Economic Geology

Vol. 95, 2000, pp. 361–390

Formation and Transformation of Clay Minerals in the Hydrothermal Deposits of

Middle Valley, Juan de Fuca Ridge, ODP Leg 169
K. S. LACKSCHEWITZ,†,*
Institut für Geowissenschaften, Universität Kiel, 24098 Kiel, Germany

A. SINGER,
Seagram Center for Soil and Water Sciences, Hebrew University of Jerusalem, 76100 Rehovot, Israel

R. BOTZ, D. GARBE-SCHÖNBERG, P. STOFFERS, AND K. HORZ

Institut für Geowissenschaften, Universität Kiel, 24098 Kiel, Germany

Abstract
Mineralogical, chemical, and isotopic characteristics of clays from Ocean Drilling Program (ODP) sites 1035,
856, 1036, and 858, drilled in Middle Valley at the northern end of the Juan de Fuca Ridge during Legs 139
and 169, were studied using X-ray powder diffraction, differential thermal analysis, infrared spectroscopy, scan-
ning electron microscopy, transmission electron microscopy, X-ray fluorescence, and inductively coupled
plasma-mass spectrometry. Oxygen isotope measurements on authigenic clay minerals provide a record of min-
eral formation temperatures. Holes 1036A and 858B are located in the present-day hydrothermally active Area
of Active Venting and reflect several stages of hydrothermal alteration. Smectite and some mixed-layer chlo-
rite-smectite clays are the dominant phyllosilicates in the upper part of holes 1036A and 858B. Nearly pure
chlorite-smectite mixed-layer clays and corrensite mainly occur more than 20 meters below the seafloor. For-
mation temperatures of these nearly pure chlorite-smectite mixed-layer clays and corrensite were calculated as
250° to 268°C. In the deepest unit, chlorite is the dominant phyllosilicate coexisting with authigenic quartz.
The high-temperature alteration stages resulted in gains (50–85%) of Mg and Zn, an almost complete loss of
K, Ca, Cu, Rb, Sr, and Ba (≥ 90%), and depletion (20–40%) in Si, Fe, Na, and Zr.
Alteration effects in hemipelagic and turbiditic sediments from hole 856H, drilled near the summit of the
Bent Hill massive sulfide deposit, are similiar to those of sediments in hole 1035H, which was drilled 350 m
south of hole 856H near the southernmost peak of the Ocean Drilling Program mound. At both holes, the clay
fraction of the sedimentary sequence between 140 and 470 m below seafloor is composed of varying amounts
of authigenic quartz and chlorite indicating high-temperature alteration. The chlorite formation temperatures
calculated from oxygen isotope data lie at 290° to 320°C in hole 1035H and 250° to 264°C in hole 856H. Min-
eralogical data from hole 1035D, drilled 75 m east of hole 856H, and hole 1035A, drilled 77 m west of hole
856H, indicate lower temperatures of hydrothermal alteration in the upper 50 m of their sedimentary se-
quences. These sequences are partly altered to chlorite-smectite mixed-layer clays coexisting with several de-
trital minerals and with a dolomite formation temperature of 74°C. Oxygen isotope evidence indicates authi-
genic chlorite in the deeper sedimentary section of hole 1035A that formed at relatively low temperatures
(180°C), whereas chlorites from hole 1035D formed at higher temperatures (277°–292°C). Chemical changes
in the strongly altered clay fractions of hole 856H and 1035D include gains (60–75%) of Fe, Mg, Zn, and Cu,
a nearly complete loss (≥ 95%) of K, Na, Ca, Rb, Sr, and Ba, and a strong depletion (30–70%) in Si, Ti, and Zr.
All of the chondrite-normalized rare earth element patterns of chlorite are characterized by negative Eu anom-
alies and varying levels of light rare earth enrichment.
The chloritization observed at the Area of Active Venting is similar to the Mg-rich chlorite and to chloritic
minerals observed at sediment-covered spreading centers in Escanaba trough on Gorda Ridge and in Guaymas
basin on the East Pacific Rise. This observation suggests intense Mg metasomatism in a mixing zone where hy-
drothermal fluids interact with seawater and sediment at temperatures above 200°C.
The similarity of the Fe-rich chlorites from the Bent Hill massive sulfide deposit to Fe-rich chlorites from
several ancient massive sulfide deposits indicates that they probably reflect hydrothermal fluid-rock interaction
at high temperatures below the zone of seawater mixing.

Introduction main areas approximately 2 km apart were drilled during Leg

THE MIDDLE VALLEY of northern Juan de Fuca Ridge is one
of the few well-studied, sediment-covered spreading centers.
It was the subject of detailed investigations during the Ocean
Drilling Program (ODP) Legs 139 and 169 (Fig. 1a). Two
† Corresponding
author: email, klacksch@uni-bremen.de
*Present address: Fachgebeit “Petrolgie der Ozeankruste,” Fachbereich
5–Geowissenschaften, Postfach 330 440, 28334 Bremen, Germany.
169: the Dead Dog vent field in the Area of Active Venting
(site 1036), and the Bent Hill massive sulfide deposit in the
Bent Hill area (sites 1035 and 856, Fig. 1b). The Bent Hill mas-
sive sulfide deposit is a 35-m-high and 100-m-wide sulfide
mound near the southern margin of the Bent Hill mound (Fig.
2). Drilling on Leg 139 established that the Bent Hill massive
sulfide deposit is an area of former discharge, where hydro-
thermal activity has led to strong alteration in the sedimentary

0361-0128/00/3038/361-30 $6.00 361

362 LACKSCHEWITZ ET AL.

A
a

52°N NORTH
AMERICAN
PLATE

48° Juan Middle

de
Valley
Bb 129°00'
Fuca
128°30' 58mm/yr
Ridge
49°

N
44°

Gorda Ridge

40°

130°W 126° 122°

48°30'
Sites 1036
Sites 1036
and 858
and 858

Sites1035
Sites 1035
and
and856856

48° 0 25

km
FIG. 1. a. Location map showing the tectonic setting of the northern Juan de Fuca Ridge and the sedimented Middle Val-
ley rift (after Davis et al., 1992). b. Tectonic interpretation of the Middle Valley with names shown for each of the major tec-
tonic elements. Arrows refer to sites 1036, 858, 1035, and 856. Stippled areas show recent volcanic flows (modified from
Davis and Villinger, 1992).

0361-0128/98/000/000-00 $6.00 362

 FORMATION & TRANSFORMATION OF CLAY MINERALS, JUAN DE FUCA RIDGE 363

The principal center of hydrothermal activity in the Middle

N Valley is the Area of Active Venting, which includes the Dead
Dog vent field (Fig. 3). The Area of Active Venting contains at
least 20 active vents with temperatures up to 276°C (Ames et
al., 1993). The vent fluid composition indicates significant in-
teraction of hydrothermal fluid with sediment. Surface sedi-
ments consist of anhydrite rubble with minor Mg smectites and
sulfide minerals, whereas the deeper sequences are character-
Bent Hill ized by unaltered to altered hemipelagic sediments interbed-
ded with turbidites. The composition of the hydrothermal de-
posits and fluids reflects chemical interaction of the
hydrothermal fluids with the sediment (Butterfield et al., 1994;
Früh-Green et al., 1994; Goodfellow and Peter, 1994). Hy-
drothermally precipitated Mg-rich phyllosilicates and carbon-
ate concretions from holes drilled in the Middle Valley area
have been studied by Buatier et al. (1994). The vertical distrib-
ution of the hydrothermal products in the sediments is charac-
Bent Hill Massive
128°41'

1035A 856H Sulfide Deposit (BHMS) terized by a transition from smectite to corrensite to chlorite.
The transformation from smectite to chlorite is especially well
48°26' 1035D expressed at site 858 in the Area of Active Venting, where clay-
size fractions show a distinct change with depth from smectite
to corrensite to smectite-chlorite mixed-layer clays and finally
1035E to chlorite. The isotopic and chemical compositions of hy-
drothermal minerals indicate steep thermal gradients and ex-
tensive sediment-fluid interaction (Buatier et al., 1995).
Leg 169 of the Ocean Drilling Program provided the op-
portunity to investigate the clay mineral genesis within the
hydrothermal deposits drilled at the Bent Hill massive sulfide
1035H

0 100 200 m N
ODP Mound

FIG. 2. Map showing the Bent Hill sediment mound, the inactive Bent
Hill massive sulfide deposit, and the Ocean Drilling Program sulfide mound
128°42.75'

approximately 330 m south of Bent Hill. The locations of holes in the area of
Bent Hill are discussed in the text.

cover and to the formation of a large body of massive sulfides 48°27.50'

(Davis et al., 1992). Mineralogical composition, textures of the
primary sulfides, and fluid inclusion homogenization temper-
atures (Goodfellow and Peter, 1994; Peter et al., 1994) sug- Dead Dog
gest that the Bent Hill deposit formed by venting of 350° to Mound
400°C fluid at or near the seafloor and are consistent with de- 261-268°C
rivation of the metals from a basaltic rock source. The Bent
Hill deposit was revisited during ODP Leg 169, mainly for 1036A
the purpose of assessing the thickness and the lateral extent
of mineralization. Hole 856H, drilled near the summit of the 858B
deposit, penetrated 93.8 m of massive sulfide during Leg 139
128°42.75'

and was deepened by 402.2 m during Leg 169 to a total depth

of 500 m. Hole 1035A was drilled to a depth of 170 meters Chowder
below the seafloor and is located about 77 m west of the cen- Hill
ter of the Bent Hill deposit. Hole 1035D was drilled 75m east 48°27.25' 276°C
of the deposit mound, whereas hole 1035E was drilled ap- 0 100 m
proximately 100 m south of the deposit. The upper sediments
consist of clastic sulfides underlain by unaltered to altered
hemipelagic sediments (Fouquet et al., 1998). In holes 1035A FIG. 3. Location map showing the Area of Active Venting (AAV), includ-
and 1035D, sediments are intercalated with massive to semi- ing the Dead Dog mound, the Chowder Hill mound, and holes discussed in
massive sulfides. the text.

0361-0128/98/000/000-00 $6.00 363

364 LACKSCHEWITZ ET AL.
deposit (site 1035) and the Dead Dog vent area (site 1036) in
Middle Valley. In this paper, we present and interpret new
major and trace element data and oxygen isotope composi-
tions of a suite of hydrothermal phyllosilicates. We report the
results of mass balance calculations with the objectives of
characterizing the authigenic clay mineral phases and quanti-
fying chemical changes due to hydrothermal alteration. Fur-
thermore, we compare our results with other sediment-
hosted sulfide deposits such as the Guayamas basin and
Escanaba trough.
Methods
Samples analyzed in this study are from holes at sites 1035,
856, and 1036 from Leg 169. For comparison, we have addi-
tionally analyzed several samples from site 858 (Leg 139).
After weighing, the freeze-dried sample was divided into a
fine (<63 µm) and a coarse fraction (>63 µm) by wet sieving.
Grain-size separation into a silt (2–63 µm) and a clay fraction
(<2 µm) was performed by the settling of particles in stand-
ing cylinders according to Stokes’ law (Moore and Reynolds,
1989).
Clay mineralogy was determined by X-ray diffractometry.
For mineral identification, a Philips X-ray diffractometer PW
1710 with monochromatic Cu-Kα radiation and a Philips
X-ray diffractometer PW 1710 with Co-Kα radiation were
used. Oriented samples were produced by vacuum filtration
through a 0.15 µm filter. Measurements were carried out on
air-dried, glycol-saturated, KCl-saturated and heated (550°–
600°C) samples. Randomly oriented powder preparations were
produced (measurement: 60°–75°2θ) to identify di- or triocta-
hedral clay minerals from 060 reflections. On the basis of the
(060) spacing, we have estimated the b parameter, which is re-
lated to octahedral Fe, Mn content (Brindley and Brown, 1980).
Using the relation between composition and peak positions
for glycol-saturated trioctahedral chlorite-trioctahedral smec-
tite given by Brindley and Brown (1980), we have estimated
the percentages of the chlorite interlayer. On the basis of
structural and chemical data, Brindley and Brown (1980) cal-
culated X-ray diffraction (XRD) profiles for trioctahedral
chlorite-(glycol) trioctahedral smectite with different propor-
tions of smectite by a computer program that treats three-
component interstratification. These authors documented the
observation that significant changes in the amount of one
component affects the position of the peaks. Thus, the care-
ful measurement of the spacing of these peaks gives the com-
positional information for chlorite-smectite mixed-layer clays
using the d-values of Brindley and Brown (1980).
Clay suspensions were deposited on a carbon copper grid
for particle morphology observation with a JEM 100 CX
transmission electron microscope (TEM) and a Philips EM
400T. Scanning electron microscope (SEM) observations
were made on Au-coated fragments of bulk samples using a
JEOL JSM-5410LV and a CamScan electron microscope.
Differential thermal analysis (DTA) was performed using a
Perkin Elmer Model 1700 System 4 apparatus. For analysis,
15 mg of the sample (<2 µm and <0.2 µm) were heated at a
rate of 10°/min. Infrared spectroscopy (IRS) was carried out
on a Nicolette FTIR apparatus. Infrared spectra were regis-
tered in the 4,000 to 400 cm–1 range using nearly monomin-
eralic samples prepared in the form of KBr disks.
0361-0128/98/000/000-00 $6.00 364
For major and trace element analysis, bulk and clay sedi-
ment samples were oven dried at 40°C and then pulverized.
The major element composition was determined using
Philips (PW 1400 and PW 1480) X-ray fluorescence (XRF)
spectrometers and a Camebax SX 50 microprobe. HTEM-
AEM analyses were performed with the JEOL JEM-2010
electron microscope. For XRF analysis, the sediments were
dried at 900°C and melted using lithiumtetraborate (Li2B4O7)
in a mixture consisting of 600 mg sediment and 3,600 mg
lithiumtetraborate. For calibration of the XRF determina-
tions, 44 international rock standards were used. Trace ele-
ments were analyzed by ICP-MS using a VG Plasma-Quad
PQ 1. Total dissolution of bulk sediment was performed by
pressurized HF-HCLO4-aqua regia attack (Garbe-Schön-
berg, 1993). The accuracy of the analytical results was con-
trolled by measuring the international standard reference ma-
terial BHVO-1.
For oxygen isotope analysis, only samples with monominer-
alic clays, dolomite, or authigenic chlorite-quartz pairs (de-
termined by means of XRD) were analyzed. Prior to isotope
analysis, free Fe and Mn oxides were removed by the method
of Mehra and Jackson (1960). Oxygen was extracted from sil-
icates using the ClF3 method (Clayton and Mayeda, 1963;
Borthwick and Harmon, 1982). The samples were transferred
to nickel reaction vessels and heated for 2 hours at 150°C. All
samples were then reacted with ClF3 at 600°C for 12 hours.
Oxygen was quantitatively converted to CO2, which then was
analyzed isotopically in a Finnigan MAT 251 stable isotope
mass spectrometer. Oxygen isotope ratios are expressed in the
conventional δ notation as deviation in per mil from standard
mean ocean water (SMOW). The oxygen isotope value of
Standard NBS-28 (Matsuhisa, 1974) was determined as δ18O
= 9.2 ± 0.2 per mil. A dolomite sample was reacted with 100
percent H3PO4 to produce CO2 for mass spectrometric analy-
sis. The δ value denotes the oxygen isotope ratio relative to
Peedee Belemnite (PDB; McCrea, 1950).
Geologic Setting and Lithostratigraphy
Sites 1035 and 856:
Bent Hill massive sulfide and ODP deposits
The Bent Hill massive sulfide deposit is located in the east-
ern part of Middle Valley, 3 km west of the eastern rift-
bounding normal fault. Extrapolation of the spreading rate
suggests that the deposit is located over oceanic crust that is
approximately 320,000 yr old (Davis and Villinger, 1992). The
sedimentary sequence in Middle Valley consists of Holocene
and Pleistocene interbedded hemipelagic and turbiditic sedi-
ments (Goodfellow and Blaise, 1988; Davis et al. 1992; Fou-
quet et al., 1998). The Bent Hill massive sulfide is a recent hy-
drothermally inactive sulfide mound near the southern
margin of Bent Hill (Fig. 2). It rises 35 m above the sur-
rounding turbidite fill of the valley and extends a minimum
distance of 60 m north-south and 90 m east-west.
A second mound of massive sulfide (referred to here as the
ODP deposit) occurs approximately 330 m farther south of
the Bent Hill massive sulfide (Fig. 2), along the trend of the
north-south scarp that bounds the western side of Bent Hill.
The northernmost mound is characterized by an active vent
with 264°C hot clear fluids (Ames et al., 1993).
 FORMATION & TRANSFORMATION OF CLAY MINERALS, JUAN DE FUCA RIDGE
The Bent Hill massive sulfide deposit was penetrated on
Leg 169 in seven holes. An additional hole (1035H) was
drilled on the ODP sulfide mound. Lithostratigraphic studies
indicate a complex history of hemipelagic and turbiditic sedi-
mentation, hydrothermal fluid flow, sulfide deposition, and
igneous activity (Fouquet et al., 1998; Zierenberg et al.,
1998). Seven lithologic units were recovered from holes
856H, 1035A, 1035D, 1035E, and 1035H investigated in this
study. Unit I consists of Holocene to Late Pleistocene
hemipelagic sediment with minor fine-grained turbidite beds.
Unit II is composed of nonindurated or weakly indurated
hemipelagic and turbiditic sediments. Unit III consists of un-
consolidated clastic sulfides. Unit V is characterized by mas-
sive to semimassive sulfides. Unit VI is a sulfide feeder zone
characterized by crosscutting sulfide veins. Units VII and
VIII are composed of basaltic sills and flows. Hydrothermal
alteration provides a record of past and present fluid flow and
is controlled by structural, stratigraphic, and sedimentary fa-
cies (Fouquet et al., 1998).
Site 1036: Dead Dog
The Dead Dog vent field lies in the Area of Active Venting
(Fig. 3), which is located about 6 km east of the current axis
of Middle Valley and about the same distance from the east-
ern valley-bounding normal faults (Fig. 1b). On the basis of
the spreading rate (58 mm/yr) and the distance to the Brunhes-
Matuyama magnetic boundary, the age of the underlying ig-
neous crust is about 350,000 yr (Davis and Fisher, 1994). The
sediment cover on the fault block in the area surrounding the
vent field is approximately 450 m and overlies a sill-sediment
complex that forms the transition to oceanic crust (Davis et
al., 1992). Active vents occur predominantly on top of 5- to
15-m-high sediment-covered mounds a few tens of meters in
diameter.
Three lithostratigraphic units are recognized in hole 1036A.
Lithologic unit I is a hemipelagic sediment with minor tur-
bidites which has been subdivided into an unaltered and an
altered section on the basis of hydrothermal-diagenetic alter-
ation (Fouquet et al., 1998). Unit II is characterized by al-
tered hemipelagic silty clays interbedded with silt to fine-sand
turbidites. Unit IV is composed of collapsed chimney-derived
sediments.
Mineralogy
X-ray diffraction investigations were conducted to identify
the mineral assemblages of the clay fraction to determine the
origin, distribution, and genesis of the various minerals.
Quartz, feldspar, and illite are considered to be detrital in all
holes and to represent the background hemipelagic sedimen-
tary supply (Davis et al. 1992; Fouquet et al., 1998). Never-
theless, scanning electron microscope observations demon-
strate that quartz is locally authigenic. Hydrothermal input is
well documented by the presence of sulfide minerals.
XRD
Area of active venting: Holes 1036A and 858B show a dis-
tinct change in silicate mineralogy with depth (Fig. 4; Table
1). The upper 20 m below seafloor are mainly composed of a
mixture of detrital quartz, illite, feldspar with smectite, and
0361-0128/98/000/000-00 $6.00 365
365
two or three types of chlorite-smectite mixed-layer clays (Fig.
5a). In contrast, the clay fraction below 20 m below seafloor
is dominated by nearly pure chlorite-smectite mixed-layer
clays showing an increase of interstratified chlorite with
depth. Reflections from the (060) peak indicate that the chlo-
rite-smectite mixed-layer clays are interstratifications of trioc-
tahedral chlorite with trioctahedral smectite. A (001) super-
lattice reflection of 29 Å, which expands to 31 Å after
glycolation, indicates a regular mixed-layered chlorite-smec-
tite mineral, i.e. corrensite (Fig. 5c). Upon heating to 600°C,
the superlattice peak in our samples collapsed to 24 to 28 Å
and the reflections at 7 Å were destroyed (see also Howard
and Fisk, 1988). The XRD patterns of some other clay sam-
ples are characterized by a pure, randomly interstratified
chlorite-smectite mixed-layer clays with 20 or 30 percent
chlorite (Fig. 5b).
Bent Hill massive sulfide deposit: Mineralogical composi-
tions were determined on the clay fractions from the holes
1035A, 1035D, 1035E, 1035H, and 856H (Table 2). The lat-
eral and vertical distribution of the mineral assemblages in
holes 1035A, 856H, and 1035D, lying on a west-east transect,
is summarized in Figure 6. The uppermost sediments of hole
1035A consist of clastic sulfides (unit III) containing siderite
as the major component in the clay fraction. Pyrite is the
major component in sample 1035A-7X-2, 44 to 45 cm from a
massive sulfide bed (unit V). In hole 1035A, quartz and chlo-
rite are the major components of the clay fraction of unit II
except sample 169-1035A-5H-5, 101 to 103 cm, which shows
dolomite as the dominant mineral. Samples from unit II con-
tain in addition feldspar, illite, and mixed-layer chlorite-smec-
tite. A 060 spacing of 1,539 to 1,543 Å and 1.52 to 1.54 Å in-
dicates that the chlorite-smectite mixed-layer clays are
trioctahedral chlorites and trioctahedral smectites, respec-
tively. In addition, weak reflections at 1,500 to 1,503 Å in
some samples indicate dioctahedral smectites, e.g, montmo-
rillonite, as well. In hole 1035A, smectite is the dominant in-
terstratified component in the mixed-layer clay from the shal-
lower samples. Deeper in the sedimentary column, chlorite
becomes more abundant as the interstratified layer. Corren-
site first appears in sample 169-1035A-5H-7, 28 to 30 cm, and
was also found in sample 169-1035A-6H-5, 64 to 65 cm and
169-1035A-7X-2, 44 to 45 cm, where it occurs with chlorite or
other chlorite-smectite mixed-layer clays. The samples below
70 m below seafloor show illite and chlorite without expand-
able layers as clay minerals. The section between 90 and 120
m below seafloor is characterized by a large amount of authi-
genic quartz. At 120 m below seafloor, the mineralogical se-
quence reflects a distinct change to a chlorite-dominated clay
fraction. The deepest sample, 169-1035A-19X-CC, 12 to 14
cm, shows the highest chlorite content (>90%) of the hole.
At hole 1035D, the clay fractions are composed of a mix-
ture of detrital and authigenic minerals (Table 2). Quartz,
feldspar, illite, and chlorite are the major components in the
upper 50 m of hole 1035D, whereas corrensite becomes
abundant between 35 and 50 m (Fig. 7a). Sediments from 50
to 106 m are characterized by massive and semimassive sul-
fides with pyrite and chalcopyrite, whereas below 106 m,
nearly pure chlorite occurs as the dominant component (Fig.
7b). A sample at 106 m consists of small amounts of talc.
366 LACKSCHEWITZ ET AL.

NW SE
Tfluid = 278°C Hole Hole T fluid= 270°C
1036A 858B
mbsf
0

C/S-mixed layer 10
+ Qz, Fs, Ill C/S T C/S= 177°C

20
TC/S= 206°C
C/S-mixed layer TC/S= 250-
T C/S= 268°C
264°C 30

Qz, Chl C/S, Chl,

Qz
40

Hemipelagic sediments with minor turbidites

Interbedded turbidites and hemipelagic sediments

Clastic sulfides

Massive sulfides

FIG. 4. Northwest-southeast geologic cross section of the southern Area of Active Venting showing the clay mineralogical
composition and distribution, and the mineral formation temperatures based on oxygen isotope analyses; Chl = chlorite; C/S
= chlorite-smectite mixed-layer clay; Fs = feldspar; Ill = illite; Qz = quartz; TC/S = mineral formation temperature of chlorite-
smectite mixed-layer clay; mbsf = meters below seafloor.

Sediments drilled at hole 1035E appear to be almost free of
sulfate and sulfide phases. The whole sedimentary sequence
downward to 46 m below seafloor has an assemblage similiar
to the upper 50 m of hole 1035D, which is characterized by
quartz, feldspar, illite, chlorite, and smectite (Table 2). Upon
heating to 600°C, the 13.9 Å peak in sample 169-1035E-3H-
4, 134 to 136 cm, and 169-1035E-4H-5, 135 to 136 cm col-
lapsed to 11.8 Å. A (060) reflection at 1.545 Å indicates a tri-
octahedral chlorite. Based on the (060) spacing, we have
estimated Fe contents of 15 to 20 percent.
Hole 856H, drilled near the summit of the Bent Hill mas-
sive sulfide deposit and which penetrated 93.8 m of massive
sulfide during Leg 139, was deepened by 406.2 m during Leg
169. The clay fraction of sediments from the sulfide feeder
zone (103.6–203.3 m below seafloor), which consists of sul-
fide veins, sulfide impregnations, and greenish gray interbed-
ded and variably altered hemipelagic and turbiditic sedimen-
tary rocks (210.6–431.7 m below seafloor), is composed of
authigenic chlorite and quartz (Table 2; Fig. 6). Only two clay
samples are characterized by nearly pure chlorite.
Hole 1035H, which was drilled 350 m south of hole 856H, in-
tersected three major massive sulfide zones. The clay fraction of
sediments from the underlying, variably sulfide-impregnated,
hydrothermally altered hemipelagic and turbiditic sedimen-
tary rocks is also composed of authigenic chlorite and quartz
(Table 2). Two clay samples contain nearly pure chlorite.
SEM and TEM
SEM and TEM observations confirm the different miner-
alogical phases determined by XRD. Samples from the upper
50 m below seafloor of holes 1035A, 1035D, and 1035E are
made up of detrital quartz and feldspar of different size and
morphology surrounded by a fine-grained, clay-rich matrix of
chlorite and illite (Fig. 8a). Hydrothermally altered sediments
from the upper 20 m below seafloor of holes 1036A and 858B
also contain detrital quartz and feldspar but the clay matrix is
dominant. From SEM images, each type of clay in hy-
drothermal layers identified by XRD has a distinct morphol-
ogy. Pure smectitic mixed-layer samples below 20 m below
seafloor are characterized by rosette-like aggregates of curved
flakes (Fig. 8b). Observations of TEM images show typical
curled edges (Fig. 8c). Corrensite samples from hole 1036A
show a typical flake-shaped morphology (Fig. 8d), which is
similar to the SEM observations on samples from holes of site
858 (Buatier et al., 1994; Buatier et al., 1995). Figure 8e-f re-
veals the typical rigid platelet morphology of the authigenic

0361-0128/98/000/000-00 $6.00 366

 FORMATION & TRANSFORMATION OF CLAY MINERALS, JUAN DE FUCA RIDGE 367

TABLE 1. Mineralogical Assemblages of Unaltered, Altered, and Strongly Altered Sediments from Holes 1036A and 858B
(as determined by X-ray diffraction analyses)

Interval Depth
Leg Hole Core section (cm) (mbsf) Unit Qz Fs Ill Talc Py Cp Sm Cor C-S (% Chl) Chl

169 1036A 1H 04 79-81 5.29 IV xx xxx xx x

169 1036A 1H 05 67-69 6.67 I xxx xx x x xx (20) xx
169 1036A 1H 06 113-115 8.63 I xx xx x xxx xxx (90,60,10) xxx
169 1036A 2H 01 116-117 10.66 I xxx xx x xxx (60,10) xxx
169 1036A 2H 05 106-107 16.56 I xxx xx x x xx (80,20) xxx
169 1036A 2H 07 10-12 18.60 I xxx xx x xx (90,10) xxx
169 1036A 3H 02 129-131 21.79 I x xxxx (20)
169 1036A 3H 06 128-130 27.78 I * xxxx
169 1036A 4H 03 13-15 30.83 II xxx * xxx (90)
169 1036A 6X CC 14-16 37.64 II xxxx * xxx
139 858B 2H 02 110-112 9.81 I xxx xx x xx xx (90,20) xx
139 858B 2H 03 90-92 11.11 IV * * xxxx (20)
139 858B 2H 04 19-21 11.90 IV xxxx (30)
139 858B 2H 05 66-68 13.87 I xxx xx xx xxxx (70)
139 858B 2H 06 33-35 15.04 I xx xx xx xxxx
139 858B 2H 07 23-25 16.44 I xxx xx xx xxxx
139 858B 5H 01 53-55 24.44 I * xxxx (70,20)
139 858B 5H 02 26-28 25.67 II xxxx
139 858B 5H 03 81-83 27.72 II xxxx (70)
139 858B 5H 04 81-83 29.22 II xxxx (60)
139 858B 6H 01 18-20 31.69 II xx (60,20) xxxx
139 858B 8X 01 50-52 33.21 II xx * xxx xxx (10) xxxx
139 858B 8X 02 10-12 33.85 II xx * xxxx xxx (10)
139 858B 8X 03 80-82 35.38 II xx x xxx (30,60)
139 858B 8X 04 68-70 36.76 II xx * xxx xxx (20)
139 858B 8X 05 57-59 38.15 II x * x xxxx xxx (20) xxx

Symbols: xxxx = very abundant, xxx = abundant, xx common, x = rare, * = traces

Abbreviations: Chl = chlorite, Cor = corrensite, Cp = chalcopyrite, C-S = chlorite-smectite mixed layer, Fs = feldspar, Ill = illite, Py = pyrite, Qz = quartz,
Sm = smectite; % Chl = percentages of chlorite interlayers in C-S; mbsf = meters below seafloor

chlorite particles. The deepest sample in hole 1036A is dom-
inated by quartz of euhedral hexagonal shape coexisting with
pure chlorite (Fig. 8g).
Differential thermal analysis (DTA)
The thermograms of the nearly pure chlorite-smectite
mixed-layer clay samples from the Area of Active Venting are
characterized by a low-temperature endothermic peak, a
well-defined endothermic peak, and a high-temperature
exothermic peak (Fig. 9). The low-temperature dehydration
endotherms occur between 90° and 119°C, due to the loss of
sorbed moisture, with most of the water probably being lost
from interlayered smectite. The second endothermic peak at
547° to 569°C corresponds to the dehydroxylation of the in-
terlayer hydroxide sheet of chlorite, followed by a higher tem-
perature peak system at 798° to 828°C resulting from the de-
hydroxylation and recrystallization of the silicate sheet
(Paterson and Swaffield, 1987).
The thermograms of all pure chlorite clays from the Bent
Hill massive sulfide show similiar features (Fig. 9). The en-
dothermic reaction took place between 525° and 550°C and
the exothermic reaction took place between 750° and 795°C.
A second weak endothermic reaction took place between 710°
and 735°C. These curves are very similar to the general form
of DTA curves for chlorites (Paterson and Swaffield, 1987).
Infrared spectroscopy (IR)
The infrared spectra of chlorite-smectite mixed-layer clays
and corrensite from the Area of Active Venting are similiar to
the IR spectra of a typical chlorite with the exception of a dis-
tinct adsorption band between 3,670 and 3,677 cm–1. This is
a diagnostic value for trioctahedral layer silicates such as talc
or smectite (Fig. 10).
The IR spectral features of the pure chlorite clays from the
Bent Hill massive sulfide show features characteristic of chlo-
rites. The strong broad doublet at 3,559 to 3,565 cm–1 and
3,431 to 3,434 cm–1 arises from OH-stretching vibrations of
the interlayer hydroxide sheet, which is distinctive for a fully
trioctahedral chlorite (Farmer, 1974; Barnhisel and Bertsch,
1989). Absorption bands near 820 and 758 to 760 cm–1 can be
associated with tetrahedral Al-O vibrations of trioctahedral
chlorites (Wilson, 1994). The strong band at 656 to 667 cm–1
are associated with Fe-OH vibrations. These adsorption
bands indicate a high Fe occupancy of the octrahedral sites
(Farmer, 1974).
The IR spectrum of samples from hole 1035H is similar to
those of hole 856H but a small band at 3,635 cm–1 suggests a
contribution from a dioctahedral chlorite. However, the
strong OH-stretching doublet at 3,550, and 3,427 cm–1 re-
flects the observation that most of the chlorites retain a trioc-
tahedral character.
Geochemistry
In order to document compositional differences among the
phyllosilicates, XRF, electron microprobe, and ICP-MS analy-
ses were performed. Representative chemical analyses with
elemental compositions from 11 nearly monomineralic clay
samples are listed in Table 3. Formulas for the trioctahedral

0361-0128/98/000/000-00 $6.00 367

368 LACKSCHEWITZ ET AL.

1036A-2H-1, 116,5 cm 1036A-3H-2, 130 cm 1036A-3H-6, 129 cm

C/S mixed layer + chlorite + C/S mixed layer Corrensite
detrital minerals
14.3Å 3.6Å
Qz
14.3Å

28.8Å
4.8Å
3.85Å
15Å
14.3Å 7.2Å
9.5Å

7.1Å 5.7Å
Fs
3.6Å
4.7Å

3.1Å 15.3Å 3.4Å

9.5Å
4.3Å

7.3Å

4.8Å
4Å 3.8Å

30.5Å
10Å 4.9Å
7.7Å

16.4Å
16.6Å
3.3Å
3.25Å
5.1Å
15.4Å
14.3Å 4.4Å
8.3Å

glycol glycol glycol

9.3Å

3.9Å
5.5Å

6.2Å
4.7Å
7Å

10 20 30 10 20 30 10 20 30
a º2ΘCuKα b º2ΘCuKα c º2QCuKa

FIG. 5. XRD patterns (CuKα) of < 2-µm-size fractions, MgCl2 solvated and MgCl2 + glycerol solvated (= Mg glycerol) for
a. chlorite-smectite (C/S) mixed-layer clay + chlorite + detrital minerals (1036A-2H-1, 116.5 cm), b. chlorite-smectite mixed-
layer clay (1036A-3H-2, 130 cm), and c. corrensite (1036A-3H-6, 129 cm).

chlorites have been calculated on the basis of the ideal
O20(OH)16 anion content, whereas formulas for the interstrat-
ified chlorite-smectite mixed-layer clays have been calculated
on the basis of the O20(OH)10 anion content. The number of
cations in the octhadral site of chlorite-smectite mixed-layer
clays is between 8 and 9, which could be consistent with a
combination of one layer of chlorite and one layer of smectite
that respectively contain a maximum of 6 and 3 cations in
their octahedral site.
Bulk compositions were determined for the uppermost
samples from holes 1035D, 1035E, and 1036A and are given
as mean values in Table 4. These samples display unaltered to
least altered hemipelagic and turbiditic sediment that contain
quartz, feldspar, smectite, illite, and chlorite. They can be
considered as a reference for unaltered background detrital
sediments in comparison with samples composed of highly al-
tered minerals. Table 5 presents the chemical composition of
two strongly altered bulk sediment samples from the Bent
Hill massive sulfide and the Area of Active Venting.
High resolution TEM analyses of unaltered clay samples
from the uppermost sediment sequence of holes 1035E and
1036A have provided chemical compositions for the various
types of clay particles. These analyses suggest that the clay
minerals are actually saponite, Al-chlorite, and illite.
The microprobe analyses of the clay samples 169-1036A-
3H-2, 129 to 131 cm, 139-858B-2H-3, 19 to 21 cm, and 139-
858B-2H-4, 66 to 68 cm containing chlorite-smectite mixed-
layer clays give rise to compositions that can be interpreted as
combinations of different proportions of chlorite and smec-
tite. Chemical variations of the chlorite-smectite mixed-layer
clays and chlorite are represented on an Al2O3-MgO-FeO
compositional plot depicting XRF and averaged microprobe
analyses (Fig. 11). The FeO contents in regularly and ran-
domly interlayered chlorite-smectite mixed-layer clays of
holes 1036A and 858B are consistently lower than those in
nearly pure chlorites of hole 856H and hole 1035D. Most of
the regularly and randomly interstratified phases of hole
1036A and hole 858B have Al2O3-MgO-FeO ratios very simi-
lar to those of chimney clay samples from the Area of Active
Venting (Percival and Ames, 1993).
Several studies have documented changes in the composi-
tional parameters and the character of mafic phyllosilicates

0361-0128/98/000/000-00 $6.00 368

 FORMATION & TRANSFORMATION OF CLAY MINERALS, JUAN DE FUCA RIDGE 369

TABLE 2. Mineralogical Assemblages of Unaltered, Altered, and Strongly Altered Sediments from
Holes 856H, 1035A, 1035D, 1035E, and 1035H (determined by X-ray diffraction analyses)

Interval Depth
Leg Hole Core Section (cm) (mbsf) Unit Qz Fs Ill Talc Dol Py Cp Si Sm Cor C-S (%Chl) Chl

169 856H 20R 01 20–24 103.80 VI xx x xx

169 856H 21R 02 60–64 115.42 VI xx x xx
169 856H 22R 01 59–62 117.69 VI xxx x x
169 856H 23R 02 10–14 125.60 VI xx x xx
169 856H 24R 01 46–50 134.16 VI xx x xx
169 856H 25R 01 116–120 144.46 VI * xxxx
169 856H 26R 01 55–59 153.45 VI x xxxx
169 856H 27R 01 77–81 163.27 VI x xxxx
169 856H 28R 01 44–47 172.54 VI * xxxx
169 856H 29R 01 52–56 182.22 VI xx xx
169 856H 30R 01 41–45 191.81 VI xxx xx
169 856H 31R 01 41–45 201.41 VI xxxx x
169 856H 33R 01 32–36 220.52 II xxxx x
169 856H 35R 01 45–49 239.85 II xxxx x
169 856H 38R 01 61–64 268.91 II x xxxx
169 856H 39R 01 5–9 278.05 II x xxx
169 856H 40R 01 16–19 287.76 II x xxx
169 856H 42R 01 35–39 307.25 II x xxx
169 856H 45R 01 0–3 335.80 II xx xx
169 856H 47R 01 38–42 355.48 II * xxxx
169 856H 50R 01 57–60 384.37 II xx xx
169 856H 53R 01 0–4 412.60 II xxx xx
169 856H 56R 01 52–56 434.82 II xx xx
169 856H 59R 01 116–118 461.86 II xx xx
169 856H 62R 01 24–27 470.54 II xx xx
169 1035A 2H 01 20–22 7.70 III xx x xx x
169 1035A 2H 01 70-72 8.21 III xxx xx
169 1035A 2H 01 88–91 8.37 III xxx xx x x x
169 1035A 2H 01 103–106 8.50 III xxx xx x x x
169 1035A 2H 02 134–136 10.24 II xxx xx x xx
169 1035A 2H 05 135–137 14.75 II xxx xx xx xx xxx
169 1035A 3H 01 95–96 17.95 II xxx xx xx xx xx (20) xx
169 1035A 4H 02 80–82 28.80 II xxx xx xx x xx (80) xx
169 1035A 4H 06 53–55 34.53 II xxx xx x xx (80) xx
169 1035A 4H CC 18–20 36.31 II xxx xx xx xx (60) xx
169 1035A 5H 03 136–138 40.28 II xxx xx x xx (10) xx
169 1035A 5H 05 101–103 42.93 II xx xxxx x
169 1035A 5H 07 28–30 45.20 II xxx xx xx xx xx (30) xx
169 1035A 6H 02 113–115 48.13 II xxx xx xx xx (40) xxx
169 1035A 6H 05 64–65 52.14 II xxx xx xx xx xxx
169 1035A 6H CC 16–18 52.74 II xxxx xxx x x
169 1035A 7X 02 44–45 56.59 V xx xxx x xx xx (80)
169 1035A 7X CC 39–40 57.06 V xxx x xx x xx (80,90) xxx
169 1035A 8X 01 54–57 61.54 II xx x x xxx
169 1035A 8X 03 48–49 64.36 II xxx xx xx x xx (80) xxx
169 1035A 9X 01 4–6 70.64 II xxx xx x xxx
169 1035A 9X 02 98–100 73.08 II xxx * x x x (90) xxx
169 1035A 9X 04 26–28 75.36 II xxx xx x x
169 1035A 9X 06 79–81 78.89 II xxxx xxx
169 1035A 10X 01 2–4 80.22 II xxx xxx x xxx
169 1035A 10X 03 78–80 83.98 II xxxx xxx
169 1035A 11X 01 3–5 89.83 II xxx xx xx xxx
169 1035A 12X 01 65–67 100.05 II xxx xxx x x
169 1035A 12X 02 129–131 102.19 II xx xx x xxx
169 1035A 12X 05 58–60 105.48 II xxxx xx xx xxx
169 1035A 13X 04 137–139 114.87 II xxx xxx x xxx
169 1035A 13X 05 40–42 115.40 II xx xx xx xxxx
169 1035A 14X 04 126–128 124.36 II xxx xx x xxxx
169 1035A 14X 06 14–16 126.24 II xxx xx x xxxx
169 1035A 15X 01 28–29 128.48 II xxx xx x xxxx
169 1035A 15X 03 145–147 132.65 II xxx xx x xx
169 1035A 15X 05 130–132 135.50 II xxx xx x xx
169 1035A 16X 02 75–77 140.05 II xxx xx x xx
169 1035A 16X 05 18–20 143.98 II xx x x xxx
169 1035A 17X CC 22–23 147.62 II xxx xx x xxx
169 1035A 19X 01 46–48 167.16 II xx x xxx
169 1035A 19X 02 117–119 169.37 II xx x xxx

0361-0128/98/000/000-00 $6.00 369

370 LACKSCHEWITZ ET AL.

TABLE 2. (Cont.)

Interval Depth
Leg Hole Core Section (cm) (mbsf) Unit Qz Fs Ill Talc Dol Py Cp Si Sm Cor C-S (%Chl) Chl

169 1035A 19X CC 12–14 169.59 VI xxx x xx xx

169 1035D 1H 01 137–139 1.37 III xxx xx xx x xx
169 1035D 1H 02 131–133 2.81 III xxx xx xx x xx
169 1035D 1H 03 139–141 4.39 III xxx xx xx x xx
169 1035D 1H 04 7–9 4.57 III xxx xx xx * x xx
169 1035D 2H 01 125–127 6.35 III xxx xx xx x x xx
169 1035D 2H 01 147–149 6.57 III xxx xx xx x xxx
169 1035D 2H 02 47–49 7.07 II xxx xx xx x xxx
169 1035D 3H 02 136–138 17.46 II xxx xx xx x xxx
169 1035D 3H 04 135–137 20.45 II xxx xx xx x xxx
169 1035D 4H 04 135–137 29.95 II xxx xx xx x xxx
169 1035D 4H 05 137–139 31.47 II xxx xx x xxx xxx
169 1035D 5H 02 135–137 36.45 II x x x xxx xxx
169 1035D 5H 03 14–16 36.74 II x xx x xxx xxx
169 1035D 7X CC 21–24 48.31 V xxxx xx
169 1035D 10X 01 77–79 67.27 VI xxx * xx
169 1035D 15X CC 13–16 106.33 VI xx xxxx
169 1035D 20X 01 129–131 145.99 II * xxxx
169 1035D 21X 01 70–73 155.00 II x xxxx
169 1035D 24X CC 34–35 173.84 VI xxxx
169 1035E 1H 01 92–94 0.92 I xxx xx xx xx xx
169 1035E 1H 03 45–47 3.45 I xxx xx xx xx xx
169 1035E 1H 05 23–25 6.23 III xxx xx xx xx xx
169 1035E 2H 02 74–76 9.74 III xxx xx xx x xx xx
169 1035E 2H 02 88–90 9.88 III xxx xx xx x xx xx
169 1035E 2H 05 136–138 14.86 II xxx xx xx xx xx
169 1035E 3H 02 131–133 19.81 II xxx xx xx xx xx
169 1035E 3H 04 134–136 22.84 II xxx xx xx xx xx
169 1035E 3H 06 128–130 25.78 II xxx xx xx xx xx
169 1035E 4H 02 135–136 29.35 II xxx xx xx xx xxx
169 1035E 4H 03 138–139 30.88 II xxx xx xx xx xxx
169 1035E 4H 05 135–136 33.85 II xxx xx xx xx xxx
169 1035E 5H 02 94–96 38.44 II xxx xx xx xx xxx
169 1035E 5H 03 11–13 39.11 II xxx xx xx * xxx
169 1035E 5H CC 21–23 45.81 II xxx xx xx xxx
169 1035H 4R 01 61–63 27.01 VI xxx xxx
169 1035H 5R 01 52–54 36.52 VI * xxx
169 1035H 6R 01 124–127 46.84 VI x xxx
169 1035H 6R 02 27–30 47.33 II x xxx
169 1035H 10R 01 43–45 84.63 VI xx xx
169 1035H 20R 01 71–73 171.81 VI xxx x
169 1035H 22R 01 11–14 190.41 VI * xxxx
169 1035H 25R 01 117–119 220.27 II x xxxx
169 1035H 25R 02 14–17 220.65 II x xxxx
169 1035H 26R 02 42–44 230.62 II * xxxx
169 1035H 27R 01 98–100 239.28 II * xxxx

xxxx = very abundant, xxx = abundant, xx = common, x = rare, * = traces

Abbreviations: Chl = chlorite, Cor = corrensite, Cp = chalcopyrite, C-S = chlorite-smectite mixed layer, Dol = dolomite, Fs = feldspar, Ill = illite, Py =
pyrite, Qz = quartz, Si = siderite, Sm = smectite, % Chl = percentages of chlorite interlayers in C-S

(e.g., smectite vs. chlorite) as a function of metamorphic
grade (e.g., Bettison and Schiffman, 1988; Schiffman and Fri-
dleifsson, 1991; Beaufort et al., 1997). Beaufort et al. (1997)
reported that Al in mafic phyllosilicates increases from smec-
tite to corrensite to chlorite. The Al vs. Si plot (Fig. 12) plot
shows that the Middle Valley phyllosilicates are scattered also
along a mixing line between the end-members saponite and
chlorite. For comparison, we have indicated four samples
representing a characteristic compositional range from the
smectitic end-member saponite to corrensite to the chloritic
end-member chlinoclore (Newman and Brown, 1987; Perci-
val and Ames, 1993; Beaufort et al., 1997). The Si content is
strongly correlated with the Al content because of the in-
creased percentage of chlorite interlayers in interstratified
chlorite-smectite mixed-layer clays. They have Al contents
that increase with increasing depth and with the transition
from chlorite-smectite mixed-layer clays in holes 1036A and
858B to discrete chlorite in holes 856H and 1035D.
The total Fe/(Fe + Mg) content in the hydrothermal layer
silicates generally decreases from chlorite to chlorite-smectite
mixed-layer clays (Fig. 13). Bent Hill chlorite exhibits the
largest range in Fe/(Fe + Mg), whereas chlorite-smectite
mixed-layer clays, corrensite, and smectite from the Area of
Active Venting show similiar Fe/(Fe + Mg) ratios of 0.05 to 0.2.

0361-0128/98/000/000-00 $6.00 370

 FORMATION & TRANSFORMATION OF CLAY MINERALS, JUAN DE FUCA RIDGE 371

W Hole E
Hole 856H Hole
1035A 1035D mbsf
0

Qz, Fs, Ill, Sm Qz,Fs,Ill,Chl,Sm

C/S-mixed layer Tin situ= 54°C
+ Qz, Fs, Ill, D
TDol= 74°C Py, Cp

100
Qz, Chl Qz, Chl
TChl= 237°C
TChl,Qz = Chl
180°C Chl

TChl= 261°C TChl= 277-

Chl 292°C 200
TChl= 180°C
Qz, Chl
Sulfide feeder zone and mineralized sediments

Turbidites and hemipelagic sediments

300
Clastic sulfide
Qz, Chl
Massive sulfide and semi-massive sulfide

Basaltic sills and sediments

400
Qz, Chl

FIG. 6. West-east geologic cross section of the Bent Hill massive sulfide deposit showing the clay mineralogical composi-
tion and distribution, and the mineral formation temperatures based on oxygen isotope analyses; Chl = chlorite; Cp = chal-
copyrite; C/S = chlorite-smectite mixed-layer clay; Fs = feldspar; Ill = illite; Py = pyrite; Qz = quartz; TD,Chl,Qz = mineral for-
mation temperature of dolomite, chlorite, and quartz; mbsf = meters below seafloor.

Trace element compositions of the chlorite samples from
the Bent Hill massive sulfide and the corrensite and chlorite-
smectite mixed-layer clays samples from the Area of Active
Venting are represented in Table 6 by location and depth. All
samples are characterized by high contents of Zn (211–333
ppm; Fig. 14). Most of them also have high contents of Cu
(133–389 ppm) and a chlorite-smectite mixed-layer clays sam-
ple from the Area of Active Venting shows a relatively high Pb
content (93.7 ppm). Furthermore, all samples are character-
ized by very low contents of Rb (<0.4 ppm), Sr (<2.4 ppm),
and Cs (<0.05 ppm). Most of the samples also show very low
concentrations of Ba (<5.2 ppm). In contrast, least-altered
samples from the Bent Hill massive sulfide and the Area of
Active Venting display much higher contents of Rb (71 and 98
ppm), Sr (52 and 190 ppm), Ba (364 and 693 ppm), Zr (89 and
164 ppm), and slightly higher contents of Sc (25 and 26.2 ppm),
Th (5.6 and 7.9 ppm), and Pb (12.8–32 ppm), respectively.
Concentrations of rare earth elements (REE) are generally
in the range of 0.1 to 56 ppm (Table 6). For comparative pur-
poses, the concentration data have been normalized to the av-
erage rare earth element concentrations of chondrites (Sun
and McDonough, 1989).
The chondrite-normalized REE patterns for the clay sam-
ples from the Bent Hill massive sulfide and Dead Dog are
shown in Figure 15. All of the samples from the Bent Hill
massive sulfide show light rare earth element (La to Gd) en-
richment demonstrated by NdN/YbN values that range from
2.0 to 7.0.

0361-0128/98/000/000-00 $6.00 371

372 LACKSCHEWITZ ET AL.
a 1035D-5H-3, 15 cm
Corrensite+chlorite+
detrital minerals 3.5Å
7Å
14.2Å
Fs
29-30Å
4.7Å
9.7Å 4.9Å
15.3Å 7.6Å
14.1Å
32Å
5.1Å
10 20
º2ΘCuKα
FIG. 7. XRD patterns (CuKα) of < 2µm-size fractions, MgCl2 solvated and MgCl2 + glycerol solvated (= Mg glycerol) for
a. corrensite + chlorite + detrital minerals (1035D-5H-3, 15 cm), and b. chlorite (1035D-24X-CC, 34.5 cm).
The most noticeable feature of the chondrite-normalized
REE patterns is the development of a marked negative Eu
anomaly in most of the authigenic chlorites. Values of
EuN/Eu*N (where Eu*N is defined as the value of EuN ob-
tained by linear interpolation between the plotted points for
SmN and GdN) are generally in the range of 0.15 to 0.39, al-
though two samples show weaker Eu anomalies and have
EuN/Eu*N values of 0.73 and 0.8. No systematic variations
with depth in the absolute REE concentrations, degree of
light REE enrichment, or magnitude of the Eu anomaly are
evident in the chlorites from the Bent Hill massive sulfide
area.
The chondrite-normalized REE patterns of the nearly
pure, randomly and regularly interstratified chlorite-smectite
mixed-layer clays from the Area of Active Venting show two
striking differences relative to those from the Bent Hill mas-
sive sulfide area. Although the NdN/YbN values of 2.8 to 7.0
are not significantly different, in the corrensite sample La ap-
pears to be depleted relative to the other light REE and the
chlorite-smectite mixed-layer clays sample shows a marked
positive Eu anomaly with a EuN/Eu*N value of 1.7.
The chondrite-normalized REE patterns of the two bulk
sediment samples from holes 1035D and 1036A (Fig. 15) in-
dicate very similar REE fractionation in the clays compared
to their bulk samples.
0361-0128/98/000/000-00 $6.00
b 1035D-24X-CC, 34,5cm
Chlorite 3.5Å
7.1Å
4.7Å
2.8Å
14.3Å
4.7Å
2.8Å
14.2Å
glycol glycol
30 10 20 30
º2ΘCuKα
Oxygen Isotopes
Formation temperatures of minerals have been calculated
from oxygen isotope data (Table 7) by using published frac-
tionation factors. Oxygen isotope data from pore water sam-
ples of holes 1035D and 1036A indicate a mean seawater
isotopic composition of –0.4 per mil and 0.1 per mil, respec-
tively, relative to SMOW (W.C. Shanks III, pers. commun.,
1998).
Oxygen isotopes in clay minerals, however, also are related
to mineral structure (Savin and Lee, 1988). Buatier et al.
(1995) compared oxygen isotope fractionation factors, pub-
lished to date, between chlorite and water. These authors also
compared fractionation data of other Mg phyllosilicates such
as saponite and corrensite. More recently, Cole et al. (1998)
presented additional oxygen isotope fractionation values for
the chlorite-water system at the temperature range of 170° to
350°C.
Overall, there exist significant uncertainties in mineral for-
mation temperatures, calculated from oxygen isotope frac-
tionation of clay minerals, as a result of the lack of good oxy-
gen isotope fractionation curve for phyllosilicates (M. Buatier,
pers. commun., 1999). However, according to all fractiona-
tion curves published to date, the uncertainties in tempera-
ture determination are minimized at high temperatures as the
372
 FORMATION & TRANSFORMATION OF CLAY MINERALS, JUAN DE FUCA RIDGE 373

a b

c d

e f

g h

FIG. 8. SEM and TEM photographs. a. Unaltered clay sample from hole 1035E, showing detrital minerals including Al-rich
chlorite (square); b. chlorite-smectite mixed-layer clay from sample 169-1036A-3H-2, 129 to 131 cm, appearing as rosettelike
aggregate of curved flakes; c. TEM photomicrograph of sample 169-1036A-3H-2, 129 to 131 cm, showing curled edges (scale
bar = 1 µm); d. corrensite from sample 169-1036A-3H-6, 128 to 130 cm, showing typical flakelike morphology; e. chlorite from
sample 169-1035D-24X-CC, 34 to 35 cm, occurring as packets of small, platy crystals; f. TEM photomicrograph of sample 169-
1035D-24X-CC, 34 to 35 cm, showing an aggregate of platy chlorite minerals with rotational moire patterns, where the lattice
of one crystal is rotated through a small angle relative to the other crystal (scale bar = 1 µm); g. SEM image of small chlorite
plates on authigenic quartz from sample 169-856H-30R; h. authigenic quartz from sample 169-1036A-6X-CC.

0361-0128/98/000/000-00 $6.00 373

374
FIG. 9. Differential thermal curves of the clay fraction of samples from
sites 1035, 1036, and 858; C/S = chlorite-smectite mixed-layer clay.
fractionation curves converge at temperatures greater than
250°C. Nevertheless, Buatier et al. (1995) pointed out that,
even at those high temperatures (e.g., low oxygen isotope
fractionations), uncertainties in formation temperature still
may be up to 50°C.
Chlorite samples from this study have measured δ18O val-
ues near 0 per mil. The pore water data also were determined
to be near 0 per mil, and therefore the oxygen isotope frac-
tionation (103 lnα chlorite-water) is also close to zero. We
used the chlorite-water fractionation data of Wenner and Tay-
lor (1971) to calculate chlorite formation temperatures (Table
7). We compared the temperature data with the modified
equation, which takes into account a correction in the exper-
imental quartz-water fractionation data of Clayton et al.
0361-0128/98/000/000-00 $6.00
LACKSCHEWITZ ET AL.
(1972). In this range of low fractionations, only minor (ca.
10°C) temperature variations are calculated. If the quartz-
water fractionation factors of Sharp and Kirschner (1994) are
used in the Wenner and Taylor (1971) equation, significantly
lower temperatures (up to 40°C) are calculated for chlorite
formation. This is the approximate (40°–50°C) uncertainty
given by Buatier et al. (1995) for chlorite formation tempera-
tures calculated from comparable, low chlorite-water frac-
tionation data in the system.
An oxygen isotope fractionation equation for mixed-layer
chlorite-smectite clay has not yet been established. Thus, oxy-
gen isotope equilibration temperatures were calculated for
mixed-layer chlorite-smectite clays of both holes 1036A and
858B using the chlorite-water oxygen isotope fraction equa-
tion after Wenner and Taylor (1971) and the smectite-water
oxygen isotope fractionation equation of Sheppard and Gilg
(1996). Isotopic temperatures were calculated from δ18O val-
ues of mixed-layer (smectite-chlorite) samples, considering
the relative amount of each mineral (Table 7). In hole 1036A,
the formation temperatures calculated from oxygen isotope
data for the smectite-rich chlorite-smectite mixed-layer clays
is 206°C, and for the corrensite sample is 268°C. It should be
noted, however, that corrensite oxygen isotope fractionation is
slightly larger than for Mg-Fe chlorites (Buatier et al., 1995).
The fractionation differences between corrensite and chlorite
are minimized at temperatures of 150° to 300°C. Moreover,
chlorite-smectite mixed-layer clays from Middle Valley ex-
hibit various contents of interlayered smectite. Thus, we ex-
pect intermediate oxygen isotope fractionation between pure
smectite and chlorite. Again, a temperature uncertainty of
50°C (or less) is indicated for clay samples containing corren-
site. In hole 858B, the smectite-rich chlorite-smectite mixed-
layer clay samples at 11.9 and 13.9 m below seafloor show for-
mation temperatures of 178° and 177°C, respectively. The
formation temperatures for smectite-rich chlorite-smectite
mixed-layer clays and corrensite samples at 25.7 and 27.7 m
below seafloor are 264° and 250°C, respectively.
For the nearly pure chlorite of holes 1035A, 1035D, and
1035H, we have calculated the oxygen isotope equilibration
temperatures after Wenner and Taylor (1971). In hole
1035D, the formation temperature of chlorite was deter-
mined to be 277° to 292°C (Table 7). The isotopic composi-
tion of the authigenic chlorites below 180 m below seafloor at
hole 1035H gives formation temperatures between 290° and
320°C, whereas a nearly pure chlorite at 170 m below seafloor
in hole 1035A gives a formation temperature of 180°C. Fig-
ure 16 shows that the δ18O values of chlorite-quartz mixtures
vary with the percentage of quartz present in the samples in
hole 1035A. Consequently, it is assumed that hole 1035A
contains variable mixtures of two end members, chlorite
(δ18Oextrapolated near 0.0‰) and quartz (δ18Oextrapolated near
15‰). Equilibrium temperatures for both authigenic chlorite
and quartz may thus be calculated. Based on the mean
δ18OH2O value of 0.2 per mil (W.C. Shanks, III, per. commun.,
1998), the authigenic quartz in hole 1035A more than 84 m
below seafloor would be in isotopic equilibrium with this fluid
at a temperature of approximately 180°C (Clayton et al., 1972;
Shiro and Sakai, 1972), identical to the chlorite-water equilib-
rium temperature of 180°C. Authigenic dolomite found at 42.9
m below seafloor precipitated at a calculated temperature of
374
 FORMATION & TRANSFORMATION OF CLAY MINERALS, JUAN DE FUCA RIDGE 375

1036A-3H02

3671 753,5

3559 667

3434

1036A-3H06 769

3677,5 990
670
446
3427
3579 1012 465
959

4000 3500 3000 1000 500

856H-25R
758
3557
986
667

3431

1035H-22R
3635
758

3551
982
3427

461
545
1035D-24XCC

755
3564
992 657
3433

4000 3500 3000 1000 500

Wavenumbers (cm ) -1

FIG. 10. Infrared spectra of the clay fraction of samples from sites 1035 and 1036.

0361-0128/98/000/000-00 $6.00 375

376 LACKSCHEWITZ ET AL.

TABLE 3. Representative Electron Microprobe Analyses and Calculated Formulae for Phyllosilicates

Leg 169 169 169 169 169 169 169 139 139 139 139
Hole 856H 856H 1035D 1035D 1035D 1036A 1036A 858B 858B 858B 858B
Core-section 25R-1 28R-1 20X-1 21X-1 24X-CC 3H-2 3H-6 2H-4 2H-5 5H-2 5H-3
Interval (cm) 116–120 44–47 129–131 70–73 34–35 129–131 128–130 19–21 66–68 26–28 81–83
Depth (mbsf) 144.5 172.5 146.0 155.0 173.8 21.8 27.8 11.9 13.9 25.7 27.72
Mineral Chl Chl Chl Chl Chl Chl-Sm Cor Chl-Sm Chl-Sm Chl-Sm Cor
Percent chlorite 95 95 95 90 100 20 50 20 30 30 50

Wt percent
SiO2 27.0 29.0 26.7 32.4 30.1 47.6 40.1 44.5 40.4 44.5 40.9
TiO2 0.1 0.2 0.1 0.1 0.09 0.01 0.01 0.1 0 0 0.02
Al2O3 20.5 21.1 19.3 16.2 22.4 5.0 13.6 5.6 6.8 7.9 1.2
FeO 27.8 24.4 27.7 18.6 25.4 4.3 5.2 9.3 7.1 4 4.0
MnO 0.3 0.5 0.36 0.3 0.3 0.04 0.21 0.14 0.23 0.23 0.23
MgO 11.8 13.9 12.8 19.8 18.8 28.6 30.5 22.8 25.1 31.1 28.2
CaO 0.06 0.08 0.1 0.1 0.0 0.14 0.02 0.19 0.02 0 0.03
Na2O 0.1 0.1 0.1 0.1 0.0 0.07 0.02 0.08 0.01 0.02 0.05
K2O 0.01 0.01 0.01 0.0 0.0 0.01 0.0 0.17 0.0 0.0 0.01
Total 90.7 92.1 87.6 89.6 90.1 81.0 89.2 78.1 79.0 87.4 85.0

No. of cations 28-oxygens 25-oxygens

Si 5.75 5.92 5.73 6.52 5.72 7.56 6.52 7.5 7.44 7.29 6.89
Al IV 2.25 2.08 2.27 1.48 2.28 0.44 1.48 0.50 0.56 0.71 1.11
Al VI 2.89 3.00 2.61 2.36 2.62 0.61 1.12 0.83 0.92 0.82 1.27
Ti 0.00 0.00 0.02 0.02 0.00 0.00 0.00 0.01 0.00 0.0 0.00
Fe 4.95 4.17 4.97 3.13 4.33 0.58 0.64 1.30 0.99 0.49 0.51
Mn 0.05 0.09 0.07 0.05 0.05 0.01 0.0 0.02 0.04 0.03 0.03
Mg 3.75 4.23 4.10 5.94 4.80 7.48 7.39 6.15 6.89 7.60 7.08
Ca 0.01 0.02 0.02 0.02 0.02 0.22 0.00 0.50 0.00 0.00 0.01
Na 0.02 0.02 0.02 0.02 0.00 0.04 0.01 0.01 0.00 0.01 0.00
K 0.00 0.00 0.00 0.00 0.00 0.01 0.00 0.01 0.00 0.00 0.00
Total 11.67 11.53 11.81 11.54 11.82 8.95 9.19 8.83 8.84 8.95 8.90

Cation total 19.66 19.52 19.81 19.54 19.82 16.95 17.19 16.83 16.84 16.95 16.90

Chl = chlorite, Ch-Sm = chlorite-smectite mixed layer, Cor = corrensite

74°C (Table 7). This temperature is similar to the tempera-
tures measured in situ and, thus, hydrothermal carbonate
precipitation preceded silicate formation.
Discussion
In order to understand the mechanisms of the formation
and/or transformation of hydrothermal products, and to char-
acterize the interaction between sediments and hydrothermal
fluids, it is necessary to document the origin of these sedi-
ments and to discriminate detrital from authigenic minerals.
The sediments from Middle Valley contain a record of
hemipelagic and turbiditic sedimentation in which hydro-
thermal activity produced several alteration zones. The min-
eralogical and geochemical composition of the sediments
reflects the interlayering and mixture of detrital and hy-
drothermal phases. Surficial, unaltered sediments contain de-
trital minerals with variable biogenic components (Goodfellow
and Blaise, 1988). Transmission electron microscope observa-
tions of the clay fraction from unaltered and/or slightly al-
tered sediments display heterogeneous particles with differ-
ent shapes. According to Buatier et al. (1994), the detrital
minerals of the uppermost unmodified sediments of several
holes from Leg 139 consist of quartz, feldspar, mica, smectite,
and chlorite which represent the products of continental
weathering. Oxygen and hydrogen isotopes of the shallowest
samples from hole 858B clearly reflect the detrital origin of
the clays (Buatier et al., 1995).
Most of the quartz identified in highly altered sediments
from the Bent Hill massive sulfide and the Area of Active
Venting, however, occurs in typical euhedral hexagonal habits
suggesting authigenic formation. The hydrothermal origin is
established by a formation temperature of approximately
270°C at 32 m below seafloor in hole 858B calculated from
stable isotope data (Früh-Green et al., 1994).
Mineralogical and chemical analysis demonstrates that the
dominant phyllosilicates in the holes are trioctahedral, ran-
domly and regularly interlayered chlorite-smectite mixed-
layer clays and chlorite with typical structural formulas. On
the basis of systematic changes of these phyllosilicates in the
holes of Leg 169 and 139, a characteristic mineralogical se-
quence with depth can be documented for the recently active
hydrothermal fields of Dead Dog and Chowder Hill in the
Area of Active Venting and the former active hydrothermal
area of the Bent Hill massive sulfide.
Fine-grained sediments below the massive sulfide deposit
at the summit of the Bent Hill mound mainly are composed
of authigenic chlorite and quartz indicating high-temperature
alteration over the entire sedimentary sequence. The upper
sediments (<30 m) at the flanks of the Bent Hill massive sul-
fide are characterized by smectite and chlorite coexisting with
detrital quartz, feldspar, and illite, which suggests a low degree

0361-0128/98/000/000-00 $6.00 376

 FORMATION & TRANSFORMATION OF CLAY MINERALS, JUAN DE FUCA RIDGE 377

TABLE 4. Chemical Composition of Unaltered and Least Altered TABLE 5. Chemical Composition of Strongly Altered Hemipelagic
Hemipelagic and Turbiditic Sediment from Bent Hill Massive Sulfide and Turbiditic Bulk Sediment Samples from Bent Hill Massive Sulfide
and Area of Active Venting and Area of Active Venting

Element Mean Standard Deviation 1035D-21X-1, 1036A-3H-6,

Element 70–73 cm 128–130 cm
SiO2 51.9 3.6
Al2O3 14.3 1.3 SiO2 31.9 35.1
MgO 5.0 1.8 Al2O3 13.6 12.7
Fe 7.0 0.60 MgO 18.2 28.4
CaO 3.3 1.7 FeO 22.8 5.9
K2O 2.4 0.50 CaO 0.25 0.17
TiO2 0.8 0.01 K2O 0.03 0.06
Na2O 3.1 0.60 TiO2 0.57 0.68
Sc 20.1 1.4 Na2O 0.24 1.5
Cr 93.6 12.6 Sc 13.7 14.3
Co 20.8 6.2 Cr 71.4 84.8
Ni 49.2 13.6 Co 36.7 18.6
Cu 46.0 8.1 Ni 35.6 48.5
Zn 116 15.7 Cu 113 11.2
Ga 19.7 2.3 Zn 629 167
Rb 78.1 26.5 Ga 23.7 22.7
Sr 249 50.6 Rb 0.64 0.56
Y 19.7 2.0 Sr 9.7 17.2
Zr 77.3 16.5 Y 11.4 25.8
Nb 10.4 2.8 Zr 27.3 52.1
Cs 4.4 1.9 Nb 6.4 8.4
Ba 1,741 1,584 Cs 0.05 0.02
La 21.7 6.7 Ba 5.8 4.1
Ce 44.3 12.9 La 12.5 29.0
Pr 5.6 1.6 Ce 28.0 83.4
Nd 21.9 5.5 Pr 3.6 13.6
Sm 4.7 1.0 Nd 14.2 56.3
Eu 1.2 0.1 Sm 2.8 10.9
Gd 4.2 0.7 Eu 0.17 0.95
Tb 0.63 0.1 Gd 2.4 7.4
Dy 3.7 0.4 Tb 0.36 0.96
Ho 0.73 0.1 Dy 2.2 5.2
Er 2.0 0.2 Ho 0.45 0.98
Tm 0.30 0.01 Er 1.2 2.6
Yb 2.0 0.3 Tm 0.17 0.35
Lu 0.29 0.01 Yb 1.1 2.2
Hf 2.3 0.5 Lu 0.15 0.30
Ta 0.57 0.3 Hf 1.1 1.54
Pb 12.8 4.8 Ta 0.38 0.50
Th 6.9 2.9 Pb 1.5 4.5
U 2.2 0.5 Th 3.1 5.1
U 0.98 11.6
Concentrations of all oxides are expressed in wt %; concentrations of all
other elements expressed in ppm Concentrations of all oxides expressed in wt %; concentrations of trace
Major element compositions determined using an X-ray fluorescence elements are expressed in ppm
spectrometer; trace elements analyzed by ICP-MS Major element compositions determined using an X-ray fluorescence
Number of samples: 5 spectrometer trace elements analyzed by ICP-MS

of hydrothermal alteration. An increase in induration and al-
teration below approximately 30 to 40 m below seafloor is as-
sociated with the disappearance of smectite. Whereas on the
eastern flank of the Bent Hill massive sulfide deposit a zone
of massive and semimassive sulfides occurs between 40 and
80 m below seafloor, sediments from the western flank are
characterized by the occurrence of chlorite-smectite mixed-
layer clays at 36 to 57 m below seafloor. More than approxi-
mately 120 m below seafloor at the western flank, chlorite is
the major authigenic clay mineral coexisting with quartz,
feldspar, and illite, reflecting an increase of hydrothermal al-
teration with temperatures ≥180°C. On the eastern flank of
the Bent Hill massive sulfide, the sedimentary sequence
more than 100 m below seafloor is characterized by the oc-
currence of nearly pure chlorite formed at temperatures
greater than 270°C, indicating high-temperature alteration.
Smectite, chlorite, and chlorite-smectite mixed-layer clays
are the dominant trioctahedral phyllosilicates coexisting with
detrital quartz, feldspar, and illite in the upper approximately
15 m below seafloor of Dead Dog. More than 15 m below
seafloor, the clay fractions show a successive change from
smectite to smectite-rich chlorite-smectite mixed-layer clays,
to corrensite, to chlorite-rich chlorite-smectite mixed-layer
clays, and to chlorite. Studies on clay minerals of hole 858B
from Buatier et al. (1995) report a similiar mineralogical
zonation although smectite, rather than smectite-rich chlo-
rite-smectite mixed-layer clays, was observed. In both holes,
nearly all detrital quartz, feldspar, and illite disappear more
than 20 m below seafloor, whereas quartz in the samples
more than 33 m below seafloor is of authigenic origin.

0361-0128/98/000/000-00 $6.00 377

378 LACKSCHEWITZ ET AL.

FeO wt %

MgO wt % Al2O3 wt %
FIG. 11. Ternary total iron as FeO-MgO-Al2O3 wt percent plot of chlorite, corrensite, and chlorite-smectite from the Area
of Active Venting (AAV) and Bent Hill massive sulfide (BHMS). Corrensite and chlorite-smectite mixed-layer clay samples
plot within or near the field of hydrothermal clay minerals from chimneys in the Area of Active Venting (vertically striped
field = analyses from Percival and Ames, 1993).

FIG. 12. Total Al vs. Si for clay samples from holes 1036A, 858B, 1035D, and 856H on the basis of O20(OH)10. The com-
positions of corrensite, chloritic mixed-layer clays with 90 percent chlorite interlayers, saponite, and clinochlore are shown
for reference.

0361-0128/98/000/000-00 $6.00 378

 FORMATION & TRANSFORMATION OF CLAY MINERALS, JUAN DE FUCA RIDGE 379

1 X-ray fluorescence and microprobe analyses provide chem-

chlorite
chlorite(856H
(856Hand
and1035D)
1035D) ical compositions of nearly pure chlorite from the Bent Hill
0,9 massive sulfide and interlayered chlorite-smectite mixed-
corrensite(1036A
corrensite (1036Aand
and858B)
858B) layer clays from the Area of Active Venting. In the Dead Dog
0,8 C/S
C/S (1036A
(1036Aand
and858B)
858B) and Chowder Hill region of the Area of Active Venting, where
hydrothermal processes are still active, corrensite and chlo-
Si/(Si+Al)

*smectite(AAV)
smectite (AAV) rite-smectite mixed-layer clays have lower Fe and higher Mg
0,7
contents than chlorite in the formerly active Bent Hill mas-
sive sulfide area. This observation indicates alteration by a
0,6 mixed seawater and hydrothermal fluid. Mg-rich smectite
minerals such as saponite occur within active anhydrite-rich
0,5 and sulfide-bearing chimneys in the Area of Active Venting
(Percival and Ames, 1993). The formation of saponite is con-
0,4 sistent with the massive dumping of Mg from seawater en-
0 0,1 0,2 0,3 0,4 0,5 0,6 trained locally into the zone of fluid upflow (Goodfellow et
al., 1993). The MgO contents of these smectites range from
Fe/(Fe+Mg)
21 to 29 wt percent and the FeO content is less than 5 wt per-
FIG. 13. Fe/(Fe + Mg) vs. Si/(Si + Al) plot (ratio of cations per formula cent, which is very similar to our results from samples of holes
unit) of chlorite from holes 856H and 1035D, corrensite from holes 1036A
and 858B, chlorite-smectite mixed-layer clays from holes 1036A and 858B, 858B and 1036A (Fig. 11). At hole 858B, the clay sequence is
and hydrothermal clays from chimneys in the Area of Active Venting (AAV; correlated with the geochemical gradient of pore water, i.e.,
open triangles = analyses from Percival and Ames, 1993). Mg content decreases at 14 m below seafloor in the pore

TABLE 6. Trace Element Analyses of Hydrothermal Strongly Altered Clay Samples from Bent Hill Massive Sulfide and Area of Active Venting

Element 1035D-20X-1 1035D-21X-1 1035D-24X-CC 1036A-3H-2 1036A-3H-6 856H-25R-1 856H-28R-1

129-131 70-73 34-35 129-131 128-130 116-120 44-47
Chl Chl Chl C-S Cor Chl Chl BHVO ICP-MS

Sc 18.8 13.2 20.3 3.0 9.4 18.4 16.3 32.0 32.0

Cr 121 89.0 181 29.9 91.0 126 239 289 300
Co 20.1 16.1 14.0 1.18 16.2 34.5 6.9 45.0 46.4
Ni 49.0 29.2 54.7 15.1 55.2 45.7 38.6 121 124
Cu 151 133 185 166 69.4 389 81.2 136 135
Zn 216 282 211 333 221 229 213 105 106
Ga 21.3 29.0 26.8 10.7 20.2 38.4 36.4 21.0 21.5
Rb 0.17 0.22 0.34 0.22 0.10 0.30 0.28 9.2 9.3
Sr 1.1 0.2 1.2 1.7 1.6 2.4 1.5 403 409
Y 11.0 16.8 4.02 10.0 19.2 23.8 21.7 27.6 24.0
Zr 17.4 49.9 16.9 50.0 59.7 11.1 4.20 179 168
Cs 0.01 0.01 0.02 0.04 0.01 0.01 0.01 0.13 0.10
Ba 5.1 2.9 5.2 72 1.0 1.7 1.9 139 135
Hf 0.679 1.37 0.470 1.02 1.51 0.410 0.210 4.38 4.64
Tl 0.009 0.005 0.010 0.036 0.034 0.000 0.010 0.058 0.060
Pb 2.7 3.6 2.3 93.7 7.6 0.71 0.84 2.06 2.04
Th 3.12 3.75 2.47 1.46 3.66 2.45 1.54 1.08 1.22
U 1.08 1.25 0.517 10.3 8.22 1.10 0.790 0.42 0.44
La 20.0 19.0 4.35 1.73 16.1 8.52 25.58 15.8 15.5
Ce 39.6 42.7 10.2 3.91 50.6 19.6 51.3 39.0 37.6
Pr 4.87 5.59 1.31 0.555 8.25 2.61 6.45 5.70 5.39
Nd 18.7 22.0 5.13 2.55 34.7 11.2 25.4 25.2 24.5
Sm 3.51 4.38 1.01 0.719 6.89 2.95 5.34 6.2 6.1
Eu 0.172 0.227 0.239 0.491 0.697 0.430 0.410 2.06 2.04
Gd 2.96 3.84 0.916 1.00 5.19 3.62 4.93 6.4 6.2
Tb 0.411 0.557 0.140 0.173 0.697 0.640 0.710 0.96 0.94
Dy 2.28 3.26 0.808 1.16 3.91 4.26 4.04 5.2 5.3
Ho 0.436 0.657 0.163 0.264 0.740 0.860 0.760 0.99 0.97
Er 1.24 1.86 0.480 0.819 2.04 2.43 2.08 2.4 2.6
Tm 0.163 0.250 0.071 0.118 0.281 0.330 0.270 0.33 0.33
Yb 0.98 1.56 0.495 0.737 1.79 2.03 1.65 2.02 1.96
Lu 0.131 0.215 0.073 0.109 0.247 0.290 0.210 0.291 0.270
NdN/Y 6.7 5.0 3.8 1.3 7.0 2.0 5.6
Eu*N 18.8 23.7 5.5 4.8 35.4 18.5 29.6
EuN/Eu*N 0.15 0.16 0.73 1.7 0.33 0.39 0.23

BHVO = international reference standard, analyzed during the ICP-MS analytical runs, for controlling accuracy of analytical results; trace elements ana-
lyzed by ICP-MS; concentrations of trace elements expressed in ppm
Subscript N = chondrite-normalized values; superscript * = value obtained by linear interpolation between adjacent elements
Abbreviations: Chl = chlorite, Cor = corrensite, C-S =chlorite-smectite mixed layer

0361-0128/98/000/000-00 $6.00 379

380 LACKSCHEWITZ ET AL.

FIG. 14. Trace element concentrations (in ppm) in clay samples from holes 856H, 1035D, and 1036A.

0361-0128/98/000/000-00 $6.00 380

 FORMATION & TRANSFORMATION OF CLAY MINERALS, JUAN DE FUCA RIDGE 381

FIG. 15. Chondrite-normalized rare earth element patterns of clay samples from holes 856H, 1035D, and 1036A; nor-
malizing values are from Sun and McDonough (1989).

0361-0128/98/000/000-00 $6.00 381

382 LACKSCHEWITZ ET AL.

TABLE 7. Oxygen Isotope Ratio and Calculated Mineral Formation Temperature of Clay Samples from Middle Valley

Isotopic
Depth Interlayered Chlorite Quartz Temperature1
Leg Hole Core Section Interval (mbsf) Minerals chlorite (%) (%) (%) δ18O (°C)

169 856H 25R 1 116–120 144.5 Chlorite 95 5 1.3 273

169 856H 28R 1 44–47 172.5 Chlorite 95 5 0.8 261
169 1035A 5H 5 101–103 42.9 Dolomite –10.2 74
169 1035A 10X 3 78–80 84.0 Quartz, Chlorite 25 75 11.7 180
169 1035A 12X 5 58–60 105.5 Quartz, Chlorite 37 63 10.2 180
169 1035A 14X 4 126–128 124.4 Chlorite-Quartz 57 43 5.8 180
169 1035A 15X 1 28–29 128.5 Chlorite-Quartz 55 45 5.1 180
169 1035A 19X 1 46–48 167.2 Chlorite 90 10 2.6 180
169 1035D 20X 1 129–131 146.0 Chlorite (+Talc) 90 2 0.2 292
169 1035D 21X 1 70–73 155.0 Chlorite 98 2 0.5 277
169 1035D 24X CC 34–35 173.8 Chlorite 100 0 0.4 280
169 1035H 22R 1 11–14 190.4 Chlorite 95 5 –0.26 320
169 1035H 27R 1 98–100 239.3 Chlorite 90 10 0.2 290
169 1036A 3H 2 129–131 21.8 C-S 20 6.0 206
169 1036A 3H 6 128–130 27.8 Corrensite 50 2.6 268
139 858B 2H 3 19–21 11.9 C-S 20 7.3 178
139 858B 2H 4 66–68 13.9 C-S 30 6.8 177
139 858B 5H 2 26–28 25.7 C-S 30 2.7 264
139 858B 5H 3 81–83 27.7 Corrensite 50 2.2 250

C-S = chlorite-smectite mixed layer; mbsf = meters below seafloor

1 Derived from Chl-H O fractionation (Wenner and Taylor, 1971) and from Sm-H O fractionation (Sheppard and Gilg, 1996)
2 2

water of deposits that are composed of Mg-rich phyllosilicates
(Davis et al., 1992). This observation indicates high fluid/rock
ratios near a highly permeable, fractured, and higher temper-
ature upflow zone (Leybourne and Goodfellow, 1994). The
thermal gradient for hole 858B is approximately 10°C/m, in-
dicating temperatures higher than 150°C below 14 m below
seafloor. The removal of Mg from seawater increases from
200° to 300°C, at which temperature it is virtually completely
removed (Seewald et al., 1990). At hole 1036A, there is a
marked increase in the Mg concentration of the pore water
between 20 and 30 m below seafloor. This is a consequence
of intense lateral flow resulting in alteration of the sediments
within the layer (Fouquet et al., 1998). Goodfellow et al.
(1993) indicated that the lateral migration of hydrothermal
fluids and the entrainment of seawater are probably tempo-
rally variable.
15
Hole 1035A
y = 0,1458x + 0,1982
12,5
R2 = 0,9105
10
δ 18O
The Bent Hill massive sulfide represents a hydrothermally
inactive upflow zone that is characterized by Fe-rich, authi-
genic chlorites. Goodfellow and Peter (1994) indicated that
Fe-rich chlorites at Bent Hill formed from an Mg-depleted,
end-member hydrothermal fluid that was probably similar to
the high-temperature (350°C), basalt-equilibrated hydrother-
mal fluids that vent along nonsedimented oceanic rifts (Ed-
mond et al., 1979; Von Damm, 1990). The small amounts of
Mg in the authigenic chlorites from the Bent Hill massive sul-
fide originated either from Mg that was present in the pre-
cursor sediment or from the entrainment of minor amounts
of seawater into the center of fluid upflow.
As shown in Figure 12, chlorite samples from the Bent Hill
massive sulfide have higher ratios of Fe/(Fe + Mg) than cor-
rensite and chlorite-smectite mixed-layer clays samples from
the Area of Active Venting. Several authors (e.g., Bevins et al.,
1991; Schiffman and Staudigel, 1995; Walker and Murphy,
1995) have indicated that the Fe content of chlorites is related
to the Fe content of the host rock. Schiffman and Staudigel
(1995) observed a general decrease in Mg/(Mg + Fe) from
smectite to chlorite in the fossil hydrothermal sytem of La
Palma. These authors suggested that the recharge and dis-
charge of cold seawater and hot hydrothermal fluids, respec-

tively, through the upper oceanic crust may significantly con-

δ18O

7,5
trol the Mg/(Mg + Fe) ratio of trioctrahedral layer silicates. Alt
5 et al. (1989) observed that chlorite-smectite mixed-layer clays
contain more Si and Mg, and less Fe and Al, than chlorite. A
2,5
similiar kind of compositional trend was shown by Shau and
Peacor (1992) for chlorite, corrensite, and saponite in hy-
drothermally altered basalts from Deep Sea Drilling Program
0
hole 504B (Leg 83), reflecting optimal solid solution. Brigatti
0 25 50 75 100
and Poppi (1984) inferred that if Fe/(Fe + Mg) were greater
% Quartz
%Quartz than 0.5, chlorite alone is favored, however, with increasing
FIG. 16. Diagram illustrating a positive correlation between δ18O and per- Mg, chlorite appears to be replaced by corrensite and triocta-
centages of quartz in clay samples from hole 1035A. hedral smectite. Velde (1977) suggested that corrensite tends

0361-0128/98/000/000-00 $6.00 382

 FORMATION & TRANSFORMATION OF CLAY MINERALS, JUAN DE FUCA RIDGE
to be more magnesian and that chlorite is favored to form in
rocks that are Fe2+-rich, based on hydrothermal experiments.
Shau et al. (1990) found that the ratio Fe/(Fe + Mg) and tem-
perature are major factors in controlling the formation of
chlorite and mixed-layer chlorite-smectite or corrensite. Bu-
atier et al. (1995) inferred that the compositional differences
between chlorite and corrensite in hole 858B reflect corren-
site as a unique phase rather than as an intermediate phase
between hydrothermal saponite and chlorite. More recently,
Beaufort et al. (1997) suggested that corrensite should be re-
garded as a true phase in the thermodynamic sense with a dis-
crete stability field and paragenetic relationships with
saponite and chlorite. These authors concluded that chlorite-
rich chlorite-smectite mixed-layer clays are intergrowths or
mixtures of corrensite and chlorite.
In the present study, chlorite-rich chlorite-smectite mixed-
layer clays are always associated with other minerals, permit-
ting detailed investigations of the corrensite to chlorite conver-
sion. However, a linear compositional trend from corrensite
to smectite-rich chlorite-smectite mixed-layer clays in hole
1036A and hole 858B is shown by the Al vs. Si plot (Fig. 12).
This implies that the transition from smectite to corrensite in
sediments from the Area of Active Venting is probably a pro-
gressive conversion. A decrease in Si and an increase in Al
content appear to be characteristic of the saponite-corrensite-
chlorite conversion sequence. In chlorites from the Los
Azufres and Salton Sea geothermal fields, tetrahedral Al and
octahedral cation vacancies both show excellent positive and
negative correlations, respectively, as a function of downhole
temperatures between 130° and 310°C (Cathelineau, 1988).
A similar linear function between Al content and temperature
has also been described for chlorite-smectite mixed-layer
clays in the Nesjavellier geothermal field of Iceland (Schiff-
man and Fridleifsson, 1991). Shau and Peacor (1992) indi-
cated that the availability of Al, as produced by albitization, is
a factor in phyllosilicate paragenesis, in addition to other fac-
tors such as temperature and the Fe(Fe + Mg) ratio of reac-
tant fluids. Therefore, bulk sediment composition, pore water
composition, and temperature are major factors which are
control the hydrothermal formation of the phyllosilicates in
Middle Valley.
The rare earth elements are also affected by hydrothermal
alteration. All chondrite-normalized REE patterns of corren-
site and chlorite are characterized by negative Eu anomalies
and light REE enrichments. The presence of a significant Eu
depletion in corrensite and chlorite reflects high-temperature
hydrothermal alteration of their precursor minerals. Because
plagioclase exhibits a strong positive Eu anomaly and the Eu
anomaly of the least altered clay samples is very small, the de-
velopment of a strong negative Eu anomaly must reflect the
release of Eu into the fluids during alteration of plagioclase.
Michard and Albarede (1986) and Klinkhammer et al. (1994)
have documented that hydrothermal solutions on the East
Pacific Rise are strongly enriched in Eu, which has been in-
terpreted to be related to reduction of Eu3+ to Eu2+ associ-
ated with high temperatures. At high pressures and tempera-
tures above 250°C, Eu2+ should predominate in most natural
systems (Sverjensky, 1984; Wood, 1990). A pure chlorite-
smectite mixed-layer clay, which is formed at lower tempera-
tures (206°C), is characterized by a positive Eu anomaly. A
0361-0128/98/000/000-00 $6.00 383
383
possible explanation for this behavior is that the chlorite-
smectite mixed-layer clay was precipitated at lower tempera-
tures and under redox conditions that resulted in Eu3+ being
the dominant species. The similarity of the Bent Hill chlorite
patterns to the East Pacific Rise fluid patterns indicate that,
with the exception of Eu, the REE are not fractionated dur-
ing the formation of chlorites. Only the corrensite of sample
169-1036A-3H-6, 128 to 130 cm displays distinct changes in
the light REE, suggesting that this corrensite was probably
formed in the presence of fluids with a different history.
Large variations in trace element concentrations in the clay
fraction, shown in Figure 14, indicate that there have been
significant enrichments and depletions of trace elements due
to hydrothermal alteration. In order to characterize the alter-
ation effects, relative elemental gains and losses were calcu-
lated on strongly altered bulk sediment samples from the
Bent Hill massive sulfide deposit (hole 1036A) and the Dead
Dog area (hole 1035D). One widely used method has been
that of Gresens (1967. He argued that, if components are
used which are likely to be related to any volume change that
has taken place, then the gains or losses of other components
can be calculated on the basis of that assumed volume
change. His equation was later modified by Grant (1986) to
one relating the concentration of a component in the altered
rock to that in the original through mass change term, such
that
C Ai = MO/MA (C 0i + ∆Ci), (1)
where C Ai is the concentration of component i in the altered
rock, MO is the mass of the original fresh rock, MA is the mass
of the altered rock, C Oi is the concentration of component i in
the original rock, and ∆Ci is the change in concentration of
component i. Therefore, if the analytical data are plotted as
C Ai vs. C Oi , those elements that are immobile (i.e., ∆Ci = 0)
will define the mass change term (MO/MA) or isocon that can
be used to calculate gains and losses of other components
through a series of simple equations.
For the calculation of elemental gains or losses during al-
teration, it is necessary to identify a precursor. We have se-
lected five representative samples of unaltered hemipelagic
and turbiditic sediment from holes 1035D, 1035E, and 1036A
to determine the precursor chemistry. Table 4 presents an av-
erage of chemical analyses and descriptive statistics of the
bulk sediment samples.
For hydrothermally altered sediments from the Bent Hill
massive sulfide and the Area of Active Venting, two bulk sed-
iment samples from holes 1035D and 1036A were analyzed
(Table 5) to determine the major gains and losses of elements
in the clay-size fraction and in the bulk sediment. For the
Bent Hill massive sulfide deposit, we selected sample 169-
1035D-21X-1, 70 to 73 cm (pure chlorite in the <2 µm frac-
tion) as the most strongly altered material containing greater
than 80 percent clay (<2 µm). The control of fine-grained au-
thigenic mineralogy on the bulk chemical composition is
given by element ratios of bulk sediment vs. clay fraction.
SiO2, Al2O3, MgO, and FeO ratios between 0,8 and 1,2 indi-
cate that these elements are bound mostly in the authigenic
chlorite. CaO and Na2O show higher ratios due to the albiti-
zation of plagioclase and the presence of talc in the coarser
384 LACKSCHEWITZ ET AL.

fraction, whereas the higher ratios of K2O and TiO2 suggest
the presence of some illite-group minerals in the silt fraction.
An isocon plot was constructed for the unaltered precursor
vs. strongly altered bulk rock sample (1035D-21X-1, 70-73
cm) from the Bent Hill massive sulfide (Fig. 17a). Because Al
is assumed to be relatively immobile during alteration
processes (Birkeland, 1984), we have selected Al2O3 to calcu-
late the mass change term (i.e., the slope of a best-fit line or
isocon through Al2O3 to the origin). Thus, the gain or loss of
a mobile component can be measured by the displacement of
its data point from the isocon. The slope of the isocon is 0.95,
which correspond to approximately 5 percent mass gain. The
alteration is characterized by a nearly complete loss (≥95%) of
K2O, Na2O, CaO, Sr, Rb, and Ba, a strong depletion (30–
70%) in SiO2 TiO2, and Zr, and a strong enrichment (60–75%)
in FeO, Zn, Cu, and MgO in the clay fraction. The calculated
mobility (total gains and losses) of elements is summarized in
Figure 17b.
The loss of K, Ti, Ba, Sr, and Rb is governed by the hy-
drothermal alteration of feldspar and illite to chlorite. The loss
of Si and Zr seems to reflect the hydrothermal alteration of zir-
con and detrital quartz, as indicated by Goodfellow and Peter
(1994) for turbiditic silt and sand of Middle Valley. Fe increase
is due to the formation of Fe-rich chlorite, which in turn sug-
gests the presence of Fe-bearing hydrothermal fluids. The en-
richment of the chalcophile elements Cu and Zn is probably
related to the hydrothermal leaching of basalts from the Bent
Hill area. Fouquet et al. (1998) showed that most igneous
rocks from the sills as well as the basement from the Bent Hill
massive sulfide are strongly depleted in these base metals.
Sample 1036A-3H-6, 128 to 130 cm was selected as a
strongly altered sediment for the Area of Active Venting
(Table 5). All detrital components in the clay fraction, which
represents greater than 80 percent of the bulk sediment, are
completely replaced by corrensite. The control of corrensite
on the bulk chemical composition is indicated by the element
ratios of bulk sediment vs. clay fraction. SiO2, Al2O3, MgO,
and FeO contents in the bulk sediment are controlled mostly
by corrensite, as reflected by ratios between 0.88 and 1.1. The
higher ratios of CaO and Na2O are related to the presence of
some anhydrite and albite minerals in the silt and sand frac-
tion, whereas the higher ratios of K2O and TiO2 suggest the
presence of some micas in the silt fraction.
An isocon plot of the unaltered percursor vs. the strongly
altered bulk rock sample (1036A-3H-6, 128-130 cm) confirms
that Al is immobile (Fig. 18a). The slope of the isocon is 0.89,
which corresponds to approximately 10 percent mass gain.
The alteration is characterized by nearly complete loss (≥90%)
of K2O, CaO, Cu, Sr, Rb, and Ba, and a depletion (20–40%)
in SiO2, FeO, Na2O, and Zr due to the hydrothermal alter-
ation of the detrital minerals. A strong enrichment (50–85%)
in MgO and Zn most likely results from the formation of the
Mg-rich corrensite. The mass-balance calculations are sum-
marized in Figure 18b. There is a contrast in alteration styles
of the chlorite sample from the Bent Hill massive sulfide de-
posit and the corrensite sample from the Dead Dog area. The
former shows strong enrichments in FeO and Cu, whereas
the latter shows strong depletions in FeO and Cu. From these
data, the types of precursor phases and the fluid composition
are inferred to have major effects on the chemical compostion

Change (mass % of percursor)

a
A Bb -100 -50 0 50 100
30 0.5Zn SiO2
SS

Depletion Enrichment
A
.M

Al2O3
ST

5
N

0.9

MgO
CO

25
=
CA

FeO
FeO
(Sample 1035D-21X-1)

20 K2O
Altered Sediment

MgO
TiO2
0.5SiO2
Na 2O
CA

15 CaO
Al2O3
Cu
Cu
10 Zn
2TiO 2
K2O Rb
5 Sr
Zr
Na 2O Sr Zr
CaO Rb 0.1Ba
0 Ba 169-1035D-21X-1, 71,5cm
0 5 10 15 20 25 30
Unaltered sediment
Co
FIG. 17. a. Isocon diagram comparing a strongly altered clay sample (Table 5, sample 169-1035D-21X-1, 70-73 cm) from the
lower sediment section with unaltered hemipelagic and turbiditic bulk sediment (Table 4) from Middle Valley. Oxides are
plotted in wt percent and trace elements are plotted in 0.1 ppm. b. Enrichment-depletion diagram showing the relative
changes in element concentrations between a strongly altered bulk sediment (sample 169-1035D-21X-1, 70-73 cm) and un-
altered hemipelagic and turbiditic bulk sediment from Middle Valley.

0361-0128/98/000/000-00 $6.00 384

 FORMATION & TRANSFORMATION OF CLAY MINERALS, JUAN DE FUCA RIDGE 385

Change (mass % of percursor)

A
a
Bb -100 -50 0 50 100
30 MgO
SiO2

SS
Depletion Enrichment

A
.M
Al2O3

ST
N
25 MgO

CO

9
0.8
(Sample 1036A-3H-6)

FeO

=
C A
Altered Sediment

20 K2O
0.5SiO2
TiO2
CA

0.5Zn
Na 2O
15 2TiO 2
Na 2O CaO
Al2O3
K 2O Cu
10 Zn
Fe O Rb
5 Zr Sr
CaO
Sr Zr
Cu
Rb 0.1Ba
0 Ba 169-1036A-3H-6, 129cm

0 5 10 15 20 25 30
Unaltered sediment
Co
FIG. 18. a. Isocon diagram comparing a strongly altered bulk sediment (Table 5, sample 169-1036A-3H-6, 128-130 cm)
with unaltered hemipelagic and turbiditic bulk sediment (Table 4) from Middle Valley. Oxides are plotted in wt percent and
trace elements are plotted in 0.1 ppm. b. Enrichment-depletion diagram showing the relative changes in element concen-
trations between a strongly altered bulk sediment (sample 169-1036A-3H-6, 128-130 cm) and unaltered hemipelagic and tur-
biditic bulk sediment from Middle Valley.

of the authigenic clay minerals in the sedimentary sequences
of Middle Valley.
In active geothermal systems, the smectite-chlorite transi-
tion is strongly temperature dependent (Kristmannsdottir,
1979; Liou et al., 1985; Harvey and Browne, 1991; Schiff-
mann and Fridleifsson, 1991). Früh-Green et al. (1994)
demonstrated that the oxygen isotope record stored in the au-
thigenic phases of the sediments at Middle Valley can be used
to calculate temperature gradients in both the active and fos-
sil hydrothermal systems.
The stable isotope ratios of clay minerals and quartz from
holes 856H, 1035A, 1035D, 1035H, 1036A, and 858B provide
a record of authigenesis associated with high heat flow and
fluid-sediment interaction in a convective hydrothermal sys-
tem. Holes 858B and 1036A are located in the immediate
vicinity of active hydrothermal vents and show high-tempera-
ture alteration and mineralization associated with the dis-
charge of hydrothermal fluids. An increase in temperature
with depth is indicated by a progressive decrease of δ18O val-
ues in the clay fractions from both holes. At hole 858B, the
calculated temperatures are 177° to 178°C at 10 to 13 m
below seafloor and 250° to 264°C at 25 to 27 m below
seafloor. The latter temperatures are similiar to the tempera-
tures calculated using oxygen isotope data from Mg-rich au-
thigenic clays and quartz (Tisot = 265°C at 32 m below seafloor
in hole 858B; Früh-Green et al., 1994). Furthermore, the cal-
culated oxygen isotope fractionation equilibrium tempera-
tures are similiar to the in-situ temperature measurements in
hole 858G (Davis et al., 1992), which is situated approximately
50 m north of hole 858B. These formation temperatures of
the authigenic phases at the central vent area of hole 858B
give an isotope temperature gradient of ≥10°C/m, which is
nearly the same gradient as indicated by Früh-Green et al.
(1994).
The isotopic temperatures of the interstratified clays from
hole 1036A increase from 206°C for chlorite-smectite mixed-
layer clays at 22 m below seafloor to 268°C for corrensite at
28 m below seafloor. These formation temperatures imply an
isotope temperature gradient of 8°C/m, which is slightly
lower than that of hole 858B. However, both holes are char-
acterized by similiarly high geothermal gradients, suggesting
that the hydrothermal fluids at site 858 and site 1036 are part
of one hydrothermal circulation system.
The isotopic composition of hydrothermal chlorites below
140 m below seafloor at hole 1035D gives calculated forma-
tion temperatures between 277° and 292°C which reflect a
paleogeothermal gradient of approximately 2°C/m, consistent
with an in situ temperature of 55°C at 25 m below seafloor
(Fouquet et al., 1998). Isotopic temperatures of chlorites
from the sulfide feeder zone at hole 856H range up to 273°C.
At hole 1035H, hydrothermal chlorites display the highest
mineral formation temperatures at 290° and 320°C, which
are slightly higher than the temperature of 264°C for
presently venting hydrothermal fluids (Ames et al., 1993).
The calculated maximum temperature of 320°C is compara-
ble to a relatively constant maximum fluid inclusion temper-
ature of around 320°C for sites 856, 857, and 858 of Leg 139
(Peter et al., 1994). The temperatures obtained from quartz-
chlorite mixtures of hole 1035A have a relatively consistent
minimum equilibrium temperature of approximately 180°C,

0361-0128/98/000/000-00 $6.00 385

386 LACKSCHEWITZ ET AL.
indicating a significant temperature reduction of the former
hydrothermal system toward the west flank of the Bent Hill
massive sulfide deposit.
The calculated formation temperatures of clay samples
from Middle Valley are comparable to those of smectite-cor-
rensite-chlorite sequences with a range of temperatures be-
tween 60° and 300°C described in other hydrothermal sys-
tems (e.g., Kristmannsdottir, 1979; Alt et al., 1986; Harvey
and Brown, 1991; Schiffman and Fridleifsson, 1991; Beaufort
and Meunier, 1994). The temperature interval over which
smectite first converts to chlorite-smectite mixed-layer clays
is strongly dependent on the geologic environment. In the
Krafla geothermal field of Iceland, chlorite-smectite mixed-
layer clays first occurs at downhole temperatures of 180° to
200°C, becoming predominant at 200° to 240°C (Kristman-
ndottir, 1979), whereas in the Nesjavellir geothermal field of
Iceland, chlorite-smectite mixed-layer clays first occurs at
temperatures greater than 150°C (Schiffman and Fridleifs-
son, 1991) and becomes predominant between 245° to
265°C. In this study we conclude that, in hole 858B, chlorite-
smectite mixed-layer clays first appears downhole, in associa-
tion with chlorite, at approximately 100°C and becomes pre-
dominant at 170° to 180°C. In hole 1036A, the first
appearance of chlorite-smectite mixed-layer clays probably
occurs also at approximately 100°C and chlorite-smectite
mixed-layer clays become dominant at an estimated temper-
ature of 206°C. Thus, the temperatures of the first occurence
and the predominance of chlorite-smectite mixed-layer clays
are somewhat lower than those reported for the geothermal
fields of Iceland. In the Wairakei geothermal system, Harvey
and Brown (1991) reported chlorite to be present at temper-
atures of 110°C, but to be stable only at temperatures above
220°C. The calculated high formation temperatures of 277°
to 290°C and 261° to 273°C at 140 to 180 m below seafloor
for the pure chlorite samples in hole 1035D and hole 856H
are close to the calculated temperatures of 275°C for pure
chlorites of the Nesjavellir geothermal field of Iceland
(Schiffman and Fridleifsson, 1991), whereas the temperature
of 240°C in the Krafla geothermal field is somewhat lower
(Kristmannsdottir, 1979).
The widely different formation temperatures for chlorite,
close to 100°C in some settings and 220° to 270°C in others,
was explained by Harvey and Browne (1991) as being due to
different permeability regimes. In poorly permeable sedi-
ments, where diffuse fluid flow prevails, a clearly defined se-
quence of mixed-layer clays occurs. These are absent where
channel flow dominates and discrete chlorite forms by direct
precipitation from fluids. According to Buatier et al. (1995),
corrensite from hole 858B was precipitated directly from sea-
water-dominated hydrothermal fluids. Therefore, in hole
1036A, where a distinct lateral fluid flow is indicated between
20 and 30 m below seafloor, the monomineralic chlorite-
smectite mixed-layer clays may also be formed by direct pre-
cipitation, whereas the chlorite-smectite mixed-layer clays
associated with other minerals at shallower depths are possi-
bly transformation products formed where diffuse fluid flow
prevails. However, the vertical distribution of chlorite-
smectite mixed-layer clays in holes 1036A and 858B is the
result of a marked thermal gradient. It is obvious that the
chlorite-smectite mixed-layer clays have been evolved toward
0361-0128/98/000/000-00 $6.00 386
the more stable phase of chlorite. Assuming that chlorite-
smectite mixed-layer clays are metastable at all temperatures,
such phases probably all have been replaced by chlorite in the
hydrothermally altered sequences of holes 1035D and 856H.
The distinct mineralogical, chemical, and isotopic changes
in the hydrothermally altered sediments from the Bent Hill
massive sulfide and the Area of Active Venting reflect the
strong control of the fluid-sediment interaction and the ther-
mal gradient on the formation of the phyllosilicate mineral
parageneses. In addition to the observed relationship with
temperature, the compositional variations may be signifi-
cantly affected by other kinetic factors, and also will be de-
pendent on the precursor minerals.
Clay deposits of hydrothermal genesis are known from sev-
eral modern sediment-covered ocean ridge segments such as
the Escanaba trough on the southern Gorda Ridge and the
Guaymas basin in the Gulf of California. In both environ-
ments, Mg-rich phyllosilicates are the predominant alteration
products. Hydrothermal deposits sampled in the Escanaba
trough are characterized by smectite and chlorite alteration.
The formation of Mg smectite proceeds primarily by direct
precipitation in the zone of mixing of hydrothermal fluid with
seawater without extensive replacement of detrital silicates
(Zierenberg and Shanks, 1994). Chloritization is character-
ized by the replacement of hemipelagic sediments by Mg-rich
chlorite resulting from intense Mg metasomatism in a mixing
zone where hydrothermal fluids interact with seawater and
sediment at temperatures above 200°C (Zierenberg et al.,
1993; Zierenberg and Shanks, 1994). Magnesium-rich phyl-
losilicates, including talc, also have been described from hy-
drothermally altered sediment from the Guaymas basin
(Drits et al., 1989). Thus, the alteration processes and prod-
ucts observed in the Area of Active Venting are comparable to
the shallow alteration zones found in the Escanaba trough
and Guaymas basin, whereas the Fe-rich chlorite of the for-
mer active Bent Hill massive sulfide mound are similar to the
Fe-rich phyllosilicates from hydrothermal mounds of the
Galapagos spreading center (Hoffert et al., 1980; Honnorez
et al., 1983) and a volcanic complex at the Tonga convergent
margin (Vitali et al., 1999). According to Honnorez et al.
(1983) and Buatier et al. (1989), the Fe-rich clay minerals
from the Galapagos spreading center are the result of the in-
teraction of hydrothermal solutions and pelagic sediments.
The occurrence of the Fe-rich clays from the Tonga conver-
gent margin yields chemical and mineralogical evidence for
thermal diagenesis (Vitali et al., 1999).
In addition, several onshore massive sulfide deposits, in-
cluding Kidd Creek, Canada (Slack and Coad, 1989) and
ophiolite-hosted deposits in Cyprus (Richards et al., 1989)
and Oregon (Zierenberg et al., 1988), have alteration zones
that are characterized by the presence of Fe-rich chlorite.
These alteration zones are interpreted to reflect hydrother-
mal fluid-rock interaction below the zone of seawater mixing,
similar to the formation of pure Fe-rich chlorite phases below
the massive sulfide deposits of the Bent Hill massive sulfide.
Conclusions
The Middle Valley is one of the best areas where recent
(<300,000 yr ago) and present-day oceanic hydrothermal
processes at sedimented ridges can be studied. One of the
 FORMATION & TRANSFORMATION OF CLAY MINERALS, JUAN DE FUCA RIDGE
main characteristics of hydrothermal deposits in Middle Val-
ley is the intensive hydrothermal alteration causing secondary
minerals. Significant differences in secondary mineralogy
occur in the clay fraction of hydrothermal deposits from ODP
sites 1035, 1036, and 856 at Middle Valley. Detailed miner-
alogical, chemical, and isotopic studies on phyllosilicates indi-
cate variations in the degree of alteration.
In the immediate vent areas of the Area of Active Venting,
at holes 1036A and 858B, hydrothermal alteration is charac-
terized by a systematic change with depth. Detrital assem-
blages of quartz, feldspar, illite, chlorite, and chlorite-smec-
tite mixed-layer clays are the major minerals in the upper 20
m. The chlorite-smectite mixed-layer clays indicate formation
temperatures of less than 180°C. An intensely altered and
mineralized zone occurs below 20 m, where pure chlorite-
smectite mixed-layer clays, corrensite, and chlorite formed at
temperatures of 200° to 270°C. The highest mineral forma-
tion temperatures are consistent with present-day vent tem-
peratures of 261° to 276°C. The Mg-rich nature of the clays
suggests that a partly reacted seawater component was pre-
sent in addition to hydrothermal fluids.
Fe-rich chlorites are the major authigenic clay minerals at
the Bent Hill massive sulfide deposit. They are generally de-
pleted in 18O, but δ18O values in holes 1035D and 856H
(<1.3‰) are lower than in hole 1035A, reflecting higher al-
teration temperatures at the summit and at the eastern flank
of the Bent Hill massive sulfide. A recently high-temperature
(270° to 290°C) alteration stage produced nearly monomin-
eralic chlorite layers. Progessive reactions at somewhat lower
temperatures (<180°C) resulted in formation of dolomite,
chlorite-smectite mixed-layer clays, corrensite, chlorite, and
quartz at the western flank of the Bent Hill massive sulfide. A
δ18O value of –0.3 per mil from hole 1035H at the ODP
mound reflects reactions with hydrothermal fluids at a tem-
perature of 320°C. The observed clay mineral sequences in
Middle Valley are similar to the smectite-corrensite-chlorite
sequences described in other geothermal fields.
The high temperature alteration stages caused enrich-
ments and depletions of metals from the clay fractions.
Chemical changes in strongly altered clay fractions from the
Bent Hill massive sulfide include increased Fe, Mg, Zn, and
Cu contents and a nearly complete loss of K, Ti, Na, Rb, Sr,
and Ba. The nearly pure chlorites are characterized by a dis-
tinct negative Eu anomaly and by light REE enrichments
with NdN/YbN values from 2 to 6.7. This suggests that signif-
icant fractionation of Eu relative to the other REE has taken
place during high-temperature hydrothermal alteration
processes because of the expected stability of Eu2+ during
these processes. A similar Eu anomaly is observed for cor-
rensite from the Area of Active Venting, which exhibits a
high formation temperature of 268°C. The formation of cor-
rensite and chlorite-smectite mixed-layer clays replacing
quartz, feldspar, and illite during hydrothermal alteration led
to increased Mg and Zn contents. As with the chlorite from
the Bent Hill massive sulfide, the corrensite and chlorite-
smectite mixed-layer clays are characterized by a nearly com-
plete loss of K, Ca, Na, Rb, Sr, and Ba, indicating similarities
in the chemical alteration processes between the present-day
active Area of Active Venting and the recently active Bent
Hill massive sulfide area.
0361-0128/98/000/000-00 $6.00 387
387
The alteration processes and products observed in the Area
of Active Venting are comparable to the shallow marine alter-
ation zones of other seafloor hydrothermal deposits on sedi-
ment-covered spreading centers in the Gulf of California and
on the southern Gorda Ridge. Otherwise, the Fe enrichment
in alteration zones underlying massive sulfide deposits of
Bent Hill is interpreted to reflect hydrothermal fluid-rock in-
teraction below the zone of seawater mixing. This type of al-
teration also has been described for several ancient massive
sulfide deposits.
Acknowledgments
The authors thank the following staff at Kiel University
(Germany) and the Hebrew University of Rehovot (Israel): O.
Krüger, J. Scholten (Kiel University), and S. Hochman (He-
brew University) for assistance with XRD and FTIR analyses;
N. Bahat (Hebrew University) and M. Czank (Kiel Univer-
sity) for assistance with TEM; C. Samtleben (Kiel University)
and N. Bahat (Hebrew University) for assistance with SEM;
A. Neaman (Hebrew University) for assistance with DTA; D.
Ackermand (Kiel University) for electron microprobe analy-
ses; C.W. Devey (Kiel University) and D. Rau (Geomar Kiel)
for XRF analyses; H. Blaschek and T. Arpe (Kiel University)
for sample preparation and ICP-MS analyses; I. Dold (Kiel
University) for sample preparation for oxygen isotope analy-
ses; U. Schuldt (Kiel University) for providing SEM photomi-
crographs; and A. Gutzmann and M. Kummetz for sample
preparation. We are grateful to E. Vladimir at the Ben Gurion
University of Negev (Israel) for HRTEM/AEM analyses. Dis-
cussions with N. Goldman (Hebrew University) and R. Mühe
(Kiel University) were most beneficial. Critical reviews by M.
Buatier and P. Schiffman significantly improved the manu-
script. Our English was edited by C.W. Devey. This work was
financially supported by the Deutsche Forschungsgemein-
schaft grant Sto 110/26.
November 30, 1998; October 22, 1999
REFERENCES
Alt, J.C., Honnorez, J., Laverne, C., and Emmermann, R., 1986, Hydrothermal
alteration of a 1 km section through the upper oceanic crust, Deep Sea
Drilling Project hole 504B: Mineralogy, chemistry, and evolution of seawater-
basalt interactions: Journal of Geophysical Research, v. 91, p. 10,309–10,335.
Alt, J.C., Anderson, T.F., Bonnell, L., and Muehlenbachs, K., 1989, Mineral-
ogy, chemistry, and stable isotopic compositions of hydrothermally altered
sheeted dikes: ODP hole 504B, Leg 111: Ocean Drilling Program Scientific
Results 111, Proceedings, College Station, p. 27–39.
Ames, D.E., Franklin, J.M., and Hannington, M.D., 1993, Mineralogy and
geochemistry of active and inactive chimneys and massive sulfide, Middle
Valley, northern Juan de Fuca Ridge: An evolving hydrothermal system:
Canadian Mineralogist, v. 31, p. 997–1024.
Barnhisel, R.I., and Bertsch, P.M., 1989, Chlorites and hydroxy-interlayered
vermiculite and smectite, in Dixon, J.B., and Weed, S.B., eds., Minerals in
soil environments: Soil Science Society of America, Madison, Wisconsin, p.
729–788.
Beaufort, D., and Meunier, A., 1994, Saponite, corrensite, and chlorite/
saponite mixed-layer minerals and saponite in the Sancerre-Couy deep drill
hole (France): Clay Minerals, v. 29, p. 47–61.
Beaufort, D., Baronnet, A., Lanson, B., and Meunier, A., 1997, Corrensite: A
single phase or a mixed-layer phyllosilicate in the saponite-to-chlorite con-
version series? A case study of Sancerre-Couy deep drill hole (France):
American Mineralogist, v. 82, p. 109–124.
Bettison, L.A., and Schiffman, P., 1988, Compositional and structural varia-
tions of phyllosilicates from Point Sal ophiolite, California: American Min-
eralogist, v. 73, p. 62–76.
388 LACKSCHEWITZ ET AL.
Bevins, R.E., Robinson, D., and Rowbotham, G., 1991, Compositional varia-
tions in mafic phyllosilicates from regional low-grade metabasites and ap-
plication of the chlorite geothermeter: Journal of Metamorphic Geology, v.
9, p. 711–721.
Birkeland, P.W., 1984, Soils and geomorphology: London, Oxford University
Press.
Borthwick, J., and Harmon, R.S., 1982, A note regarding ClF3 as an alterna-
tive to BrF3 for oxygen isotope analysis: Geochemica et Cosmochimica
Acta, v. 46, p. 1665–1668.
Brindley, D.W., and Brown, G., 1980, Crystal structures of clay minerals and
their X-ray identification: London, Mineralogical Society, 495 p.
Brigatti, M.F., and Poppi, L., 1984, Crystal chemistry of corrensite: A review:
Clay and Clay Minerals, v. 32, p. 391–399.
Buatier, M., Honnorez, J., and Ehret, G., 1989, Fe-smectite-glauconite tran-
sition in hydrothermal green clays from the Galapagos spreading center:
Clay and Clay Minerals, v. 37, p. 532–541.
Buatier, M.D., Karpoff, A.-M., Boni, M., Früh-Green, G.L., and. McKenzie,
J.A., 1994, Mineralogic and petrographic records of sediment-fluid interac-
tion in the sedimentary sequence at Middle Valley, Juan de Fuca Ridge,
Leg 139: Ocean Drilling Program Scientific Results 139, Proceedings, Col-
lege Station, p. 133–154.
Buatier, M.D., Früh-Green, G.L., and Karpoff, A.M., 1995, Mechanism of
Mg-phyllosilicate formation in a hydrothermal system at a sedimented
ridge (Middle Valley, Juan de Fuca): Contributions to Mineralogy and
Petrology, v. 122, p. 134–151.
Butterfield, D.A., McDuff, R.E., Franklin, J., and Wheat, C.G., 1994, Geo-
chemistry of hydrothermal vent fluids from Middle Valley, Juan de Fuca
Ridge: Ocean Drilling Program Scientific Results 139, Proceedings, Col-
lege Station, p. 395–410.
Cathelineau, M., 1988, Cation site occupancy in chlorite and illites as a func-
tion of temperature: Clay Minerals, v. 23, p. 471–485.
Clayton, R.N., and Mayeda, T.K., 1963, The use of bromine pentafluoride in
the extraction of oxygen from oxides and silicates for isotopic analysis: Geo-
chemica et Cosmochimica Acta, v. 27, p. 43–52.
Clayton, R.N., O’Neil, J.R., and Mayeda, T.K., 1972, Oxygen isotope ex-
change between quartz and water: Journal of Geophysical Research, v. 77-
17, p. 3057–3067.
Cole, D.R., Riciputi, L.R., and Horita, J., 1998, Stable isotope exchange equi-
libria and kinetics in mineral-fluid systems, in Arehart, G.B., and Hulston,
J.R., eds., Water-Rock Interaction: Rotterdam, Balkema, 827–830.
Davis, E.E., and Fisher, A.T., 1994, On the nature and consequences of hy-
drothermal circulation in the Middle Valley sedimented rift: Inferences
from geophysical observations, ODP Leg 139: Ocean Drilling Program Sci-
entific Results 139, Proceedings, College Station, p. 695–720.
Davis, E.E., and Villinger, H., 1992, Tectonic and thermal structure of the
Middle Valley sedimented rift, northern Juan de Fuca Ridge: Ocean
Drilling Program Scientific Results 139, Proceedings, College Station, p.
9–41.
Davis, E.E., Mottl, M.J., Fisher, A.T., et al., 1992, Ocean Drilling Program
Scientific Results 139, Initial Report, College Station, Proceedings, p.
1–1026.
Drits, V.A., Khvorova, I.V., Sokolova, A.L., and Voronii, B.I., 1989, Clay min-
erals in deep water hydrothermal structures of the Guayamas depression
(Gulf of California): Lithology and Mineral Resources, v. 24, p. 10–16.
Edmond, J.M., Measures, C., McDuff, R.E., Chan, L.H., Collier, R., and
Grant, B., 1979, Ridge crest hydrothermal activity and the balances of
major and minor elements in the ocean: The Galapagos data: Earth and
Planetary Science Letters, v. 46, p. 1–19.
Farmer, V.C., 1974, The infrared spectra of minerals: Mineralogical Society,
Monograph 4, 536 p.
Fouquet, Y., Zierenberg, R.A., Miller, D.J., et al. 1998, Ocean Drilling Pro-
gram Scientific Results 169, Initial Report, Proceedings, College Station, p.
Früh-Green, G.L., McKenzie, J.A., Boni, M., Karpoff, A.M., and Buatier, M.,
1994, Stable isotope and geochemical record of convective hydrothermal
circulation in the sedimentary sequence of Middle Valley, Juan de Fuca
Ridge, Leg 139: Ocean Drilling Program Scientific Results 139, Proceed-
ings, College Station, p. 291–305.
Garbe-Schönberg, D., 1993, Simultaneous determination of 37 trace ele-
ments in 28 international rock standards by ICP/MS: Geostandards
Newsletter, v. 17, p. 81–97.
Goodfellow, W.D., and Blaise, B., 1988, Sulfide formation and hydrothermal
alteration of hemipelagic sediment in Middle Valley, northern Juan de Fuca
Ridge: Canadian Mineralogist, v. 26, p. 675–696.
0361-0128/98/000/000-00 $6.00 388
Goodfellow, W.D., and Peter, J.M., 1994, Geochemistry of hydrothermal al-
tered sediment, Middle Valley, northern Juan de Fuca Ridge: Ocean
Drilling Program Scientific Results 139, Proceedings, College Station, p.
207–290.
Goodfellow, W.D., Grapes, K., Cameron, B., and Franklin, J.M., 1993, Hy-
drothermal alteration associated with massive sulfide deposits, Middle Val-
ley, northern Juan de Fuca Ridge: Canadian Mineralogist, v. 31, p.
1025–1060.
Grant, J.A., 1986, The isocon diagramæa simple solution to Gresen’s equation
for metasomatic alteration: ECONOMIC GEOLOGY, v. 81, p. 1976–1982.
Gresens, R.L., 1967, Composition-volume relationships of metasomatism:
Chemical Geology, v. 2, p. 47–65.
Harvey, C.C., and Browne, P.R., 1991, Mixed-layer clay geothermometry in
the Wairakei geothermal field, New Zealand: Clays and Clay Minerals, v.
39, p. 614–621.
Hoffert, M., Person, A., Courtois, C., Karpoff, A.M., and Trauth, D., 1980,
Sedimentology, mineralogy, and geochemistry of hydrothermal deposits
from holes 424, 424A, 424B, and 424C (Galapagos spreading center): Ini-
tial Reports of the Deep Sea Drilling Project, Washington, D.C., U.S. Gov-
ernment Printing Office, v. 54, p. 339–376.
Honnorez, J., Karpoff, A.M., and Trauth Badaut, D., 1983, Sedimentology,
mineralogy and geochemistry of green clay samples from the Galapagos hy-
drothermal mounds, holes 506, 506C, and 507D, Deep Sea Drilling Project
leg 70: Initial Reports of the Deep Sea Drilling Project, Washington, D.C.,
U.S. Government Printing Office, v. 70, p. 221–224.
Howard, K.J., and Fisk, M.R., 1988, Hydrothermal alumina-rich clays and
boehmite on the Gorda Ridge: Geochimica et Cosmochimica Acta, v. 52, p.
2269–2279.
Klinkhammer, G.P., Elderfield, H., Edmond, J.M., and Mitra, A., 1994, Geo-
chemical implications of rare earth element patterns in hydrothermal flu-
ids from mid-ocean ridges: Geochimica et Cosmochimica Acta, v. 58, p.
5105–5113.
Kristmannsdottir, H., 1979, Alteration of basaltic rocks by hydrothermal ac-
tivity at 100°–300°C, in Mortland, M., and Farmer, V., eds., Developments
in sedimentology: Amsterdam, Elsevier, p. 359–367.
Leybourne, M.I., and Goodfellow, W.D., 1994, Mineralogy and mineral
chemistry of hydrothermally altered sediment, Middle Valley, Juan de Fuca
Ridge: Ocean Drilling Program Scientific Results 139, Proceedings, Col-
lege Station, p. 155–206.
Liou, J.G., Seki, Y., Guillemette, R.N., and Sakai, H., 1985, Composition and
parageneses of secondary minerals in the Onikobe geothermal system:
Chemical Geology, v. 49, p. 1–20.
Matsuhisa, Y., 1974, 18O/16O ratios for NBS-28 and some silicate reference
samples: Geochemical Journal, v. 8, p. 103–107.
McCrea, J.M., 1950, On the isotopic chemistry of carbonates and a pale-
otemperature scale: Journal of Chemical Physics, v. 18, p. 849–857.
Mehra, O.P., and Jackson, M.L., 1960, Iron oxide removal from soils and
clays by dithionite-citrate system buffered with sodium bicarbonate: Clays
and Clay Minerals, v. 7, p. 317–327.
Michard, A., and Albarede, F., 1986, The REE content of some hydrother-
mal fluids: Chemical Geology, v. 55, p. 51–60.
Moore, D.M., and Reynolds, C.R., Jr., 1989, X-ray diffraction and the identi-
fication and analysis of clay minerals: New York, Oxford University Press,
322 p.
Newman, A.C.D., and Brown, G., 1987, The chemical constitution of clays,
in Newman, A.C.D., ed., Chemistry of clay and clay minerals: Mineralogi-
cal Society of London, Monograph 6, p. 480.
Paterson, E., and Swaffield, R., 1987, Thermal analysis, in Wilson, M.J., ed.,
A handbook of determinative methods in clay mineralogy: New York,
Chapman and Hall, p. 99–132.
Percival, J.B., and Ames, D.E., 1993, Clay mineralogy of active hydrothermal
chimneys and an associated mound, Middle Valley, northern Juan De Fuca
Ridge: Canadian Mineralogist, v. 31, p. 957–971.
Peter, J.M., Goodfellow, W.D., and Leybourne, M.I., 1994, Fluid inclusion
petrography and microthermometry of the Middle Valley hydrothermal
system, northern Juan de Fuca Ridge: Ocean Drilling Program Scientific
Results 139, Proceedings, College Station, p. 411–428.
Richards, H.G., Cann, J.R., and Jensenius, J., 1989, Mineralogical zonation
and metasomatism of the alteration pipes of Cyprus sulfide deposits: ECO-
NOMIC GEOLOGY, v. 84, p. 91–115.
Savin, S.M., and Lee, M., 1988, Isotopic study of phyllosilicates, in Bailey,
S.W., ed., Hydrous phyllosilicates (exclusive of muscovite): Mineralogical
Society of America, Review of Mineralogy, v. 19, p. 189–223.
 FORMATION & TRANSFORMATION OF CLAY MINERALS, JUAN DE FUCA RIDGE
Schiffman, P., and Fridleifsson, G.O., 1991, The smectite-chlorite transition
in drillhole NJ-15, Nesjavellir geothermal field, Iceland: XRD, BSE and
electron microprobe investigations: Journal of Metamorphic Geology, v. 9,
p. 679–696.
Schiffman, P., and Staudigel, H., 1995, The smectite to chlorite transition in
a fossil seamount hydrothermal system: The basement complex of La
Palma, Canary Islands: Journal of Metamorphic Geology, v. 13, p. 487–498.
Seewald, J.S., Seyfried, W.E., and Thornton, E.C., 1990, Organic-rich sedi-
ment alteration: An experimental and theoretical study at elevated temper-
atures and pressures: Applied Geochemistry, v. 5, p. 193–209.
Sharp, Z.D., and Kirschner, D.L., 1994, Quartz-calcite oxygen isotope ther-
mometry: A calibration based on natural isotopic variations: Geochimica et
Cosmochimica Acta, v. 58, p. 4491–4501.
Shau, Y.-H., and Peacor, D.R., 1992, Phyllosilicates in hydrothermally altered
basalts from DSDP hole 504B, Leg 83æa TEM and AEM study: Contribu-
tions to Mineralogy and Petrology, v. 112, p. 119–133.
Shau, Y.-H., Peacor, D.R., and Essene, E.J., 1990, Corrensite and mixed-
layer chlorite-smectite in metabasalt from northern Taiwan: TEM/AEM,
EMPA, XRD, and optical studies: Contributions to Mineralogy and Petrol-
ogy, v. 105, p. 123–142.
Sheppard, S.M.F., and Gilg, H.A., 1996, Stable isotope geochemistry of clay
minerals: Clay Minerals, v. 31, p. 1–24.
Shiro, Y., and Sakai, H., 1972, Calculation of the reduced partition function
ratios of alpha-beta quartz and calcite: Japan Chemical Society Bulletin, v.
45, p. 2355–2359.
Slack, J.F., and Coad, P.R., 1989, Multiple hydrothermal and metamorphic
events in the Kidd Creek volcanogenic massive sulfide deposit, Timmins,
Ontario: Evidence from tourmalinites and chlorites: Canadian Journal of
Earth Sciences, v. 26, p. 694–715.
Sun, S.S., and McDonough, W.F., 1989, Chemical and isotopic systematics of
oceanic basalts: Implications for mantle composition and processes: Geo-
logical Society of London, Special Publication, v. 42, p. 313–345.
Sverjensky, D.A., 1984, Europium redox equilibria in aqueous solution:
Earth and Planetary Science Letters, v. 67, p. 70–78.
Velde, B., 1977, Clays and clay minerals in natural and synthetic systems:
Amsterdam, Elsevier, 218 p.
Vitali, F., Blanc, G., Largue, P., Duplay, J., and Morvan, G., 1999, Thermal
diagenesis of clay minerals within volcanogenic material from the Tonga
convergent margin: Marine Geology, v. 157, p. 105–125.
0361-0128/98/000/000-00 $6.00 389
389
Von Damm, K.L., 1990, Seafloor hydrothermal activity: Black smoker chem-
istry and chimneys: Annual Annual Review of Earth and Planetary Sci-
ences, v. 18, p. 173–204.
Walker, J.R., and Murphy, M., 1995, Chloritic minerals from zeolite and
prehnite pumpellyite facies rocks of the Winterville Formation, Aroostock
County, Maine: Geological Society of America Special Paper 296, p.
141–156.
Wenner, D.B., and Taylor, H.P., 1971, Temperatures of serpentinization of
ultramafic rocks based on 18O/16O fractionation between coexisting serpen-
tine and magnetite: Contributions to Mineralogy and Petrology, v. 32, p.
165–185.
Wilson, M.J., 1994, Clay mineralogy: Spectroscopic and chemical determina-
tive methods: London, Chapman and Hall, 367 p.
Wood, S.A., 1990, The aqueous geochemistry of the rare-earth elements and
yttrium, 1. Review of available low-temperature data for inorganic com-
plexes and the inorganic REE speciation of natural waters: Chemical Ge-
ology, v. 82, p. 159–186.
Zierenberg, R.A., and Shanks, W.C., III, 1994, Sediment alteration associ-
ated with massive sulfide formation in Escanaba trough, Gorda Ridge: The
importance of seawater mixing and magnesium metasomatism: U.S. Geo-
logical Survey Bulletin, v. 2022, p. 257–278.
Zierenberg, R.A., Shanks, W.C., III, Seyfried, W.E., Jr., Koski, R.A., and
Strickler, M.D., 1988, Mineralization, alteration and subseafloor metamor-
phism of the ophiolite-hosted Turner-Albright sulfide deposit, southwest-
ern Oregon: Journal of Geophysical Research, v. 93, p. 4657–4674.
Zierenberg, R.A., Koski, R.A., Morton, J.L., and Bouse, R.M., 1993, Genesis
of massive sulfide deposits on a sediment-covered spreading center, Es-
canaba trough, southern Gorda Ridge: ECONOMIC GEOLOGY, v. 88, p.
2069–2098.
Zierenberg, R.A., Fouquet, Y., Miller, D.J., Bahr, J.M., Baker, P.A., Bjergard,
T., Brunner, C.A., Duckworth, R.C., Gable, R., Gieskes, J., Goodfellow,
W.D., Gröschel-Becker, H.M., Guerin, G., Ishibashi, J., Iturrino, G., James,
R.H., Lackschewitz, K.S., Marquez, L.L., Nehlig, P., Peter, J.M., Rigsby,
C.A., Schultheiss, P., Shanks, W.C., III, Simoneit, B.R.T., Summit, M., Tea-
gle, D.A., Urbat, M., and Zuffa, G.G., 1998, The deep structure of a sea-
floor hydrothermal deposit: Nature, v. 392, p. 485–488.
0361-0128/98/000/000-00 $6.00 390