ which in a reservoir simulator corresponds to a total material balance error. The result vector contains the diagonal elements of cols~m(I126-~u~ E3a-12.E2B -1u2 + E 3a -1 u....... ...)B = A -1 ( B + u. applications which give rise to a symmetric coefficient matrix..... B-A..) D o .. Similar considerations apply to the evaluation of y.....3a-lu3) which is required in the calculation of y......... using u = a -1 (8 - P9 ..... and from the previous plane. and the outermost layer of the factorisation (equation (16)) may be omitted.) (24) This is not recursive....Y ul) . as E3p involves the solution from the previous plane.. The evaluation of ~ o l s u m ( E ~ B ....... At the outermost level. is achieved + by multiplying a column vector whose elements are all unity by the transpose of (E26-1~2+ k3a-1243).. these equations can be solved a plane at a time.....403 By combining equations 16-19.. The object of this choice of y is to force the sum of the residuals. Within each plane. I f the iteration is started from the ..u 3 ) 1 - colsum(.... to zero...43)... within each plane.. leaving B = ( B + E.... (23) Because the matrix a is block diagonal....... this approximation is used as a preconditioning matrix for a truncated For conjugate gradient algorithm of the type described by Vinsome(6).~ u 2P.. (21) The method reduces to an exact Cholesky decomposition for one dimensional problems -1 B = (Y + = A (...... The calculation proceeds a plane at a time and.. (22) In PORES... os() The procedure for evaluating B-l8 is hierarchical.... (25) where once again..... the previous line in the current plane.. we obtain the following expression for B B = A + (E28-1u2 + E3a .is zero. which must be done before the iteration begins.. only known solution variables appear on the right-hand side. a (20) For two dimensional systems.. we solve block triangular matrix equations of the form (a + E3)p = B ...the calculation involves elements of y from the previous cell in the current line...) colsum(!L2B -1U2) + E2B -1U 2 - . a line at a time......

ehd('.O. Whilst it is convenient when describing the algorithm to view it as a series of nested LDU factorisations (equations 16-18) a significantly more efficient implementation is achieved by combining the D and U factors... . once established.A) = 0. the evaluation of B ' 8 for an arbitrary nvector 8 requires about 22n floating multiplications for 3D problems (10n for 2D problems). remains valid throughout the iteration('). An important feature of the nested factorisation algorithm is its very modest ' 8 requires storage for only one diagonal storage requirements. O (28) (29) + i2) (I -1 u.. and u2 bands).*. . .. and the smallest on the 13 and 74 bands. . The fact that residuals sum to zero within each plane may also be used as a test on the correctness of the program. B = (a + E3) (I where a = (B + + a u3) ' f3 .. and which led to the recent upsurge of interest in other multi-grid mtos') The method described above avoids the worst of these problems by eliminating the lowest frequency eigenvectors (those with a non-zero residual sum) from the outset.. the residuals sum to zero independently within each plane (or.- D ... Moreover. . appears to be to align the axes so that the largest off. it is low frequency eigenvectors which are most persistent.3 bands.eD. The figure for 2D problems can be reduced to 16n by combining the calculation of B-% with that of AB-10. The total for each preconditioned conjugate gradient iteration is 33n(19n) if A is symmetric. 9 .. O . it was this observation which prompted the additive correction methods of Wts' at() and of Settari and Azi~'~). D . ~. Indeed. within each line) and that this condition also holds throughout the i terat ion ~ It is well known that in most iterative methods for solving linear equations. Algorithms displaying similar properties have been described by Gustafsson(ll) and Cheshire et a l ( ' ) although only the latter noted the importance of starting the iteration from an initial solution with zero residual sum (equation 26). . the next largest on the E2 and 72 1 . this choice is made automatically. Given this form of the algorithm..404 initial solution 5 = B ' b . a - (26) then the initial residual ro is given by r = b .. ..A x and the sum of the components of ro is zero if colsum(B . for 2D systems.. numerical experiments. It is easily shown that this condition. ~.D.) Oe.OO. it may be seen that because the error matrix is block diagonal. deduced from a series of connected by the 11. and which ultimately determine the rate of convergence. e s D . An important consideration in the implementation of this algorithm is the orientation of lines (cells connected by the &1 and u1 bands) and planes (lines The best strategy. In the PORES implementation.. o ... Evaluation of B . . 4 1 diagonal elements are on the 11 and u1 bands. . . or somewhat more for a Vinsome type truncated conjugate gradient algorithm.

which is t h e only one which e x h i b i t s t h e s t r o n g d i r e c t i o n a l i t y On t h i s problem c h a r a c t e r i s t i c of c r o s s s e c t i o n and 3D r e s e r v o i r simulations. ( i t would be p o s s i b l e t o reduce t h i s f i g u r e t o 16n on t h e s e problems). Numerical T e s t s of t h e Nested F a c t o r i s a t i o n Algorithm D The n e s t e d f a c t o r i s a t i o n algorithm was t e s t e d on a series of s i n g l e phase 2 test problems described by S e t t a r i and Aziz('). t h e n e s t e d f a c t o r i s a t i o n algorithm i s f a s t e s t by a f a c t o r of t h r e e . and ICCG3(7). The corresponding f i g u r e s for t h e o t h e r methods a r e 16n f o r ICCGO and ICCGO with colsum. and competes e f f e c t i v e l y with ICCG3 which i s more complex and r e q u i r e s more s t o r a g e . The 2 r e s u l t s discussed above correspond only t o a s i n g l e n e s t i n g (B = a ) and D do not demonstrate t h e f u l l p o t e n t i a l of t h e technique a r i s i n g from t h e second n e s t i n g for 3D problems. The importance of f o r c i n g t h e sum of t h e r e s i d u a l s t o zero by choosing B i n such a way t h a t colsum(B-A) is zero is shown by t h e r e s u l t s from ICCGO and ICCGO (with colsum).l ) i n a d d i t i o n t o t h e elements of A. a u v a r i a n t of ICCGO i n which t h e column s m of t h e e r r o r matrix i s forced t o zerocl). w e have used a symmetric conjugate g r a d i e n t algorithm t o a c c e l e r a t e convergence. Perhaps t h e m o s t s i g n i f i c a n t r e s u l t is t h a t on problem number 2. 22n for ICCC3 and 19n f o r n e s t e d f a c t o r i s a t i o n . The algorithm is implemented i n PORES for 1. required by t h e conjugate g r a d i e n t algorithm. w e have shown t h e computational work required t o reduce t h e l a r g e s t normalised r e s i d u a l t o The u n i t of work is taken a s a SIP i t e r a t i o n (about 22n f l o a t i n g p o i n t m u l t i p l i c a t i o n s ) . t o g e t h e r with r e s u l t s obtained on t h e same problem using SlP(12). Because t h e s e problems give rise t o symmetric c o e f f i c i e n t m a t r i c e s . 2 and 3 phase simultaneous problems and o u r experience t o d a t e on l a r g e complex 3 D s i m u l a t i o n s a r i s i n g from p r a c t i c a l North Sea s t u d i e s h a s been very encouraging. The r e s u l t s a r e shown i n Table 3 . In each c a s e . I C C d 7 ) . . TABLE 3 Some s t o r a g e is a l s o COMPARISON O NESTED FACTORISATION F WITH OTHER ITERATIVE M T O S EH D COMPUTATIONAL WORK (SIP ITERATIONS) Problem Number SIP* ICCGO I CCGO (WITH COLSUM) ICCG3 NESTED FACTORISATION 13 6 13 1 2 3 4 5 28 20 38 28 > 50 50 32 26 36 31 34 27 19 20 22 20 21 17 20 29 22 20 16 12 21 15 6 1 1 The r e s u l t s show t h e nested f a c t o r i s a t i o n method t o be f a s t e s t of t h e methods t e s t e d on a l l t h e problems. The l a t t e r converges s i g n i f i c a n t l y f a s t e r i n every case.405 matrix ( i n general y .

t h e l i n e a r i s e d equations become more d i f f i c u l t t o s o l v e by i t e r a t i v e methods. 4. l i n e a r i s e d f i n i t e d i f f e r e n c e equations. A new technique has been developed t o c o n t r o l t i m e t r u n c a t i o n e r r o r s i n r e s e r v o i r simulation. water o r gas) = average r e l a t i v e permeability over a t i m e s t e p = r e l a t i v e permeability of water = r e l a t i v e permeability of o i l kr kr krw kro II = lower band of t h e Jacobian matrix = mass accumulation v e c t o r . A new n e s t e d f a c t o r i s a t i o n method i s described f o r t h e s o l u t i o n of t h e 2. By comparison with e x i s t i n g published r e s u l t s i t is shown t h a t t h e n e s t e d f a c t o r i s a t i o n method i s highly e f f i c i e n t f o r simple two dimensional problems. each element of d is a 3x3 matrix i n simultaneous 3 phase s i m u l a t i o n s = e s t i m a t e of t i m e t r u n c a t i o n e r r o r v e c t o r E F F f fW = flow v e c t o r . water or gas = f r a c t i o n a l flow of water = r e l a t i v e permeability ( o i l . Because t h e nested f a c t o r i s a t i o n method i s highly r e c u r s i v e i t makes minimal a d d i t i o n a l demands on computer s t o r a g e . a s l a r g e r time s t e p s a r e taken. The nested f a c t o r i s a t i o n method has been incorporated i n PORES and our experience t o d a t e i n d i c a t e s t h a t t h e f u l l power of t h e method i s most apparent on l a r g e d i f f i c u l t t h r e e dimensional s t u d i e s a r i s i n g i n North Sea applications N MNL T R O E CAUE 6. 3. each element r e p r e s e n t s t h e sum of flows i n t o a c e l l from neighbouring c e l l s and wells = flow v e c t o r obtained by r e p l a c i n g a l l r e l a t i v e p e r m e a b i l i t i e s a t t h e end of a t i m e s t e p by t i m e averaged r e l a t i v e p e r m e a b i l i t i e s = f r a c t i o n a l flow of o i l . A B = Jacobian matrix a r i s i n g i n f i n i t e d i f f e r e n c e c a l c u l a t i o n s = an approximation t o t h e Jacobian m a t r i x A Colsum = diagonal matrix formed by summing t h e elements of a matrix i n columns b d = right-hand s i d e of t h e l i n e a r equatian = diagonal elements of A .406 CONCLUSIONS 1. The method can be incorporated i n t o e x i s t i n g s i m u l a t o r s with r e l a t i v e e a s e and has been demonstrated on one. 5. each element of M r e p r e s e n t s t h e mass of a phase i n a c e l l = change i n M over a timestep A t = number of elements i n d M Ah! n r = residual vector . I t i s shown t h a t . two and t h r e e dimensional t e s t problems. 7.

. CROZIER... J. APPLEYARD.K.E. 1981) 33.D. Three Dimensional. and HIRASAKI. HARPER. A. 1980). P. the British National Oil Corporation and the British Gas Corporation for permission to publish this work. VINSOME.J. REFERENCES 1. 515-530 ~ 2.407 S = saturation = change in saturation during a time step = time = duration of a simulation time step = time truncation correction = upper band of the Jacobian matrix = solution of a linear equation.Pet.J. England.A. paper SPE 5729 presented at the SPE-AIME Fourth Symposium on Numerical Simulation of Reservoir Performance.K. E. 3.. G. 1979). Three Phase Reservoir Simulator".. Ca. RUBIN. 1976). Nev. Soc. 4. BEHIE. paper SPE 9302 presented at the SPE-AIME 55th Annual Fall Meeting of the SPE. R. P. ODEH. A. (Jan. and ODEH. Dallas. McDONALD. . M .. "Orthomin. "A Strongly Coupled.S.. 325-336. . R .. 1972).S.M.W-.. D. and VINSOME. BANKS.Eng.M.. (Feb. J. Las Vegas. AU. 1980).K. "An Efficient Fully Implicit Simulator".A. state vector = change of state vector during a Newton iteration = matrix for two dimensional systems = matrix for one dimensional systems = diagonal matrix As t At TTC U X Ax a B Y 6 0 = 0 or 1 corresponding to an invading or displacing phase = constant = pressure dependent component of the flow vector = right-hand side of a linear equation = solution of a linear equation x Q U ACKNOWLEDGEhlENTS The authors wish to thank the United Kingdom Department of Energy. A. O'DELL. and HOLMES.. an Iterative Method for Solving Sparse Banded Sets of Simultaneous Linear Equations". Texas (Sept. 6.. MORELAND... "Techniques for Fully Implicit Reservoir Simulation". paper SPE 8329 presented at the SPE-AIME 54th Annual Fall Meeting of the SPE. TODD. B.W. 13-25. A.. "Comparison of Solutions to a Three-Dimensional Black-Oil Reservoir Simulation Problem". (Sept. "Methods for Increasing Accuracy in Numerical Reservoir Simulators". (Dec.Tech. P.. J. P. paper EUR 179 presented at the European Offshore Petroleum Conference and Exhibition. (Oct. J. Fully Implicit. CHESHIRE.R. BANSAL.Pet. I. London. 5.P.L.J.E. Los Angeles.

SIAM J. SETTARI. 9. J. "An Iterative Matrix Solution Method Suitable for Anisotropic Problems". 1968) 5 . I. 1977) 2. BRANDT. 10.7 . Mathematics of Computation (April 1977) 3l. (March 1971) 11. WATTS.47-51.Ana1..Pet.A. . 12. 142-156.. 11.Numer. 333-390. (Sept. "An Iterative Solution Method for Linear Systems of which the Coefficient Matrix is a Symmetric MMatrix". and AZIZ.J.. H.Eng.Numer. "Iterative Solution of Implicit Approximations of Multidimensional Partial Differential Equations".. H. and VAN DER VORST. J.W. GUSTAFSSON. "A Generalisation of the Additive Correction Methods for the Iterative Solution of Matrix Equations". A. (June 1973) 2.. SIAM J.L.A. "Multi-Level Adaptive Solutions to Boundary-Value Problems".Ana1. BIT (April 1978) 18. 8 . 148-162. 530-558. S0c. "A Class of First Order Factorisation Methods".Mathematics of Computation (Jan. A. STONE. 506-521. MEIJERINK. K O .

A further improvement in the efficiency with which the CFTE code can be applied has been achieved by incorporating a line successive overrelaxation iterative solver (LSOR) into the code. These LSOR options run about twice as quickly as the direct inversion on the IBM scalar machine and.NUMERICAL METHODS 409 SOME CONSIDERATIONS CONCERNING THE EFFICIENCY OF CHEMICAL FLOOD SIMULATIONS R. INTRODUCTION The Chemical Flood Ternary Equilibrium Simulator (CFTE Ref I ) is one of several EOR codes being used at AEE Winfrith. Initially written for a scalar machine. as an alternative to the direct inversion method. This mesh generation code has also proved useful in visualising flow patterns and consequent mesh requirements for studying possible pilot applications of surfactant flood under real reservoir conditions. This paper briefly presents some of the experience associated with the use of a special mesh generation code. they run up to 6 5 times faster than the original code on the IBM. has enabled suitable grid patterns to be generated for use in the CFTE code. During the period for which CFTE has been available. the work of Martin and Wagner supports the application of conformal grids. For the low tension. . Dorset. W. DT2 8DH ABSTMCT This paper discusses some aspects of improving computational efficiency and accuracy in surfactant flood calculations. FOULSER AEE Winfn’th. its application on the CRAY vector machine has led to the development of alternative LSOR codes using different grid block orderings.S. it has become necessary to develop supplementary programs to aid the generation of input for the studies. when vectorised on the CRAY. which applies the results of potential theory. As more difficult assessment calculations have been undertaken.Dorchester. The development at Winfrith of the PASL code. that has been written as one of the support programs. PASL. low concentration surfactant processes being considered. and also the steps that have been taken to improve the efficiency of CFTE when used on the CRAY vector processor. Computational time is also becoming a significant factor in the use of the CFTE code in large assessment calculations. Thus the vectorised LSOR approach has proved to be very powerful. one of which has been reported elsewhere by Fayers et a1 (Ref 2). The use of curvilinear grids in chemical flooding simulation can reduce grid orientation effects and speed up the calculations. a number of assessments studies have been performed.

4 10 The PASL mesh generation code has been used as a tool in preliminary studies to determine broad characteristics of flow patterns within possible pilot areas of an oil field. Sone investigations of this type of surfactant flood for North Sea applications have been discussed in Ref 2. it was decided to incorporate an iterative solver into the code. No-flow boundaries associated with linear fault lines have also been included in the formulation which uses the basic results of potential theory. The preliminary flood pattern studies provide a basis for identifying computing mesh requirements in subsequent CFTE calculations. The recent availability of a CRAY-I vector processor in association with the IBM 3033 at Harwell. the matrix inversion time begins to dominate the total execution time. The basic . such as SIP. One of the first examples studied using a grid generated bv PASL. Once the streamlines and equipotentials have been computed. For a system of wells the velocity potential and stream function are given by: where mi is the strength of well i. The code provides output of curvilinear mesh dimensions in a format suitable for direct input to CFTE. was Concerned with the problem of mesh orientation effects during the calculation of a simple surfactant flood in a 5 s p o t geometry with a uniform residual oil saturation. Because of this. The LSOR method was chosen.to obtain a continuous line fault and the method of images is then used to provide an equivalent system of wells in an infinite domain. The CFTE code was used to model a low concentration surfactant flood in a 1/8 th symmetry sector of the 5-spot pattern. using parallel and diagonal Cartesian grids with curvilinear grids. and the resulting implicit pressure matrix problem was originally solved by direct methods. The PASL code may also be used to generate curvilinear co-ordinate systems for use in CFTE. THE DETERMINATION AND USE OF CURVILINEAR GRIDS IN SURFACTANT FLOODING STUDIES The generation of curvilinear grids and the study of flow fields for arbitrary arrangements of injection and production wells has been implemented in the PASL code. The use of curvilinear co-ordinates in surfactant flood calculations allows a considerable improvement to be obtained in calculation efficiency as well as a reduction in the errors caused by grid These improvements are quantified by comparing the results orientation effects. The faults are dealt with by applying conformal transformations. but also because this iterative method is more readily vectorised than some of the more complicated strategies. The solution method then uses the classical equations for representing a distribution of wells in ) infinite space (Ref 3 . This has been especially useful in examining alternative well patterns in a pilot flood and also as a means of studying pattern confinement strategies. The CFTE code uses the IMPES formulation. where the computing for this project is undertaken. partly because of its robustness and its relatively simple form for progranrming. emphasised the need to choose a method which could be readily vectorised. The second part of this paper is concerned with an investigation of the performance of the LSOR method in chemical flooding calculations and also with the aspects of vectorisation of this particular method. With about 400 grid blocks in three-dimensional calculations. inverse conformal transformations are used to map the results back to the original co-ordinate system.

411 d a t a adopted a r e s i m i l a r t o those described i n t h a t reference. t h e r e l a t i v e permeability of t h e o i l and water phases move progressively from t h e i r i n i t i a l o i l .OS and behind t h e bank t h e m o b i l i t y of t h e s u r f a c t a n t s o l u t i o n i s somewhat g r e a t e r than 5. A t t h i s s t a g e . A l l t h r e e c a l c u l a t i o n s show low concentrations around t h e i n j e c t i o n w e l l w i t h a f a i r l y s t e e p rise i n o i l s a t u r a t i o n i n t o t h e o i l bank where t h e o i l s a t u r a t i o n reaches about 40%.w a t e r to their limiting straight l i n e forms a t low c a p i l l a r y numbers (26 x forms as the c a p i l l a r y number i n c r e a s e s t o 2 x The r e s u l t s of the c a l c u l a t i o n s a r e shown i n Figure 1 where the +spot region has been covered by 90 g r i d blocks f o r t h e p a r a l l e l and c u r v i l i n e a r g r i d systems and 91 blocks f o r t h e diagonal g r i d system. These diagrams show t h e o i l d i s t r i b u t i o n a f t e r 0. The cumulative o i l production curves. i t has only been p o s s i b l e t o study problems f o r s p e c i a l l y chosen s e c t o r s . From a computational p o i n t of view. It i s these connecting g r i d blocks which tend t o c o n t r o l t h e timestep s i z e t h a t can be taken i n the e x p l i c i t s o l u t i o n of t h e c o n c e n t r a t i o n and s a t u r a t i o n equations. The s u i t a b i l i t y of c u r v i l i n e a r g r i d methods f o r more g e n e r a l a p p l i c a t i o n i s a l s o being s t u d i e d . t h e p a r a l l e l g r i d shows a p r e f e r e n t i a l flow along t h e d i r e c t path between t h e w e l l s .25 mD/cp. This i s an example of a f i e l d problem f o r which s e l e c t i o n of a s a t i s f a c t o r y Cartesian mesh would have been very d i f f i c u l t . running some 5 times f a s t e r than t h e o t h e r s . A v a r i a t i o n of r e l a t i v e permeability w i t h c a p i l l a r y number similar t o t h a t used by Todd and Chase (Ref 1) has been assumed where. thus supporting use o f c u r v i l i n e a r two-dimensional g r i d s i n CFTE c a l c u l a t i o n s with t h i s type of system. Thus t h e e q u a l l y mobile two phase mixture i n t h e o i l bank and the watered o u t zone ahead are being d i s p l a c e d by a much more mobile f l u i d . For maximum e f f i c i e n c y i n t h e CFTE code t h e e q u i p o t e n t i a l s are u s u a l l y chosen t o produce equal volume g r i d blocks along t h e s h o r t e s t stream tube. The i n t e r s e c t i o n of s t r e a m l i n e s and e q u i p o t e n t i a l s have been used i n CFTE c a l c u l a t i o n s as t h e g r i d block c o r n e r s of a conformal c u r v i l i n e a r co-ordinate system. f r o n of t h e o i l bank h a s a I n t h e p a r t i c u l a r example s t u d i e d the water m o b i l i t y of 1. t h e c u r v i l i n e a r c a l c u l a t i o n was by f a r t h e most e f f i c i e n t . (4b) + . The shaded region i n t h e Western s e c t o r of t h i s f i g u r e has been used i n t h i s way t o generate co-ordinate s y s t e m f o r t h e problem on which CFTE performance estimates are reported i n t h e second h a l f of t h i s paper. a n example of t h e unfavourable mobility r a t i o e f f e c t s i n s u r f a c t a n t flooding. n o t Shawn. Martin and Wagner (Ref 4) suggest t h a t t h i s type of problem i s amenable t o f i x e d stream tube methods. Note t h a t t h e f a u l t s a r e s t r e a m l i n e s of t h e flow p a t t e r n and t h a t t h e e q u i p o t e n t i a l s c r o s s t h e steamlines perpendicularly. This occurred because the connections with t h e wells i n t h e c u r v i l i n e a r c a l c u l a t i o n s was v i a 5 f a i r l y l a r g e g r i d b l o c k s . The p r e d i c t i o n of t h e c u r v i l i n e a r g r i d s i m u l a t i o n l i e s between t h e s e two r e c t a n g u l a r g r i d r e s u l t s . b u t i n t h e v i c i n i t y of two i n t e r s e c t i n g impermeable f a u l t s . r a t h e r than v i a t h e s i n g l e smaller g r i d block i n t h e Cartesian systems. Figure 2 shows flow p a t t e r n s generated from t h e PASL code. when a p p l i e d t o an i n v e r t e d 5-spot operated i n i s o l a t i o n . A l l t h r e e c a l c u l a t i o n s a l s o show t h e o i l bank pinching o f f a small region where t h e s a t u r a t i o n has not y e t r i s e n above t h a t l e f t a f t e r waterflooding. i n t h i s case. t h e t o t a l m o b i l i t y + i n t h e o i l bank i s 1 . while t h e diagonal g r i d allows t h e less a c c e s s i b l e corner region t o be swept more e a s i l y . G e n e r a l i s a t i o n of t h e mesh c o n n e c t i v i t y arrangements w i l l be s t u d i e d i n t h e f u t u r e i n connection w i t h t h e LSOR method discussed i n t h e next s e c t i o n . A s would be expected.5 PV of s u r f a c t a n t s o l u t i o n i n j e c t i o n which i s not long a f t e r breakthrough of the o i l bank. This i s . a l s o i n d i c a t e s t h a t t h e c u r v i l i n e a f g r i d gives r e s u l t s i n t e r m e d i a t e t o t h e C a r t e s i a n mesh arrangements. because of l i m i t a t i o n s i n t h e c o n n e c t i v i t y of mesh which can be handled by t h e CFTE code. t h e r e f o r e .

.I COMPARISON OF LW TENSION SURFACTANT FLOOO COMPUTED a L MSTRlBUTlONS AFTER 0.5 PV OF FLUB INJECTION.- -1 -- DIAMNAL GRID FIG.4 12 PARALLEL GRID .

and its simplicity for vectorisation on a CRAY were further reasons for choosing this method.4 13 quipatmtid rrtnomlir lnpcrmeoblc foul1 FIG. the blocks are chosen to correspond to lines of grid blocks.2 EXAMPLE OF APPLICATION OF POTENTIAL FLOW THEORY TO ROW FAlYDN VISUALISATION IN AN INVERTED 5-SPOT CONFINED BY INTERSECTING FWLTS. K (3) i where xi is an n-tuple of variables (the block sire) and Aki and Q are the correspodding sub-matrix of co-efficient8 and the n-tuple right-hand side. The robustness of the convergence behaviour of LSOR. An iterative sequence of vectors sf: may then be defined - . it was decided to investigate the application of the LSOR iterative procedures as analternative to the existing direct inversion method in the CFTE code. In line successive over relaxation. and the practical difficulties imposed by generalised connectivities between meshes introduced by such schemes. Outline of LSOR In block successiveaver relaxation the system of equations to be solved can be written in the form: C h i y = q k I. its ease of programning.. . THE PERFORMANCE OF SCALAR AND VECTOR VERSIONS OF THE LSOR METHOD IN CHEMICAL FLOOD CALCULATIONS In view of the need to speed up the inversion process for large problems with curvilinear meshes...

t h e n w i s r e l a t e d t o the s p e c t r a l r a d i u s !J of the matrix (Aii)A-I by: Since the matrix (&d)A-I i s dependent only on t h e o r i g i n a l matrix and on the block s e l e c t i o n ( i e t h e d i r e c t i o n of the l i n e s i n LSOR). provided the blocks hi are ordered c o n s i s t e n t l y . w l i s t h e asymptotic r a t i o of successive i t e r a t e s of t h e eigenvector elements (p287 Ref 6 ) .where B is a relaxation factor.I The e s s e n t i a l p o i n t s of LSOR a r e thus: (i) To choose t h e d i r e c t i o n of t h e l i n e s ( i e the Akk) 80 a s t o maximke the rate of convergence given by equation (9). For a seven p o i n t approximation t o the three-dimensional Laplacian o p e r a t o r i t can be s h m n t h a t . . This may be w r i t t e n i n matrix form: The convergence of the i t e r a t i v e scheme describe.1 so t h a t equation (7) may be r e w r i t t e n as: This shows t h a t the optimum r e l a x a t i o n f a c t o r can be determined by powering the i t e r a t i o n m a t r i x E-’F assuming B = 1 . The s p e c t r a l r a d i u s of t h e i t e r a t i o n matrix with t h e optimal r e l a x a t i o n f a c t o r can be shown t o be w = 6-1 and t h i s g i v e s a value f o r the asymptotic r a t e of convergence of: I = L -log (. !J depends only on the block arrangement of t h e matrix and on t h e r e l a x a t i o n f a c t o r 6. In r e s e r v o i r s i t u a t i o n s t h i s almost i n v a r i a b l y means choosing the v e r t i c a l d i r ect i o n f o r the l i n e s. As i n p o i n t successive r e l a x a t i o n t h e optimal r e l a x a t i o n f a c t o r B can b e shown t o be given by: By p u t t i n g BPI i n equation ( 6 ) . i t may be seen t h a t p2 = w.by equation (5) depends on the s p e c t r a l r a d i u s w of t h e i t e r a t i o n m a t r i x E-’F (Ref 5 ) .

001. The t o t a l s c r a t c h s t o r a g e space needed by the s c a l a r LSOR code i s compared i n Figure 4 with t h e s t o r a g e r e q u i r e d by t h e D4 d i r e c t e l i m i n a t i o n code o r i g i n a l l y i n CFTE. and a f t e r t h i s t h e previous i t e r a t e of t h e pressure d i s t r i b u t i o n i s s t b r e d during the pressure solution. S c a l a r Code E i t h e r t h e x. This ordering i s c o n s i s t e n t . A necessary and s u f f i c i e n t condition f o r an o r d e r i n g t o b e c o n s i s t e n t i s given on page 245 of Ref 7. (iii) (i v ) The convergence t e s t f o r the p r e s s u r e d i s t r i b u t i o n t h a t has been used i n t h i s application is: (10) and E has been taken t o be 0. t h i s i n f a c t r e p r e s e n t s t h e c u r v i l i n e a r co-ordinate system f o r t h e c u r v i l i n e a r s t r u c t u r e shown i n Figure 2. i e w i s close t o u n i t y . The eigenvector a r r a y r e p r e s e n t s about h a l f t h e s c r a t c h s t o r a g e requirement. To use equation (5) t o i t e r a t e on the unknown v e c t o r zn+lu n t i l convergence i s achieved. Although the g r i d block arrangement i s displayed a s a square. t h e l i n e s of g r i d blocks are ordered i n a n a t u r a l o r d e r i n the h o r i z o n t a l plane. This s t e p i n the procedure can be q u i t e c o s t l y unless a l l t h e eigenvector i s s t o r e d f o r use as a n i n i t i a l guess a t subsequent t i m e s t e p s . Checks of the m a t e r i a l balance i n chemical flooding c a l c u l a t i o n s have shown t h i s t o be an adequate test. However.415 (ii) To choose an o r d e r f o r r e v i s i n g the a t h a t i s c o n s i s t e n t . The o t h e r h a l f of t h e s c r a t c h s t o r a g e i s used t o s t o r e t h e previous eigenvector i t e r a t e during the s p e c t r a l norm c a l c u l a t i o n . A n o r d e r i n g of a 7 x 7 arrangement of g r i d block l i n e s i s shown i n Figure 3. This t e s t ensures t h a t successive i t e r a t e s must be more t i g h t l y converged when t h e convergence i s slow. To power t h e m a t r i x E-lF with f b 1 so a s t o determine w and then use equation (8) t o determine t h e optimum r e l a x a t i o n f a c t o r . y o r z d i r e c t i o n may be chosen f o r the l i n e d i r e c t i o n i n a l l t h r e e v a r i a n t s of t h e LSOR code discussed i n t h i s paper. . Usually the eigenvector changes r e l a t i v e l y slowly d u r i n g a displacement c a l c u l a t i o n and only a few a d d i t i o n a l i t e r a t i o n s are needed a t each timestep. This involves t h e i d e n t i f i c a t i o n of the eigenvector corresponding t o the maximum eigenvalue of t h e i t e r a t i o n matrix. I n the s c a l a r version of t h e code. f o r convenience i t i s r e f e r r e d t o as t h e v e r t i c a l d i r e c t i o n . where t h e wells a r e connected t o the g r i d blocks along the l e f t a n d r i g h t e d g e s . s t a r t i n g i n one c o m e r of t h e a r e a modelled and working up t o t h e o p p o s i t e corner. I n o r d e r t o determine the i t e r a t i o n matrix eigenvalue the matrix i s powered. The p r e s s u r e d i s t r i b u t i o n i s found using equations (5) and ( 8 ) and the convergence c r i t e r i o n i n equation (10). The advantage of i t e r a t i v e schemes i n t h i s r e s p e c t i s w e l l known.

3 GRID BLOCK ORDERIN6 SCHEMES USED TO INVESTIGATE LSOR PERFORMANCE.1 43 7 l " 1 " 2 " 3 I 4 5 6 2 1 2 Tt. 11 12 13 M 2 3 4 5 6 7 1 1 NOLEN'S RED-BLAtK OROERNG USED FOR VECTOR PROGRAM DIAGONAL GRoupHt USB) FOR VBCTOR PROGRAM PRINCIPLE OF ORMRlNG S O ABOVE HAS VETOR LENCTH OF 7 HW DEVELOPMENT S E W HAS VECTOR LENGTH OF 21 FIG. .t.

The canputing time taken t o i n v e r t t h e pressure matrix using LSOR v a r i e s with time according t o t h e d i f f i c u l t y of the problem ( i e i n i t i a l guess and the matrix s p e c t r a l r a d i u s ) . Figure 5 shows t h e t i m e taken f o r the f i r s t 150 timesteps of the sample problem.4 COMPARISON OF SCRATCH STORAGE REQUIREHENT FOR LsoA MATRIX INVERSKW STRATEGIES J x J x 5 PROBLEM.3.3.and t o introduce some a x i a l v a r i a t i o n s .10. i n v e r s i o n r o u t i n e a v a i l a b l e i n the CFTE . Also shown f o r comparison purposes is t h e time taken by the d i r e c t E4 m a t r i x . the p e r m e a b i l i t i e s of t h e 5 + e r t i c a l l a y e r s have been varied with values of 10. This influences t h e s p e c t r a l norm of the i t e r a t i o n matrix.10 mD r e s p e c t i v e l y .Words 10 1 C / I I Oirrt l t i n D mno . 10~10x5(500 g r i d blocks) and 14x14~5(980 g r i d blocks). Pressure constrained w e l l models were assumed i n which 15% of the t o t a l pressure drop a t t h e i n i t i a l conditions was l o s t i n t h e w e l l completion f a c t o r . The same s u r f a c t a n t system as described i n the s e c t i o n on c u r v i l i n e a r g r i d s has been used. A number of t e s t s have been made using t h i s version of the code. The n m b e r of g r i d blocks varying from 7 x 7 ~ 5(245 g r i d blocks). Ordwig LSOR 1 S lo 6 J-Dimcnrion FIG.

-.-.. .-:& -...... -.-.... .. .- . :a Gaussian etiminatim with 01.5 EXAMPLES O TIME SPENT IN MATRIX INVERSION USING SCALAR VERSION F OF LSOR Co[E AND CURVILINEAR GRlO BUXK REPRESENTATION OF WESTERN QUADRANT OF INVERTED 5-SPOT SHOWN IN FIG..4 18 0.. * . . -. ordering - 7 x 7 x 5 GRIDBLOCKS I .2 .. ..# lOxlDx5 G R D W C K S Y t r l C x S GRIDBWCKS RG. --.. .2 6 . r. * c- . ...

even f o r the smallest problem (245 g r i d blocks). The i n c r e a s e i n i t e r a t i o n matrix s p e c t r a l r a d i u s .674 0. however. The o v e r a l l p a t t e r n of behaviour i n these c a l c u l a t i o n s . The i n c r e a s e i n i n v e r s i o n t i m e a t 0. V e c t o r i s a t i o n of LSOR f o r CRAY . which more than halved t h e running t i m e r e l a t i v e t o code generated by a non-op t imi sing compiler . with i n c r e a s i n g problem s i z e i s w e l l known.30 (9) Typical average convergence rate 0.23 0. This occurs because of t h e r e l a t i v e l y rapid pressure changes t h a t occurred making the eigenvector and pressure d i s t r i b u t i o n s a t t h e previous timestep less good as i n i t i a l guesses. This i s less than the f a c t o r of four expected on t h e b a s i s of the r e l a t i v e clock times of t h e machines. This i s summarised i n Table I be low : TABLE 1 COMPARISON OF THEORETICAL AND CALCULATIONAL CONVERGENCE RATES Mesh s i z e I t e r a t i o n matrix s p e c t r a l r a d i u s Asymptotic convergence rate 7X7X5 IOxlOx5 0 796 0. Over t h e timescales represented. LSOR i s more e f f i c i e n t than the d i r e c t method.854 0.39 0. In micellarfpolymer floods t h e most pronounced peaking has been noted when a highly viscous micro-emulsion breaks through i n t o the producerwell. a l s o of i n t e r e s t i s the f a c t t h a t t h e w e l l f a c t o r s can considerably i n f l u e n c e t h e convergence p r o p e r t i e s of t h e i t e r a t i o n scheme. i s t h a t of a decrease i n i n v e r s i o n t i m e from the i n i t i a l conditions. .2 PV of f l u i d i n j e c t i o n and then a gradual decrease. followed by an i n c r e a s e a t about 0. I8 14X14X5 0.4 19 code. A t long timescales the c a l c u l a t i o n a l time i n LSOR becomes i n s i g n i f i c a n t as the pressure d i s t r i b u t i o n changes very l i t t l e during t h e timesteps. I t i s believed t h a t t h i s r e s u l t s from t h e optimising c a p a b i l i t y of t h e IBM compiler used. It i s of i n t e r e s t t o compare t h e achieved convergence rates with the asymptotic rates given by equation (9).16 0. t h e average convergence r a t e achieved i s somewhat slower than t h e asymptotic rate. and thus t h e decrease i n convergence rates. and thus i n c r e a s i n g t h e coupling between t h e c a l c u l a t i o n a l mesh and the f i x e d well bore pressures.2 PV co-incides with t h e breakthrough of the o i l bank a t t h e producer w e l l . most c l e a r l y seen i n t h e l a r g e s t problem. as would be expected.Algorithm Performance The s c a l a r v e r s i o n of LSOR described i n t h e previous s e c t i o n w a s t r a n s f e r r e d t o t h e CRAY without modification and some m a t r i x i n v e r s i o n timings were made t o compare the CRAY and IBM performance. Making t h e w e l l f a c t o r s very l a r g e . The CRAY c a l c u l a t i o n s without v e c t o r i s a t i o n were about twice as f a s t as the IM equivalent B c a l c u l a t i o n s . speeded up convergence by 20% i n t h e examples discussed above.12 A s may be seen.

This was prompted by the observation that the red-black scheme resembles a two-step line over relaxed Jacobi scheme for the initial iterations. when there is an odd number of lines.NY-I)/2 for the red and black groups respectively. Killough has described an attempt to vectorise the SIP algorithm (Ref 9 . The choice of lines which are solved simultaneously is important for two reasons. Nolen (Ref 8) has identified one possible ordering scheme in which the lines are ordered with a chequer board arrangement. unfortunately the vectorisation method ) degraded the convergence of the algorithm so as to outweigh the gain in speed due to vectorisation. The alternative scheme devised corresponds to . The concept has been generalised slightly fromthat reported by Nolen to allow an odd number of grid block lines to be used. the red lines are solved first and then the black lines. The generalisation to an odd number of grid block lines in the plane makes the arithmetic expressions for referencing neighbouring grid blocks more complicated than in the less general situation. the choice must lead to a grid block referencing systemwhichallows the gridblocks in a line. However. and therefore the average convergence properties might be less than for other possible ordering schemes.NY/2 for problems with an even number of grid blocks. An example of the line ordering is shown in Figure 3. The required condition is that any chosen ordering must be consistent. Following Buzbee et a1 (Ref 8) the method employed to overcome this has been to solve for a number of lines simultaneously. An alternative to the red-black ordering scheme has also been investigated. for the preferred curvilinear co-ordinate method the inactive grid block facility would not normally be required. This was due to three causes: (i) (ii) (iii) the use of conditional testing to eliminate inactive grid blocks. The attraction of the LSOR method is that the lines for simultaneous solution can be chosen so that the asymptotic convergence behaviour is preserved. the extensive use of indirect addressing to reference neighbouring grid blocks.NY+I)/Z and (NX. For Cartesian grid approximations to flow patterns in repeated patterns.420 It should also be noted that very little of the original LSOR coding could be vectorised automatically. Firstly. the recursive nature of the Thomas' algorithm which is the kernel sub-program in the LSOR method. Most of the conditional testing and indirect addressing could be removed by suppressing the elimination of inactive grid blocks. and thegridblocks in neighbouring lines. as well as even. and (NX. to be easily referenced. Each line of grid blocks gives rise to a tridiagonal system of operations which cannot be vectorised by the CRAY due to the recursions that occur in performing the forward eliminations and back substitutions. Secondly. The vector length for the simultaneous tridiagonal solver is NX. the removal of this facility incurs a significant work penalty since unnecessary matrix solutions are undertaken for grid blocks which are isolated from the active region by imposing zero transmissibilities. the ordering of the grid blocks must not degrade the basic solution strategy of the algorithm.

but this has not been investigated. allows neighbouring grid blocks to be referenced by simple arithmetic expressions. and the order in which the grid block line problems are solved in this strategy.38 0. The basic arrangement is illustrated in the top middle diagram of Figure 3 This scheme . On the CRAY the vector performance becomes awre advantageous the longer the vector length.0N grid points along the right boundary are directly coupled to the well on the right in this iteration.14 . The lower diagram in Figure 3 indicates one of these combinations. Thus. the influence of the well connected to the left hand edge of the grid pattern is transmitted in the first iteration to the lower triangular zone of 0. so one pass is equivalent to three iterations. To understand the possible differences in initial convergence behaviour further.22 0. The ordering remains consistent despite these combinations.18 0. Thus for slowly convergent problems the schemes should behave identically.12 I). Each diagonal is considered in turn until all the grid block eigenvector elements or pressures have been revised.5N grid points are coupled to wells on each side in the red sweep (for N even). it is necessary only to consider curvilinear calculations with a symmetric arrangement of NxN grid blocks in the horizontal plane. The average rates of convergence achieved by the alternative strategies are s h a m in Table 2: TABLE 2 IH COMPARISON OF CONVERGENCE RATES WT VARIOUS ORDERINGS Mesh size Standard ordering Red-black ordering Diagonal line ordering 7X7X5 0.37 1 ox 10x5 14X14X5 0. and I. Thus the diagonal arrangement can on average transmit boundary effects across the area faster. since all three are consistently ordered. but only the line of 1. which for small problems may not be sufficient to take full advantage of the vector machine. In the particular case shown each pass of the program revised each line three times.16 0. for small problems. With the diagonal grouping.ONgrid points in the black sweep. but is asymmetric in its behaviour.421 grouping the lines in diagonals so as to obtain a number of line problems that can be solved simultaneously. The asymptotic convergence rates of the two schemes must be identical with the standard ordering. the diagonals have been combined to increase the vector length. Symmetry could be introduced in the diagonal scheme by reversing the sweep order on successive iterations.5 (N2+N) mesh points. These calculations correspond to the initial steps in a chemical flood simulation and as such represent the calculation of an almost symmetric problem. The diagonal grouping leads to vector lengths of NY for the tridiagonal solver. The calculations already reported for the scalar code have been repeated using code modifications employing the red-black and diagonal grouping ordering.22 0. In the red-black arrangement. 0.30 0. subject to the limitation of filling the 64 element registers.

with the resulting storage requirements shown in Figure 4. and thus the convergence rates are identical. TABLE 3 EXAMPLE VECTOR-LENGTHS ACHIEVED WITH DIFFERENT ORDERING SCHEMES Operation Red-Black Ordering 980 Diagonal Grouping 980 70 70 Updating previous iterate Setting up right-hand sides for Thomas' algorithm Setting up left-hand sides Thomas' Algorithm Updating solution (relaxation step) Convergence test 5 490 98 490 980 56 70 980 Table 4 gives the CRAY cpu time required by the various vectorised options to perform one iteration in either the eigenvalue calculation or the calculation of the pressure distribution. the implementation of the two alternative ordering schemes entailed completely rewriting the routines concerned with powering of the iteration matrix. so the convergence in this case is slightly slower. and the asymmetric propagation is more marked. Initial attempts to vectorise the above algorithms involved the extensive use of additional scratch storage associated with use of the CRAY SCILIB routines GATHER and SCATTER (Ref 10). the red-black and diagonal group orderings cu-kcidedue to the combining of groups to increase the vector length. Most of the execution time is expended in the iterating routines. four diagonals are grouped together. and iterating the pressure distribution. In the 14x14~5case. and this explains why this scheme has the poorest average convergence rate. has also been completely eliminated from the routines. The diagonal grouping in the 7 x 7 ~ 5 problem leads to a slight asymmetry in the initial revision pattern. Vectorisation of LSOR for CRAY - Coding Performance While the overall strategy in the vectorised scheme is the same as in the original code. .422 For some small problems. such as the 10x10~5calculation. except when testing for convergence. thus the reduced convergence compared with the red-black scheme. Table 3 indicates the vector lengths achieved in the major sections of these routines when applied to the 14x14~5sample problem. In all cases the standard ordering leads to an initial asymmetric behaviour even more pronounced than the diagonal grouping. However. an inefficient computing task. program refinements eliminated the need for these routines and the additional scratch storage. Conditional testing.

This has been demonstrated by using the same inversion routines on the IBM machine.69 0.49 14. By reverting to the restricted case where the number of grid blocks in the plane is known to be even.3 were observed.04 3. it may be possible to make the two vectorised options comparable. Direct inversion employed in the solution of the implicit formulation of the pressure equation in the CFTE program leads to computer speed limitations for large curvilinear mesh problems.77 1. The last two arrangements have been shown to be amenable to selection of long vector lengths on a CRAY computer so that diagonal line formulation in the revised code runs up to 19 times faster than the standard ordering in LSOR. The reduction in mesh orientation errors resulting fromthe use of curvilinear co-ordinates has been demonstrated for a 5-spot pattern with surfactant flooding.88 0. CONCLUSIONS This paper has discussed advantages which can be derived in utilising curvilinear mesh co-ordinate systems in surfactant flood calculations.47 0.51 As may be seen the red-black coding is 4 or 5 times faster than the original code and the diagonal grouping is 10 to 16 times faster. The curvilinear geometry gives advantages in choice of mesh blocks adjacent to wells.90 2. This allows a computational geometry to be adopted broadly consistent with the anticipated flow patterns of a problem. To overcome this.78 2.34 I .73 0. where enhancements by factors of 1.423 TABLE 4 CRAY CPU TIME IN MILLI-SECONDS TO EXECUTE ONE ITERATION Eigenvalue calculation Mesh size Pressure distribution 7 x 7 ~ 5 10x10~5 14x14~5 7x7~5 10x10~5 14~14x5 Standard ordering Red-Black ordering Diagonal grouping 3. the LSOR method has been programed and tested using the standard ordering.35 0. such as that adopted in the CFTE code for simulation of surfactant floods. A small part of the gains achieved above are associated with simply reducing the generality of the inversion programming and the generation of better fortran coding.2 and 1 .85 08 . and also the shorter vector length associated with setting up the right-hand side column vector. Relative to D4-direct inversion on the IBM 3033 . The difference between the last two options is almost certainly due to the slightly more complex arithmetic needed in the red-black code to identify the neighbouring grid blocks. as well as a red-black and a diagonal line consistent ordering.9 0.67 0.51 11. which in turn gives a superior time step capability in an IMPES formulation. Implementation of flow stream geometry utilising the code PASL has been illustrated for a generalised field problem with sealing fault lines.

Elsevier. presented at the Fifth Symposium on Reservoir Simulation. 1960. presented at the Fifth Symposium on Reservoir Simulation. Paper SPE 7673. VARGA R . and IiASOW W .R. REFERENCES 1 TODD M. BOLEY D. ACKNOWLEDGEMENT The work reported i n t h i s paper has been funded by the UK Department of Energy. . 1937. BUZBEE B L . Prentice Hall. J.. presented a& the Fifth Symposium on Reservoir Simulation. and WAGNER R. MUSKAT M. Thus these improvements have placed field computation closer to practical realisation in terms of computer costs. "Some Aspects of the Potential Application of EOR Processes in North Sea Reservoirs". CRAY-I Library Reference Manual. Pet. Denver 1979. The total code performance is dependent on problem size and for the largest example discussed here of 980 mesh blocks. "Fundamentals of Numerical Reservoir Simulation". "Applications of Block Relaxation" . "Matrix Iterative Analysis". . McGraw Hill. "Flow of Homogeneous Fluids Through Porous Media". 6 7 8 9 10 SR-0014. 1980.I. and MATTHEWS J .E. .E. W T I N J. Paper EUR 194.C..W. presented at the European Offshore Petroleum Conference and Annual Exhibition. SOC. needs to be considered in the future . and CHASE C. Eng.. 1962. (October 1979) pp313-323. 2 5 PEACEMAN D. John Wiley. 1977. FORSYTHE G. and PARTERS. Denver 1979.424 machine. "Numerical Solution of Multiphase TwoDimensional Incompressible Flow Using Stream Tube Relationships".E.A. KILLOUGH J. Much larger problems are needed for real field studies where the inversion aspect becomes completely dominant. D . direct inversion required about 60% of the overall running time. "Finite Difference Methods for Partial Differential Equations". "A Numerical Simulator for Predicting Chemical Flood Performance".. and with consequent difficult patterns of off-diagonal non-zero elements. S . FArdRS P . "The Use of Vector Processors in Reservoir Simulation". LSOR diagonal line inversion runs some 65 times faster on the CRAY.V.. R .. Denver 1979. Paper SPE 7672. .. Consideration of further generalisation of the LSOR method to curvilinear meshes with a wider range of connectivities. . Paper SPE 7689. HAWES R. J .

i n c l u d i n g t h e p r e s e n t one. e q u a t i o n o f s t a t e . C.NUMERICAL METHODS 425 CONTROL OF NUMERICAL DISPERSION IN COMPOSITIONAL SIMULATION D. Since then.p o i n t and m i d . Our r e s u l t s show a s i g n i f i c a n t r e d u c t i o n i n t h e number of g r i d b l o c k s r e q u i r e d t o achieve t h e same numerical accuracy.0 INTRODUCTION The f i r s t compositional s i m u l a t o r s appeared i n t h e l a t e 1960's. T h i s weighting scheme l a g e n e r a l i n i t s a p p l i c a t i o n i n f i n i t e d i f f e r e n c e models f o r r e d u c i n g d i s p e r s i o n i n convective parameters W show how t h i s scheme c a n b e e s u c h a s s a t u r a t i o n and c o n c e n t r a t i o n . numerical d i s p e r s i o n r e m a i n s a m a j o r problem. c o m p o s i t i o n a l model. f o r reducing numerical d i s p e r s i o n on f u l l y i m p l i c i t compositional simulators. g r i d s i z e l i m i t a t i o n s a n d are p r i m a r i l y .p o i n t weighting schemes. i m p l e m e n t e d on a n i m p l i c i t . 1. a l b e i t i n a l e s s o b v i o u s m a n n e r . London SW 7 2BP ABSTRACT T h i s p a p e r p r e s e n t s a technique. T h i s technique is n o t even robust enough f o r 1-D problems. 2 . s o l u t i o n t e c h n i q u e s and model g e n e r a l i t y . TAN. P. Numerical examples i n c l u d i n g m u l t i p l e c o n t a c t (MCM) and n e a r m i s c i b l e (NM) p r o b l e m are used t o compare i t s performance w i t h two published compositional s i m u l a t o r s which u t i l i s e f u l l u p s t r e a m w e i g h t i n g . and i m p l i c i t backward (Van Quy ( 6 ) ) d i f f e r e n c e e q u a t i o n s . CASINADER Deparmzent of Mineral Resources Engineering. WILSON. T.C. In common w i t h B-simulators.p o i n t . w i t h l a r g e r c e l l s e l s e w h e r e . This i s too expensive f o r practical application. tremendous p r o g r e s s h a s been achieved i n t h e t r e a t m e n t of f l u i d p r o p e r t i e s . S i m p l e g e o m e t r i c a l a n a l y s i s of curve prof i l e s and r e e x a m i n a t i o n of v a r i o u s weighting schemes l e a d t o t h e development of a dynamic w e i g h t i n g t e c h n i q u e t h a t e x p l o i t s t h e optimum f e a t u r e s o f 1 . s u i t a b l e f o r multidimensional a p p l i c a t i o n . The most d o m i n a n t a s p e c t of n u m e r i c a l e r r o r s o c c u r i n t h e composltfonal f i e l d . Imperial allege. s a t u r a t i o n d i s p e r s i o n and e r r o r s i n t h e p r e s s u r e f i e l d remain. McFarlane e t a1 (1) used smaller cells i n t h e r e g i o n of m a x i m u m compositional change. C. P r i c e e t a1 (2) proposed t h e u s e of time and space d i s c r e t i s a t i o n i n 1-D. They r e q u i r e s e v e r e t i m e s t e p . As i n B-models. The works o f Peaceman ( 3 ) and L a n t z ( 4 ) on s t a b i l i t y and t r u n c a t i o n e r r o r a n a l y s i s l a i d down t h e foundation f o r many subsequent works. Various t e c h n i q u e s have been reviewed and are d i s c u s s e d below. such t h a t t h e n u m e r i c a l d i f f u e i v i t y i s of t h e same order as t h e physical d i f f u s i v i t y . It i s t h u s p o s s i b l e t o u s e a n a r t i f i c i a l d i f f u s i o n term t o c a n c e l o u t t h e numerical d i f f u s i v i t y i n e x p l i c i t backward (Chaudhari (S)).

T h i s p a p e r g i v e s some simple a n a l y s i s of v a r i o u s r e p r e s e n t a t i v e p r o f i l e s o f convective parameters i n t h e l i g h t of e x i s t i n g s t a b i l i t y and t r u n c a t i o n e r r o r analysis. r e p o r t e d t h e u s e o f 2 . b u t is w i t h i n t h e c o n s t r a i n t s of n u m e r i c a l S t a b i l i t y . which is assumed t o be c o n t i n u o u s and l i n e a r i n space and t i m e (Fig. T h i s l e a d s t o t h e i d e n t i f i c a t i o n of s e v e r a l weaknesses i n t h e 2 .0 THEORETICAL DEVELOPMENT The following s e c t i o n s d i s c u s s t h e t h e o r e t i c a l b a s i s f o r our model. The i n t e g r a l average of C i+Q over t h e This analysis time s t e p is t h e arittnoetic mean of CM1 I+% and Chr I+& a g r e e s w i t h t h e l i n e a r i s e d t r u n c a t i o n e r r o r a n a l y s i s f o r t h e convection equation in which mid-point weightings in space and t i m e are found t o b e t h e most a c c u r a t e (Appendix A l ) . Field application. It appears t h a t t h e e a r l y work of Gardner e t a1 ( 8 ) using t h e method of c h a r a c t e r i s t i c s is of g r e a t e r u t i l i t y i n m i s c i b l e flooding.4 i+I For i n t e r b l o c k f l o w a t i+3. where c o m p o s i t i o n s a r e t h e o n l y v a r i a b l e s s o l v e d .b l o c k v a l u e Ci+k is n o t a simple a r i t h m e t i c mean between tn and tn*'. v a r i a b l e d i s t a n c e weighting scheme h a s been developed which o p t i m i s e s t h e b e s t f e a t u r e s o f t h e 1 . however. i n s t e a d . The d e v e l o p m e n t is e m p i r i c a l . i n n a t u r e . t h e compositional f i e l d is decoupled from t h e c o n s e r v a t i o n e q u a t i o n used f o r s o l v i n g t h e p r e s s u r e .p o i n t u p s t r e a m w e i g h t i n g scheme is t h e most w i d e l y q u o t e d d i s p e r s i o n c o n t r o l t e c h n i q u e (Todd e t a 1 ( 9 ) ) . however.p o i n t and t h e h i t h e r t o unused m i d . and extended f o r i m p l i c i t t r e a t m e n t (Wheatley (11)).p o i n t u p s t r e a m weighting scheme f o r a n IMPES compositional s i m u l a t o r . However. and t h e b e s t m e t h o d o f e x p l o i t i n g mid-point weighting. C o n s e q u e n t l y . Cyit . An o b s e r v e r a t I+* w o u l d n o t i c e a continuous change in t h e value of C from Cn I+$ a t t h e start of a t i m e s t e p . 2 . C v a r i e s non-lineazly. I L-l n+I i . Nghtem e t a 1 ( 1 2 ) . W r e p o r t s u c c e s s f u l a p p l i c a t i o n s on d i f f i c u l t numerical problems.p o i n t u p s t r e a m w e i g h t i n g scheme. incompressible systems. i t must be found by a time-integration over t h i s range. A v a r i a b l e time l e v e l . . and may become d i s c o n t i n u o u s . t h e i n t e r . The e x p l i c i t 2 . In p r a c t i c e . I t is a p p l i c a b l e f o r m i s c i b l e . e and is g u i d e d by t h e a v a i l a b l e knowledge on t r u n c a t i o n e r r o r s . to a t t h e end of t h e time s t e p . Laumbach ( 7 ) d e v e l o p e d a t r u n c a t i o n c a n c e l l a t i o n procedure which removed t h e time s t e p and g r i d s i z e l i m i t a t i o n s by c a n c e l l i n g a p o r t i o n of t h e e r r o r i n t h e convection term w i t h t h a t i n t h e a c c u m u l a t i o n term. 2. r e q u i r e s both p r e s s u r e and c o n c e n t r a t i o n s o l u t i o n s . 2. and so d o e s n o t s t r i c t l y o b s e r v e t h e c o n s e r v a t i o n p r i n c i p l e . 1). It h a s r e c e n t l y been improved (Banks e t a1 ( L O ) ) .426 s u i t a b l e f o r miscible d i s p l a c e m e n t .p o i n t schemes.1 I n t e r p o l a t i o n Methods C o n s i d e r a c o n v e c t i v e p a r a m t e r C. t h e p a r a m e t e r C f o r t h e f i n i t e d i f f e r e n c e e q u a t i o n must be e v a l u a t e d a t I+&.p o i n t .

t h e mobile range of water s a t u r a t i o n i s between Swc and (1-Sor). (1) A continuous l i n e a r ( o r near l i n e a r ) curve between t h e p i v o t a l points.2 Curve Anal sis C o n s i d e r 4 arYbitrary p r o f i l e s a t a f i x e d t i m e l e v e l ( F i g . Before doing t h i s . A d e s c r i p t i o n of our proposed theory now follows.w ~ c ~ ~ ' ~ + ~ ~ . and that some i n t e r m e d i a t e weighting f a c t o r must be determined. gradient. equal g r i d spacing is assumed.~. X).but does not include t h e s e a c t u a l values.X i &~ " Xw+ . w i t h values between 0 and 1. . example. t h e e x t r a p o l a t e d values can be found on t h e upstream p o r t i o n of t h e c u r v e s . This implies that t h e use of upstream e x t r a p o l a t i o n i s e f f e c t i v e l y upstream weighted ( f c w < l ) . a n d . following d e s c r i p t i o n . (2 1 * * cx = e [ ~ c ~ + ~ + ~ ~ . 2 ) . and i n t e r p o l a t i o n b e t w e e n i a n d i + l .427 S i n c e t h e p a r a m e t e r C p r o p a g a t e s w i t h time. w h i l e t h e i n t e r p o l a t e d v a l u e s l i e o n t h e downstream p a r t of t h e curves with r e f e r e n c e t o I+$. 2. T h i s shows t h a t n e i t h e r s i n g l e p o i n t u p s t r e a m n o r midstream w e i g h t i n g i s t o t a l l y c o r r e c t f o r t h e non-linear problem. a r e possible.i n t e g r a l a v e r a g e v a l u e of C on t h e h i s t o r y curve a t L*. gradient. On Curves 1 and 2 . rr+l Consider t h e general l i n e a r i n t e r p o l a t i o n formula:tn4t b t c. concave convex convex concave W a r e i n t e r e s t e d i n a p p r o x i m a t i n g t h e value of C at I*. (2) The p i v o t a l p o i n t s must be "mobile". i t f o l l o w s t h a t t h e a b o v e i n t e g r a l must be equal t o a d i s t a n c e i n t e g r a l between x I+$ and some upstream X p o i n t X Thus:- - and. (3 ) In the -c PROFILE Figure2a 1 I 1FigureZb I 1 I I I I r J igure 2d Figure (2a) Curve (2b) Curve ( 2 c ) Curve (2d) Curve 1 2 3 4 : increasing : increasing : decreasing : decreasing gradient. e Both upstream e x t r a p o l a t i o n from 1-1 and 1. By t h i s we mean t h a t C should be For i n t h e mobile range bounded by t h e maximum and minimum p o s s i b l e v a l u e s . gradient. i t i s necessary t o examine t h e r e p r e s e n t a t i v e curves which a r e t o be i n t e r p o l a t e d .w ~ c e X[ i where 8 and W are t h e t i m e and d i s t a n c e weighting f a c t o r s . W o c o n d i t i o n s must be s a t i s f i e d f o r l i n e a r i n t e r p o l a t i o n t o b e valid.+r =.c >~ + ~ ~ . Ci++ C ( % t 7 dx] /(x- "'+&I .by t h e must correspond t o some point i n t h e i n t e r v a l (xi++. Our a i m i s t o u s e e q u a t i o n ( 3 ) t o p r e d i c t t h e v a l u e of an i n t e r b l o c k c o n v e c t i v e p a r a m e t e r such t h a t i t l i e s c l o g e t o t h e t r u e t i m e .[ Mean Value Theorem.

e The same a n a l y s i s can be extended t o t h e h i s t o r y c u r v e s a t a f i x e d point. F u r t h e r a n a l y s i s of upstream e x t r a p o l a t i o n and m i d . i n t h e l i n e a r i n t e r p o l a t i o n equation. c a n be incorporated i n t o an i m p l i c i t simulator t o s u b s t i t u t e f o r i n t e r b l o c k convective parameters o r t h e i r dependent f u n c t i o n s . 3a. a n upstream e x t r a p o l a t i o n i s r e q u i r e d t o provide low numerical d i s p e r s i o n . e x t r a p o l a t i o n is e f f e c t i v e l y downstream weighted. The 4 b a s i c c u r v e t y p e s ( F i g 2. b). t h e r e v e r s e c o n d i t i o n s o c c u r . and h e n c e u n s t a b l e .xi- (4) ci-l i e F i q r e 3. The u s e of u p s t r e a m On C u r v e s 3 a n d 4. and v i c e v e r s a . (2) The h i s t o r y curve a t t h e block i n t e r f a c e over t h e t i m e s t e p i s i n t h e same c u r v e g r o u p as t h e immediate upstream p r o f i l e a t t i m e l e v e l n (see next s e c t i o n ) . and Curve 2 can r e p r e s e n t t h e water S a t u r a t i o n . Based on t h e previous curve a n a l y s i s . [\Gi\ -~GIL-. They are camplementary i n t h e i r f u n c t i o n s . a n d on C u r v e 4. 8 . The (1) same d i s t a n c e weighting f a c t o r s can be used a t two f i x e d t i m e l e v e l s . Thus on Curve 3 . I d e n t i f i c a t i o n is p o s s i b l e through g r a d i e n t testing. W. This o c c u r s when t h e p i o v o t ( s ) become "immobile". w h i l e maintaining numerical s t a b i l i t y . c o n s t r u c t i o n s show how t h i s can b e done. S u f f i c e i t t o s a y h e r e t h a t both e x t r a p o l a t i o n and i n t e r p o l a t i o n have t h e i r advantages and l i m i t a t i o n s . i t c r e a t e s o v e r s h o o t . such t h a t t h e l i n e a r i n t e r p o l a t i o n formula.3 Firmre 3b /lE t 2 . [ G i = q+l-c. e q u a t i o n ( 3 ) . It i s w e l l known that t h e u s e of mid-point weighting c r e a t e s o v e r s h o o t u n d e r such circumstances. Xk.3 A Dynamic Weighting Scheme The p u r p o s e o f t h i s d e v e l o p m e n t i s t o f i n d a method of e v a l u a t i n g e x p l i c i t l o c a l weighting f a c t o r s .0 ( e q u a l g r i d s p a c i n g ) . The use of i n t e r p o l a t i o n i s e f f e c t i v e l y downstream weighted (o< W < i ) . W are i n t e r e s t e d i n t h e h i s t o r y curve a t 1 o v e r t h e time s t e p . t h e following s t r a t e g y i s adopted. t h e i n t e r p o l a t i o n f a c t o r on t h e h i s t o r y p r o f i l e . In s u c h s i t u a t i o n s upstream weighting i s t h e b e s t s t a b l e approximation. (Pigs.p o i n t I n t e r p o l a t i o n i s given i n t h e Appendix A2. T h i s s i t u a t i o n occurs j u s t behind an immiscible Thus Curve 1 c a n displacement. Under c e r t a i n c o n d i t i o n s t h e y a r e b o t h m i s t a b l e . 2.ci xi . When a Group 1 c u r v e i s d e t e c t e d .) can be subdivided i n t o 2 groups. When e x t r a p o l a t i o n i s u n s t a b l e . D can be foundon t h e chord BF.428 hence s t a b l e .xi .while full upstream weighting r e s u l t s i n a n undershoot o f t h e d i s p l a c i n g phase ( s a t u r a t i o n d i s p e r s i o n ) . r e p r e s e n t t h e o i l s a t u r a t i o n . . n and n+l. 5 and 1. u p s t r e a m e x t r a p o l a t i o n c r e a t e s u n d e r s h o o t . T h i s c a n b e i n v o k e d on t h e l i n e a r i n t e r p o l a t i o n f o r m u l a by choosing weighting Simple g e o m e t r i c a l f a c t o r s between 0 . i n t e r p o l a t i o n i s s t a b l e . a h i s t o r y curve a t a f i x e d p o i n t over a t i m e s t e p i s r e l a t e d t o t h e p o r t i o n of t h e d i s t a n c e p r o f i l e i m m e d i a t e l y u p s t r e a m of t h e f i x e d p o i n t a t t h e beginning of t h e t i m e s t e p . Due t o t h e % c o n v e c t i v e n a t u r e of C.where a shock i s p r e s e n t . 0 I The b. o r i s b u i l d i n g up.\] > < = 0 . a n d Group 2 ( c u r v e s 3 a n d 4 ) h a s d e c r e a s i n g g r a d i e n t s . Therefore. i s assumed t o be e q u a l t o t h e d i s t a n c e i n t e r p o l a t i o n f a c t o r .ssic assumptions are:Group 1 curves do n o t evolve i n t o Group 2 c u r v e s over a time s t e p . . E i t h e r Group 1 or Group 2 are p r e s e n t l o c a l l y a t a f i x e d t i m e . Group 1 (curves 1 and 2 ) h a s i n c r e a s i n g g r a d i e n t s i n t h e flow d i r e c t i o n .p i n t upstream extraplation.=$ S) crouP &-car ' 0 4 Group 2}.

which is given by:- Case 1 V(C) constant v T h i s assumption is approximately v a l i d i f t h e t i m e s t e p is smal1. F i g u r e 4 b shows t h e same p r o f i l e w i t h r e s p e c t t o X . Assuming that t h e p r o f i l e X(C) and t h e v e l o c i t y of p r o p a g a t i o n V ( C ) a r e known. t h i s f a c t o r s are c a l c u l a t e d by s e t t i n g R i l l . t h e d i s t a n c e measured upstream from t h e i n t e r f a c e . CI being i t s value a t t h e i n t e r f a c e between b l o c k s i and i+l.which c r o s s e s t h e i n t e r f a c e o v e r t h e t i m e . i s narrow. and hence t (C) belongs t o t h e same group of curves as X (C). is derived below. E f f e c t i v e f u l l upstream.4 R e l a t i o n s h i p Between Time Weighting and Distance Weighting Figure 4a Figure 4b Figure 4c F i g u r e 4a shows t h e p r o f i l e of a t y p i c a l parameter C a t t i m c t“ . It is required t o f i n d t h e e x t r a p o l a t e d p o i n t D on BF. g i v e s Wi-ai-0. c2). I f t h e g r i d spacing is miform. = a + (I. The i n t e r p o l a t i o n scheme proposed is t h e r e f o r e dynamic i n nature. + S c r e e n i n g must be a p p l i e d t o e x c l u d e t h e use of t h e i n t e r p o l a t i o n formula under 2 i n v a l i d conditions:(1) Gradient r e v e r s a l (R i is n e g a t i v e ) (2) E i t h e r . f u l f upstream weighting a f f o r d s t h e best s t a b l e a l t e r n a t i v e available. . BF is a downstream chord. The necessary weighting f a c t o r f o r t h e i n t e r p o l a t i o n formula.of t h e p i v o t s (%. o r both.5. 2-point upstream e x t r a p o l a t i o n .so that t h e band of v a l u e s ( c l . i (54) s u b s t i t u t i n g equation (5c) i n t o equation (5b) g i v e s (54 w b = I (ai-i)/Q The w e i g h t i n g When a Group 2 curve is d e t e c t e d .429 A B is an upstream chord. S i n c e V is p o s i t i v e . t ( c ) (Figure 4 c ) . o r mid-stream i n t e r p o l a t i o n w i t h v a r y i n g d e g r e e of i m p l i c i t n e s s c a n be invoked l o c a l l y via t h e same l i n e a r i n t e r p o l a t i o n formula. a t t h e i n t e r f a c e .x . 2. i t is required t o determine t h e shape of t h e h i s t o r y curve. ) / x . t ” has t h e same sign a s X”.a L ) ( x . i n t e r p o l a t i o n is s u p e r i o r . c ) are ”immobile” Once t h e s e c o n d i t i o n s are d e t e c t e d .s t e p . - -. This point is D’. For such a case. t(c) - - = x(4 V and hence t ”(C) X “(C)/V. W.

we cannot sign. Group 1).(VX' X '*70. I f .e. The "immobile" c o n d i t i o n s f o r t h e s e l e c t e d canponents must a l s o be d e f i n e d . The c o n c e n t r a t i o n p r o f i l e o f t h e most " s e n s i t i v e " component i s u t i l i s e d t o e v a l u a t e t h e weighting f a c t o r s . we o b t a i n : t J (vx'= tJ'= [v(vx"- xv')/v2. - 30 . OVERALL APPLICATION TO AN IMPLICIT COMPOSITIONAL SIMULATOR There is a unique dependence of t h e o v e r a l l component f r a c t i o n a l flows on t h e o v e r a l l composition.4 30 Case 2 V(C) i s v a r i a b l e .1 Temporal and S p a t i a l Weightin Figure 5 A t any p o i n t i n t h e system. a t t i m e t*(where tnS t% tn+' 1.~ V / ( V X ' . due t o t h e second be c e r t a i n w h e t h e r or n o t t " w i l l c h a n g e term being p o s i t i v e . and t h e e n t i r e e x p r e s s i o n w i l l be p o s i t i v e . we have xv'/) .1 3. It i s t h e r e f o r e a p p r o p r i a t e t o f i n d t h e dynamic w e i g h t i n g f a c t o r s b a s e d o n t h e l o c a l o v e r a l c o n c e n t r a t i o n p r o f i l e s . s i n c e X ' and V' have o p p o s i t e s i g n s . f u l l upstream weighting i s used i f t h e upstream and downstream blocks do not have t h e same number of hydrocarbon phases. For p r o p a g a t i o n a l s t a b i l i t y . C. o n l y d i s c u s s t h e model where i t h a s b e e n modified t o take i n t o account t h e preceding d i s c u s s i o n . t h e v a l u e of a v a r i a b l e u. i t follows t h a t t h e second term of X V ' ) ] i s always p o s i t i v e .2 v . and t h e c u r v e s w i l l b e l o n g t o t h e same g r o u p ( 1 . W w i l l .X V ' ) ] / V 3 Now. f o r a l l c a s e s . X Z O (1. N e v e r t h e l e s s . e t h e r e f o r e . however. e . These i d e a s are i l l u s t r a t e d i n t h e n u m e r i c a l examples. a n d . i f 2 phases e x i s t . t h e n V Y 0. can be r e l a t e d t o i t s v a l u e s a t tn and tn+' by:- u = Of. * eK n+ I +(I-e)Z = kn+ eSu"" (10%) 2'- U " = SU* = e Sun+'- 00 4) . t h e n . Furthermore. f o r p r a c t i c a l purposes. W f u r t h e r assume t h e l o c a l e x i s t e n c e e of e i t h e r t h e Group 1 c u r v e s . C o n s t r u c t i o n of t h e Model The c o m p o s i t i o n a l model used i n t h i s s t u d y i s based on a n e q u a t i o n of s t a t e . a n d f o l l o w s t h e i m p l i c i t f o r m u l a t i o n p r e s e n t e d by C o a t s (14). They a r e used f o r a l l components i n b o t h h y d r o c a r b o n p h a s e s . i f t h e above e x p r e s s i o n [ . o r t h e Group 2 c u r v e s . we s h a l l assume that. D i f f e r e n t i a t i n g ( 6 ) w. Group 2 ) . S e l e c t i o n of t h e most " s e n s i t i v e " component i s important t o a v o i d t h e need t o c h o o s e t h e most s t a b l e w e i g h t i n g f a c t o r s evaluated f r a n a l l t h e c o n c e n t r a t i o n p r o f i l e s . d i f f e r e n t i a t i n g y e t a g a i n . X and t belong t o t h e same group. c o n c e n t r a t i o n v e l o c i t i e s a t a f i x e d p o i n t i n s p a c e and t i m e a r e e q u a l ( H e l f f e r i c h (13)).r. 3. and so X" and t " w i l l have t h e same s i g n .t. To account f o r phase d i s c o n t i n u i t y .1.

c e n t r e s . involving components: C1. s u b s t i t u t i o n then l e a d s to: 4 n+l ?'S -n+( ! = ewsKL. 3.. ) I n c r e m e n t s i n t h e t r a n s m i s s i b i l i t y term T are c a l c u l a t e d by means of partial derivatives w.431 so.1 Displacement 1 (Coats (14)).8 sLn+l 4 U C+l ( 4 c+: % - * * u = wu. I f . +(l-w)uL+l " .r.n-C4 and n-C 10..t. Un). I n t h e s i m u l a t i o n . 4. r e s p e c t i v e l y . u i s also evaluated a t a p o i n t o t h e r than t h e b l o c k .1. The boundary between t h e f i r s t two zones w i l l be r e f e r r e d t o a s t h e g a s f r o n t . i n a d d i t i o n t o t h i s intermediate time-level. In MCM problems. however. T h i s r e t a r d a t i o n of t h e m i s c i b l e f r o n t i s c o n s p i c u o u s in C o a t s ' r e s u l t s . t h e complete set of v a r i a b l e s (U. Thus:- st$ .2 Expansion of t h e "Flow" term The f l o w between two n e i g h b o u r i n g b l o c k s i a n d i + i can be expressed i n t h e form:- T+(C. and Smith and Yarborough ( 1 6 ) . The d a t a f o r t h e s e r u n s were t a k e n from Coats (14). and i s displaced by a r i c h gas.. and. i n o r d e r t o demonstrate t h e a p p l i c a t i o n of t h e above theory i n a v a r i e t y of s i t u a t i o n s . 4 O m d 80 b l o c k s show a p r o g r e s s i v e . t h r e e zones can be i d e n t i f i e d : a downstream zone containing undersaturated o i l . and f i n a l l y an upstream miscible zone containing a s i n g l e dense f l u i d whose composition changes from t h e c r i t i c a l composition t o t h a t of t h e i n j e c t i o n gas. t h e n i t h a s t o be r e l a t e d t o t h e block-centre values by means of t h e d i s t a n c e weighting formula. while t h a t between t h e latter two w i l l be c a l l e d t h e miscible f r o n t .( l l b ) and (12a). Leach and Y e l l i g ( 1 5 ) .. a middle zone comprising two p h a s e s whose c o a p o s i t i o n s c o n v e r g e i n t h e u p s t r e a m d i r e c t i o n . we o b t a i n t h e expansion of t h e flaw term. where t h e u s e of 20. and t h e cumulative change.-: P . f i n a l l y . T h i s i s a n MCM problem. . t h e use of s i n g l e p o i n t upstream weighting l e a d s t o s e v e r e c o m p o s i t i o n a l d i s p e r s i o n which causes s u b s t a n t i a l delay i n t h e attainment of miscibility. + e(i-w)Sai+. t h i s implies:- u ) denote t h e change i n u over t h e . f o r each i t e r a t i o n .0 DISCUSSION OF RESULTS T h r e e d i f f e r e n t d i s p l a c e m e n t p r o b l e m were chosen. r e s p e c t i v e l y . . as follows:- 4.. using equations ( ~ O C ) . The system e x i s t s i n i t i a l l y as a n undersaturated l i q u i d . The symbols % A I' (=cC K ) and SK (=V i t e r a t i o n 1.

Thus C 1 was a r b i t r a r i l y chosen f o r t h i s purpose. c . 0 - a -"0 CO"TI10L + U m 0.S DAYS . . . 0 Figure 6 D i s p l a c e m e n t 1 (MCH) : 1-D c o m p a r i s o n o f d y n a m i c weighting with f u l l upstream weighting. d ) . w i t h t h e i n j e c t i o n and p r o d u c t i o n w e l l s l o c a t e d i n two diagonally-opposite Once a g a i n . . . c. c o n c e n t r a t i o n p r o f i l e ( d ) o i l recovery and GOR vs time.2- .D M C M PROBLEM 6 1 . following i t s breakthrough t o t h e producer. t h e f d r s t of t h e s e b e i n g d e s i g n e d t o s i m u l a t e FCN d i s p l a c e m e n t .being displaced by d r y g a s (Cl). I n t h e a b s e n c e of a n a n a l y t i c ' a l s o l u t i o n .and simulating t h e displacement i n t h e s u p e r c r i t i c a l region ( a t 3000 psi). . The system used i n t h i s displacement was a binary mixture of C 1 and nC5. ( F i g u r e s 7a. e v a l u a t i o n of weighting f a c t o r s can be c a r r i e d out on e i t h e r component. T h i s was achieved by assuming a n i n i t i a l composition of 50% C 1 and 50% n-CS. The scheme has a l s o been t e s t e d i n 2D. d show a comparison between t h e use of t h i s t e c h n i q u e and s i n g l e . .D M C M P R O B L E M COAT S I . The use of t h e proposed technique c l e a r l y results In a f a s t e r advance of t h e m i s c i b l e f r o n t .2. Since t h i s is a p e r f e c t piston-type displacement. I n t h i s case.D M C M P R O B L E M 1.p o i n t u p s t r e a m weighting. b. COAT S I . b. w i t h 2 0 b l o c k s . The i n t r o d u c t i o n of t h e dynamic w e i g h t i n g scheme described i n t h i s paper produces a marked improvement. 4.432 8. . . (a) Sg p r o f i l e . s i z e of t h e miscible zone can be observed. Tbo runs were perfomed on t h i s system. a n improvement i n t h e c o r n e r blocks. (c) C4 - i n c r e a s e i n i t s speed of p r o p a g a t i o n . Figures 6a. a n s w e r s w h i c h a r e of c o m p a r a b l e a c c u r a c y t o C o a t s ' 40-block s o l u t i o n . @ z _I 0. which is confirmed by t h e e a r l y and s t e e p rise i n GOR. i t is j u s t i f i a b l e t o assume t h a t t h e 80-block s o l u t i o n is t h e nearest t o real i t y. COAT S I . using a C a r t e s i a n g r i d of 9x9 b l o c k s .2 Displacement 2 (Smith and Yarborough (16)). and has enabled u s t o o b t a i n .5 COAT S I . using our technique. 5 DAYS z 0 L 200 -NO c o u r m DINAMC WEICIIIWC TIME.D M C M PROBLEM TlMEn2lO D A Y S 20 BLKS OlMAX=T. ( b ) Advance of m i s c i b l e f r o n t .0' TIME=210 DAYS 20 BLKS D l M A X = 7 .1 3 I- 4 W U . t h e a n a l y t i c a l s o l u t i o n c o n s i s t s of a s t e p change i n c o m p o s i t i o n from t h e i n j e c t i o n c o m p o s i t i o n t o t h e i n i t i a l c o m p o s i t i o n .DAYS 0 U 0.

c 7 d) T IVEI PV o 1-240 INJ -0.5% n-C5 was c h o s e n .240 PV INJ ~0. .5% C 1 and 12. F i g u r e s 8a and 8 b show t h e C 1 p r o f i l e a t 210 days.637 = O . ( b ) C4 c o n c e n t r a t i o n map. ( c ) Gas s a t u r a t i o n map. T h i s p o s t u l a t e d a b s e n c e of c o m p o s i t i o n a l d i s p e r s i o n i s v e r i f i e d by t h e numerical results shown i n Figures 8 c and Ed. f u l l u p s t r e a m weighting. with t h e l i q u i d assumed t o be immobile. and the n-C5 concentration i n t h e e f f l u e n t a s a f u n c t i o n of time. f o r some problems ( s u c h a s of t h i s type) t h e amount of compositional dispersion i s negligible. a n i n i t i a l c o m p o s i t i o n of 87. f u l l upstream weighting. i n both of which t h e results of u s i n g s i n g l e p o i n t upstream weighting are v i r t u a l l y i d e n t i c a l t o t h o s e obtained with t h e p r e s e n t technique. dynamic weighting. C 1 w s i n j e c t e d . which was a l s o t h e p r e s s u r e a t which t h e s i m u l a t i o n was c o n d u c t e d . so a s t o y i e l d an i n i t i a l condensate l i q u i d of 7% s a t u r a t i o n a t 1525 p s i . ( d ) C4 c o n c e n t r a t i o n map. and t h e problem was run i n t h e 2-phase mode.564 0 A \ \ I \ Figure 7 . a Agein. dynamic weighting.637 0 OIL R E C ~ 0 . a l t h o u g h b o t h show a n a p p r e c i a b l e compositional d i s p e r s i o n r e l a t i v e t o t h e a n a l y t i c a l s o l u t i o n . I n t h e second run. The weighting technique shove better r e s u l t s than t h e "full u p s t r e a m " case.?\ 433 '''' T I VF I 0 1. The purpose of t h i s r u n was t o d e m o n s t r a t e t h a t . 5 6 4 \ v. ( a ) Gas s a t u r a t i o n map.Displacement 1 (MCM) : 2-D comparisons.

by CO2. The above f e a t u r e s have a l s o been v e r i f i e d i n t h e NM r u n .04.3 Displacement 3 (Leach and Y e l l i g ( 1 5 ) ) . MCM and NM). with merely 20 blocks.385 PV INJECTED ': . The rJeighting t h e one which had t h e least i n i t i a l c o n c e n t r a t i o n technique h a s enabled u s t o match t h e l a b o r a t o r y results t o a good a c c u r a c y .FEET CONOENSATE REVAPORIZATION PV CH4 INJECTED CONDENSATE REVAPORIZATION I Q+IO- 0. f o r example. Leach e t a1 (15) presented l a b o r a t o r y results c o v e r i n g t h e various displacement t y p e s (PCM. C o n s i d e r i n g t h e MCM r e s u l t s f i r s t . of a s y n t h e t i c crude o i l . and advance of "dry f r o n t " v s t i m e (re-vaporization). The component which we s e l e c t e d f o r " g r a d i e n t t e s t i n g " was namely C6. and t h e s t e e p change i n t h e GOR and t h e e f f l u e n t composition. This m e a study of t h e mechanisms involved i n t h e displacement. and t h e s t e e p e r C02 p r o f i l e r e s u l t i n g f r a n t h e use of t h i s technique. To t e s t our technique. by showing t h e delayed breakthrough of C02.- 0.FEET PV METHANE INJECTED Figure 8 Displacement 2 : 1-D c o m p a r i s o n of dynamic w e i g h t i n g with f u l l upstream weighting. DRY GAS DISPLACING RlCH G A S 210 D A Y S ( 3 4 S T E P S ) .434 0 = DRY GAS DISPLACING RICH GAS ' ' ' ' ' ' ''' ' '''' 1.08-- 0.06-- 0. l=W=DYN 50 100 150 200 250 DISTANCE . and a n NM d r i v e (Run 7). i n t h e s i g n i f i c a n t sharpening - . (d) C5 c o n c e n t r a t i o n i n e f f l u e n t . using 100 blocks. ( c ) C1 and o i l s a t u r a t i o n p r o f i l e s (re-vaporieation). a s c a n be s e e n . ( b ) C5 c o n c e n t r a t i o n i n e f f l u e n t v s t i m e (FCN). p e r h a p s t o a g r e a t e r e x t e n t . two rune were chosen: an MCM d r i v e (Run 61. F i g u r e 9 a demonstrates t h e f a s t e r advance of t h e m i s c i b l e f r o n t . - 4.02-- DISTANCE . Figures 9b and 9c f u r t h e r support our d i s p e r s i o n c o n t r o l method. ( a ) C1 c o n c e n t r a t i o n p r o f i l e (FCN).1 i 4 ' ' ' ' ' ' . and aleo simulated t h e s e on t h e i r compositional model.

. LEACH ETAL DATA RUN 7 . .!.6- . 0. 1 1 (b) HCPV COZ INJECTED LEACH ETAL DATA ( R U N 6 ) DINAMC wuGnnNG I ' N CONfRCX O I. .0 . I 1. a ..(I I 3 0.I DAY RUN 6 0.. .I- u . . . 1 1... . .. (a)-(d). . 4 HCPV C02 INJECTED [L OVNAHC 20 ELKS wmcnn" .2 . .4 . -. .-.. .-0.2 .435 LEACH ETAL DATA 20 BLK DT=. Capuon overleaf.8 A..O0 TIME ( D A Y S ) (d) O 'I ' ' DISTANCE FROM INJECTION END ( F E E T ) l!o' (C) '' bj2: bi4' O ~ l i . .. . .? . L J a - . .6 0. -. . . HCPV C 0 2 INJECTCD Fg9 i. ' 'O. 0. . .

(e) Norullred concentrrtlon of CO2 and C 1 i n effluent v.6 0.4 13. and t h e consequent h i g h e r recovery r e s u l t i n g from t h e u s e of t h i s technique. It needs t o be mentioned. HCPV injected (f) COR and 011 recovery va HCPV injected (NH). - - 5.p o i n t u p s t r e a m s c h e m e s . however.I D p 10. rod CZ-C6 in profller (MCM). 0 0.0 CONCLUSIONS The work d e s c r i b e d i n t h i s p a p e r l e a d s u s t o t h e f o l l o w i n g main conclusions : (1) The u s e of single-point upstream weighting c a u s e s s e v e r e c o m p o s i t i o n a l d i s p e r s i o n . b a s e d o n t h e p r o p e r t i e s of t h e p r o f i l e .5 m \ - 12. MCM and NM displacements. > w V W I I ( I 0 \ =' 0 4. due t o t h e weighting scheme ( F i g u r e 9d). C L c 0 L u m 9.4 0. - .5 I I .- Q I. p a r t i c u l a r l y when s i m u l a t i n g FCM.4 (el HCPV C02 INJECTED 'O. (2) A dynamic w e i g h t i n g scheme h a s b e e n d e v e l o p e d . t o determine t h e optimum weighting f a c t o r s i n t i m e and space.5 z u L I W W = 0 c a > a 6. and s i n g l e . I t e x p l o i t s t h e c l a s s i c a l f e a t u r e s of mid-stream.2 0. a n d y i e l d s r e s u l t s which. The p r o d u c t i o n h i s t o r y ( F i g u r e 9 e ) and t h e e f f l u e n t C 1 and C02 c o n c e n t r a t i o n s ( F i g u r e 9 f ) confirm t h e delayed a r r i v a l of t h e 2-phase z o n e . - which o c c u r s i n t h e Sg and C02 p r o f i l e s . (4) The method i s supported by g e o m e t r i c a l arguments. i f t h e i r a c c u r a c y i s t o be reproduced on a " f u l l y upstream" model. 7. would r e q u i r e s e v e r a l times a s many g r i d blocks.p o i n t u p s t r e a m .8 1. (3) The t e c h n i q u e h a s been s u c c e s s f u l l y t e s t e d o n M C M . injected (ncn). b e f o r e t h e r e s u l t s o f Leach e t a1 ( 1 5 ) could be s u c c e s s f u l l y reproduced. t w o .2 1. t h a t a c r i t i c a l g a s s a t u r a t i o n o f 1 5 % had t o be i n t r o d u c e d t o t h e r e l a t i v e p e r m e a b i l i t y t a b l e .5- 20 ELK DT=. L n 0 3 . (a) Sg and COZ concentration (b) Normalized concentration of 032.0 1. (Nn).5 "8 .436 15. ( c ) COR and o i l recovery vs HCPV effluent vr HCPV injected (MCW).Q0 Figure 9 Dirplacement 3 : 1-D comparison of dynamic VOightlng vlth full uprtrerm Wlghting.o-- a 0 =I I . FCM a n d NM d i s p l a c e m e n t s . (d) Sg and COZ concentration profilea (Nn). which u t i l i s e s t h e p r o f i l e of t h e v a r i a b l e concerned. and can beimplemented e a s i l y i n multi-dimensional s i m u l a t o r s .

and CORTEVILLE. SIMANWUX. D r ..437 ACKNOWLEDGEMENTS The a u t h o r s would l i k e t o t h a n k t h e UK D e p a r t m e n t of Energy and Imperial College of S c i e n c e and Technology f o r s u p p o r t i n g t h i s r e s e a r c h .. M. J .M. AIME. K. SPE P a p e r 9306. "Methods f o r I n c r e a s e d Accuracy i n Numerical R e s e r v o i r Simulators". P. (March. (1971).D. S o c i e t y o f P e t r o l e u m E n g i n e e r s J . E l s e v i e t . S o c i e t y o f P e t r o l e u m E n g i n e e r s J . z. 26-36 9. A. D.J.... AIME. I. S o c i e t y of Petroleum Engineere J. Multiphase Displacement I n Porous Media".H. 2.K. 1979. "Compositional Modelling w i t h An E q u a t i o n o f S t a t e " . . and POLLARD. 12. NGHIEM. 251 6.C. HELFFERICH. MUELLER.R. 315-320. 363-377 A.H. AIME. 261 S o c i e t y of Petroleum Engineers J.A.. MILLER.. s o c i e t y of PetroleumEngineers J. "A Numerical Study of Diphasic Multicomponent Flow".. VAN QUY. . Texas (September 1980) 13. MCFARWE.X. R. S o c i e t y o f P e t r o l e u m Engineers J.. D.W. COATS. Wall..D i f f e r e n c e O i l R e s e r v o i r S i m u l a t i o n " . D. Green o f ERC f o r h e r p a t i e n c e i n typing t h e v a r i o u s d r a f t s of t h i s paper.I.. REFERENCES 1. 1. 20. 99-203 11. ODELL. " G e n e r a l Theory of Multicomponent. CHAUDHARI. 'CHESHIRE. WHEATLEY. "An Equation of S t a t e Compositional Model". P E A C W . AIME 253 7. paper presented a t S o c i e t y of Petroleum Engineers 5 t h Symp. Amsterdam. TODD. 171-184. "An Improved N u m e r i c a l T e c h n i q u e f o r S o l v i n g M u l t i .. D. ( 1 9 7 7 ) . S o c i e t y o f Petroleum Engineers J.A.G.M. LAUMBACH. Society of J. P. R. G. N. SPE Annual F a l l Meeting. Denver. 7677 12.S. M.. and P E A C W .B. N.W.. Trans. 8. 14. K. 277-284. (February 1981). " Q u a n t i t a t i v e Evaluation of Numerical D i f f u s i o n (Truncation Error)".S t a t e D i s t r i b u t i o n s o f F l u i d C o m p o s i t i o n s i n Two-Phase O i l R e s e r v o i r s U n d e r g o i n g Gas I n j e c t i o n " . S o c i e t y . "Isothermal Displacement P r o c e s s e s w i t h I n t e r p h a s e Mass Transfer". F i n i t e D i f f e r e n c e T e c h n i q u e f o r T r e a t i n g t h e Convection-Diffusion Equation". Trans. l . ( 1 9 7 5 ) -5 517-531 1 . D.D i m e n s i o n a l M i s c i b l e D i s p l a c e m e n t E q u a t i o n s " . and POZZI. Dallas.O. 2. and HIRASAKI.. and AZIZ. 61-74.D. 1967). . T. (1972). ( A p r i l 1972).. Dublin (June 1980). Trans. P r o c e e d i n g s of BAIL Conference. "A Version of Tvo P o i n t U p s t r e a m W e i g h t i n g For Use i n I m p l i c i t N u m e r i c a l R e s e r v o i r Simulators".. and DONOHUE. R. 2 of Petroleum Engineers J. GARDNER. "A Technique f o r C o n t r o l l i n g Numerical D i s p e r s i o n i n F i n i t e . 11. Dam f o r t h e i r c o n t i n u i n g i n t e r e s t . PetroleumEngineers 3.. (October 1980). "A H i g h A c c u r a c y . ( 1 9 7 7 ) 65-82 4. (19641.. PRICJ3.J... A. (June 1967) 1. Trans.. F. a n d Hiss M.G. BANKS.K.T. S o c i e t y o f P e t r o l e u m E n g i n e e r s J. SPE Paper No. "Fundamentals of Numerical R e s e r v o i r Simulation". FONG. 251 5 . F. R. "Numerical C a l c u l a t i o n o f M u l t i d i m e n s i o n a l M i s c i b l e D i s p l a c e m e n t by t h e Method of C h a r a c t e r i s t i c s " . P r o f e s s o r C. 115-130. L.G. 515-530 10. D. LANTZ.M. " U n s t e a d y . On R e s e r v o i r Simulation.

79-91. L. " E q u i l i b r i u m R e v a p o r i z a t i o n of Retrograde Condensate by Dry Gas I n j e c t i o n .2. T r u n c a t i o n Errors and Numerical D i s p e r s i o n The n o n l i n e a r c o n v e c t i o n e q u a t i o n is: ax 2. AIM. SPE P a p e r 8368. S o c i e t y of P e t r o l e u m E n g i n e e r s J. - 243 17.F. w e are. Peaceman ( 1 7 ) showed t h a t a n o n l i n e a r s t a b i l i t y a n a l y s i s g a v e t h e same p r a c t i c a l c r i t e r i a for a f u l l upstream d i f f e r e n c e scheme (W-1). 2 ) shows a l e a d i n g t r u n c a t i o n e r r o r term of t h e form: (A I ."f'. .3) V / A ~ [w-+) vd+g(e--:)]. SPE Annual F a l l Meeting.. %&* ax+ at (A 1 .and is terned numerical d i s p e r s i o n . ( F e b r u a r y 1 9 7 7 ) .5 t o 1 (Equal g r i d s p a c i n g ) . The a p p r o x i m a t e s t a b i l i t y subdomain i n which t h e dynamic weighting scheme is o p e r a t i n g is more r e s t r i c t i v e t h a n t h a t permitted by t h e l i n e a r i z e d S t a b i l i t y analysis. A Schematic I l l u s t r a t i o n of The Numerical S t a b i l i t y Domains S t a b i l i t y Domain Domain i n whichdynamic weighting schemeoperates Conditional S t a b i l i t y Domain I n c r e a s i n g Numerical S t a b i l i t y and Truncation Errors la 0 0 w- I . S t a b i l i t y . SMITH. " 87-94 Trans. D.ac..P. i n convection e q u a t i o n of t h e form: e f f e c t . W. solving a d i f f u s i o n ank*a+ 2% DalLm = 1 + - - a%. (1968). M. and YARBOROUGH.. This c r e a t e s a r t i f i c i a l d i f f u s i o n of C. (A\. The r e s u l t s of t h e l i n e a r i z e d s t a b i l i t y a n a l y s i s a r e summarised i n t h e diagram shown below.438 LEACH. "Compositional Model S t u d i e s : Co2 O i l Displacement Mechanisms".:. P E A C W . L .I m p l i c i t Mobility". By s o l v i n g t h e d i f f e r e n c e e q u a t i o n of e q u a t i o n A1. AIME 259 15. "A Nonlinear S t a b i l i t y A n a l y s i s f o r D i f f e r e n c e Equations Using S e m i .R. 4 ) . W and 8 a r e t h e d i s t a n c e and time weightinn f a c t o r s f o r C i n t h e d i f f e r e n c e equation. Las Vagas. Nevada (September 1979) 16. and YELLIG. 17. L i n e a r i s e d s t a b i l i t y a n a l y s i s shove that t h e numerical s o l u t i o n s a r e s t a b l e i f t h e w e i g h t i n g f a c t o r s l i e i n t h e r a n g e 0. Trans.\) T r u n c a t i o n e r r o r a n a l y s i s o n t h e f i n i t e d i f f e r e n c e a p p r o x i m a t i o n of t h e l i n e a r i s e d equation (A I .W. APPENDICES Al.

T h i s i s a Group 2 curve s i t u a t i o n c r e a t e d by t h e use o f r e l a t i v e p e r m e a b i l i t y (Kr) e x t r a p o l a t i o n . I n t h i s example. (a) Spivak d a t a (SPEJ. However. . . LANGSRUD DA TA F i g u r e LO Comparison of w a t e r s a t u r a t i o n p r o f i l e s f o r v a r i o u s m o b i l i t y e v a l u a t i o n schemes ( b l a c k o i l model). . . SPEJ June 1972). ( b ) Langsrud d a t a (Nolen and Berry. f o r a water f l o o d i n g problem.7 i n s t e a d of t h e s p u r i o u s n e g a t i v e value. w h i c h i s l e s s c o n s i s t e n t t h a n s a t u r a t i o n e x t r a p o l a t i o n on t h e following grounds:(1) S a t u r a t i o n s a t t h e block interface d i c t a t e t h e i n t e r b l o c k flow. b e l o n g t o t h e Group 2 curves category.according t o our p r e s e n t a n a l y s i s . (2) I t c r e a t e s .439 A.2 A d d i t i o n a l Notes on 2-Point Upstream Weighting and Mid-Point Schemes Weighting The p r e v i o u s c u r v e a n a l y s i s shows t h a t 2-point upstream weighting cannot be a p p l i e d on Group 2 curves. - . (3) T y p i c a l l y . Krw a t 1124 would be 0. Case 1: Todd e t a 1 showed a n example of u n i t m o b i l i t y . a Group 1 s a t u r a t i o n p r o f i l e exists. February 1977). Group 2 s a t u r a t i o n p r o f i l e s e x i s t above t h e shock f r o n t s a t u r a t i o n value.but not f u l l y upstream weighted on t h e a c t u a l curve p r o f i l e . . Had s a t u r a t i o n e x t r a p o l a t i o n been used. . which is g r e a t e r t h a n 1 by 2-point upstream eXtraFQlatiOn. t h e e x t r a p o l a t e d r e l a t i v e p e r m e a b i l i t i e s w i l l not correspond t o a t o t a l s a t u r a t i o n of 1. Case 2: Todd showed t h a t a s p u r i o u s e x t r a p o l a t i o n e r r o r would o c c u r n e a r a s h a r p WOC or GOC. a Group 2 p r o f i l e (which i s n o t amenable t o l i n e a r e x t r a p o l a t i o n ) . o r a c c e n t u a t e s t h e c r e a t i o n o f .aS shown. A m i d s t r e a m i n t e r p o l a t i o n is t h e b e s t alternative. T o d d e t a1 recommended s e t t i n g t h e s p u r i o u s e x t r a p o l a t e d value t o t h e maximum of t h e 2 bounding v a l u e s . .1 . m i s c i b l e d i s p l a c e m e n t It i s p o s s i b l e t o c a l c u l a t e o i l r e l a t i v e permeability o f o i l by s o l v e n t . Our c u r v e a n a l y s i s i n d i c a t e s t h a t t h i s i s e f f e c t i v e l y downstream weighting and could create a n undershoot of t h e o i l phase i f i t i s approaching zero s a t u r a t i o n . Todd e t a 1 ( 9 ) showed 2 c a s e s which. . I t i s e f f e c t i v e l y upstream. The corresponding K r w p r o f i l e s 1.

using 2 . i s i l l u s t r a t e d i n Figure lob. This supports t h e evidence that mid-point weighting i s unstable. Group 1 curves (which are not amenable t o i n t e r p o l a t i o n ) a r e p r e s e n t around t h e flood front.44 0 have s t r o n g e r Group 2 c h a r a c t e r i s t i c s . An e x p l i c i t v e r s i o n of t h e dynamic weighting scheme. The f a c t that e x p l i c i t 2-point Kr e x t r a p o l a t i o n d o e s n o t c a u s e i n s t a b i l i t y is p r o b a b l y d u e t o t h e non-sharpening n a t u r e of t h i s s a t u r a t i o n range. Figure 1Oa i l l u s t r a t e s some ueaknesees of u s i n g Kr e x t r a p o l a t i o n . The s a t u r a t i o n r a n g e b e l o w t h e s h o c k f r o n t s a t u r a t i o n i s self-sharpening. n o r m a l l y n o t o b s e r v a b l e without imposing f r o n t a l c o n t r o l . On t h e o t h e r hand. thereby a g g r a v a t i n g t h e weakness of i n t e r p o l a t i o n .p o i n t s a t u r a t i o n e x t r a p o l a t i o n a t t h e c o n t r o l l e d f r o n t and m i d s t r e a m weighting everywhere else. .

convection and &EI transfer rates. . the greater computational . Pb or Re.BANKS and D. c h a r a c t e r i s e d by the treatment of j u s t two hydrocarbon components.Enghnd ABSTRACT Mass t r a n s f e r may be described i n black o i l s i m u l a t o r s by allowing o i l and gas t o e x i s t i n both l i q u i d and vapour phases. mainly f r o m t h e s t a n d p o i n t of a f u l l y implicit formulation. Sw. A t p r e s e n t . have t r a d i t i o n a l l y been more concerned w i t h d i 8 p l a c e m n t mechanisms t h a t the PVT dominated processes of EOR 8 t u d i e s . A v a r i a b l e s u b s t i t u t i o n method for simultaneouslymodellingofp and rs v a r i a t i o n s i n an implicit b l a c k s i m u l a t o r is d e s c r i b e d i n S e c t i o n 2. S The equation6 determining P and r8 would involve d f f f u s i o n . Pd or re. A n q f f i c i e n t f u l l y implicit method of simultaneously modelling bubble and dew p o i n t is described. A partial re-solution algorithm o p t i o n is d e s c r i b e d which I n t e r p o l a t e s bettween total. The ‘ o i l ’ and ‘gaa’ M y be any two groups of . The vapour oil-gru ratio is g e n e r a l l y preferable.re-solution logic. or the rrolution g a a . mil%thi8 cannot y i e l d the d e t a i l e d d a c r i p t l o n obtrined from a t r u e multi-coaponent compositional s i m u l a t o r .K.and no. Pb. or t r u e stock tank o i l and gaa.NUMERICAL METHODS 441 INTERPHASE MASS TRANSFER EFFECTS IN IMPLICIT BLACK OIL SIMULATORS D. d i n g two o f t h e v a r i a b l e s dependent on t h e primary ones. The three equation p i c t u r e may be r e s t o r e d by employing bubble and dew p o i n t models. B l a c k o i l s i m u l a t o r s . Pa. aa it enabfes thx vapour t o be described i n r e g l o n s o f low p r e s s u r e and re. hydrocarbon components.PONTING Atomic Energy Research Establishment. t h e extra c a p l t a t i o n a l effort r e q u i r e d t o s o l v e the f o u r t h and f i f t h e q u a t i o n s is p r o h i b i t i v e . Hanuell.o i l ratio. Black o i l models g e n e r a l l y d e s c r i b e the c o n c e n t r a t i o n o f d i s s o l v e d gaa i n the reservoir l i q u i d by t h e bubble p o i n t p r e s s u r e . P I n a g e n e r a l b l a c k o i l model t h e r e are therefore f i v e independent v a r i a b l e s per cell: . or t t e vapour oil-gaa ratio r or R . I n t h i s paper we d i s c u s s a n u m b e r o f v p e c t s ofma8stransferinblackoilsimulator8.23. The d l s p r e i o n o f di88olved gaa and vapourlsed o i l is discussed. A s u b t r a c t e d t o t a l gaa formulation is found t o combine t h e advantages of t h e free and total gas approaches.[1. for which no dew p o i n t exists. C-sltional effects are modelled simply by mass t r a n a f e r between l i q u i d and vapour phases. Oxfordshire. R The q u a n t i t y o f o i l i n t h e vapour is described by t h e dew p o i n t p r e s s u r e .

Using typical d i f f u s i o n c o e f f i c i e n t s . F a c i l i t i e s such as f a u l t connections and d i r e c t i o n a l r e l a t i v e permeabilities are t h e n a v a i l a b l e for s t u d i e s involving m i l d l y v o l a t i l e o i l s or dew p o i n t t r a n s i t i o n s . i f anything. A subtracted t o t a l gaa method which overcomes t h i s is described i n S e c t i o n 3. and is similar t o the t r a p p i n g f r a c t i o n approach. and e s s e n t i a l l y implies i n s t a n t a n e o u s flow of d i s s o l v e d gaa through o i l . where an i n t i m a t e gas-oil c o n t a c t e x i s t s . r e s u l t i n g i n the gas bypassing some of the o i l . NO r e . The q u a n t i t y o f g a s e x i s t i n g i n s o l u t i o n may be t y p i c a l l y 50-1000 times g r e a t e r t h a n t h a t e x i s t i n g i n t h e vapour phaae. This can s t i l l be f u l l y expanded i n a three v a r i a b l e formulation. and w i l l o v e r e s t i m a t e g a s cap sizes. These problem8 are. No. Those are the no. A t o t a l gaa formulation avoids t h i s .s o l u t i o n and total. or j u s t the f r e e component. so that a layer of s a t u r a t e d o i l w i l l immediately b u i l d up at a gas-oil i n t e r f a c e and p r e v e n t f u r t h e r solution. I n a sense. Such methods may be important f o r compositional s i m u l a t o r s aa the i n c r e a s i n g number of e q u a t i o n s r e n d e r s f u l l y simultaneous s o l u t i o n methods i m p r a c t i c a l l y expensive.s o l u t i o n or e q u i l i b r i u m f r a c t i o n for each cell. This is due t o the non-specification of the d e t e d n i n g d i f f u s i o n rates. although other methodg are possible i f t h e time step l e n g t h is restricted. Free g a s may t h e n pass through a cell without Coqpletely s a t u r a t i n g the o i l . however.res o l u t i o n assumes that d i s s o l v e d gaa does not d i f f u s e through o i l . vapour invading o i l is l i k e l y t o f i n g e r or channel. rather than c a u s i n g it. t w o a l t e r n a t i v e s are a v a i l a b l e f o r the t r e a t m e n t o f g a s s o l u t i o n . and the rate at which s o l u t i o n o c c u r s . I n o u r experience v a r i a b l e s u b s t i t u t i o n is t h e o n l y method of modelling mass t r a n s f e r which does not l i m i t the a b i l i t y of the s i m u l a t o r t o t a k e l a r g e time steps.442 e f f i c i e n c y enables detailed f u l l f i e l d s t u d i e s t o be performed. O r i g i n a l l y a free g a s formulation w a a used i n PORES [lo]. the f r a c t i o n o f the l i q u i d hydrocarbon i n a cell i n c o n t a c t w i t h the vapour. and other black o i l s i m u l a t o r s .r e . a conservation c o n d i t i o n v i o l a t e d when i n t e r p h a s e maas t r a n s f e r occurs. b u t the re-solutionof gasdepends on t h e p r e s e n c e o f g a s i n c o n t a c t w i t h l i q u i d o i l . it can e a s i l y be shown t h a t a Inm d r o p l e t w i l l reach 991 of its u l t i m a t e d i s s o l v e d g a s c o n c e n t r a t i o n i n less than 24 hours. more s e v e r e for vapourised o i l . so that changes i n R and r are M i a t e l y p r o m a t e d across cells. Experiment [7] i n d i c a t e s that. t h e artificial'cell k u n d a r i e s introduced by t h e s i m u l a t o r prevent d i s p e r s i o n f r o m b e i n g t o t a l . so that free g a s may o n l y exist w i t h s a t u r a t e d o i l . I n PORES. t h e column s m c o n d i t i o n r e p r e s e n t s an u attempt t o c o n s e n e free gas.s o l u t i o n logic has an e q u i v a l e n t problem i n t h a t g a s is evolved from undersaturated o i l when flow .s o l u t i o n o p t i o n s .r e . I n practice. I n s o l v i n g t h e material c o n a e r v a t i o n e q u a t i o n s . a c o l u m n s u m c o n d i t i o n i s i m p o s e d w h i c h a t t e m p t s t o z e r o the s m of errors on d i a g o n a l p l a n e s of cells w i t h i n the reservoir model. The determining factor i n g a s s o l u t i o n is the rate a t which g a s d i f f u s e s through l i q u i d o i l . Gas must come o u t o f s o l u t i o n when the o i l p r e s s u r e crosses the bubble p o i n t . leaas t o p o o r l y conditioned e q u a t i o n s which t h e s e q u e n t i a l method f r e q u e n t l y fails t o s o l v e . and numerical d i s p e r s i o n may be limited by the use of amall g r i d blocks. e q u i l i b r i u m is established on a timescale short compared t o those t y p i c a l l y involved i n reservoir engineering. For a free g a s formulation. T h i s u is p a r t i c u l a r l y important i n the s e q u e n t i a l method of s o l v i n g t h e l i n e a r matrix equations. due t o the l a r g e d i s s o l v e d g a s c o n t r i b u t i o n . Total re-solution l o g i c assumes that i n t e r p h a s e e q u i l i b r i u m always exists i n each cell. While t h i s o p t i o n is l o g i c a l l y c o n s i s t e n t . however. The q u e s t i o n arises aa t o whether t h e maas conservation equation for g a s should involve a l l t h e g a s . Simulators which permit g a s s o l u t i o n have d i f f i c u l t i e s w i t h the d i s p e r s i o n of d i s s o l v e d g a s and cell size dependence. I n S e c t i o n 4 we d e s c r i b e a partial re-solution o p t i o n which enables the engineer t o set a r e . it is u n r e a l i s t i c f o r r e s i d u a l o i l droplets. b u t . T h i s aSSumptiOn of instantaneous e q u i l i b r i u m is u s u a l l y also made i n compositional s i m u l a t o r s .

6 ] .. 1. The omission of a p p a r e n t l y minor terme from the Jacobian can l i m i t convergence of t h e non l i n e a r e q u a t i o n s unacceptably. gas 8 a t u r a t i o n . w i t h b g and rS-r s (Po+Pcog( 0 ) ) . S -0 Po... 5 ] .. v a r i a b l e a t a l l times.. Sw and S as s o l u t i o n varia%. Swandr a r e s o l u t i o n v a r i a b l e s .. I n t h e u n d e r s a t u r a t e d o i l case t h e gas e q u a t i o n is b e i n g converged t o a known s o l u t i o n . Po. rs-rs ( Po+Pcq?kg fi (iii) Liquid only.. such methods run i n t o t r o u b l e &en a d r y vapour or d i s s o l v e d g a s s l u g is propogated. the bubble p o i n t is implicit. P g variables. and v a r i a b l e s u b s t i t u t i o n has proved s u p e r i o r . and a d j u s t s R (Pb) t o s a t i s f y phase e q u i l i b r i u m . Depending on the c o n d i t i o n s i n a cell. b u t n o t f u l l y expanded. 9 - V a r i a b l e s u b s t i t u t i o n does n o t attempt t o r e t a i n gas S a t u r a t i o n as the t h i r d bubble p o i n t or vapour oil-gas ratio. If g a s i a e c t i o n o c c u r s i n t o u n d e r s a t u r a t e d 0 1for example... Both methods have been used i n PORES. may be the primary s o l u t i o n v a r i a b l e .. as t h e y modify the s l u g shape by sharpening t h e l e a d i n g edge. R. (1) r sat( P ) is t h e curve d e s c r i b i n g the oil-gas ratio for vagour in e q u i l i b r i u m w i t h lfquid 811. Unless t h i s is done e x a c t l y . w i t h Pb-Po. t h e exact c a l c u l a t i o n of i n t e r b l o c k and w e l l flows a t the advanced time level.. more p r e c i s e l y .es. as t h e p r e s s u r e changes which occur when free gas goes i n t o s o l u t i o n d i s t u r b the i n t e r b l o c k flows i n a manner n o t i n c o r p o r a t e d i n t o the Jacobian of the newtonian i t e r a t i o n . although it i n v o l v e s o r g a n i s a t i o n a l d i f f i c u l t i e s i n keeping trackof whether t h e t h i r d s o l u t i o n v a r i a b l e is S P o r r The cell by cell b 8' technique r e t a i n s Po. as i n the pseudo s o l u t i o n gas method. w i t h S -1-S andPb-Po s o w S and S are Solution ( ii ) Liquid and vapour hydrocarbon A t e s e n t . It is possible t o c o n t r o l t h i s d i s p e r s i o n for t h e simple case of fret g a s invading o i l by o n l y allowing R t o rise due t o c o n t a c t w i t h f r e e vapour. r .. Assuminginterphase e q u i l i b r i u m .443 o c c u r s across an R g r a d i e n t .. s and sat P are s o l u t i o n v a r i a b l e s . is knoun 8nd included i n the Jacobian.... However..I is i n c r e a s e d .. S 0 .. b u t r u n s i n t o convergence problem on l o n g t i m e steps. p r e v e n t s i n s t a b i l i t i e s which can occur using first order approximations t o the implicit flows [ll]. I n each case it is c r u c i a l that the f u n c t f o n a l dependence o f the secondary variables. These correspond t o s a t u r a t i o n p r e s s u r e and flash technique8 i n compositional s i m u l a t i o n [ 5 . which is o b t a i n e d w i t h i n c r e a s i n g accuracy as the Newtonian i t e r a t i o n converges. This y i e l d s a working scheme.. there areonlythreepossibilities f o r t h e state of a cell:- (i) Vapouronly. . and one cell methods i n which cell properties are modified t o be c o n s i s t e n t w i t h t h e s o l u t i o n i n terms of a f i x e d set of v a r i a b l e s [3]. MODELLING BUBBLE AND DEW POINT VARIATIONS Two main approaches e x i s t t o modelling mass t r a n s f e r i n black o i l s i m u l a t o r s : the v a r i a b l e s u b s t i t u t i o n method [ 4 . This can be overcome by u s i n g a one cell Newtonian i t e r a t i o n toexact m a t e r i a l b a l a n c e t o f i x t h e Rs change p r e c i s e l y .. a n e g a t i v e gas s a t u r a t i o n is o b t a i n e d on t h e subsequent i t e r a t i o n and g e n e r a l l y c a u s e s m a t e r i a l b a l a n c e errors.. However. and of f u n c t i o n s of t h e s e secondary bariables. ( s u c h as Pb and r i n a cell i n which S is the primary v a r i a b l e ) ...

if rs)rs a t (Po+P (1-5 ) ) .. S e t S -O.Sg-l-Sw. an EOR type process expected t o some extent i n most reservoirs with gaa injection. S e t Pb-PO... (2) j-l.. aa c e l l s i n which r (4) is the Solution variable would o t h e h s e be repeatedly solved for a constane gas saturation of zero..444 The maas conservation equations take the form Rj - 1 TAT -[m... as Elements of the residual.. " d g. as follows:- From s t a t e s (i). cog w 0 9 .. cog g c h a n g e t o (i) (ii).. The extra cost of aodelling r variation is s m a l l . S -6.. water and free gaa flows given by Darcy's way. derivatives being taken w i t h respect t o the primary solution variables for each c e l l ... where fo Law (3) fw i n t%' us& and fg are the free o i l . on the -is of the c u r r e n t approximation t o the advanced time step solution.. m s terms... i f Pb)Po...w are the corresponding w e l l terms. change t o ( l i ) . W e re is not the prirary variable... . J AT - mT] j - qJ TAT - & fnj n T+AT I .... S e t rg-rs sat (Po+P ( S ))-c. 9 g change t o ( i i l ) From s t a t e (iii).. effects such as the vapourisation of residual o i l into re-injected gas. aa w e l l as gaa solution and a primary recovery waterflood. w e l l term and flows have a three vector formr- Rj qj W + R s q Ow 1 j ... the effeck is merely t o add extra hn terme t o the Jacobian... N the number of c e l l s .N . i f S < O . R.......Pb-Po-c..... This is essentially a combination of the methods proposed by Cook et a1 [ 2 ] and Spivak and Dixon [l]. Transitions may occur between t h e three states of (l).... A The equatioagiven by R ( X w A T a are solved by Newtonian iteration.. This enable.Sg-~... S e t rs-r s a t ... t o be followed. c h a n g e t o (ii) Prom s t a t e From s t a t e (ii).. i f So<O...

. dofinad by 9 9 J 9. and acme of the convergence properties of the f u l l Newton method are lost. Both it...p-DJg. The J a c o b i a n is t h e n n o t t h e derivative of the r e s i d u a l . a r e e t q u i v a l e n t t o those o b t a i n e d from a total gaa r e s i d u a l . The a l t e r n a t i v e is t o u s e a s t r a i g h t f o r w a r d total gas formulation. of two large c a n c e l l i n g term. Rxg I R8R ... which. (6) where Rtg is approximate... such that i t e r a t i o n i 8 X -6( As ).. water and total gaa. w h i l e i n a total g u &-lation R is g i v e n by a approrirate m a t r i x %tini%n Ro-D( Rf +R R ) If2the ersors i n t h e l i q u i d and vapour flaw are camQarab1e.g 9 X . i n v o l v e s t h e matrix d e c o l p o s i t i o n r - IJgrp Ix3 J 0.. p)-bo-D(Rf<RsXe I f a free gas f o r m u l a t i o n is used.9-Jo. g i v e n by R X : - J ( Jo .. The dissolved gaa c o n t r i b u t i o n swampa the free component i n t h e e q u a t i o n determining t h e thd '.. i n the t o t a l gaa caae.. This c o n s t r a i n t ..... which g e n e r a l l y speeds convergence. It is possible n o t t o d i f f e r e n t i a t e t h e R t e r m i n the r e s i d u a l ..p J O 919 I ... 9.R t g 9 x ..... This is p o s n i b l e for simultaneous s o l u t i o n methods..Y - 9 .. - and D is a d i a g o n a l m a t r i x .&$ l l n e a r solver and s e q u e n t i a l method u s e column sum methods to preserve zero r e s i d u a l s u r .. I n t h e free gaa cam R f g c o n s i s t s f a n approxisate m a t r i x a c t i n g o n R -D R . gaa s a t u r a t i o n ..p J 019 Jg.s f f e c t i v e l y material balance on d i a g o n a l planes of cells.. The corresponding free gaa r e s i d u a l is?'-R "-R R A free gaa f o m l a t i o n was o r i g i n a l l y used i n PORES. (5) S R The elements of Re are g i v e n by ( R ) The change i n ASg ov&iaN&Anfat.. and the r e s u l t i n g v a l u e s of As are less a c c u r a t e than i n the f n e QMC a m " . . i f s o l v e d exactly.gl -I 1 O JolP-DJglp llJg.. The S e q u e n t i a l method of s o l v i n g t h e linear eqUatiOM.445 TBE SUBTRACTED 'IDTAL GAS PO-TION The r e s i d u a l i n ( 3 ) i n c l u d e s t e for t o t a l o i l ... as free gaa r e s i d u a l s w does not correspond t o material b a l a n c e i f s a t u r a t e d o i l is p r e s e n t . b u t t h e s e q u e n t i a l method fails completely. haa l i t t l e v a l u e i f a free gas f o r m u l a t i o n is used.... one of which i 8 agproriute...... X 9 is g i v e n b~ Zr or8 arise i n t h e e v a l u a t i o n of t h e right-hand aide of (b).g. l e a d i n g t o a eet of e q u a t i o n s .k 8 O( 10 t R 10 ). i n the simple case of a noncondensate gaa o i l system...R . The right-hand s i d e of t h e e q u a t i o n d e f i n i n g gaa s a t u r a 6 o M cowk8.

c a u s i n g r e s i d u a l s which the s i m u l a t o r converges o u t a c c u r a t e l y .... all THE S O u l T I O N OF GAS I N O I L Black o i l s i m u l a t o rs h a v e tended n o t t o p r o v i d e the e n g i n e e r w i t h v e r y comprehensive facilities for i n v e s t i g a t i n g gaa S o l u t i o n effects.. So- 1-s -S w g' and the iteration The advantages of both f o r m u l a t i o n s may be combined i n a subtracted t o t a l g a s formulation... The best eema 80 be$. The s e q u e n t i a l method can s t i l l f a i l when a l a r g e i n i t i a l R gradient exists across t h e reeerpoir.. It seem clear that r e s i d u a l o i l droplets w i l l e q u i l i b r a t e q u i c k l y .P 9 0 ... For a time step from T t o *AT. the degree of e q u i l i b r i u m between phases w i l l be d i f f e r e n t for r e s i d u a l o i l i n a g a s cap from that a t t a i n e d i n t h e case o f g a s i n j e c t i o n i n t o u n d e r s a t u r a t e d o i l ... b u t a l metre diameter area of o i l w i l l take over a Y e a r t o s a t u r a t e i f f r e e gaa channels past it. that is r e m v e d is the b u l k of the i n i t i a l d i s s o l v e d g a s which otherwise c a u s e s the gaa e q u a t i o n t o be a n e a r m u l t i p l e of the o i l equation. The continuous updating of R~ a d seem e s s e n t i a l .. t h e partial r e .s o l u t i o n o p t i o n r a y be sunmurised as.. i n which case a f u l l y simultaneous s o l u t i o n k t h o d must be used.. Pb is s o l u t i o n v a r i a b l e I f PT < Pb T+AT <Po T+AT then sT+ATIOl g pT+AT_pT bmin If bmin T+ATlpT+AT bmin then S T+ATIO 9 I P bmin b Saturated o i l . however. o f the o i l i n a c e l l which is i n close c o n t a c t w i t h vapour. P a r t l y . If t h i s is a major effect. as i n t h e Odeh test problem [9].( 8 ) The choice o f R sub is rather critical. f .. Providing no.. S g is s o l u t i o n v a r i a b l e T T+AT T+AT T+ATIMin I f S >O t h e n Pb -P .r e .. and these processes may occur i n t h e same s t u d y . The PORES partial r e .*..and total.. s t a n d o u t o v e r the R s u b t r a c t i o n .446 Errors i n S are fed back i n t o the o i l r e s i d u a l v i a o f t e n diveqges..- Undersaturated o i l .s o l u t i o n o p t i o n s e n a b l e s t h e s e n s i t i v i t y of t h e problem t o g a s s o l u t i o n effects t o be e s t a b l i s h e d .. updated by the predictor a t t h e start of each tim step... The An USC d R rather t h a n z s can i n c r e a s e run times by 50%. i n which the g a s e q u a t i o n r e s i d u a l is Rs t g - Rg - R~ 's " R 0 . I n p a r t i c u l a r .... s u c h a s those due t o gaa i n j e c t i o n . t h i s is due t o a lack of knowledge concerning the processes involved..*. there are no facilities for h i s t o r y matching. sub Anadvantage o f t h i s m e t h o d i s t h a t i s o l a t e d d i s s o l v e d gaa changes.s o l u t i o n o p t i o n a l l o w s t h e u s e r t o d e f i n e a r e . and s e v e r a l a l t e r n a t i v e s have been t r i e d .s o l u t i o n or e q u i l i b r i u m f r a c t i o n .

b u t it seem u n r e a l i s t i c t o ascribe d i f f e r e n t behaviours t o gas at j u s t under and over c r i t i c a l s a t u r a t i o n .. and is c o m u n i c a t e d t o its neighbours by o i l flows. o i l s a t u r a t i o n s remain less t h a n critical.. I n cases of h i g h throughput ratio. such as re-solution i n r e s i d u a l o i l . due t o g r a v i t y segr e g a t i o n .s o l u t i o n scheme is less t h a n ideal I ( i ) I f p r e s s u r e drops through the bubble p o i n t of t h e by-passed o i l . The e q u i v a l e n t of g u i n j e c t i o n r a r e l y arises. for example. as d i f f u s i o n is limited t o one block per time etep.PT+AT-P TtAT .. i n t h e same time step.PT+AT_Min b 0 bmin T .. o f f . This c o u l d be allowed. However. however. which e f f e c t i v e l y impose an upper l i m i t of AX/AT on t h e speed a t which d i s s o l v e d g a s or vapourised o i l may propopate. I n a d d i t i o n . There are s e v e r a l respects i n which t h i s partial r e .. For gas i n j e c t i o n .. it would be expected that t r a n s v e r s e s a t u r a t i o n would o c c u r behind the f r o n t . O t h e r possibilities exist for partial r e ..s o l u t i o n . t h i s is e q u i v a l e n t t o t o t a l r e . gas appearing I f Pb 9 s e t t o E. t h e n it w i l l not r e . d i s c o n t i n u o u s changes i n the f u n c t i o n a l form of t h e r e s i d u a l can s l o w convergence. and gas comes o u t of s o l u t i o n a t less t h a n m o b i l e s a t u r a t i o n . This problem might be expected t o be r a t h e r less s e v e r e i n IHPES type s i m u l a t o r s . due t o convection and maas t r a n s f e r u s u a l l y show d i s p e r s i o n effects. The r v a l u e can be followedfir t h e vapour as described i n S e c t i o n 2 .... When. This model y i e l d s t o t a l and no-resolution as the limits of f-1 and 0 r e s p e c t i v e l y . such a s i m u l a t o r i e l i k e l y t o t a k e more steps than a f u l l y implicit one t o s o l v e a given problem. t o t h e next cell. I n most condensate s t u d i e s . There is also t h e d i f f i c u l t y of i d e n t i f y i n g t h e r e s i d u a l o i l i f the cell is subsequently f l u s h e d w i t h mobile o i l . when gaa displacement is stable.p r e s e u r i s a t i o n .. . This does n o t o c c u r i n t h i s model. DISPERSION PROBLElls I N M S TRANSPER AS V a r i a t i o n s i n Rs and Is. (ii) When f i n g e r i n g or c h a n n e l l i n g occurs. r e s i d u a l o i l s a t u r a t i o n is r a r e l y a t t a i n e d . a c o n s i d e r a b l e f r a c t i o n of t h e reservoir r a y need t o be s a t u r a t e d before free g a s appears i n the i n j e c t i o n cell. the bubble p o i n t rises i n a cell due t o gas s o l u t i o n t h i s rise is assumed t o o c c u r evenly throughout the e n t i r e cell. so t h a t e q u i l i b r i u m is a r e a s o n a b l e assumption.s e t t i n g t h e advantage of lover d i s p e r s i o n per time step.s o l u t i o n o p t i o n s ... ( 9 ) i s t h e minimum bubble p o i n t a t t a i n e d by t h e c e l l d u r i n g the s i m u l a t i o n .... The r e s u l t i n g rise i n R for the neighbouring cell s is t h e n passed.447 B u b b l e p o i n t t r a n s i t i o n .d i s s o l v e i n t h e non-equilibrium f r a c t i o n or r e . The q u a n t i t y of d i s s o l v e d g a s i n t h e cell and i n t e r b l o c k flows is o b t a i n e d as a weighted average u s i n g Pban a c t i n g as bubble p o i n t for the 'trapped' or non-equilibrium o i l f r a c t i o n . A l l function& d e r i v a t i v e s can be expanded i n t h e JaCobianInlkrms of t h e prvariables. I t would be possible t o have a similar o p t i o n f o r o i l v a p o u r i s a t i o n . However.

Front sharpening methods w i l l t e n d to d i s t o r t such s l u g s by sharpening the l e a d i n g edge. However. where g a s i n j e c t i o n may be c o n t r o l l e d by production rates and s a l e a c o n t r a c t s . t h i s method of d i s p e r s i o n c o n t r o l r u n s i n t o problems when the convection of s l u g s is considered. . so t h a t a n i n c r e a s e d R v a l u e w i l l not appear a t the downstream i n t e r f a c e u n t i l the cell is s a t u r a t a . CASE 2. 1. the r e s u l t i n g r d i s t r i b u t i o n s may not have simple shapes. of such a technique on the c a m 2 odeh problem are shown i n f i g .448 For simple g a s i n j e c t i o n i n t o undersaturated o i l . The result. P a r t i c u l a r l y i n the came of condensate reservoirs. it is f a i r l y e a s y t o p r e v e n t t h i s d i s p e r s i o n by assuming t h a t g a s p a s s i n g i n t o a cell s a t u r a t e o i l o n l y as it o v e r r i d e s it. GOR VS T I M E cy 0 U 0 0 1 2 3 4 5 6 7 8 9 10 T I M E I N YEARS Figure 1. I n t e r b l o c k flow schemes other t h a n upstreamlng w i l l y i e l d unphysical r e s u l t s i f f l o w o c c u r s fromacellof zero re t o o n e o f f i n i t e r b u t upatreaming c a u s e s 8' unacceptable a r t i f i c i a l d i f f u 8 i o n .

Minor o i l vapouris a t i o n and condensate effects w i l l occur i n many s t u d i e s . a c o n t i n u o u s l y modified subtracted total gas formulation is preferable t o either total or f r e e g a s formulationn. CONCLUSIONS ( i ) It is possible t o model g a s condensate and bubble p o i n t e f f e c t s simult a n e o u s l y i n an e f f i c i e n t g e n e r a l purpose black o i l s i m u l a t o r . The b a s i c a l l y c o n v e c t i v e n a t u r e of r and R transport s u g g e s t s a p o i n t following algorithm.449 W e can p r e s e n t no simple s o l u t i o n t o the d i s p e r s i o n problem.and total. A r e . and it may be t h a t it is i n h e r e n t i n the concept of u s i n g a dew or bubble p o i n t model r a t h e r t h a n a separate e q u a t i o n . d e f i n e d aa ~ e V l u m e p b P P I /P b xb. phase p P : The . P . ( i v ) A l l bubble and dew p o i n t models are a poor excuse f o r s o l v i n g a d i s s o l v e d gas or vapourised o i l e q u a t i o n . b u t d i s p e r s i o n c o n t r o l methods may d i s t o r t t h e shape of convected s l u g s . b Ir ~ a ~ e ~ ~ ~ r f O ~ u ~ ~ factoro for phase p. where p is f l u i d d e n s i t y Pr f nj to ' p rock j I The flow rate. measured i n term of s u r f a c e volume. ( i i ) There is a need t o p r o v i d e e n g i n e e r s w i t h a more flexible method of matching g a s s o l u t i o n e f f e c t s . from w e l l w t o cell j r e s i d u a l of the maas convervation e q u a t i o n for cell j .t y p e e q u a t i o n t o t h e t r a d i t i o n a l black o i l p i c t u r e t h a n t o go t o t h e number o f cells r e q u i r e d t o reduce d i s p e r s i o n t o an acceptable l e v e l . although those s u g g e s t e d t o date are n o t f u l l y implicit. It may be cheaper.s o l u t i o n o p t i o n s are o b t a i n e d as l i m i t i n g caees. to add an extra t r a n s p o r t . No. however. such as d r y gaa invading u n d e r s a t u r a t e d o i l . and these can be included a t l i t t l e extra cost. measured i n terms o f s u r f a c e volume.r e . and has a simple f u n c t i o n a l form which is p a r t i c u l a r l y s u i t a b l e for i m p l i c i t s i m u l a t o r s . (iii) For s e q u e n t i a l methods. This can be c o n t r o l l e d i n Simple cases.s o l u t i o n f r a c t i o n approach e n a b l e s o i l by. from cell n cell N : The I numbsr of a c t i v e cells i n the reservoir P P bj dj T e bubble p o i n t p r e s s u r e of cell j h : The I dew p o i n t p r e s r u r e o f cell j P Pj S The p r e s s u r e of phase p i n cell j Pj : The : s a t u r a t i o n o f phase p i n cell j P qjw R j The rate of flow. numerical d i s p e r s i o n o f vaporised o i l and d i s s o l v e d g a s can be s i g n i f i c a n t . This may be p a r t i c u l a r l y important for g a s i n j e c t i o n i n t o condensate reservoirs. I n p a r t i c u l a r .pass and c h a n n e l l i n g e f f e c t s t o be included. of phase p.

"Numerical simulation of compositional phenomena i n petroleum reservoirs". and DIXON. "Comparison o f solutions t o a three d i m e n s i o n a l black-oil reservoir s i m u l a t i o n problem". SPE R e p r i n t S e r i e s . European O f f s h o r e Conference and E x h i b i t i o n .J. t h r e e p h a s e reservoir s i m u l a t o r . (March 1965). milas.14.Eng.. ed. Soc. D.Eng. " ~ n efficient f u l l y implicit simulator". ODEE.. J "A beta-type reservoir S i m u l a t o r for approximating compositional e f f e c t s d u r i n g g a s i n j e c t i o n " . NGEEIU. on Numerical S i m u l a t i o n o f Reservoir Performance. Soc. and RAWZSH.. P. 62-76. "An e q u a t i o n of state c o m p o s i t i o n a l Podel". and E x h i b i t i o n o f SPE.Pet. P. J. EERIE. Proc. 5 5 t h Ann. 1 . U. COOIC. 1974 porous Trans.K. 6. K. and E x h i b i t i o n of SPE. 8. (1973). and NIELSEN. "Reservoir simulation: A g e n e r a l model f o r m u l a t i o n and associated 1 p h y s i c a l / n u m s r i c a l sources o f i n s t a b i l i t y " . 5. Soc.D. K. 13-25 CBESHIRC. (06. A.. f u l l y implicit.H.V a<iaSj x Pj : T e pth p r i m a r y h solution v a r i a b l e for cell j s Rs :The minimum Re v a l u e i n the reservoir model : The mean Rs value i n the reservoir model REFERENCES 1. 51-59. A. SPE9306. 7. "Compositional modelling w i t h an e q u a t i o n of state". Uiller. Proc. L.. Bode P r e s s . "A s t r o n g l y coupled.H. AIM! 234 SAWDRu.. and VINSOM!.. A. COATS. 1974). COATS. Las Vegas. i n j e c t i o n " . and m C S . .Pet. (Jan.J. 1980. 3rd symp. milas. P r o c . BAILl Conf. F a l l Conf. 1980. P a l l Conf. . 1 9 8 l ) .N. . June 1980. 11. A.. 4. T e c h n i q u e s for f u l l y -licit resemoir s i m u l a t i o n " . No.P.. P a l l Conf.Tech. AU.K. 1980. P r o c . 471-481 3. JACOBI... " s i m u l a t i o n of g a s condensate reseervoirs". R.Co. 55th Ann. BANSAL. 269. 1. 1974. et al.. Houston. NO=.J.4 50 J 'j : The element of the Jacobian.Eng. 8 . R E .. RUBIN. T. Dublin. K.. Proc.B. P r o c .H.A.. London.H. SPE8329. e t al. 1980 ). J. and AZIZ. SPIVAII. under g a s 9. 1973. RAIIIOM)I. PONG. P. 363-376 (Oct.Pet. 2. R.. SPE4271. J. "naes transfer between phases i n a R A0 medium: A s t u d y of equilibrium".X. " ~ y n d c s of petroleum reservoir. -79. A. R. and E x h i b i t i o n o f sPE.J.Pet. 325 10. 5 4 t h Ann. K. t h r e e d i m e n s i o n a l .S. Gulf Pub.

The m i x e r was tested w i t h o i l from German o i l r e s e r v o i r . The appa- r a t u s p e r m i t s t h e c o n d u c t i n g of f l o o d e x p e r i m e n t s w i t h l i v i n g o i l w i t h i n a p r e s s u r e r a n g e b e t w e e n 1 and 600 bar a t f l o o d i n g r a t e s from 1 t o 50 cm3.h-l. The c o n c e n t r a t i o n s of t h e components C 0 2 . The f l o o d i n g e x p e r i m e n t s are t o be c o n d u c t e d w i t h t h e t r a n s i t i o n zon e s p r e v i o u s l y e s t a b l i s h e d . a new e x p e r i m e n t a l s e t . B e c a u s e t h e p e r f o m a n c e of core f l o o d i n g e x p e r i m e n t s i s f e a s i b l e onl y w i t h i n a c e r t a i n l e n g t h l i m i t . T h i s d e s i g n i s based on a p u b l i c a t i o n by WATKINS. which cons i s t s of i n c l i n e d p i p e s w i t h a t o t a l l e n g t h o f a b o u t 6 m . With t h e h e l p of t h e s e e x p e r i m e n t s . a n a l y s i s of s t o c k t a n k o i l up t o C26. w h e r e a s t h e d e v e l o p m e n t of t h e t r a n s i t i o n z o n e n e a r t h e m i n i m a l miscibility pressure requires a f l o o d d i s t a n c e of a t l e a s t 1 m . g a s c h r o m a t o g r a p h i c a l g a s a n a l y s i s u p t o .u p h a s been During t h e e x p e r i m e n t . CORE FLOODING VOLKER MEYN Institut f i r Tiefbohrkunde und Erdolgewinnung der TU clmrsthal ABSTRACT A newly d e v e l o p e d core f l o o d i n g a p p a r a t u s i s described. The f o l l o w i n g d a t a a r e t h e r e b y mea- tested. i t c a n be d e m o n s t r a t e d t h a t s u c h a mixer i s cap a b l e of p r e p a r i n g a p h a s e whose c o m p o s i t i o n s i m u l a t e s t h a t i n t h e r e a l t r a n s i t i o n z o n e . The e s t a b l i s h m e n t of t h e t r a n s i t i o n z o n e d u r i n g the f l o o d p r o c e s s i s s i m u l a t e d i n a t h r e e . were recorded gas-chroma- t o g r a p h i c a l l y a t t h e o u t p u t of t h e m i x i n g d e v i c e .EXPERIMENTAL TECHNIQUES 451 A NOVEL DEVICE FOR CO. number of moles of g a s p r o d u c e d .s t a g e m i x i n g d e v i c e . p r e s s u r e d i f f e r e n c e . e v e n i n t h e v i c i n i t y of t h e minimal miscib i l i t y pressure. water c u t p r o d u c e d . . t h e mass f l o w of C02 a t t h e i n p u t i s h e l d c o n s t a n t . C 7 . s u r ed : o i l p r o d u c t i o n . a s w e l l a s C1 t o C Z 6 .

u p i s d e s i g n e d f o r a p r e s s u r e u p t o 6 0 0 bar and a t e m p e r a t u r e u p t o 150OC. I n order t o a l l o w a measurement of t h e u n i t d i s p l a c e m e n t e f f i c i e n c y . e v e n i n " s h o r t " r e s e r v o i r models. It c o m p r i s e s a pumping u n i t f o r i n j e c t i n g t h e C 0 2 . F o r t h i s r e a s o n a c e l l of t h e Hassler t y p e c o n s i s t i n g o n l y of T e f l o n (PTFE) and s t a i n l e s s s t e e l h a s b e e n d e v e l o p e d . s i n c e t h e c o n s t r u c t i o n of t h e t r a n s i t i o n z o n e r e q u i r e s .452 INTRODUCTION L a b o r a t o r y i n v e s t i g a t i o n s a r e b e i n g p e r f o r m e d i n t h e c o u r s e of a p r o j e c t /1/ c o n c e r n i n g t h e p o s s i b i l i t i e s of C 0 2 f l o o d i n g i n West Germany. t h e embedding. o r g a n i c s e a l i n g m a t e r i a l s t e n d t o s w e l l and b l i s t e r u n d e r t h e s e c o n d i t i o n s . s l i m cores a r e b e i n g employed d u r i n g a n i n i t i a l p h a s e . However. I n order t o d e m o n s t r a t e s u c h a p o s s i - b i l i t y .p r e s s u r e C 0 2 and t h e t e m p e r a t u r e s u p t o 12OoC. F o l l o w i n g t h i s c o n c e p t . I n g e n e r a l . o n l y a s h o r t l e n g t h is n e c e s s a r y f o r c o n s t r u c t i n g t h e t r a n s i t i o n zone i n t h i s c a s e . A r e q u i r e m e n t f o r t h e u s e of black o i l s i m u l a t o r s i s t h a t t h e t r a n - s i t i o n z o n e i s r e s t r i c t e d t o a s i n g l e c e l l . a length of a t least 1 m /2/. m i x e r tests and c o m p a r a b l e s l i m t u b e e x p e r i m e n t s h a v e been conducted. Such a r e q u i r e m e n t c a n n o t be s a t i s f i e d i n t h e c a s e of c o r e f l o o d e x p e r i m e n t s w i t h p u r e C 0 2 . e s p e c i a l l y of s l i m c o r e s . and t h e a n a l y t i c a l e q u i p m e n t . a d e v i c e which s h o u l d be s u i t e d f o r b o t h s l i m t u b e tests and c o r e f l o o d e x p e r i m e n t s h a s b e e n d e v e l o p e d . t h e f l o o d t u b e . WATKINS / 3 / h a s d e m o n s t r a t e d t h a t t h e r e s i d u a l o i l s a t u r a t i o n c a n be s u b s t a n t i a l l y r e d u c e d . . EXPERIMENTAL SET-UP The s e t . A main o b j e c t i v e i s t o p r o v i d e e x p e r i m e n t a l d a t a f o r a simulation study. by t h e u s e of a p r e m i x i n g v e s s e l . i m p o s e s d i f f i c u l t i e s b e c a u s e of t h e h i g h . C o n s e q u e n t l y . t h e u s e of a p r e m i x e r s h o u l d p e r m i t d i s p l a c e m e n t by a medium whose c o m p o s i t i o n i s s i m i l a r t o t h a t of t h e t r a n s i t i o n z o n e . A f u r t h e r d i f f i c u l t y a r i s e s from t h e i n v a s i o n of t h e core by t h e a d h e s i v e . For t h e f l o o d e x p e r i m e n t s . For t h i s r e a s o n t h e t r a n s i t i o n z o n e must be s h o r t .

i n o r d e r t o keep t h e c o m p r e s s i b i l i t y low. I n t h e c a s e of t h e s l i m t u b e t e s t s . 1) is d e s i g n e d such t h a t t h e C 0 2 mass f l o w a t t h e i n l e t i s m a i n t a i n e d c o n s t a n t .rage v e s s e l .T h e pumping r a t e c a n be v a r i e d m w i t h i n a r a n g e from 0. C 0 2 is d i s p l a c e d by mercury a t a c o n s t a n t f l o w r a t e from a s t o r a g e vessel ( 5 ) . 3 m i x e r tank. The t u b e bundles can be e a s i l y emptied and t h e r e f o r e r e u s e d o f t e n . According t o p r e v i o u s e x p e r i e n c e . The aim is t o a c h i e v e a t o t a l 1 l e n g t h of 2 m . 5 C02-sto. tee f i t t i n g s were i n s t a l l e d a f t e r e v e r y 4 m of l e n g t h . The p r e s s u r e i n t h e s t o r a g e vessel i s h e l d c o n s t a n t by means of t h e .453 8 mesitylene F i g . For t h i s p u r p o s e . 6 d i s p l a c e m e n t pumps The set-up ( f i g . b a c k . 2 back p r e s s u r e r e g u l a t o r .p r e s s u r e r e g u l a t o r ( 1 ) . For a n a l y t i c a l r e a s o n s a c o m p a r a t i v e l y l a r g e d i a m e t e r of 0 . I n o r d e r t o f a c i l i t a t e t h e packing of t h e f l o o d t u b e . 1: Set up of t h e f l o o d d e v i c e : 1 . The f l o o d t u b e i s immersed i n a t h e r m o s t a t i c o i l b a t h which c a n accomodate a t o t a l l e n g t h of 30 m . 4 f l o o d t u b e . c o r e s up t o 3 . 8 7 5 c m was chosen f o r t h e t u b i n g . For t h e c o r e f l o o d e x p e r i m e n t s of t h e f i r s t p h a s e . i n which t h e p r e s s u r e and t e m p e r a t u r e a r e m a i n t a i n e d con- stant. t h e f l o o d t u b e ( 4 ) c o n s i s t s of s t r a i g h t t u b e s e c t i o n s 2 m i n l e n g t h connected by elbows.4 t o 5 0 c / h . T h e s t o r a g e v e s s e l is t h e r m o s t a t e d a t a t e m p e r a t u r e below t h e c r i t i c a l v a l u e . a d i a m e t e r of a b o u t 1 mm was s e l e c t e d .

b r i n e p r o d u c t i o n and g a s p r o d u c t i o n . The d r i l l i n g of l o n g e r c o r e s p r e s e n t s d i f f i c u l t i e s w i t h t h e s a n d s t o n e u s e d h e r e . i n order t o attach a pressure transducer. The c o r e s e c t i o n s a r e i n s e r t e d i n t o a Teflon sleeve. i t i s p o s s i b l e t o d r i l l a h o l e t h r o u g h t h e p a c k e r down t o t h e core a f t e r t h e a s s e m b l y . To s i m p l i f y t h e a s s e m b l y .454 200 mm i n l e n g t h and w i t h a d i a m e t e r of 11 mm c a n be d r i l l e d w i t h o u t d i f f i c u l t y . 3 ) i s i n t e n d e d t o c o l l e c t s u c h d a t a a s o i l p r o d u c t i o n . S i n c e a s i n g - l e p a c k e r s e a l s a t two p o s i t i o n s . 2: Packer assembly: 1 c o r e . t h e t u b e c o n s i s t s of f o u r p a r t s . Each f l a n g e i n c l u d e s a p o r t f o r c o n n e c t i n g a p r e s s u r e transducer. 3 p a c k e r I n order t o p r e v e n t l e a k a g e d u e t o t h e f l o w of t h e T e f l o n a t e l e v a t e d t e m p e r a t u r e s . A n a l y t i c a l equipment The a n a l y t i c a l e q u i p m e n t ( f i g . 2). t h e packer is d e s i g n e d as an a u t o m a t i c g a s k e t . The number of moles of g a s i s d e t e r m i n e d by means of a p r e s s u r e . This s l e e v e is i n s t a l l e d i n t h e s t a i n l e s s s t e e l body by means of p a c k e r s ( f i g . a s w e l l a s c o m p o s i t i o n of t h e g a s u p t o C. and of s t o c k t a n k O i l / b r i n e i s s e p a r a t e d from t h e g a s i n a s m a l l p a c k e d column (1): t h e gas i s s u b s e q u e n t l y withdrawn i n t o e v a c u a t e d v e s s e l s . 2 PTFE s l e e v e . F i g . o i l up to C26.

I n order t o m a i n t a i n t h e p r e s s u r e i n t h e s e p a r a t o r r e g u l a t o r h a s been i n s t a l l e d . An e l e c t r o s t a t i c o i l / b r i n e s e p a r a - t o r w a s t h e r e f o r e i n s t a l l e d ( f i g . The c h a n n e l s a r e p r e d o m i n a n t l y 5 x 5 mm i n s i z e . I n p r e l i m i n a r y t e s t s i t became e v i d e n t t h a t e m u l s i o n s c a n be p r o d u c e d w i t h t h e o i l employed. A bypass w i t h a sample v a l v e ( 3 ) t o a g a s chromatograph ( P e r k i n E l m e r Sigma 1) i s i n c l u d e d . B b r i n e / w e i r . 1 p e r c e n t a f t e r t h e s e p a r a t i o n . A l o c a t i o n of t h e b r i n e / water c o n t a c t . 4 : S k e t c h of t h e o i l / b r i n e s e p a r a t o r 0 o i l / w e i r . a back-pressure a d v a n t a g e o f s u c h a s e t . 3 sample valve. 4 ) . 3: A n a l y t i c a l e q u i p m e n t : 1 g a s / o i l s e p a r a t o r . Gor - . The c o n s t a n t .u p i s t h e f a c t t h a t t h e measurement is l a r g e l y i n d e p e n d e n t of t h e p r o d u c t i o n r a t e and t h a t t h e method o f measurement is c u m u l a t i v e . 2 b r i n e l o i l s e p a r a t o r . The water c o n t e n t i n t h e s t o c k t a n k o i l produced w a s less t h a n 0 .455 F i g . T h i s s e p a r a t o r h a s been milled from P l e x i g l a s . 4 evacuated v e s s e l s measurement.GOS F i g . The v o l t a g e of 60 V o v e r t h e p l a t e s i s s u f f i c i e n t for s e p a r a t i o n .

means of t h e c h a n g e i n c o n d u c t i v i t y when t h e o i l / b r i n e c o n t a c t p a s s e s an electrode.b r i n e s e p a r a t o r i s n o t f e a s i b l e b e c a u s e of t h e l a r g e dead volume and o f t h e r e s u l t i n g r e m i x i n g . . g a s chrom a t o g r a p h i c a l a n a l y s e s a r e p e r f o r m e d on b o t h t h e g a s and t h e o i l . t h e q u a n t i t i e s of b r i n e and o i l i n t h e s e p a r a t o r must be k e p t c o n s t a n t w i t h i n L 0 . T h e r e f o r e t h e s e p a r a t o r i s a u t o m a t i c a l l y t i l t e d when t h e o i l b r i n e c o n t a c t a t A d e v i a t e s from t h e s e t p o i n t . I n order t o c i r c u m v e n t t h e r e m a i n i n g p r o blems w i t h t h e bond stress ( c r e e p i n g of o i l . r e s p e c t i v e l y . c a r r i e d o u t s i m u l t a n e o u s l y i n t h e g a s c h r o m a t o g r a p h . The o i l and b r i n e which emerge from t h e s e p a r a t o r a r e weighed i n c o l l e c t i n g bottles. S a m p l i n g b e h i n d t h e G a s and o i l a n a l y s e s a r e o i l .t a n k the gas-oil o i l a r e w i t h d r a w n from t h e bottom of s e p a r a t o r by means of a s y r i n g e . s i n c e t h e bond stress b e t w e e n t h e media and t h e w a l l m a t e r i a l c a n c h a n g e .456 The q u a n t i t i e s of o i l and b r i n e i n t h e s e p a r a t o r a r e g o v e r n e d by t h e d i f f e r e n c e i n h e i g h t s o f t h e w e i r s 0 and B . and t h e e d g e of t h e b r i n e w e i r was p r o v i d e d w i t h a c e l l u l o s e rider. However. The p o s i t i o n i s d e t e r m i n e d by 3 . ANALYTICAL PROCEDURE c a l c u l a t i o n of t h e l o c a l c o n c e n t r a t i o n u n d e r r e s e r v o i r cond i t i o n s from t h e measured d a t a is p o s s i b l e o n l y i f t h e o i l anal y s i s is c o m p l e t e . t h e h e i g h t of t h e o i l / b r i n e c o n t a c t a t A must be h e l d c o n s t a n t w i t h i n 5 0. f o r m i n g of a hemis- p h e r e a t t h e water w e i r ) a V i t o n i n s e r t was u s e d a s o i l weir. f o r e x a m p l e . I n order t o a c h i e v e s u f f i c i e n t a c c u r a c y . For t h e g a s p h a s e . T h i s p r o c e d u r e i m p l i e s t h a t a compromise must be r e a c h e d b e t w e e n t h e a n a l y t i c a l r e q u i r e m e n t s imposed by t h e c a p i l l a r y column and t h e packed column. T h i s is n o t f e a s i b l e f o r a c o n s t a n t s e t t i n g of t h e weirs. e s p e c i a l l y i n t h e t r a n s i t i o n zone. The s a m p l e s of s t o c k . The sample s t o r a g e r e q u i r e d f o r a v o i d i n g such a procedure a p p e a r s impractical A . a P o r a p a c k Q/S-packed 1 / 8 " column 6 m i n l e n g t h is e m p l o y e d . I n order t o d e t e r m i n e t h e o v e r a l l c o n c e n t r a t i o n of t h e i n d i v i d u a l o i l components. t h e o i l i s a n a l y s e d i n a s i l i c o n e r u b b e r c a p i l l a r y column 40 m i n l e n g t h . The s e p a r a t o r s a r e t h e r m o s t a t e d a t a t e m p e r a t u r e of 2OoC. 1 c m T h a t i s . The s i m u l t a n e o u s e x e c u t i o n of b o t h a n a l y s e s i s n e c e s s i t a t e d by t h e l o n g d u r a t i o n o f 2 h.2 cm. t h e a n a l y s i s e x t e n d s o n l y u p t o C26.

and d r y i n g by low p r e s s u r e C 0 2 . i n t h i s c a s e m e s i t y l e n e . Hence. A c c o r d i n g t o t h e f o l l o wing e q u a t i o n . t h e f l o w r a t e c a n be c a l c u l a t e d from t h e known C02-mass f l u x and t h e measured p r e s s u r e v a l u e s : -d-V dt - The main p u r p o s e of t h i s p r o c e d u r e i s t o correct f o r i n t e r f e r e n c e d u e t o i n s u f f i c i e n c i e s i n t h e p e r f o r m a n c e of t h e b a c k . From t h e g a s chromatogram. From t h e known f l o w r a t e i n t h e r e s e r v o i r model. t h e c o n c e n t r a t i o n i s o b t a i n e d d i r e c t l y from t h e m o l a r f l u x . l per cent was o b t a i n e d a f t e r f l u s h i n g . I n order t o d e t e r m i n e t h e amount of r e s i d u a l o i l i n t h e e f f l u e n t . MIXER TANK WATKINS / 3 / h a s u s e d premixed media f o r d i s p l a c e m e n t e x p e r i m e n t s . The f l u s h i n g p r o c e s s w a s c h e c k e d f o r t h o r o u g h n e s s by means of p r e l i m i n a r y tests.1. I n order t o determine t h e r e s i d u a l o i l . For t h e o i l s u s e d h e r e . and t h u s t h e r e s i d u a l o i l mass c a n be c a l c u l a t e d w i t h good a c c u r a c y . a t a g g i n g compound. SLIM TUBE EXPERIMENTS The s l i m t u b e e x p e r i m e n t s were p e r f o r m e d p r i m a r i l y f o r o b t a i n i n g a n estimate of t h e minimal m i s c i b i l i t y p r e s s u r e .p r e s sure regulator a t the outlet. For t h i s p u r p o s e s a n d from t h e i n l e t and o u t l e t was a n a l y s e d f o r t o t a l c a r b o n . t h e major p a r t o f t h e s o l v e n t was d i s t i l l e d o f f . t h e a c c u r a c y of i n j e c t i o n does n o t s u f f i c e f o r calc u l a t i n g t h e m o l a r f l u x of t h e components. f o r example t o l u e n e . H e allowed C 0 2 t o b u b b l e t h r o u g h t h e o i l from below i n a n a u t o - . a mass c o n t e n t of c a r b o n l e s s t h a n 0 . n-hexane is added t o t h e resit h e mass r a d u e a t a r a t i o o f 1 : . and t h e d i s t i l l a t e w a s analysed gas-chromatographically. t i o of n-hexane t o s o l v e n t i s d e t e r m i n e d .457 F u r t h e r m o r e . o n l y a small q u a n t i t y of o i l is p r e s e n t i n t h e d i s t i l l a t e . Of c o u r s e . Thus i t i s p o s s i b l e t o c a l c u l a t e t h e m o l a r f l u x of a l l d e t e c t e d components from t h e g a s chromatograms. i s a d d e d t o t h e o u t f l o w a t cons t a n t r a t e b e f o r e t h e p r e s s u r e r e d u c t i o n . S i n c e t h e major s h a r e of t h e r e s i d u a l o i l l i e s beyond C Z 6 and t h e r e f o r e c a n n o t be analysed gas-chromatographically. t h e flood tube w a s flushed with a solv e n t .

The q u a n t i t y which was p r o d u c e d p r i o r t o t h e C 0 2 b r e a k . From t h e s e two f i n d i n g s i t c a n be c o n c l u d e d t h a t t h e e q u i l i b r a t i o n i s r a t h e r good i n t h e i n d i v i d u a l s t a g e s . For b o t h s l i m . With t h e u s e of s u c h a p r o c e d u r e . b e c a u s e o f t h e u n f a v o u r a b l e d i a m e t e r . 8 and 9 ) a 1 : 1 m i x t u r e of s a n d and silca p d e r was used (permeability: 3. a good m i x i n g p e r f o r m a n - ce c a n n o t be e x p e c t e d .08 D was u s e d . 6 a s a n d p a c k w i t h a p e r m e a b i l i t y of 3.t u b e e x p e r i m e n t s described h e r e . w i t h f l o o d r e s u l t i n g i n a v e l o c i t y of 5 .o i l mixtures. .03 D ) .f i l l e d m i x e r t a n k c o n s i s t i n g of s l i g h t l y i n c l i n e d tubes ( f i g . For t h e f i r s t s l i m t u b e t e s t shown i n f i g . Hence a d i f f e r e n t method was u s e d . f l u x of 4. 5 ) . 3 m . t h e s t o c k . filter p l a t e 1 F i g . i n which t h e d e n s e r l i q u i d p h a s e i s d i s p l a c e d by a g a s p h a s e . I f t h e amount of o i l d i s p l a c e d p u r e l y by s w e l l i n g i s c a l c u l a t e d from t h e PVT d a t a f o r C 0 2 . s u b d i v i d e t h e m i x e r i n t o t h r e e s t a g e s . 1 0 ) shows t h a t o n l y p u r e o i l f l o w s o u t a t f i r s t .t h r o u g h was 1 3 7 c m 3 A f t e r 1 6 1 cm3 had b e e n p r o - .m a s s l e n g t h of 6 . The f i r s t e x p e r i m e n t ( f i g .9 c and a l e n g t h of 2 m . 5 : S k e t c h of t h e m i x e r t a n k The t u b e s h a v e an i n n e r diameter o f 0. 3 m / d . t h e d e n s i t y d i f f e r e n c e between C02r i c h and C02-poor o i l p r o v i d e s f o r a d e q u a t e c i r c u l a t i o n .s ( u n d e r r e s e r v o i r c o n d i t i o n s ) was employed. Within t h e i n d i v i d u a l stages. 6 and 7 ) was p e r f o r m e d a t a mean p r e s s u r e of 219 b a r and a C 0 2 . The f l o o d i n g of t h e t a n k ( f i g . a value of 1 7 0 c m 3 is o b t a i n e d . RESULTS AND DISCUSSION E x c l u s i v e l y recombined o i l s were u s e d f o r t h e i n v e s t i g a t i o n . whereby t h e d i f f u s i o n p a t h s a r e s h o r t b e c a u s e o f t h e s m a l l diameter-tolength ratio. o i l w i t h a v i s c o s i t y of 6 mPa.t a n k o i l was o n l y s l i g h t l y c o l o u r e d . m The v e r t i c a l t u b e s e c t i o n s .458 c l a v e . C 0 2 i s i n j e c t e d i n t o an o i l .77 g / h .t o l e n g t h r a t i o . For p r e p a r i n g t h e s a n d pack f o r t h e second t e s t ( f i g . d u c e d .

t h e f o l l o w i n g v a i u e is o b t a i n e d f o r t h e " d e n s i t y " : dm -= dV 0 0 . If i t is assumed t h a t t h e volume e f f e c t d u e t o d i s s o l u t i o n d u r i n g t h e f l o o d p r o c e s s is comparable w i t h t h e e f f e c t o b s e r v e d i n s i n g l e .t a n k o i l produced. 7 8 g/cm 3 The d e c l i n e of d m o / d V ( f i g .805.3 1. 111 nsol lrngth C02-moss flux mean prrssurc 6.c o n t a c t PVT m e a s u r e m e n t s . 1 7 5 g/cm3. c u r v e I1 shows t h e amount of s t o c k .' 1. The m e a n 8 e n s i t y " b e f o r e C02 break-through dmO amounts t o 0 . 6 ) . m m . 7 ) . p r e s s u r e d i f f e r e n c e Ap. 6 : S l i m t u b e experimeh't a t 219 bar: moles& g a s p r o d u c e d n i n j e c t e d PV PV D e a d o i l mass mo p r o d u c e d .3 m L. 7 and 8 ) a f t e r t h e C 0 2 b r e a k . t h e ' $ e n s i t $ ' of t h e oilu n d e r r e s e r v o i r c o n d i t i o n s c a n be c a l c u l a t e d . 9 p l o t t e d v e r s u s p o r e volumes From t h e p r o d u c e d mass and t h e f l o w r a t e . 1 I . The c a l c u l a t e d val u e s a r e shown by c u r v e I ( f i g .t h r o u g h r e m a i n s l i n e a r o v e r a c e r t a i n r a n g e . and c u r v e I11 shows t h e q u a n t i t y of g a s p r o d u c e d ( f i g . Hence i t a p p e a r s p l a u s i b l e .11 g/h 219 bar AF n bar "9 2.459 Curve I shows t h e measured p r e s s u r e d i f f e r e n c e o v e r t h e e n t i r e l e n g t h .2 a' I m . The e x p e c t e d v a l u e i s 0. I 31 0 F i g . 1. and i f t h e tot a l volume s h r i n k a g e is e s t i m a t e d from t h e l e n g t h of t h e t r a n s i t i o n z o n e .1 0. T h i s d e v i a t i o n c o r r e s p o n d s t o a volume e f f e c t d u e t o t h e d i s s o l u t i o n of C 0 2 i n t h e o i l .

9 10 .1 F i g . and t h e i n c r e a s e i n m e - . . The maximum of t h e n o r m a l i z e d C1-gas-concentration cc i n t h e e f f l u e n t amounts to 1 . The p r g s s u r e i s a p p r o x i m a t e l y e q u a l to t h e MMP.C 4 and n-C4 u n d e r r e s e r v o i r c o n d i t i o n s are p l o t t e d . C 2 . F i g u r e s 8 and 9 show a s e l e c t e d i n t e r v a l of an e x p e r i m e n t a t 1 8 9 bar and a C 0 2 mass f l u x of 1 .-. It c a n be s e e n t h a t v e r y pronounced c o n c e n t r a t i o n maxima o c c u r f o r t h e lower a l k a n e s .--.-. I n f i g u r e 9 t h e o v e r a l l c o n c e n t r a t i o n d n i / d v of C1. t h e f l o w r a t e dV/dt. N o s u c h d i f f e r e n c e is r e c o g n i z a b l e h e r e . g a s concentration + .-..5.6/ t h e b e h a v i o u r of m e t h a n e s h o u l d d i f f e r from t h a t o f t h e o t h e r l i g h t a l k a n e s ..77 g/h 219 bar I 1. k& 111 2 3 0.0 I \ 0. Moreover.l 0.3 m L.. Methane d o e s n o t show a l e a d .. 07 . dV r e s i s t a n c e 3 p l o t t e d v e r s u s pore volumes at 'I x p e r i m e n t a t 219 b a r : i n j e c t e d P\ to d e f i n e t h e l e n g t h of t h e t r a n s i t i o n z o n e by means of t h e i n t e r c e p t s of t h e s t r a i g h t l i n e w i t h t h e a b s c i s s a and w i t h t h e h o r i z o n - t a l p o r t i o n of t h e c u r v e . .. 9 8 g / h .--. i .-. PV 3. 08 . t h e maxima a r e a l l s i t u a t e d a t t h e same p o s i t i o n .-.5 06 .-a -.-._. 0. and t h e gas-chrog m a t o g r a p h i c a l l y measured c o n c e n t r a t i o n .-- --_.5 --.4 60 II dm0 I bP length COjmass flux mean pressure in ! ! l ! !x x)2 dV n mi c i 3 - dV m 9/cm3 6.4. 0.. 7: S l i m t u b e '' Dens i t y flow dn L O -. T h e s e v a l u e s h a v e b e e n c a l c u l a t e d from t h e gas p r o d u c t i o n d n / a t .-.. A c c o r d i n g to t h e l i t e r a t u r e /2.1 0.2 03 .-. 0 6 o n l y .

9 1 g / h and a p r e s s u r e of 202 bar. g a s c o n c e n t r a t i o n 9 l o t t e d v e r s u s p o r e volumes i n j e c t e d PV dV dV t h a n e c o n c e n t r a t i o n i s of t h e same order of m a g n i t u d e as t h a t f o r t h e o t h e r components shown. cially striking. The s c a l e i s so c h o s e n t h a t 1 P c o r r e s p o n d s t o t h e m i x e r volume V m3 of 424 c The o b s e r v a t i o n t h a t t h e c o n c e n t r a t i o n maxima f o r C1. F i g . I t c a n be s e e n t h a t t h e s t a r t of c o n c e n t r a t i o n i n c r e a s e i s a t t h e same p o s i t i o n f o r m e t h a n e and c a r b o n dioxide.. 1 0 shows t h e c a l c u l a t e d C 0 2 and C1 c o n c e n t r a t i o n u n d e r res e r v o i r . i . F i g .c . o n d i t i o n s . C 2 . 8 : P a r t of a s l i m t u b e e x p e r i m e n t a t 1 8 9 bar: dn p D e n s i t y dmo . . T h i s e x p e r i m e n t was conV d u c t e d a t a C 0 2 mass f l u x of 9 .8 m 1.98 g/h 189 bar P .461 om0 dV - 1 Iength in gcm3 COZ-moss flux mean pressure 10. t h e c o n c e n t r a t i o n c a l c u l a t e d i n t h e same manner from t h e m i x e r o u t f l o w is p l o t t e d aga i n st i n j e c t e d P i n f i g u r e 11.C 4 and n-C4 o c c u r a t t h e same time for t h e m i x e r test t o o is e s p e - . To d e m o n s t r a t e t h e s i m i l a r i t y b e t w e e n t h e f l o o d i n g r e s u l t s obt a i n e d w i t h t h e m i x e r and w i t h t h e s l i m t u b e .

C2. In view of the fact that the maximal value of the concentration is given only by one gas chromatogram.61 . n-C4 3. . c2 2. Table: Maximal concentrations in mol/cm 3 c1 Slim tube Mixer 1.462 Fig. i-C4 1. i-C4. the agreement between the values is remarkably good. .60 3.03 1. n-C4 under reserd voir. it must be taken into consideration that the relative change in flow rate due to dissolution of C02 in the oil is certainly different for the slim tube and mixer tests. .38 . For both tests it should be emphasized that increase in concentration by a factor of about 5 occurs for the lower alkanes considered here. Furthermore.14 1. 9:Part of a slim tube experiment at 189 bar: dnVi Concentration -.13 .conditions plotted versus pore volumes PV injected The maximal concentrations for the slim tube and mixer tests are presented in the following table. .42 - .28 5.of C1.

.8 1.o 0. 11: Mixer test PV ' Dead o i l mass produced mo.. C1 -.98 g/h 189 bar 1.5 . ' 50 . 02 0.*./-a\. .7 W n F i g . 10 0.* .8 m 1.. .. moles of gas produced n plotted 9' Versus pore volumes PV i n j e c t e d .l. 100. . of C1 and C 0 2 under r e s e r v o iirr conditions p l o t t e d versus pore volumes PV i n j e c t e d 1 1 10 In 9 Mixer.11) flux 10s mean pressure 11 9. . ..o 1....463 c1 5.S . . . 'n9 45 'inmd m.9 1.. *.b 0.0..6 0.../- 0.0 11 ..test C02.8 0.91 202 g/h bar I *:.--. 10: Part of a s l i m tube experiment a t 189 bar: conConcentration d n .3.2 Fig..103 dV in m mi3 d length CO2-mass flux mean pressure 10. .

91 g/h 202 bar 10 . n-C4 under reservoir conditions plotted v ersus pore volumes injected Conclusions On the basis of the results obtained so far.test C O y m s r flux mean pressure 9. it allows conclusions concerning the composition under reservoir conditions.5 1. cores can be flooded with media corresponding to a transition zone.5 W Fig. Nomenclature P P(p) k V = inlet pressure = C02-density = C02-mass flux = volume under reservoir conditions . C2. By means of this device. 12: Mixer test dn.5 0. In particular. 1 Concentration of C1. The use of the three-stage mixer tank for core flooding experiments appears promising.0 1. i-C4. 0. the apparatus developed appears to be suited for C02 core flooding experiments.Mixer.

J. Leach Yelling - . W. Pet. Metcalfe L. Pusch Laboruntersuchungen zum KohlendioxidFluten (PVT-Verhalten und Flutversuche) ET 3048 A A Laboratory Investigation of Miscible Displacement by Carbon Dioxide SPE 3483 (1971) A 2. Eng.P.F. R.h Nghiem Fong K. Pusch. Hassing 3. 19811. L. W. 89 5. Rathmell F. Meyn G. Watkins Technique for the Laboratory Measure- ment of Carbondioxide Unit Displacement Efficiency in Reservoir Rock SPE 7474 (1978) 4. Yarborough The Effect of Phase Equilibria on the C02 Displacement Mechanism SOC.S.J. G .465 v2 vO = volume filled by C02 = volume filled by oil -y0 = compressibility of oil = compressibility of C02 = X1 m 0 n g t mass of stock tank oil time = moles of gas = C = gas-concentration of methane in the effluent Acknowledgements The autor is grateful to the Federal Ministry of Research and Technology (BMFT) for financial support of the project as well as to Prof. J.~. 242 Compositional Modelling with a Equation of State SPE 9306 (1980) Compositional Model Studies C02 OilDisplacement SOC. for his advice and support. V. Pet. M.X. C. Eng. Stalkup R. Aziz 6. J. J. (Feb. (1979) p. R. References 1. D. the project leader.

This Page Intentionally Left Blank .

Of the many experiments proposed. Examples are provided of t h e phenomena observed d u r i n g t h e course of t h e experiments which may be used t o a s s e s s t h e n a t u r e of t h e displacement process. D e t a i l s are given of t h e way i n which t h e s l i m tube displacement experiments are performed and a r e operated with automated data a c q u i s i t i o n . INTRODUCTION During t h e l a s t decade a p r o f u s i o n of d a t a has been generated i l l u s t r a t i n g t h e phenomena a s s o c i a t e d w i t h dynamic m i s c i b l e displacement A s h o r t p e r u s a l of t h e l i t e r a t u r e quickly shows t h a t t h e displacement gas used i s almost always carbon dioxide and t h a t most o f t e n the o i l i s a West Texas. By analogy w i t h t h e minimum dynamic m i s c i b i l i t y p r e s s u r e concept a minimum dynamic m i s c i b i l i t y composition may be defined which may be used t o a s s e s s t h e s u i t a b i l i t y o f a l t e r n a t i v e i n j e c t i o n gasses. The displacement i s confined w i t h i n t h e w a l l s of a .EXPERIMENTAL TECHNIQUES 467 THE USE OF SLIM TUBE DISPLACEMENTEXPERIMENTS IN THE ASSESSMENT OF MISCIBLE GAS PROJECTS BERNARD J. SKILLERNE DE BRISTOWE British Petroleum Company Limited ABSTRACT Slim tube displacemeht experiments can b e employed t o optimise dynamic m i s c i b l e displacement p r o j e c t s with r e s p e c t t o both p r e s s u r e and composition. To do t h i s . The r e s u l t s are discussed i n r e l a t i o n t o s u p e r c r i t i c a l e x t r a c t i o n phenomena which provides an e x p l a n a t i o n f o r t h e r e s i d u a l o i l phase l e f t behind i n t h e s l i m tube a t t h e end of the experiment. t h e s l i m tube displacement experiment o f f e r s t h e c l o s e s t analogue t o t h e processes t h a t occur w i t h i n ' the r e s e r v o i r while a t the same t i m e i t can be performed s u f f i c i e n t l y r a p i d l y for repeated experiments t o be performed under varying o p e r a t i n g c o n d i t i o n s . t h e experiments performed should be designed i n such a way t h a t they w i l l form a r a t i o n a l b a s i s f o r investment d e c i s i o n s . Permian Basin Crude. For an o p e r a t o r whose i n t e r e s t s l i e f a r from West Texas and who needs t o e v a l u a t e many p o t e n t i a l p r o s p e c t s i t i s imperative t h a t the maximum amount o f information i s generated i n t h e most economical way p o s s i b l e . time consuming and consequently extremely expensive t o perform. I n using t h e method the assumption is made t h a t t h e p r o c e s s e s which r e s u l t i n dynamic m i s c i b l e displacements a r e independent of t h e n a t u r e of the h o s t porous m a t r i x and depend e x c l u s i v e l y on t h e composition and p h y s i c a l p r o p e r t i e s of t h e f l u i d s involved and on t h e temperature and p r e s s u r e a t which t h e displacement takes place. Despite t h i s r e s t r i c t i o n i n comp o s i t i o n l i t t l e consensus of opinion has been reached as t o what s u i t e of l a b o r a t o r y experiments must be performed i n o r d e r t o e v a l u a t e a given f i e l d p r o j e c t (1) except t h a t they w i l l be e x t e n s i v e .

C O ~ I C .I U O E L 1 01611AL O E N S l l Y Y E I E I VENT ? I0 DACK .hA5 utim Figure 1. S C P A R A l O R A N 0 BALANCE 1 l . N e v e r t b e l e s s . s i n c e a v a r i e t y of evidence now e x i s t s t o suggest t h a t the p e t r o p h y s i c a l p r o p e r t i e s of t h e rock a r e of secondary importance i n determining the displacement e f f i c i e n c y (2) i t may be used t o provide valuable q u a l i t a t i v e information about t h e elementary displacement mechanism. O I L CYL!NOEI % CO1IC~CYLINDCR & PRESSURE 1RANSOUCCI 7. Schematic Diagram of the Slim Tube Apparatus. +VENT I I L . S L I Y . I t s major use t h e r e f o r e i s i n t h e i n i t i a l assessment o f a given p r o j e c t where the compatability of the i n j e c t e d gas with t h e r e s e r v o i r o i l i s sought.468 narrow tube so t h a t the flow i s e s s e n t i a l l y one dimensional. I t can thus provide no information regarding t h e gross f l u i d movements w i t h i n a r e s e r v o i r which a r e s u b j e c t t o hydrodynamic i n s t a b i l i t i e s s u c h as viscous f i n g e r i n g and g r a v i t y s e g r e g a t i o n . C A S S A Y P L e VALVE PLOW IS.P l E S S U I 1 E I L G U L A I O I 1l. The g e n e r a l experimental arrangement i s shown schematically i n Figure 1 and a block diagram of t h e computer c o n t r o l l e d d a t a a c q u i s i t i o n system i n Figure 2.LIDUIO S A Y ? L E V A L V E 1 1 .VALVC IL. .-* __---KLROSINP CYLINDCR VISUAL CELL ------ KLY 1. R E F I L L CYLINDER L 1. The purpose of t h i s paper i s t o show t h a t once t h i s can be done t h e process can be optimised with r e s p e c t t o e i t h e r the p r e s s u r e o r t h e composition of the i n j e c t e d gas. The key t o t h i s i s i n i d e n t i f y i n g when t h e dynamic m i s c i b l e displacement process i s o p e r a t i n g . THE SLIM TUBE DISPLACEMENT EXPERIMENT I n o r d e r t o improve the e f f i c i e n c y with which t h e s l i m tube experiments can be performed while making t h e b e s t use of t h e information a v a i l a b l e t h e d a t a g a t h e r i n g p a r t of the experiment h a s been automate