Patten 2016

Available online at www.sciencedirect.
com
ScienceDirect
Geochimica et Cosmochimica Acta 193 (2016) 132–159
www.elsevier.com/locate/gca
Sulphide mineral evolution and metal mobility during

alteration of the oceanic crust: Insights from ODP Hole 1256D
C.G.C. Patten a,⇑, I.K. Pitcairn a, D.A.H. Teagle b, M. Harris c
a
Department of Geological Sciences, Stockholm University, Stockholm, Sweden
b
Ocean and Earth Science, National Oceanography Centre Souhtampton, University of Southampton, United Kingdom
c
School of Geography, Earth and Environmental Sciences, Plymouth University, United Kingdom
Received 8 January 2016; accepted in revised form 4 August 2016; available online 10 August 2016
Abstract
Fluxes of metals during the hydrothermal alteration of the oceanic crust have far reaching effects including buffering of the
compositions of the ocean and lithosphere, supporting microbial life and the formation of sulphide ore deposits. The mech-
anisms responsible for metal mobilisation during the evolution of the oceanic crust are complex and are neither fully con-
strained nor quantified. Investigations into the mineral reactions that release metals, such as sulphide leaching, would
generate better understanding of the controls on metal mobility in the oceanic crust. We investigate the sulphide and oxide
mineral paragenesis and the extent to which these minerals control the metal budget in samples from Ocean Drilling Program
(ODP) Hole 1256D. The ODP Hole 1256D drill core provides a unique sample suite representative of a complete section of a
fast-spreading oceanic crust from the volcanic section down to the plutonic complex. The sulphide population at Hole 1256D
is divided into five groups based on mineralogical assemblage, lithological location and texture: the magmatic, metasomatised,
high temperature hydrothermal, low temperature and patchy sulphides. The initiation of hydrothermal alteration by down-
ward flow of moderate temperature (250–350 °C) hydrothermal fluids under oxidising conditions leads to metasomatism of
the magmatic sulphides in the sheeted dyke and plutonic complexes. Subsequent increase in the degree of hydrothermal alter-
ation at temperatures >350 °C under reducing conditions then leads to the leaching of the metasomatised sulphides by rising
hydrothermal fluids. Mass balance calculations show that the mobility of Cu, Se and Au occurs through sulphide leaching
during high temperature hydrothermal alteration and that the mobility of Zn, As, Sb and Pb is controlled by silicate rather
than sulphide alteration. Sulphide leaching is not complete at Hole 1256D and more advanced alteration would mobilise
greater masses of metals. Alteration of oxide minerals does not release significant quantities of metal into the hydrothermal
fluid at Hole 1256D. Mixing of rising high temperature fluids with low temperature fluids, either in the upper sheeted dyke
section or in the transitional zone, triggers local high temperature hydrothermal sulphide precipitation and trapping of Co,
Ni, Cu, Zn, As, Ag, Sb, Se, Te, Au, Hg and Pb. In the volcanic section, low temperature fluid circulation (<150 °C) leads to
low temperature sulphide precipitation in the form of pyrite fronts that have high As concentrations due to uptake from the
circulating fluids. Deep late low temperature circulation in the sheeted dyke and the plutonic complexes results in local pre-
cipitation of patchy sulphides and local metal remobilisation. Control of sulphides over Au, Se and Cu throughout fast-
spreading mid-oceanic crust history implies that the generation of hydrothermal fluids enriched in these metals, which can
eventually form VMS deposits, is strongly controlled by sulphide leaching.
Ó 2016 Elsevier Ltd. All rights reserved.
Keywords: magmatic sulphide; sulphide leaching; metal mobilisation; hydrothermal alteration; gold
⇑ Corresponding author.
E-mail address: clifford.patten@geo.su.se (C.G.C. Patten).
http://dx.doi.org/10.1016/j.gca.2016.08.009
0016-7037/Ó 2016 Elsevier Ltd. All rights reserved.
C.G.C. Patten et al. / Geochimica et Cosmochimica Acta 193 (2016) 132–159 133
1. INTRODUCTION Patten et al., 2015). These metals are depleted at different

crustal levels ranging from the upper sheeted dykes down
The cycling of metals in the oceanic crust is an impor- to the plutonic section implying that multiple mineral reac-
tant process which buffers the composition of the litho- tions may control metal mobility.
sphere and oceans, contributes to development of This investigation aims to determine the paragenesis and
microbial life, and leads to the formation of hydrothermal metal contents of the sulphide and oxide minerals in the
ore deposits such as volcanogenic massive sulphide ODP Hole 1256D crust in order to identify the mineral
(VMS) deposits. Investigations from modern-day oceanic reactions that release metals into the hydrothermal fluids.
crust (Nesbitt et al., 1987; Alt et al., 1989, 1996, 2010; We employ in situ analytical methods including electron-
Alt, 1995; Gillis et al., 2001; Bach et al., 2003; Teagle probe microanalysis (EPMA) and laser ablation-
et al., 2006; Heft et al., 2008; Coogan and Dosso, 2012; inductively coupled plasma mass spectrometer (LA-ICP-
Patten et al., 2015) and ophiolites (e.g. Richardson et al., MS) analysis to enable quantification of major, minor
1987; Schiffman et al., 1987; Jowitt et al., 2012) suggest that and trace elements in sulphide and oxide minerals present
mobilisation of metals from the lower sheeted dyke com- in the Hole 1256D drill core. We compare our data with
plex during on-axis high temperature hydrothermal alter- published whole-rock concentrations for the same samples
ation of the oceanic crust is a systematic phenomenon. from Hole 1256D (Patten et al., 2015) in order to identify
The high temperature (>350 °C) hydrothermal fluids that the host minerals for this suite of metals and the mineral
leach metals from the lower sheeted dyke section rise buoy- reactions that liberate metals into hydrothermal fluids dur-
antly through upflow zones towards the seafloor where ing alteration.
venting occurs (Alt et al., 1989, 2010; Hannington, 2014).
During this ascent metals can be lost to mineralised hori- 2. GEOLOGICAL SETTING
zons within the upper oceanic crust (Honnorez et al.,
1985; Alt et al., 1998, 2010; Bach et al., 2003; 2.1. ODP Hole 1256D lithologic units
Hannington, 2013), trapped within VMS deposits (e.g.
Fouquet et al., 1988; Herzig and Hannington, 1995; The oceanic crust at drilling site Hole 1256D is a 15 Myr
Hannington et al., 1998; Wohlgemuth-Ueberwasser et al., old crust generated from a superfast spreading ridge
2015) or vented in hydrothermal plumes and associated sed- (200 mm/yr) and is located in the Cocos Plate (6.736° N,
iments (e.g. Feely et al., 1994a,b; Hannington, 2013). 91.934° W; Wilson et al., 2003). The oceanic crust can be
Mobilisation of metals in the lower sheeted dykes occurs divided into four main lithological units: the volcanic sec-
at high temperature and under reduced conditions (>350 ° tion, the transitional zone, the sheeted dyke complex and
C; e.g. Alt et al., 2010) when solubility of metals is optimal the plutonic complex (Fig. 1). The volcanic section is
(Seewald and Seyfried, 1990). It is generally assumed that capped by a lava pond on the first 100 m that was formed
this mobilisation is partly achieved through leaching of during off-axis volcanism events (Wilson et al., 2003) under-
magmatic sulphides by hydrothermal fluids (e.g. Keays, lined by phyric to aphyric sheeted flows, aphyric massive
1987; Alt, 1995; Jowitt et al., 2012). The extent, however, units, pillow lavas and hyaloclastites (Wilson et al., 2003;
to which this mineral reaction controls the release of base Teagle et al., 2006). The transitional zone is mainly com-
metals and other trace metals such as Au, As, Sb, Se and posed of aphyric sheeted flows with presence of breccias
Te into the hydrothermal fluids is neither constrained nor characterised by angular aphyric cryptocrystalline basaltic
quantified. Reactions involving oxide and silicate minerals clasts cemented by chalcedony, saponite, carbonate, albite,
have also been shown to mobilise metals including Zn anhydrite and sulphides (Teagle et al., 2006). The sheeted
(Doe, 1994; Jowitt et al., 2012). The mineral reactions that dyke complex comprises massive aphyric basalts with com-
produce metal-rich fluids in the modern-day oceanic crust mon sub-vertical intrusive contacts (Teagle et al., 2006).
are not well constrained and systematic investigation of Due to intrusion of underlying gabbros, high-temperature
the sulphide population evolution during hydrothermal recrystallisation occurs in the lower part of the sheeted dyke
alteration would greatly improve our understanding of complex giving a granoblastic texture (Teagle et al., 2006).
metal mobilisation in the oceanic crust. The plutonic complex is characterised by two gabbro intru-
Located in the Cocos Plate, Pacific Ocean, ODP ( Ocean sive bodies (Teagle et al., 2006; Koepke et al., 2008; Wilson
Drilling Program) Hole 1256D is an ideal location to inves- et al., 2006). The contacts between the gabbro bodies and
tigate the mobility of metals in a fast-spreading mid-oceanic the dykes are intrusive, with dyke fragments occurring in
ridge (MOR) crust. The 1256D drill core recovers a com- the margin of the lower gabbro body (Alt et al., 2010).
plete section of the oceanic crust down to the plutonic com-
plex (Wilson et al., 2003; Teagle et al., 2006; Wilson et al., 2.2. Hydrothermal system and metal mobility
2006). The hydrothermal system in the crust at Hole 1256D
and the resulting alteration are well defined (Teagle et al., The oceanic hydrothermal system at Hole 1256D can be
2006; Alt et al., 2010; Alt and Shanks, 2011; Violay et al., divided into two simplified domains: an upper low
2012), providing a solid framework for investigation of temperature domain and a deeper high temperature
mineral reactions leading to metal mobility. Recent investi- domain (Teagle et al., 2006; Alt et al., 2010). The low
gations on Hole 1256D have shown depletions in whole temperature domain occurs mostly in the volcanic section
rock of Cu, Zn and Pb and also of Au, As, and Se concen- (250–1004 mbsf) and is characterised by alteration formed
trations in the deeper parts of the core (Alt et al., 2010; from circulation of low-temperature (50–185 °C) oxidised
134 C.G.C. Patten et al. / Geochimica et Cosmochimica Acta 193 (2016) 132–159
Sulphide in
Depth Subdivision Mg# veins (vol %) Temperature °C
Sap
Chl
Hbl
Act
(mbsf) 20 50 70 0 0.5 200 400 600 800
Lava pond
400
Inflated
flows
Sheet and
massive
800 flows
Transitional
zone
1200 Sheeted
dyke
complex
Plutonic
complex
Sheet flow Massive flow Pillows Dykes Gabbro Granoblastic dykes
Fig. 1. Lithostratigraphy at IODP 1256D with Mg# and sulphide proportions in veins. Major secondary minerals and temperature are also
shown. Modified from Teagle et al. (2006) and Alt et al. (2010).
fluids (Fig. 1, Alt et al., 2010). The associated alteration to represent the pathways of rising high temperature fluids
style is a relatively low-intensity background alteration from the lower sheeted dyke section and the plutonic com-
(2–20% recrystallisation) where the primary mineralogy is plex (Alt et al., 2010). Below 1300 mbsf actinolite is more
partially replaced with saponite, celadonite, iron common than chlorite, indicating an increase in the temper-
oxyhydroxides and chalcedony (Teagle et al., 2006; Alt ature of alteration (Alt et al., 2010), which also corresponds
et al., 2010). Minor pyrite and trace chalcopyrite are asso- to an increase in the extent of metal mobilisation in the
ciated with this alteration and occur as disseminated grains crust (Cu, Zn, Pb, Au, As and Se; Alt et al., 2010; Patten
in the groundmass or in veins. Substantial As enrichment of et al., 2015). The transitional zone and the upper sheeted
the volcanic section is associated with this alteration (Patten dyke section correspond to the zone of interaction of the
et al., 2015). Pyrite fronts occur on the borders of oxidised two fluid domains (Fig. 1). The large difference in
alteration halos (Alt and Shanks, 2011). Significant As and temperature between the upper and lower fluid domains
Sb (20%) enrichments are associated with these alteration triggers sulphide mineralisation characterised by a 2.8 m
halos (Patten et al., 2015). mineralised breccia at 1028 mbsf enriched in Zn, Au, As,
The high temperature domain is characterised by Sb, Se and Te (Teagle et al., 2006; Alt et al., 2010; Patten
alteration formed from circulation of high-temperature et al., 2015). The sulphide assemblage associated with this
(300–>650 °C; Fig. 1) reduced fluids and occurs in the mineralisation is pyrite and sphalerite with trace chalcopy-
sheeted dyke and plutonic complexes (1061–1507 mbsf; rite. A number of similar, but smaller, mineralised veins
Alt et al., 2010). The pervasive background alteration is also occur in the upper sheeted dykes (Teagle et al., 2006;
characterised by chlorite, actinolite, albite, titanite and pyr- Alt et al., 2010).
ite which correspond to sub-greenschist and greenschist
facies conditions (Teagle et al., 2006; Alt et al., 2010). 3. SAMPLING AND ANALYTICAL METHODS
Mobilisation of Au, As and Se associated with the back-
ground alteration occurs in the upper part of the sheeted 3.1. EPMA analyses
dyke complex (Patten et al., 2015). Veins of quartz, chlorite,
epidote, pyrite, chalcopyrite, magnetite and sphalerite over- Rock chips from the Hole 1256D drill core were
print the background alteration and have been interpreted obtained from the collection at the National Oceanographic
Centre (NOC) at Southampton, U.K. Optical microscopy Sb, Se and Te (USGS, Wilson et al., 2002; Appendix
by transmitted and reflective light was carried out on 45 A.1). Additionally, MASS-1, NiS3 and PGE_8a were also
polished thin sections representative of the whole crust sec- used as monitors (Appendix A.2). Accuracy and precision
tion. Of the 45 thin polished sections selected, 14 had suit- of calibrants and monitors are shown in Appendix A.1
able sulphide and oxide minerals for EPMA and LA-ICP- and A.2. Isotopes monitored are falling within 10% of the
MS analyses. Electron microprobe analyses were carried published values except for 121Sb and 197Au, which are
out at Uppsala University, Sweden, with a JXA-8530F 32% and 25% higher respectively than MASS-1. This dis-
JEOL SUPERPROBE. For sulphide analyses an accelerat- crepancy could be due to oxide interferences due to the
ing voltage of 20 kV and a beam current of 20 nA were used water present in MASS-1 (Wohlgemuth-Ueberwasser
for a 5 lm beam, whereas for oxide analyses an accelerating et al., 2014). No appropriate reference materials for
voltage of 15 kV and a beam current of 10 nA were used. monitoring 65Cu, 66Zn, 107Ag, 125Te, 200Hg and 208Pb were
Acquisition time of 20 s and 10 s background was used available and the implications on the study are detailed in
for sulphide analyses and acquisition time of 10 s and 5 s Appendix B. Data reduction and calculation of trace metal
background was used for oxide analyses. A total of 314 concentrations were performed off-line through IOLITE
spot analyses were carried out for sulphide minerals on software (Hellstrom et al., 2008; Paton et al., 2011). Detec-
pyrite, chalcopyrite, pyrrhotite, sphalerite, pentlandite, mil- tion limits were calculated by IOLITE software at three
lerite and marcasite (Tables 1–5), whilst 180 spot analyses times the square root of the blanks. Trace metal signals in
were carried out on oxide minerals on magnetite, titano- sulphide minerals are inclusion free which suggests no nug-
magnetite and ilmenite (Table 6). One sigma detection lim- get effect.
its for sulphide analyses are 35 ppm for S, 43 ppm for Fe, Analyses by LA-ICP-MS of oxide minerals were carried
76 ppm for Cu, 77 ppm for Zn, 65 ppm for Ni and out with the same equipment as sulphide analyses. A spot
58 ppm for Co. For oxide minerals detection limits are size of 35 lm was used with a laser pulse frequency of
104 ppm for Si, 64 ppm for Al, 64 ppm for Mg, 329 ppm 10 Hz. Each analysis consisted of 50 s background, 40 s
for Mn, 161 ppm for Ti, 83 ppm for Ca and 119 ppm for ablation and 30 s wash-out. Data from EPMA analyses of
56
Fe. The Fe2O3 content of oxide minerals are calculated Fe were used for an internal standard. Coarse oxide
using the methods described in Carmichael (1966) and grains in the plutonic section show fine exsolution textures
Lepage (2003). and their ablation by LA-ICP-MS results in a mixed signal
of magnetite and ilmenite. The true Fe concentration is cal-
3.2. LA-ICP-MS analyses culated following the Dare et al. (2014) method and using
EPMA data for both ilmenite and magnetite. The analyses
Analyses by LA-ICP-MS were carried out at the are thus likely to represent the composition of the oxide
PetroTectonics facility at Stockholm University using a before sub-solidus exsolution (Dare et al., 2012). The fol-
NWR-213 nm solid state laser coupled to a ThermoFisher lowing isotopes were monitored: 23Na, 29Si, 31P, 33S,
44
XSeries II quadrupole mass spectrometer. Spot sizes of 30 Ca, 47Ti, 49Ti, 51V, 52Cr, 56Fe, 57Fe, 60Ni, 65Cu, 66Zn,
and 60 lm were used with a laser pulse frequency of 7 Hz 75
As, 82Se, 121Sb, 125Te, 195Pt, 197Au, 208Pb. Dwell times
and a laser energy density of 8.4 J cm2 for sulphide analy- were 2 ms/peak for 23Na, 29Si and 44Ca; 10 ms/peak for
51
ses. A total of 337 spot analyses were carried out on pyrite, V, 52Cr, 60Ni, 65Cu, 66Zn and 208Pb; 15 ms/peak for
75
chalcopyrite, pyrrhotite, sphalerite and marcasite (Tables As; 20 ms/peak for 82Se and 121Sb; 25 ms/peak for 47Ti,
1–5). Sulphide grains smaller than 30 lm including pent- 49
Ti and 125Te; 40 ms/peak for 33S; 45 ms/peak for 195Pt;
landite and millerite could not be efficiently analysed. Each and 60 ms/peak for 197Au. Calibration was performed using
analysis consisted of 60 s background, 40 s ablation and NIST 610, a fused glass doped with sixty-one trace elements
30 s wash-out. After wash-out, Se isotopes showed long- (Jochum et al., 2011, Appendix A.3). Accuracy and preci-
lived residual signals and additional wash-outs were carried sion of calibration was checked by monitoring the isotopes
out until the residual signals faded away. Data from EPMA using BCR-2G a natural basaltic glass (Jochum et al., 2005;
analyses of 33S were used as an internal standard. The fol- Jochum and Nohl, 2008; Jochum et al., 2014; Appendix
lowing isotopes were monitored: 33S, 57Fe, 60Ni, 65Cu, 66Zn, A.3). Matrix match standards were not available for oxide
75
As, 76Se, 77Se, 78Se, 82Se, 105Pd, 106Pd, 107Ag, 108Pd, 109Ag, analyses and NIST 610 and BCR-2G were used following
121
Sb, 123Sb, 125Te, 193Ir, 195Pt, 197Au, 200Hg, 202Hg, 208Pb the methods of Dare et al. (2012) and Nadoll et al.
and the preferred isotopes used for calculations are listed (2014). Good accuracy and precision are reached for 47Ti,
in Appendix A.1. Dwell times were 10 ms/peak for 33S, 49
Ti, 51V, 52Cr, 65Cu, 66Zn, 121Sb and 208Pb (Appendix
57
Fe, 200Hg and 202Hg; 20 ms/peak for 60Ni, 65Cu, 66Zn, A.3). Sample analyses for 82Se, 125Te, 195Pt and 197Au are
76
Se, 106Pd, 108Pd and 208Pb; 30 ms/peak for 82Se, 105Pd, below detection limits. Sulphur was monitored in order to
107
Ag, 109Ag, 125Te, 193Ir and 195Pt; and 40 ms/peak for identify possible sulphide inclusions, and Si, Ca and P were
121
Sb, 123Sb and 197Au. Calibration was carried out using monitored to control matrix silicate contamination or titan-
three reference materials: Trans-1, a synthetic Fe-Ni sul- ite alteration of the oxides. Arsenic analyses give signals
phide doped with 17–25 ppm As, Sb, Se and Te close to the background with high standard deviation,
(Wohlgemuth-Ueberwasser et al., 2014); NiS3, a fused and furthermore, count per second (CPS) values correlate
nickel sulphide doped with 20–24 ppm Au, Pt, Pd, Ir, Os, with Ca suggesting a possible interference of 43Ca16O2 on
75
Ru and Rh (Gilbert et al., 2013); and MASS-1, a pressed As. The 75As data are thus considered to be below the
pellet (Fe–Zn–Cu–S) doped with 50–70 ppm As, Ag, Pb, detection limit. The 65Cu values are likely to be a real Cu
136
C.G.C. Patten et al. / Geochimica et Cosmochimica Acta 193 (2016) 132–159
Table 1
Major and trace metals composition of magmatic sulphides.
S Fe Co Ni Cu Zn As Se Ag Pd Sb Te Ir Pt Au Hg Pb Total
wt.% wt.% wt.% wt.% wt.% ppm ppm ppm ppm ppm ppm ppm ppm ppm ppm ppm ppm wt.%
EPMA EPMA EPMA LA-ICP-MS LA-ICP-MS LA-ICP-MS LA-ICP-MS LA-ICP-MS LA-ICP-MS LA-ICP-MS LA-ICP-MS LA-ICP-MS LA-ICP-MS LA-ICP-MS LA-ICP-MS LA-ICP-MS LA-ICP-MS
Po-Cpy intergrowth Mean 39.24 58.0 0.07 0.21 0.58 1054 5.12 56.00 3.44 <D.L. 0.24 2.23 <D.L. <D.L. 0.066 <D.L. 409.5 98.2
EPMA n = 11 s.d. 0.7 1.9 0.02 0.02 0.47 987 1.34 15.49 0.93 0.12 0.26 0.049 680.6
LA-ICP-MS n = 8 Median 39.55 59.0 0.08 0.21 0.69 855 4.90 54.50 3.51 0.24 2.20 0.066 35.1
Range Min 37.6 55.3 0.03 0.18 7.9E-03 47 3.90 36.00 2.12 0.09 2.00 0.031 8.1
Max 39.9 59.6 0.10 0.24 1.28 2900 8.00 89.00 4.80 0.41 2.50 0.100 2100.0
Pyrrhotite Mean 39.57 58.4 0.14 2.18 0.02 9.9 0.95 45.52 0.21 <D.L. <D.L. 1.25 <D.L. <D.L. 0.023 <D.L. 0.49 100.3
EPMA n = 17 s.d. 0.3 0.8 0.27 1.49 0.04 18.1 0.11 8.38 0.11 0.34 0.019 0.22
LA-ICP-MS n = 18 Median 39.60 58.7 0.07 1.67 1.5E-03 3.8 0.95 44.60 0.17 1.20 0.019 0.57
Range min 38.5 56.0 0.03 1.15 1.9E-04 1.9 0.87 32.80 0.10 0.67 0.005 0.10
max 40.0 59.2 1.20 7.50 0.14 78.0 1.02 58.80 0.42 1.81 0.053 0.82
EPMA EPMA EPMA LA-ICP-MS EPMA LA-ICP-MS LA-ICP-MS LA-ICP-MS LA-ICP-MS LA-ICP-MS LA-ICP-MS LA-ICP-MS LA-ICP-MS LA-ICP-MS LA-ICP-MS LA-ICP-MS LA-ICP-MS
Chalcopyrite Mean 35.0 31.7 0.28 0.80 31.6 1169 <D.L. 49.64 5.98 <D.L. <D.L. 2.82 <D.L. <D.L. 0.027 <D.L. 3.29 99.5
EPMA n = 14 s.d. 0.2 2.0 0.04 0.99 3.8 1748 18.71 3.38 1.36 0.017 3.64
LA-ICP-MS n = 9 Median 35.1 30.9 0.30 0.68 33.4 160 46.80 6.76 2.80 0.025 1.30
Range Min 34.6 29.8 0.24 0.01 23.0 2.4 31.50 0.37 0.85 0.012 0.28
Max 35.3 36.8 0.30 3.03 33.7 5000 89.00 10.20 4.40 0.045 10.60
EPMA EPMA EPMA EPMA
Pentlandite Mean 42.0 19.8 16.5 21.5 <D.L. <D.L. N.D. N.D. N.D. N.D. N.D. N.D. N.D. N.D. N.D. N.D. N.D. 99.8
(n = 6) s.d. 0.4 6.4 3.9 4.0
Range Min 41.5 11.9 9.5 16.1
Max 42.3 27.6 21.0 28.5
The range of composition is reported in the text if the standard deviation is higher than the average; po = pyrrhotite; cpy = chalcopyrite;. s.d. = standard deviation; D.L. = detection limit;
N.D. = not determined.
signal rather than due to 49Ti16O+ interference, as the
Total
100.0
wt.%
99.6
99.9
oxides with high 65Cu CPS values show no correlation with
LA-ICP-MS
LA-ICP-MS
LA-ICP-MS
Ti oxide content. Interference of 50Ti16O+ on 66Zn is inter-
preted to be negligible due to the low natural abundance of
16.99
19.38
94.00
11.48
37.80
N.D.
ppm
8.93
1.60
8.27
9.85
1.10
50
Pb
LA-ICP-MS Ti (5.25%).
LA-ICP-MS
LA-ICP-MS
4. RESULTS
11.80
12.20
18.00
N.D.
ppm
3.60
2.05
3.10
1.40
6.70
5.84
6.40
Hg
4.1. Sulphide classification

LA-ICP-MS LA-ICP-MS LA-ICP-MS

0.002
N.D.
ppm
0.07
0.09
0.04
0.31
0.13
0.18
0.07
0.01
0.66
The sulphide population in Hole 1256D is classified
Au
using assemblage, texture, grain size and distribution. Three

main processes can account for the majority of sulphide
<D.L.
N.D.
mineral formation: (1) magmatic crystallisation, (2) high
ppm
1.65
1.62
1.16
0.28
4.00
Pt
temperature fluid circulation, and (3) low temperature fluid

circulation.
<D.L.
<D.L.
N.D.
ppm
Ir
4.1.1. Magmatic sulphides

LA-ICP-MS
LA-ICP-MS
LA-ICP-MS
Magmatic sulphides consist of pyrrhotite, chalcopyrite

and pentlandite that are characteristic of high temperature
51.00
14.00
N.D.
ppm
4.87
8.52
2.27
0.50
3.55
3.72
2.03
0.93
Te
magmatic sulphides (Fig. 3A and B; Czamanske and

LA-ICP-MS
LA-ICP-MS
LA-ICP-MS LA-ICP-MS
Moore, 1977; Mathez, 1979; Peach et al., 1990; Barnes

et al., 2006; Patten et al., 2012). They have been observed
11.80
N.D.
ppm
0.77
1.75
0.25
0.03
0.55
1.00
0.15
0.02
3.71
in two samples from the volcanic section and the plutonic

Sb
complex at 445 mbsf and 1493 mbsf respectively (Fig. 2)

LA-ICP-MS
LA-ICP-MS
and represent only a minor proportion of the Hole 1256D

<D.L.
25.00
N.D.
ppm
4.39
8.48
1.16
0.21
sulphide population. In the volcanic section, magmatic

Pd
sulphides occur as 10–50 micron-wide blebs composed of

LA-ICP-MS
LA-ICP-MS
LA-ICP-MS
pyrrhotite and chalcopyrite intergrowths representative of

38.00
29.50
N.D.
quenched monosulphide solid solution (MSS) and interme-

ppm
2.34
5.35
0.97
0.08
9.78
7.37
7.40
1.46
Ag
diate solid solution (ISS; Fig. 3A; e.g. MacLean, 1977;

LA-ICP-MS
LA-ICP-MS
LA-ICP-MS
Czamanske and Moore, 1977; Patten et al., 2012). Similar

sulphide droplets are observed in fresh MORB glass
N.D.
ppm
108
470
150
79
86
76
36
72
12
Se
(Czamanske and Moore, 1977; Mathez, 1979; Francis,

LA-ICP-MS
LA-ICP-MS
LA-ICP-MS
1990; Patten et al., 2012) and from other oceanic crust drill
core material (MacLean, 1977; Alt et al., 1989). In the plu-
48.40
14.40
N.D.
ppm
6.58
9.67
3.45
0.42
3.35
3.64
1.86
0.42
As
tonic complex magmatic sulphides do not have systematic

LA-ICP-MS
LA-ICP-MS
s.d. = standard deviation; D.L. = detection limit; N.D. = not determined.
shape but are larger than in the volcanic section (up to hun-
dreds of microns in width), and are composed of pyrrhotite,
EPMA
<D.L.
1870
ppm
79.1
299
138
500
1.2
7.3
Zn
chalcopyrite and pentlandite (Fig. 3B). The latter occurs as

76
17
27
Major and trace metals composition of metasomatised sulphides.
a lamellae exsolution within the pyrrhotite, which is

LA-ICP-MS
characteristic of slowly cooled magmatic sulphides that

3.5E-04
EPMA
EPMA
<D.L.
wt.%
0.16
0.25
0.04
0.91
32.1
33.2
14.2
34.3
have exsolved from MSS and ISS (Fig. 3B; Kelly and
Cu
3.6
Vaughan, 1983; Alt and Anderson, 1991; Barnes et al.,

LA-ICP-MS
LA-ICP-MS
2006; Dare et al., 2010; Djon and Barnes, 2012). Mineral

1.8E-03
3.5E-03
EPMA
wt.%
proportions by visual estimation are 90% pyrrhotite with

0.51
0.67
0.17
2.50
0.75
1.25
0.14
4.10
51.3
12.3
33.4
61.6
Ni
trace exsolution pentlandite and 10% chalcopyrite.

EPMA
EPMA
EPMA
0.106
0.084
0.063
0.019
0.280
wt.%
Quenched sulphide blebs from the volcanic section are a

0.60
1.32
0.08
0.01
6.00
0.02
14.8
Co
6.4
7.6
mixture of MSS and ISS and have different composition

EPMA
EPMA
EPMA
wt.%
from the magmatic sulphides in the plutonic complex

45.8
46.2
40.5
46.9
31.2
30.7
29.6
40.6
1.3
1.9
4.0
1.9
2.5
7.6
Fe
(Table 1). The sulphide blebs from the volcanic section have
EPMA
EPMA
EPMA
wt.%
the Fe and S composition of pyrrhotite but show high and

53.0
53.3
50.2
54.4
35.5
35.1
34.5
48.9
38.2
35.2
42.5
0.9
2.4
3.5
S
variable Cu, Zn, Pb and trace metals concentrations due to

Max
Max
Max
Range Min
Range Min
Range Min
contamination from small chalcopyrite inclusions (Table 1).

LA-ICP-MS n = 82 Median
LA-ICP-MS n = 20 Median
Mean
Mean
Mean
Large pyrrhotite grains from the plutonic complex contain

s.d.
s.d.
s.d.
trace amounts of Co, Ni, Cu, Zn, As, Se, Ag, Te, Au and
Pb. Concentrations of Sb, platinum group elements
EPMA n = 58
EPMA n = 35
(PGE; Pt, Pd, Ir, Os, Ru and Rh) and Hg are below
Chalcopyrite
Table 2
detection limits. Chalcopyrite from the plutonic complex

Millerite
(n = 6)
Pyrite
has higher concentrations of Zn (1169 ppm, 2–5000 ppm),

138
Table 3

Major and trace metals composition of high temperature sulphides.
wt.% wt.% wt.% ppm ppm ppm ppm ppm ppm ppm ppm ppm ppm ppm ppm ppm ppm wt.%
All pyrites Mean 53.8 46.6 0.11 124 44.1 46.0 23.94 66.76 2.45 0.17 0.61 1.63 <D.L. <D.L. 0.12 1.89 69.72 100.6
EPMA n = 62 s.d. 0.5 0.5 0.09 205 60.8 187.6 19.93 170.07 2.40 0.10 0.78 2.22 0.10 1.76 97.73
LA-ICP-MS n = 105 Median 53.9 46.8 0.09 30.0 25.9 5.9 23.70 31.50 2.39 0.13 0.33 1.09 0.12 1.24 14.60
Range Min 52.3 44.5 0.01 0.95 0.7 0.6 0.67 1.27 0.09 0.10 0.04 0.46 0.02 0.64 0.50
Max 54.6 47.8 0.32 1225 390.0 1600.0 93.00 1243.00 12.10 0.29 4.07 14.20 0.58 7.60 408.00
Transitional zone pyrite Mean 53.7 46.8 0.10 50 34.7 52.0 31.11 88.50 2.49 0.11 0.67 2.45 <D.L. <D.L. 0.12 1.71 104.63 100.7
EPMA n = 28 s.d. 0.5 0.3 0.11 111 28.7 222.9 18.68 231.45 2.18 0.02 0.80 4.14 0.10 1.71 107.52
LA-ICP-MS n = 36 Median 53.8 46.8 0.05 19.3 29.6 5.7 32.15 34.15 2.57 0.11 0.47 1.17 0.12 1.02 68.80
Range Min 52.5 46.3 0.01 0.95 0.7 0.6 0.67 1.27 0.09 0.10 0.04 0.51 0.02 0.64 0.50
Max 54.4 47.8 0.32 578 136.0 1600.0 93.00 1243.00 9.59 0.13 4.07 14.20 0.58 7.60 408.00
Upper sheeted dyke pyrite Mean 53.9 46.5 0.12 236 60.4 35.0 6.15 42.41 2.38 <D.L. 0.07 1.33 <D.L. <D.L. <D.L. 3.45 8.13 100.6
EPMA n = 34 s.d. 0.5 0.6 0.04 260 91.9 94.6 8.59 35.13 2.81 0.03 0.72 1.91 11.05
LA-ICP-MS n = 69 Median 53.9 46.7 0.11 174.5 21.3 6.5 3.44 30.80 1.58 0.06 1.09 3.45 4.22
Range Min 52.3 44.5 0.06 11.80 0.9 0.9 0.69 12.90 0.12 0.04 0.46 2.10 0.69
Max 54.6 47.1 0.18 1225 390.0 419.0 41.40 187.00 12.10 0.13 2.88 4.80 52.60
EPMA EPMA EPMA LA-ICP-MS EPMA (wt.%) LA-ICP-MS LA-ICP-MS LA-ICP-MS LA-ICP-MS LA-ICP-MS LA-ICP-MS LA-ICP-MS LA-ICP-MS LA-ICP-MS LA-ICP-MS LA-ICP-MS LA-ICP-MS
Chalcopyrite Mean 35.0 30.2 <D.L.237 33.5 1701 1.73 160.25 5.95 <D.L. 0.31 1.04 <D.L. <D.L. <D.L. <D.L. 11.49 98.9
EPMA n = 29 s.d. 0.4 0.8 203 0.5 3755 0.89 197.92 4.56 0.28 0.44 8.99
LA-ICP-MS n = 12 Median 35.1 30.4 197 33.7 714 1.90 67.90 6.20 0.17 1.10 8.97
Range Min 34.1 28.5 49.0 32.5 2.2 0.37 11.00 0.37 0.12 0.49 1.53
Max 35.5 31.3 591 34.0 12,900 2.50 566.00 14.50 0.80 1.50 28.70
EPMA EPMA EPMA LA-ICP-MS LA-ICP-MS EPMA (wt.%) LA-ICP-MS LA-ICP-MS LA-ICP-MS LA-ICP-MS LA-ICP-MS LA-ICP-MS LA-ICP-MS LA-ICP-MS LA-ICP-MS LA-ICP-MS LA-ICP-MS
Sphalerite Mean 33.7 8.5 0.03 39.1 6683 55.2 12.24 17.42 3.59 <D.L. 22.42 <D.L. <D.L. <D.L. 0.02 95.42 65.45 98.1
EPMA n = 19 s.d. 0.3 3.1 0.01 35.6 12,071 2.9 8.38 11.13 3.49 17.47 0.01 113.16 58.45
LA-ICP-MS n = 10 Median 33.6 9.8 0.03 30.3 1650 54.2 11.50 14.40 1.73 20.90 0.02 55.50 52.55
Range Min 33.2 2.1 0.02 6.0 126 51.9 0.57 4.30 1.46 0.06 0.01 10.50 8.49
Max 34.2 12.3 0.06 110.0 37,900 61.3 23.30 29.50 12.00 52.90 0.03 310.00 190.00

Table 4
Major and trace metlas composition of low temperature sulphides.
wt.% wt.% wt.% wt.% ppm ppm ppm ppm ppm ppm ppm ppm ppm ppm ppm ppm ppm wt.%

Pyrite/Marcasite Mean 52.7 46.3 0.06 0.20 9.6 3.3 102 37 0.20 <D.L. 0.38 <D.L. <D.L. <D.L. <D.L. <D.L. 1.37 108.9
EPMA n = 28 s.d. 0.6 0.5 0.08 0.30 11.2 2.8 67 14 0.10 0.24 1.26
LA-ICP-MS n = 28 Median 52.9 46.5 0.04 0.06 6.0 3.3 98 38 0.17 0.33 0.9
Range Min 51.1 45.1 0.01 0.01 1.0 0.7 9.1 18 0.10 0.09 0.12
Max 53.6 47.0 0.26 1.07 57.0 14 271 61 0.42 1.03 4
Table 5
Major and trace metals composition of patchy sulphides.
wt.% wt.% wt.% wt.% wt.% ppm ppm ppm ppm ppm ppm ppm ppm ppm ppm ppm ppm wt.%
Pyrite Mean 53.2 46.2 0.28 0.34 0.022 9.34 43.60 34.44 0.60 <D.L. 2.67 3.12 <D.L. <D.L. 0.028 <D.L. 9.58 100.0
EPMA n = 31 s.d. 0.9 1.0 0.34 0.61 0.088 18.37 35.32 26.87 0.72 1.62 1.68 0.027 14.80
LA-ICP-MS n = 39 Median 53.4 46.5 0.15 0.11 0.001 2.22 43.00 28.00 0.28 2.82 3.00 0.015 4.01
Range Min 50.4 42.2 0.01 4E-03 8E-05 0.65 0.65 3.10 0.09 0.09 1.50 0.003 0.30
Max 53.9 47.2 1.26 3.11 0.530 89.00 118.80 104.40 2.54 6.64 4.86 0.081 75.70
EPMA EPMA EPMA LA-ICP-MS EPMA LA-ICP-MS LA-ICP-MS LA-ICP-MS LA-ICP-MS LA-ICP-MS LA-ICP-MS LA-ICP-MS LA-ICP-MS LA-ICP-MS LA-ICP-MS LA-ICP-MS LA-ICP-MS
Chalcopyrite Mean 34.8 30.4 0.55 0.14 32.72 1170 <D.L. <D.L. 2.01 <D.L. <D.L. <D.L. <D.L. <D.L. <D.L. <D.L. 20.75 98.7
EPMA n = 4 s.d. 0.3 0.9 0.76 0.17 0.98 1556 0.01 16.19
LA-ICP-MS n = 2 Range Min 34.2 29.4 0.01 0.02 31.25 69 2.00 9.30
Max 34.8 31.6 1.08 0.26 33.28 2270 2.01 32.20
139
Table 6
Composition of oxide minerals at ODP Hole 1256D; r = standard deviation.
Magmatic Skeletal Magmatic Coarse Magmatic Coarse Hydrothermal
Titanomagnetite Ilmenite Magnetite Magnetite
EPMA (wt.%)
SiO2 Average 0.49 0.12 0.17 0.59
r 0.41 0.07 0.08 0.65
Al2O3 Average 1.62 0.07 1.13 0.63
r 0.47 0.04 0.47 0.36
MgO Average 0.24 0.22 0.07 0.15
r 0.10 0.16 0.04 0.23
MnO Average 0.68 0.92 0.11 0.11
r 0.43 0.31 0.06 0.09
TiO2 Average 17.83 48.75 3.44 1.73
r 6.49 1.45 1.41 1.45
CaO Average 0.17 0.06 0.05 0.25
r 0.17 0.06 0.04 0.28
FeO Average 46.5 42.6 34.3 32.3
r 5.70 1.26 0.83 1.18
Fe2O*3 Average 29.2 6.2 57.9 61.9
r 12.51 2.53 2.13 3.24
Total Average 97.4 99.4 98.9 98.1
r 1.87 0.67 0.44 0.88
LA-ICP-MS (ppm)
Magmatic Coarse Mixed signal
V Average 4204 3959 2666
r 1905 2925 610
Median 4505 2710 2580
Min 88 1142 2110
Max 8850 10,100 3700
Cr Average 16 174 107
r 27 238 119
Median 8 58 70
Min 2 6 16
Max 139 1008 314
Ni Average 101 229 676
r 73 482 499
Median 84 112 477
Min 24 51 224
Max 310 3390 1420
Cu Average 55 31 44
r 76 64 41
Median 21 9 33
Min 5 2 8
Max 316 277 102
Zn Average 1582 251 118
r 764 195 82
Median 1517 158 86
Min 37 72 68
Max 3120 732 262
Sb Average 0.26 0.17 >d.l.
r 0.23 0.01
Median 0.14 0.17
Min 0.12 0.16
Max 0.72 0.18
Pb Average 2.62 2.30 1.76
r 4.00 4.65 0.70
Median 1.31 0.43 2.00
Min 0.08 0.11 0.73
Max 19.60 23.50 2.30
r = standard deviation. Fe2O*3 is calculated is calculated using Carmichael (1966) and Lepage (2003) method.
Lo T. ed
h ch
ig a
et ic
Depth
H ./Le
M mat
Pyrite Chalcopyrite Pyrrhotite Sphalerite Marcasite Pentlandite Millerite
hy
Pa T.
a
(mbsf)
tc
w
ag
4321
M
200
Volcanic Section
1000
1060
T.Z.
Sheeted Dyke
Complex
1400
Plutonic
1500 Complex
Fig. 2. Sulphide distribution in ODP Hole1256D. The scale corresponds to sulphide proportions with 1 = trace, 2 = discrete, 3 = common
and 4 = major. Yellow boxes are samples with sulphides analysed by EPMA and LA-ICP-MS. Meta. stands for metasomatised sulphides,
High T. stands for high temperature sulphides and Low T. stands for low temperature sulphides. (For interpretation of the references to
colour in this figure legend, the reader is referred to the web version of this article.)
Ag (5.38 ± 3.38), Te (2.82 ± 1.36 ppm), and Pb (3.29 ppm, Millerite generally occurs at the contact between pyrite
0.28–10.6 ppm) than pyrrhotite and pentlandite, and and chalcopyrite and is the smallest phase of the assem-
contains trace amounts of Co, Ni, Se and Au (Table 1). blage with a size that does not exceed a few microns
Concentrations of As, Sb, PGEs and Hg are below detec- (Fig. 3F). Magnetite occurs as small micron-sized grains
tion limit in chalcopyrite. Selenium and Au concentrations associated with pyrite, often forming rims surrounding
in pyrrhotite and chalcopyrite are similar (45.2 ± 8.2 ppm the sulphide assemblage (Fig. 3C–E).
and 49.6 ± 18.7 ppm for Se respectively, and 0.023 Metasomatised pyrites contain traces of Co, Ni, Cu, Zn,
± 0.019 ppm and 0.027 ± 0.017 for Au respectively, As, Sb, Te, Hg, Au and Pb (Table 2). Few grains show sig-
Table 1). Pentlandite, unlike pyrrhotite and chalcopyrite, nals above the detection limit for Pd and Pt, and other
does not show stoichiometric composition due to high PGEs are below detection limits. Trace metal concentra-
concentrations of Co (16.5 ± 3.6 wt.%; Table 1) that can tions in metasomatised pyrites are highly variable and pos-
substitute for Fe (19.8 ± 6.4 wt.%) and Ni (21.5 ± 4.0 wt. itively skewed as highlighted by the discrepancy between
%; Riley, 1977). The pentlandite Cu and Zn concentrations the average and median values (Table 2). This discrepancy
are below the detection limit for EPMA and analyses is due to localised enrichment of trace metals during meta-
by LA-ICP-MS were impossible due to too small grain somatism and leaching (see discussion). Selenium and As
size. can substitute for S in pyrite (e.g. Genna and Gaboury,
2015) and the Se/As ratio is a useful parameter to determine
4.1.2. Metasomatised sulphides pyrite genesis in magmatic or hydrothermal systems (Duran
Metasomatised sulphides consist of pyrite, chalcopy- et al., 2015). Metasomatised pyrites are characterised by
rite, millerite and magnetite, an assemblage that is charac- high Se/As, suggesting a magmatic origin (Duran et al.,
teristic of recrystallised magmatic sulphides (Fig. 3C–F; 2015), and intermediate to high Ni content (Fig. 4).
Table 2; Alt et al., 1989). This group occurs mostly in Chalcopyrite contains traces of Co, Ni, As, Se, Ag, Sb,
the sheeted dyke and the plutonic complexes with a few Te, Hg, Au and Pb (Table 2) whereas PGEs are below
occurrences in the volcanic section (Fig. 2). Pyrite is the detection limit. Trace metal concentrations in
most abundant mineral of the group and occurs as grains metasomatised chalcopyrite are also variable but not as
up to hundreds of microns in size that are commonly pronounced as in the metasomatised pyrites (Table 2).
anhedral in shape and can be porous in places Millerite has low Ni (51.3 ± 12.3 wt.%) relative to
(Fig. 3D). Chalcopyrite is homogeneous and occurs either stoichiometric value, which can be accounted for by the
as grains associated with the pyrite or as isolated grains. It high concentration of Co and Fe (4 ± 1.9 wt.% and
ranges in size from a few microns up to tens of microns. 6.4 ± 7.6 wt.%, respectively).
Fig. 3. Reflected light photomicrographs of the magmatic and metasomatised sulphides. Magmatic sulphides: (A) sulphide blebs preserved in
the volcanic section composed of pyrrhotite (Po) and chalcopyrite (Cpy). The Po represents the MSS and the Cpy the ISS, see discussion for
details. (B) Base metal sulphides pyrrhotite-chalcopyrite-pentlandite (Pn) from the plutonic section. Pentlandite is present as lamellae
exsolution within the Po which is specific of MSS. Metasomatised sulphides: (C) spherical blebs composed of Py and magnetite (Mag). (D)
and (E) Assemblage from the plutonic section composed of Py-Cpy-Mag; note that the Mag preferentially form a rim. (F) Assemblage of Py–
Cpy and millerite (Mi) from the plutonic section.
4.1.3. High temperature hydrothermal sulphides (Fig. 5A) whereas in the transitional zone it is euhedral to
High temperature hydrothermal sulphides represent the subhedral (Fig. 5C). This distribution of sulphide phases
largest sulphide occurrence of the Hole 1256D. They occur is similar to that reported in the mineralised zone in Hole
mostly in the upper sheeted dyke section and the transi- 504B (Honnorez et al., 1985), although the sulphide miner-
tional zone (Fig. 2) within mineralised breccias, quartz- alisation in Hole 1256D is less well developed than in Hole
chlorite dominated veins or as disseminated grains 504B.
(Fig. 1). They are characterised by euhedral to subhedral High temperature pyrites contain trace amounts of Co,
grains of pyrite, chalcopyrite and sphalerite. Pyrite grains Ni, Cu, Zn, As, Sb, Te, Hg, Au and Pb (Table 3). Few
vary in size from small (tens of microns) grains that form grains show signals above the detection limit for Pd and
porous aggregates, to large (hundreds of microns) grains other PGEs are below detection limit. Relative to metaso-
(Fig. 5A–C). Chalcopyrite is common in the upper sheeted matised pyrites, high temperature pyrites have higher
dyke section but scarce in the transitional zone (Fig. 2). median values of As (23.7 ppm), Ag (2.39 ppm) and Au
Conversely, sphalerite occurs in trace amounts in veins (0.12 ppm) but lower Ni (30.0 ppm), Cu (25.9 ppm) and
from the upper sheeted dyke section but is the second most Pb (14.6 ppm); other trace metals are within the same order
common sulphide mineral in the transitional zone miner- of magnitude (Tables 2 and 3). The variability in trace
alised breccias (Fig. 2). In the upper sheeted dyke section metal concentrations in high temperature pyrites is most
sphalerite commonly shows the chalcopyrite disease texture likely due to different trapping efficiency during sulphide
10 3
High Temp (upper SD) Low Temp
terised by pyrite, marcasite and trace chalcopyrite
Metasomatised
High Temp (TZ) Patchy (Fig. 2). Micron sized pyrite and chalcopyrite grains are
commonly disseminated in the ground mass associated with
10 2 the background alteration, but the bulk of the low temper-
ature sulphide occurs as ‘‘pyrite fronts” composed of pyrite
and marcasite (Andrews, 1979). They occur as narrow
10 1
zones hundreds of microns wide and centimetres long in
which they replace the silicate matrix and fill pores and
Se/As
cracks (Fig. 5D–F). These pyrite fronts occur on the border

of alteration halos (Wilson et al., 2003; Alt and Shanks,
10 0
2011) such as oxidation zones (Andrews, 1979).
Distinction between pyrite and marcasite grains during
in situ analyses was not possible and the data in Table 4
10 -1 represents average content of both minerals. They contain
trace amounts of Co, Ni, Cu, Zn, As, Se, Ag, Sb and Pb
(Table 4). Tellurium, PGEs, Au and Hg are below detection
10 -2 limits. Low temperature pyrites show the highest As con-
10 -1 10 0 10 1 10 2 10 3 10 4 10 5 centrations of the different sulphide groups (99 ± 66 ppm)
Ni (ppm) but have low Cu (9.6 ppm, 1.01–57 ppm), Zn (3.3
Fig. 4. Variation of the Ni content versus the Se/As of the pyrites
± 2.8 ppm), Ag (0.20 ± 0.10 ppm) and Pb (1.37
from the different sulphide groups. Metasomatised pyrites are ± 1.26 ppm) relative to metasomatised and high tempera-
characterised by high Se/As and high Ni, high temperature pyrites ture pyrites (Table 4). Because of high As concentrations,
by variable Se/As and low Ni and low temperature and patchy low temperature pyrites show low As/Se and are also char-
pyrites by low Se/As and high Ni. See text for details. acterised by high Ni content (Fig. 4). Rare chalcopyrite
grains were too small for in situ analysis.
precipitation. Differences in composition between high tem- 4.1.5. Patchy sulphides

peratures pyrite from the transitional zone and the upper Patchy sulphides occur in the lower volcanic section,
sheeted dyke section can be observed (Table 3). The former transitional zone, sheeted dyke and plutonic complexes
shows higher As (31.1 ± 18.7 ppm), Sb (0.67 ppm, 0.04– (Fig. 2). They are characterised by a spherical web-like net-
4.07 ppm), Au (0.12 ± 0.10 ppm) and Pb (104.6 ppm, 0.5– work up to one centimetre wide that replaces the silicate
408 ppm) where the latter shows higher Ni (236 ppm, matrix (Fig. 5H and I). Silicate phenocrysts, oxides and
11.8–1225 ppm). Other trace metals are within the same other sulphides are also affected. The patches do not seem
order of magnitude in the two different groups. High tem- to be associated with any specific type of silicate alteration.
perature pyrites are overall characterised by low Ni concen- Pyrite is the most common mineral and is generally homo-
tration and large Se/As (Fig. 4). Chalcopyrites contain trace geneous and anhedral, but can also be porous. Minor chal-
amounts of Ni, Zn, As, Pd, Sb, Te, Au and Pb whereas Co, copyrite also occurs and one sample from the plutonic
PGE and Hg are below detection limit (Table 3). Zinc con- complex contains patches of marcasite.
centrations in sphalerite are low (55.2 ± 2.8 wt.%) due to Metal concentrations in the patchy sulphides (Table 5)
variable concentration of Fe (8.5 ± 3.1 wt.%). The Fe/Zn are similar to the low temperature sulphides and are low
ratio of inclusion-free sphalerite can be used to calculate relative to metasomatised and high temperature sulphides
the minimum fluid temperature from which sphalerite pre- (Tables 1–5). Pyrites contain trace amounts of Co, Ni,
cipitates (Keith et al., 2014). An average temperature of Cu, Zn, As, Se, Ag, Sb, Te, Au and Pb (Table 5). Platinum
378 ± 24 °C is calculated for sphalerite from the transi- group elements and Hg are below detection limits. Similarly
tional zone from EPMA data. Calculation for sphalerite to pyrite fronts, patchy pyrites are characterised by low Se/
from the upper sheeted dyke section cannot be carried As and high Ni content (Fig. 4) which suggest similar par-
out because of systematic chalcopyrite disease texture genesis. A few in situ analyses of chalcopyrite show trace
(Keith et al., 2014). This texture also explains the high Cu amounts of Co, Ni, Zn, Ag and Pb (Table 5).
variability in sphalerite (up to 3.8 wt.%; Table 3) especially
in the upper sheeted dyke section. Sphalerite contains trace 4.2. Oxide classification
amounts of Co, Ni, Cu, Zn, As, Se, Ag, Sb, Hg and Pb.
Within the high temperature hydrothermal sulphides, pyr- Oxide minerals are common in Hole 1256D and are clas-
ites have the highest Co, As, Te and Au concentrations, sified into three main groups based on their texture, size,
whereas Ni, Se and Ag concentrations are higher in chal- lithological occurrence and mineralogical assemblage:
copyrite, and Sb, Hg and Pb in sphalerite (Table 3; e.g. skeletal magmatic titanomagnetite, coarse magmatic
Monecke et al., 2016). magnetite-ilmenite and hydrothermal-related magnetite.
Skeletal titanomagnetites are the most common oxide min-
4.1.4. Low temperature sulphides eral and occur in the volcanic section and the sheeted dyke
Sulphides from the low temperature sulphide group are complex. They are present as micrometric disseminated
common throughout the volcanic section and are charac- skeletal grains or as dendritic chains (tens of microns,
Fig. 5. Reflected light photomicrographs of the high temperature, low temperature and patchy sulphides. High temperature sulphides: (A)
pyrite (Py) aggregate in quartz–chlorite vein from the upper sheeted dyke section. The core is formed of small Py grains whereas the rim is
formed by large euhedral grains. Presence of chalcopyrite (Cpy) and sphalerite (Sp) with the Cpy disease. (B) Large euhedral Py grains in the
transitional zone. (C) Py and Sp from the mineralised breccia in the transitional zone. Low temperature sulphides: (D) pyrite front in border of
an alteration halo. (E) details of the pyrite front with Py and marcasite (Mrc) filling a crack. (F) Euhedral Py filling a vesicle. (G) Py and trace
Cpy in saponite dominated vein. Patchy sulphides: (H) and (I) Pyrite replacing the silicate matrix and partly the silicate minerals resulting in a
patchy aspect.
Fig. 6A; Wilson et al., 2003). Their presence in the ground The third group corresponds to secondary magnetite
mass is ubiquitous (Wilson et al., 2003) and is generally grains that are associated with hydrothermal fluid circula-
confined to interstitial areas (Fig. 6A; Kempton et al., tion. They either occur as precipitated euhedral grains
1985). Disseminated micrometric ilmenite and magnetite within veins, as disseminated subhedral grains or as replace-
grains are also locally observed. Titanomagnetites ment of primary silicate and sulphide phases
have TiO2 > 2 wt.%, average 17.8 ± 6.5 wt.% TiO2 and (Fig. 6C and D). They are present mostly in the sheeted
have the highest V, Cu, Zn, Sb and Pb concentrations of dyke complex. Hydrothermal magnetites have low TiO2
the three oxide groups (Table 6). concentrations (1.7 ± 1.5 wt.%) and are characterised by
The second group of oxide minerals in Hole 1256D are the highest Ni content and the lowest V, Zn and Pb content
magmatic related coarse grained magnetite and ilmenite (Table 6). The magnetite rims associated with the metaso-
(few mm) occurring in the plutonic complex (e.g. oxide gab- matised sulphide are part of the hydrothermal oxide group
bro; Teagle et al., 2006). They are characterised by coarse but, given their small size, only two grains were large
magnetite with trellis patterns of ilmenite exsolution or by enough to be analysed by LA-ICP-MS, with one of the
ilmenite with magnetite exsolution (Fig. 6B; e.g. Kent two showing considerable sulphide contamination.
et al., 1978). Homogeneous grains of magnetite or ilmenite
are also present. Ilmenites average 49.7 ± 6.9 wt.% TiO2 5. DISCUSSION
and magnetites average of 3.4 ± 1.4 wt.% TiO2 (Table 6).
Trace element concentrations analysed by LA-ICP-MS The sulphide classification described above indicates
are a mixed signal of both magnetite and ilmenite and are that sulphides are sensitive to the different styles of alter-
listed in Table 6. Coarse magnetite–ilmenites have the high- ation in the oceanic crust. In situ analyses reveal significant
est Cr and the lowest Cu content of the three oxide group variations in trace metal content between the different
(Table 6). sulphide mineral groups. The comparison between trace
Fig. 6. Reflected light photomicrographs of the oxide population in Hole 1256D. (A) Skeletal titanomagnetite (T-Mag) in the volcanic section
(group 1). M.S. = metasomatised sulphides (B) Coarse grain magmatic oxide composed of magnetite (Mag) with ilmenite (Ilm) exsolution
(group 2); note the trellis pattern and grain texture. Po = pyrrhotite; Cpy = chalcopyrite. (C) Hydrothermal related oxide corresponding to
magnetite replacing a silicate phase (group 3). (D) Hydrothermal related oxide corresponding to magnetite replacing a sulphide phase (group
3); see discussion. Py = pyrite.
element concentrations in the different sulphide groups with temperature sulphide assemblages. In the plutonic complex,
whole rock metal concentrations allows some assessment of however, where sulphide liquid is preserved from quenching,
the degree to which the sulphide minerals control the trace the observation of pyrrhotite with pentlandite exsolution as
metal budget in these rocks. This mass balance indicates lamellae implies that the sulphides cooled below 650 °C,
which mineral reactions are responsible for metal mobilisa- enabling the MSS and the ISS to exsolve to the observed
tion during the hydrothermal alteration of the oceanic phases (Fig. 3A and 7; Kelly and Vaughan, 1983; Patten
crust. et al., 2012).
The composition of magmatic sulphides from Hole
5.1. The magmatic assemblage 1256D (Table 1) differs slightly from previous reports of
fresh MORB sulphide droplets through having lower chal-
Some primary sulphides and oxides have been sufficiently cophile element concentrations (Co, Ni, Cu, Se, Ag, Te, Au;
shielded from later hydrothermal fluid circulation that their e.g. Patten et al., 2013). Differences in concentrations are
magmatic characteristics have been preserved. Magmatic attributed to different primary magmatic composition
sulphides are, however, scarce in the Hole 1256D due to rather than effect of alteration. Arsenic can be used as an
the extensive hydrothermal alteration. Magmatic S concen- indicator for fluid interaction with sulphides as it is a highly
trations in Hole 1256D average 1250 ± 200 ppm (Alt and mobile element in fluids (e.g. Smedley and Kinniburgh,
Shanks, 2011), which is high enough for a MOR basaltic 2002). Pyrrhotite grains from the plutonic section have
melt to be sulphide-saturated before eruption (e.g. low As concentration (0.95 ± 0.11 ppm, Table 1) suggesting
Mathez, 1979; Li and Ripley, 2005). Magmatic sulphides limited interaction with circulating fluids whereas sulphide
observed in Hole 1256D were most likely present as sulphide blebs in the volcanic section have relatively high As concen-
liquid within the magma prior to crystallisation, similar to tration (5.12 ± 1.34 ppm) suggesting possible contamina-
the sulphide droplets observed in fresh MORB glass tion by low temperature fluids. The magmatic sulphides
(Mathez, 1979; Peach et al., 1990; Patten et al., 2013). These from the plutonic complex can be interpreted to have con-
sulphide droplets are quenched from 1200 °C to seawater served their true primary magmatic composition most likely
temperature in a few minutes (Moore, 1975), resulting in due to limited interaction with hydrothermal fluids as impli-
their crystallisation as MSS and ISS intergrowths character- cated by the low water–rock ratio (<1) below 1300 mbsf
istic of magmatic temperatures (1200–1000 °C, (Gao et al., 2012).
Fig. 3A and 7; e.g. Czamanske and Moore, 1977; Mathez, In the volcanic section, disseminated titanomagnetites
1979; Patten et al., 2012). Magmatic sulphide droplets in have skeletal texture characteristic of a magmatic origin
the volcanic section are considered to represent these high- and occur in interstitial areas suggesting late-stage crystalli-
1 4b 4a
Py+Mc
Po Py+Cpy
Cpy
10 um 1 mm 1 mm
Hydrothermal
vent
Hypothetic Hole
1256D location
Depth (mbsf)
0
Sediments
200
3b Py+Sp
Low temperature
Volcanic fluid circulation
Section
2 mm
Sulfide
mineralisation 3a Py+Cpy
+Sp
1000 Moderate
T.Z.
1100 temperature fluid
recharge
Sheeted Dyke
Complex 1 mm
2b
1400
High temperature rising
hydrothermal fluid Cpy
Plutonic
Complex
Py
Mt
100 um
5 1 2a
Py Pn Mi
Cpy Cpy
Py
Po
Mt
1 mm 100 um 100 um
Fig. 7. Schematic diagram illustrating the paragenesis of the sulphide population at ODP Hole 1256D. (1) Magmatic sulphides forming
spherical sulphide droplets in the volcanic section and large sulphide aggregates in the plutonic complex. (2) Metasomatised sulphides showing
different degrees of sulphide leaching from (a) pristine to b) extensively leached. (3) High temperature hydrothermal sulphides precipitated
from rising hydrothermal fluids in (a) the upper sheeted dykes and (b) the mineralised breccia at 1028 mbsf. (4) Low temperature sulphides
occurring (a) in background altered samples due to low temperature fluid circulation and (b) on borders of alteration halos forming pyrite
fronts. (5) Patchy sulphides formed from deep low temperature fluid circulation in the lower part of Hole 1256D. Po = pyrrhotite,
Pn = pentlandite, Cpy = chalcopyrite, Py = pyrite, Mag = magnetite, Mi = millerite.
sation (Kempton et al., 1985). Trellis pattern of ilmenite– (Table 6). Samples hosting skeletal titanomagnetites from
magnetite exsolutions in coarse grain oxides in the plutonic the volcanic section are slightly more evolved than those
complex represent sub-solidus oxy-exsolution during hosting coarse magnetite–ilmenites from the plutonic sec-
magma cooling (Dare et al., 2012), also indicating its mag- tion (Fig. 1). The primary metal content of the magma,
matic origin. The trace element composition of the skeletal however, is unlikely to account for the different metal con-
titanomagnetite composition indicates temperatures tents of the two oxide groups as Cu, Zn, Pb and Sb show
around 500 °C (Fig. 8D). Magmatic oxide minerals host variable behaviour during magmatic differentiation with
minor Cu and Zn, traces of Pb and Sb and no As, Se, Te Cu for example, being enriched in more primitive melts
or Au. This elemental distribution can be partly explained and Pb, Zn and Sb being enriched in more evolved melts
by the Goldschmidt rule for cation substitution within the (Jenner and O’Neill, 2012). A more likely alternative is that
oxide structure, with Cu2+ and Zn2+ being substituted with the metal content of the magmatic oxides is related to the
Fe2+ in the octahedral sites, whereas other metallic cations cooling rate of the magma and the presence of sulphide
cannot be substituted with Fe cations (e.g. Dare et al., 2012; with which competition for metal trapping occurs (Dare
Nadoll et al., 2014). Skeletal titanomagnetites are richer in et al., 2014). In the volcanic rocks hosting skeletal titano-
Cu, Zn, Sb and Pb than the coarse magnetite–ilmenites magnetite, the quick cooling rate of the magma at eruption
A Coarse Mag-Ilm 0 1 B 0
Titanomagnetite
Hydrothermal magnetite
0.2 0.8
400
Depth (mbsf)
0.4 0.6
TiO
2
O
Fe
Ilmenite 800
2
0.6 0.4
1200
0.8 Ti 0.2
tan
om
ag
ne
tite Magnetite
1 0
1600
-4 -3 -2 -1 0 1 2
0 0.2 0.4 0.6 0.8 1 10 10 10 10 10 10 10
Fe2O3 Si+Ca/Ti
C 106 D 10
1
Magmatic
field
ºC
105
00
~5
100
Al+Mn (wt. % )
Ti (ppm)
Hydrothermal
104
field
ºC
00
10-1
~3
103
Decreasing
temperature
102 10-2
10-2 10-1 100 101 102 103 10-2 10-1 100 101 102
Ni/Cr Ti+V (wt. % )
E 104 F 104
103
103
Zn (ppm)
Cu (ppm)
102
102
1
10
100 101
101 102 103 104 105 100 101 102 103 104 105
Si+Ca/Ti Si+Ca/Ti
G H
102 103
101 102
Sb (ppm)
Pb (ppm)
100 101
10-1 100
10-2 10-1
101 102 103 104 105 100 101 102 103 104 105
Si+Ca/Ti Si+Ca/Ti
Fig. 8. Oxide minerals trace elements. (A) Ternary FeO2, Fe2O3 and TiO2 diagram. Mag = magnetite, Ilm = ilmenite. (B) Content of Si and
Ca in oxide minerals versus depth. Silica and Ca contents are used as a proxy for the degree of alteration. (C) Discrimination diagram for
magmatic or hydrothermal origin of magnetite and titanomagnetite with TiO2 < 20 wt.%; modified from Dare et al. (2014). Note that skeletal
titanomagnetite and coarse magnetite plot in the magmatic field whereas hydrothermal magnetite plots in the hydrothermal field. (D)
Discrimination diagram showing and titanomagnetite with TiO2 < 20 wt.% estimated formation temperature; modified from Nadoll et al.
(2014). (E)–(H) Trace metal content in oxide minerals versus the degree of alteration. See text for details.
may have inhibited effective diffusion of metals to the sul- 5.2.2. Sulphide leaching
phide blebs (low R-factor; Barnes and Lightfoot, 2005) During on-axis hydrothermal fluid circulation, equili-
allowing metals to be partly incorporated into oxides, with bration of the ‘‘recharging” moderate temperature
the converse occurring in the plutonic rocks that hydrothermal fluids with the surrounding rocks in the
host coarse magnetite-ilmenite and coarse magmatic sheeted dyke and plutonic complexes increases their tem-
sulphides. perature (>350 °C) and acidity, and causes reduction of
the fluids (e.g. Alt et al., 1986, 2010). These buoyant, acidic
5.2. Effects of hydrothermal fluid circulation in the lower and reduced high-temperature hydrothermal fluids eventu-
oceanic crust ally rise within the sheeted dyke and plutonic complexes
towards the seafloor following the similar pathways to
5.2.1. Sulphide metasomatism those taken by the recharging fluids (Alt et al., 2010;
Hydrothermal fluid circulation in the lower portion of Harris et al., 2015). Under these conditions, metal solubility
the oceanic crust leads to sub-solidus recrystallisation of in hydrothermal fluids significantly increases (Seewald and
the magmatic sulphides into metasomatised sulphides (Alt Seyfried, 1990) and sulphide minerals undergo leaching
et al., 1989). The pyrrhotite, chalcopyrite and pentlandite and dissolution (Fig. 7). These conditions correlate with
are recrystallised to secondary pyrite, chalcopyrite, millerite the onset of depletion of metals from the Hole 1256D whole
and magnetite. Two processes can trigger the recrystallisa- rock samples (Alt et al., 2010; Patten et al., 2015). The local
tion of pyrrhotite to secondary metasomatised pyrite: preservation of magmatic and metasomatised sulphides in
reduction of seawater sulphate at high temperature Hole 1256D indicates, however, that sulphide leaching is
(>250 °C) or oxidation of the magmatic sulphides. Seawa- not complete (Figs. 2 and 9). Leaching of sulphides is likely
to be dependent upon the degree of interaction with the
ter sulphate (SO24 Þ reduction in the oceanic crust provides high temperature fluids on a grain-scale as highlighted by
S as sulphide to the rock, leading to pyrrhotite (FeS) recrys-
the preservation of variably leached sulphide grains within
tallisation to secondary pyrite (FeS2 ; Shanks and Seyfried,
the same sample. Degrees of leaching range from full
1987):
preservation of pristine magmatic and metasomatised sul-
phides (Fig. 3E), to extensively leached sulphides (Fig. 9).
7FeS þ 8Hþ þ SO2
4 ¼ 4FeS2 þ 4H2 O þ 3Fe
þ
ð1Þ
In the extensively leached sulphides, chalcopyrite grains
The isotopic sulphur signature of metasomatised pyrite are more abundant than pyrite (Fig. 9A, C and D) which
suggests that this reaction occurs in the sheeted dyke com- suggests that chalcopyrite is less reactive to high tempera-
plex at Hole 504B (Alt et al., 1989). Alternatively, oxidation ture fluid circulation than pyrite (Fig. 10). This is supported
of pyrrhotite and pentlandite ðFe4 Ni5 S8 Þ can lead to the by experimental studies that show Cu is less soluble than
formation of pyrite, magnetite ðFe3 O4 Þ and millerite (NiS) other base metals in hydrothermal fluids at temperature
at moderate temperature (up to 350 °C, Craig and >300 °C in reduced conditions (e.g. Hemley et al., 1992).
Kullerud, 1969; Craig, 1973; Barnes et al., 2009; Djon Pyrite occurs either as porous patches or isolated grains
and Barnes, 2012): within the magnetite rim, and these textures indicate a sig-
nificant sulphide volume loss (Fig. 9B–D). This volume loss
FeS þ Fe4 Ni5 S8 þ 2O2 ¼ 2FeS2 þ Fe3 O4 þ 5NiS ð2Þ
is due to progressive release of S and Fe to the high temper-
The textures of the metasomatised sulphides in Hole ature fluids (Crerar et al., 1978):
1256D support the latter process as they are characterised
by metasomatised pyrite surrounded by a magnetite rim 1
FeS2s þ 2Hþ
aq þ H2 Oaq ¼ Feaq þ 2H2 Saq þ O2aq
2þ
ð3Þ
which represents an oxidation front where fluids have 2
reacted with the pyrrhotite (Fig. 3C–E). Formation of Behaviour of trace metals during sulphide leaching is
metasomatised pyrite by sulphate reduction is also likely dependent on their solubilities in the high temperature flu-
to occur in Hole 1256D but the almost ubiquitous presence ids, which are controlled by temperature, pressure, pH, fO2,
of magnetite associated with metasomatised sulphides sug- fS2 and fluid composition (e.g. Cl; Crerar et al., 1978;
gests that the oxidation is the dominant reaction. Trace ele- Hannington and Scott, 1989; Seewald and Seyfried, 1990;
ment concentrations of the magnetite rims suggest Hemley et al., 1992; Gibert et al., 1998).
oxidation reaction at temperatures 300 °C (Fig. 8B;
Nadoll et al., 2014). These temperatures and the mineralog- 5.2.3. Trace metal behaviour during metasomatism and
ical assemblage of metasomatised sulphides suggest that leaching
metasomatism is most likely driven by deep influx of Major and trace metal concentrations vary considerably
recharging hydrothermal fluids at moderate temperature during metasomatism and leaching of the magmatic sul-
(200–350 °C; Alt et al., 1986; Alt and Chaussidon, 1989; phides (Tables 1 and 2). Using median values, metasoma-
Coumou et al., 2008; Harris et al., 2015) into the lower part tised pyrite shows higher concentrations of Co (+7%), Cu
of Hole 1256D during on-axis hydrothermal alteration (+2528%), Zn (+334%), As (+265%), Se (+91%), Ag
(Tolstoy et al., 2008; Coumou et al., 2008). These fluids (+469%), Te (+89%), Au (+115%) and Pb (+1466%) rela-
penetrate into the sheeted dyke complex along dyke mar- tive to pyrrhotite (Tables 1 and 2). Only Ni shows lower
gins (Harris et al., 2015) most probably during the early concentrations (21%; Tables 1 and 2). Although the mod-
stages of hydrothermal alteration (Alt et al., 2010; Harris erate temperature hydrothermal fluids that drive sulphide
et al., 2015). metasomatism probably contain some trace metals, is it
Fig. 9. Reflected light photomicrographs of metasomatised sulphides showing different degree of sulphide leaching. (A) Shadow of magmatic
sulphide blebs in the volcanic section with a magnetite (Mag) rim and residual chalcopyrite (Cpy) inside. (B) Porous pyrite (Py) surrounded by
Mag and trace Cpy in the surroundings. (C) and (D) Trace leached sulphides with the Mag remaining. (B)–(D) are from the plutonic complex.
highly unlikely that they systematically enrich the metaso- (21%) in metasomatised pyrites cannot be explained solely
matised sulphides in trace metals to such extents. Increases by sulphide volume loss and we suggest that diffusion of
in trace metal concentrations in metasomatised sulphides these elements between the different sulphide phases present
are most likely due to residual enrichment during successive in the assemblage occurs. Nickel is likely to diffuse into the
sulphide volume loss and local metal diffusion between min- millerite from both pyrrhotite and pentlandite (Tables 1
eral grains. During metasomatism, diffusion of Fe from and 2; Fig. 10) whereas Cu is likely to diffuse from the chal-
pyrrhotite forms the magnetite rim (Eq. (2)). This process copyrite to the pyrite during metasomatism, and Pb from
leads to sulphide volume loss and recrystallisation of pyr- surrounding silicates (Tables 1 and 2; Fig. 10). Changes in
rhotite to pyrite with a corresponding S increase of the trace metal distribution between magmatic and metaso-
+34.3% and Fe decrease of 21.4% of the sulphide assem- matised sulphides suggests that localised grain-scale metal
blage (Tables 1 and 2, Fig. 10). The residual metal enrich- mobility occurs during metasomatisation (Fig. 10), but bulk
ment caused by the conversion of pyrrhotite to pyrite, mobilisation of metals from the rock does not occur until
however, only partially explains the large enrichments of the onset of sulphide leaching. Ultimately, if sulphide leach-
trace metals observed in metasomatised pyrite. Additional ing goes to completion then all trace elements hosted in
concentration of trace metals occurs during leaching where metasomatised sulphides will be mobilised into the high
preferential mobilisation of Fe and S into the hydrothermal temperature fluids (Fig. 10).
fluids leads to further sulphide loss (Eq. (3)). The behaviour
of Se clearly highlights this two-step process as
demonstrated by the variation in the S/Se ratio during 5.2.4. Oxide alteration by hydrothermal fluid circulation
metasomatism and leaching (Fig. 11A). During metasoma- Although oxide minerals show some evidence of alter-
tism the S/Se is constant as S and Se are equally concen- ation they are less affected by high temperature fluid circu-
trated during Fe diffusion to magnetite, but as leaching lation than sulphide minerals. Laverne et al. (2006) report
occurs the S/Se decreases as S is preferentially mobilised breakdown of titanomagnetite to hydroschlormite during
by the hydrothermal fluids along with Fe (Fig. 11A). The low temperature alteration. Additionally Alt et al. (2010)
high concentrations of Co, Zn, Ag, Se, Te and Au in pyrite report increasing titanomagnetite alteration below
are most likely explained by the combined effects of 660 mbsf. The Si + Ca/Ti ratio increases with depth indi-
metasomatism and leaching (Fig. 11). Although Sb is below cating greater alteration (Fig. 8B). No systematic variation
detection limit in pyrrhotite the range of concentrations in of metal concentrations with the Si + Ca/Ti ratio can be
metasomatised pyrites suggest similar behaviour (Table 2). observed in oxides indicating that, the contrary to sul-
The extreme concentration increases of Cu (+2528%) and phides, alteration of these minerals in Hole 1256D does
Pb (+1466%), and the decrease in Ni concentrations not release metals (Fig. 8E–H).
Cpy
ISS (Cpy)
Cu M* Cu M*
Pb? Cu
Ni Mi
Ni Co
MSS Py
(Po+Pn)
+S +Zn +As +Se
+Ag +Sb +Te +Au Fe
Fe Zn As Se Ag Te Au
Pb
Mag
(A) Magmatic assemblage (B) Metasomatism

High temperature 1200-650ºC Oxidation by moderate temperature hydrothermal
fluids (250-350ºC)
Cpy Cpy
Cu M* Cu M*
Mi
Py Py
++Zn ++As ++Se
++Ag ++Sb ++Te ++Au
-S -Fe -Zn
Mag -S -Fe Mag -As -Se -Ag -Sb
-Te -Au
(C) Partial sulphide leaching (D) Extensive sulphide leaching

Leaching by high temperature hydrothermal Leaching by high temperature hydrothermal fluids
fluids (>350ºC) (>350ºC)
Fig. 10. Sulphide evolution and metal content variation during metasomatism and leaching. (A) Magmatic assemblage characterised by
monosulphide solution (MSS) and intermediate solid solution (ISS). During subsolidus reaction MSS exsolve to pyrrhotite (Po) and
pentlandite (Pn), and ISS exsolve to chalcopyrite (Cpy). M* = trace metals hosted in Cpy, their behaviour during sulphide evolution is not
determined in this study. (B) Metal diffusion during metasomatism. Py = pyrite, Mag = magnetite, Mi = millerite. (C) and (D) Metal
mobilisation during partial and extensive sulphide leaching respectively. See text for details.
5.3. Mineral precipitation during high temperature fluid Sb, Au and Pb-bearing pyrite (Hannington and Scott,
circulation 1989; Metz and Trefry, 2000; Genna and Gaboury, 2015;
Keith et al., 2016; Monecke et al., 2016) and a similar
In the upper sheeted dyke and transitional zone, on-axis zonation is observed in Hole 1256D between the upper
mixing of ascending high temperature fluids with the low sheeted dyke section and the transitional zone (Table 3).
temperature fluids prevailing in the volcanic section triggers The Se/As ratio tends to be lower in pyrites from the tran-
significant sulphide precipitation forming a 2.8 m miner- sitional zone relative to those in the upper sheeted dyke sec-
alised breccia at 1028 mbsf (Teagle et al., 2006; Alt et al., tion (Fig. 4) which can be attributed to an increased input
2010; Fig. 7). Fluid inclusion microthermometry suggests of low temperature fluids leading to temperature decrease
that fluid mixing and sulphide precipitation occurred at (Genna and Gaboury, 2015), with possible As contamina-
temperatures between 320 °C and 450 °C (Alt et al., 2010) tion from these fluids. The low Ni content and variable
and at high fluid–rock ratios (Harris et al., 2015). Such tem- Se/As ratios of hydrothermal pyrites are distinctly different
peratures are consistent with the minimum hydrothermal from metasomatised pyrites highlighting the different origin
fluid temperature of 378 ± 24 °C estimated from Fe/Zn in between the two sulphide groups. Although high tempera-
sphalerite (Keith et al., 2014). The mineralogical assem- ture hydrothermal sulphides have significant trace metal
blages and trace metal concentrations in pyrite vary concentrations (Table 3) their impact on the bulk fluxes
between the transitional zone and the upper sheeted dykes of metals from the lower sheeted dykes is minimal
suggesting different mineralisation in these two areas. Tem- (Nesbitt et al., 1987; Patten et al., 2015). Hydrothermal
perature controlled metal zonation is common in VMS magnetites are also interpreted to be generated from the
deposits: precipitation of chalcopyrite and Ni-bearing pyr- circulation of high temperature fluids. Their composition,
ite occur at higher temperatures than sphalerite and As, characterised by low Ti content (Fig. 8C), and texture
105 102 102

Pyrrhotite
Median pyrrhotite
Metasomatised and leached pyrite
Calculated metasomatism increase
Median metasomatised pyrite
101 101
2.Leaching
Te (ppm)
As (ppm)
1.Metasomatism
S/Se
4 1.Metasomatism
10
2.Leaching
2.Leaching 1.Metasomatism
100 100
103 10-1 10-1

101 102 103 101 102 103 101 102 103
Se (ppm) Se (ppm) Se (ppm)
2
102 100 10
101
10-1
Zn (ppm)
Ag (ppm)
Au (ppm)
2.Leaching
101 2.Leaching
100 2.Leaching 1.Metasomatism
1.Metasomatism
10-2
10-1
1.Metasomatism
10-2 10-3 100

101 102 103 101 102 103 101 102 103
Se (ppm) Se (ppm) Se (ppm)
Fig. 11. Trace metal variations in pyrrhotite and metasomatised pyrite during metasomatism and leaching. Selenium is used as proxy to
highlight trace metal behaviours both during metasomatism and leaching. Increase concentrations of Se, Te, As, Ag, Au and Zn in
metasomatised pyrites are due to residual enrichment by progressive sulphide volume loss first during metasomatism and second by sulphide
leaching. See text for details.
confirm their hydrothermal origin and suggests that they halos and the pyrite fronts (Patten et al., 2015) causing
formed at similar temperatures (300–500 °C) that the high a low Se/As ratio (Fig. 4). In sulphide-free samples,
temperature sulphides (Fig. 8D). which also show As enrichment (Patten et al., 2015),
As is likely to be hosted in Fe-oxyhydroxide minerals
5.4. Sulphide formation during low temperature fluid (Wolthers et al., 1998; Smedley and Kinniburgh, 2002;
circulation Han et al., 2012). High Ni concentrations in pyrite fronts
(Table 4; Fig. 4) can be explained by Ni incorporation in
In the volcanic section pyrite fronts locally occur on pyrite at low temperature and under anoxic conditions
the margins of significantly oxidised zones referred to (Morse and Luther, 1999). Along with As, significant
as alteration halos (e.g. Andrews, 1979; Alt and whole-rock enrichment of Sb occurs in the volcanic
Shanks, 2011; Fig. 7). Andrews (1979) proposes that section (Patten et al., 2015), but this enrichment is asso-
the pyrite fronts form from leaching of magmatic ciated with secondary silicates rather than the low tem-
sulphides from the alteration halos by low temperature perature sulphides (Fig. 12; Jochum and Verma, 1996).
fluids. The presence of marcasite indicates acidic solution In the sheeted dyke and plutonic complexes the patchy
(<5 pH) and partly oxidised soluble species such as sulphides share numerous characteristics with low tempera-
thiosulphate (Murowchick and Barnes, 1986). Alt and ture sulphides although they are not associated with any
Shanks (2011), however, argue that there is insufficient specific alteration. They have similar mineralogical assem-
sulphide in the alteration halos to account for that in blages of pyrite, minor marcasite and trace chalcopyrite
the pyrite fronts, and additional S is required from (Fig. 2), and replacement textures (Fig. 5). Pyrites from
microbial seawater sulphate reduction. The formation the patchy sulphide group are characterised by low Se/As
of alteration halos and pyrite fronts occurs at the latest ratio and high Ni and falls within the field of the low tem-
stages of fluid circulation during off-axis low temperature perature pyrite (Fig. 4). These similarities between the two
alteration (Alt et al., 2010; Harris et al., 2015) at higher sulphide groups suggest similar paragenesis, implying that
water–rock ratio than the background alteration (Teagle the patchy sulphide group formed during late-stage off axis
et al., 2006; Gao et al., 2012). Significant As enrichment low temperature fluid circulation in the sheeted dyke and
from low temperature fluids occurs in the alteration plutonic complexes. Patchy sulphides have moderate trace
Magmatic sulphide group Metasomatised sulphide group
7 16 10
44
56
94 84 90
~100 ~100
~100 ~100
Cu Zn As Se Cu Zn As Se
5
14
35 44 44 35
65 56 56
65
~100 95 86
Te Au Pb Sb Te Au Pb
High temperature sulphide group Low temperature sulphide group
29 25 25
38 32 40
62
71 68 75 60 ~100 ~100 65
Cu Zn As Se Cu Zn As Se
20
50 50
80 ~100 ~100 ~100 ~100
Sb Te Au Pb Sb Pb
Patchy sulphide group

MSS+ISS
16
43 38
57
Pyrite
~100 84 62
Cu Zn As Se Chalcopyrite
7
16
30 27 Sphalerite
84 70 73 93
Silicates and oxides

Sb Te Au Pb
Fig. 12. Pie charts illustrating the semi-quantitative results of the mass balance calculations. Charts show the proportion to which metals are
hosted by different sulfides in%. Magmatic sulphide assemblage is shown as a combination of monosulphide solid solution (MSS) and
intermediate solid solution (ISS) because of the too few data available for each sulphide phases in this group. Mass balance calculation cannot
be determined for Te and Au in the low temperature sulphides and for Sb in the magmatic sulphides. See Appendix A.4 and B for mass
balance calculation results and uncertainties.
metal concentrations (Table 5), suggesting metal remobili- 2015). The total mass of sulphide and the compositions of
sation from the surrounding rocks during off axis fluid the different phases in the sulphide group are then used to
circulation. calculate the hypothetical whole rock compositions, which
are then compared to the real whole rock compositions to
5.5. Mass balance give the% of each metal that is hosted as sulphide (Appen-
dix B). Several significant errors are associated with the
A mass balance calculation has been carried out to mass balance calculation, such as the problem of comparing
investigate the degree to which sulphide minerals control observations from a 2D thin section surface with analyses
the whole-rock metal concentrations in Hole 1256D sam- of a 3D sample volume, and error propagations are dis-
ples. The mass balance has been carried out on samples cussed in Appendix B. Because of these errors, the mass
containing one dominant sulphide group. Hypothetical balance is considered to be semi-quantitative but the results
whole rock compositions were calculated for each sample still enable evaluation of the extent to which the sulphides
assuming that all the metals present in the sample are solely control the whole rock metals budget in Hole 1256D. The
hosted by sulphides. The total mass of sulphide is calcu- results are classified into >50%, 10–50% and <10% sulphide
lated using whole rock S concentration (Patten et al., hosting.
Although based on a low number of samples (n = 2), the whereas Cu is hosted in chalcopyrite (Fig. 12) which is more
mass balance for samples hosting magmatic sulphides sug- resistant to leaching (Fig. 9A, B and 10). The depletions of
gests that Cu, Se, Te and Au are strongly controlled (>50%) As, Zn and Pb in Hole 1256D (27%, 8% and 44%
by sulphides (Fig. 12) whereas Zn, As, Sb and Pb are mostly respectively) are neither associated with sulphide leaching
controlled by silicate and oxide minerals (sulphide < 10%, (Fig. 12) nor oxide alteration (Fig. 8F) but most likely with
Fig. 12; Gurney and Ahrens, 1973; Doe, 1994; Alt, 1995). alteration of silicate minerals (Doe, 1994). Oxide minerals
Similarly to magmatic sulphides, metasomatised sulphides host significant Zn and some Sb (Table 6; Gurney and
strongly control Cu, Se, Te and Au (>50%), partly control Ahrens, 1973; Doe, 1994) and increased alteration of these
As and Pb (10–50%) and weakly control Zn and Sb (<10%; minerals could lead to greater depletion of these metals.
Fig. 12). The similarity in mass balance between the two Antimony and Te show no significant whole rock variation
sulphide groups suggests that during metasomatism, metals in Hole 1256D during high temperature hydrothermal alter-
undergo local grain-scale diffusion but are not extensively ation (Patten et al., 2015) and more intensive alteration
mobilised by the moderate temperature hydrothermal fluids would be required to mobilise these elements (Fig. 12).
(250–350 °C; Fig. 10). Samples hosting high temperature The different metal behaviour during sulphide leaching,
sulphides, located in the upper sheeted dykes and the as summarised in Fig. 10, and the control of oxide and sil-
transitional zone, are the only ones where sulphides control icate mineral alteration on mobility of some elements indi-
most of the whole rock metal budget (Fig. 12). In cates that peak mobilisation for each metal occurs at
these samples pyrite hosts most of the Se, Te, Au and Pb different stages, and under different conditions. Trace met-
whereas Cu is controlled by chalcopyrite and Zn by als hosted in pyrite are likely to be released during initiation
sphalerite (Fig. 12). Arsenic and Sb are only partly of leaching, whereas Cu and trace metals in chalcopyrite
controlled by sulphides in these samples. In the volcanic would be released at higher temperature during advanced
section, mass balance calculations of samples hosting low sulphide leaching leading to significant fractionation of
temperature sulphides suggest that Cu and As are strongly metals in the hydrothermal fluid. This process may affect
controlled and Se partly controlled by sulphides, whereas the relative metal contents of the high temperature
Zn, Sb and Pb are poorly controlled by sulphides hydrothermal fluids through time.
(Fig. 12). Finally, the mass balance calculation of the
patchy sulphide group suggests that only Cu is strongly 5.7. Metal-rich hydrothermal fluids and formation of VMS
hosted by these sulphides, whereas As, Se, Sb, Te and Au deposits
are partly controlled and Zn and Pb are poorly controlled
by sulphides (Fig. 12). At ODP Hole 1256D, the masses of metals mobilised
during thigh temperature hydrothermal alteration of the
5.6. Mineralogical controls on metal mobilisation sheeted dyke and plutonic complexes is sufficient to form
modern-day low tonnage mafic type VMS deposits
In the oceanic crust, significant mobilisation of metals (Patten et al., 2015) such as those found at the EPR 13° N
occurs from the lower sheeted dyke section during high tem- (Fouquet et al., 1988) or the TAG VMS deposit (e.g.
perature hydrothermal fluid circulation (e.g. Hole 504B: Hannington et al., 1998). Different mineral reactions pro-
Nesbitt et al., 1987; Alt et al., 1996; Bach et al., 2003; Hole vide different metals to the fluids with sulphide leaching
1256D: Teagle et al., 2006; Alt et al., 2010; Patten et al., providing the Cu, Se and Au, and silicate reactions provid-
2015; Pito deep: Heft et al., 2008). In Hole 1256D, Au, As, ing the As, Sb, Zn and Pb (Figs. 12 and 13). Lack of known
Se, Cu, Zn and Pb are variably depleted relative to primary VMS deposits in the vicinity of Hole 1256D, however, high-
crustal compositions (46%, 27%, 27%, 10%, 8% lights the importance of metal trapping mechanisms for for-
and 44% respectively; Patten et al., 2015). As magmatic mation of VMS deposits. The preservation of both
and metasomatised sulphides have strong control over Cu, magmatic and metasomatised (i.e. not fully leached) sul-
Se and Au budgets (Fig. 12) at the time of high temperature phides in Hole 1256D crust implies that sulphide leaching
hydrothermal alteration, sulphide leaching is clearly a key has not gone to completion. The preservation of these sul-
mineral reaction for mobilisation of these metals. This is phides in Hole 1256D is most likely due to the relative low
strongly supported by the large whole-rock metal depletions water–rock ratio in the lower part of Hole 1256D (w/r < 1;
in samples hosting metasomatised and leached sulphide Gao et al., 2012) which may be a feature of fast-spreading
minerals (i.e. in samples below 1061 mbsf; Fig. 13). Addi- ridge oceanic crust compared to other tectonic settings such
tionally, variability in whole rock metal depletions can be as ophiolites which are more extensively altered (Alt and
partly explained by the variable behaviour of the metals dur- Teagle, 2000). More extensive hydrothermal alteration
ing sulphide leaching. Copper, Se and Au are strongly would lead to total sulphide leaching and advance oxide
hosted by magmatic and metasomatised sulphides but Cu leaching, increasing the quantity of metal mobilised from
shows lower whole rock depletions (10%) than Se and the oceanic crust and the likeliness to form VMS deposits.
Au (27% and 46% respectively) in the sheeted dyke Epidosite zones observed in ophiolitic systems and inter-
and plutonic complexes. Selenium and Au are principally preted as the source areas for metals enriched in associated
hosted in pyrite (Fig. 12) which is reactive to sulphide leach- VMS deposits (Jowitt et al., 2012) are likely to represent
ing (Fig. 9) explaining the higher whole rock depletions, this advanced stage of sulphide leaching.
ΔCu ΔZn ΔAs ΔSe

-100 0 1001000 -100 0 100 -100 0 100 -100 0 1001000
1000 1000 1000 1000
Depletion Depletion Depletion Depletion
SD SD SD SD
1100 1100 1100 1100
Depth (mbsf)
Depth (mbsf)
Depth (mbsf)
1200
Depth (mbsf)
1200 1200 1200
1300 1300 1300 1300
1400 1400 1400 1400

PC PC PC PC
1500 1500 1500 1500
ΔSb ΔTe ΔAu ΔPb

-100 0 100 -100 0 100 -100 0 100 -100 0 100
1000 1000 1000 1000
Depletion Depletion Depletion Depletion
SD SD SD SD
1100 1100 1100 1100
Depth (mbsf)
Depth (mbsf)
Depth (mbsf)
1200
Depth (mbsf)
1200 1200 1200
1300 1300 1300 1300
1400 1400 1400 1400

PC PC PC PC
1500 1500 1500 1500
Primary magmatic Metasomatised High Temperature Patchy
Fig. 13. Mass variations in the sheeted dyke and plutonic complexes and the associated sulphide population. Mass variations are from whole
rock data (Patten et al., 2015). The main sulphide group observed in thin section is shown. No thin sections are available for several samples.
Samples hosting metasomatised sulphides show significant metal depletion. See text for details. SD = sheeted dyke complex; PC = plutonic
complex.
6. CONCLUSIONS conditions leads to metasomatism of the magmatic sul-

phides. Recrystallisation of pyrrhotite, chalcopyrite
In situ analyses of the sulphide and oxide minerals pre- and pentlandite to form secondary pyrite, chalcopyrite,
sent in the ODP Hole 1256D drill core enables determina- millerite and magnetite drives significant diffusion of
tion of their paragenesis during the evolution of the metals between the sulphide phases and the surrounding
oceanic crust and evaluation of the extent to which they silicates and oxides, although little bulk metal mobilisa-
control bulk metal behaviour in the crust (Fig. 7). The main tion to the circulating hydrothermal fluids occurs at this
outcomes of this study are: stage.
Increase in hydrothermal alteration temperature
Five main groups of sulphides occur in ODP Hole (>350 °C) under reduced conditions leads to progressive
1256D: magmatic, metasomatised, high temperature, leaching and dissolution of the metasomatised sulphides.
low temperature and patchy sulphides. Three groups Semi-quantitative mass balance calculations suggest that
of oxides occur: magmatic skeletal titanomagnetite, sulphide leaching is responsible for the release of most
magmatic coarse magnetite-ilmenite and hydrothermal Cu, Se and Au during the hydrothermal alteration of
related magnetite. the lower sheeted dyke section and the plutonic complex.
Initiation of hydrothermal fluid circulation in the lower The different behaviour of trace metals during sulphide
sheeted dyke section and the plutonic complex at moder- leaching suggests that they have different timing for peak
ate temperature (300 °C) and relatively oxidising mobilisation. Chalcopyrite is less reactive to leaching
than pyrite implying that Cu and the metals hosted in whole rock composition as if all the investigated elements
chalcopyrite are released during the highest temperature were solely hosted by sulphide minerals, and second, com-
alteration conditions. The mobilisation of Zn, As, Sb pare the hypothetical whole rock composition with the
and Pb is controlled by silicate reactions rather than sul- actual whole rock composition. Mass balance calculation
phide leaching. is carried out for each sulphide group from the average
A fraction of the metal mobilised by the high temperature metal whole rock composition of the samples hosting the
hydrothermal fluids are trapped within mineralised veins specific sulphide group (data from Patten et al. (2015))
and breccias as high temperature sulphides in the upper and from the average metal content of the sulphide phases
sheeted dyke section and the transitional zone during mix- analysed by in-situ analyses (Table 1–5). Although pent-
ing with low temperature fluids. The high temperature landite and millerite grains are too small for LA-ICP-MS
pyrite, sphalerite and chalcopyrite have high and variable analyses their impacts on the mass balance calculation of
trace metal concentrations which highlight different beha- the elements of interest can be considered negligible as they
viour during sulphide precipitation and metal trapping. represent only a fraction of the sulphide population in Hole
Late off-axis circulation of low temperature fluids 1256D. Calculation from average values rather than for
(<150 °C) in the oceanic crust leads to the local formation each individual sample is preferred because in-situ analyses
of pyrite fronts in the volcanic section and to patchy sul- of all samples was not possible. The whole rock values of
phides in the sheeted dyke and plutonic complexes. High the investigated elements, the abundance of each sulphide
As enrichments in the pyrite fronts suggests significant phase and the concentration of each element in each sul-
uptake from the circulating low temperature fluids in phide phase are required to carry out the mass balance.
the volcanic section whereas the moderate trace metal
concentrations in the patchy sulphides suggests late metal B.1. Hypothetical whole rock composition
remobilisation within the sheeted dyke and plutonic
complexes. To determine the hypothetical whole rock composition,
The hydrothermal fluids that form during alteration of the sulphide abundance must first be determined. It is calcu-
the lower sheeted dyke units in Hole 1256D are suffi- lated from the whole rock S concentration, and the propor-
ciently metal-rich to form low tonnage VMS deposits tions and the average S concentrations of each sulphide
on the seafloor. The metal enrichments in the fluids phase present:
are driven by different mineral reactions with Cu, Se
and Au mobilised by sulphide leaching, and Zn, As, Sulphðw:f :Þ ¼ S wr =ðF SA S SA þ F SB S SB þ Þ ðB:1Þ
Sb and Pb mobilised by silicate reactions. An increase
in alteration would likely lead to total sulphide leaching with S (w.f.) the sulphide weight fraction in the rock, Swr
and metal release leading to higher metal budget of the the whole rock average S concentration of the selected sam-
hydrothermal fluids increasing the likeliness of VMS ples, FSA the fraction of the sulphide phase A, SSA the S
deposits formation. concentration of the sulphide phase A, FSB the fraction of
the sulphide phase B and SSB the S concentration of the sul-
phide phase B. Whole rock S concentrations are from
ACKNOWLEDGEMENTS Patten et al. (2015) and S concentrations SSA and SSB are
average S from EPMA data shown in Tables 1–6. Little
This work was funded by Stockholm University and by the anhydrite has been observed in the investigated samples
Swedish Research Council (PRG 621-2007-4539). The authors and it is assumed that most of the S is present as sulphide.
would like to thank Cora Wohlgemuth-Ueberwasser and Jarek A few samples host multiple types of sulphide groups, in
Majka for in situ analyses. Analyses by EPMA were funded by which case calculations are made using the average compo-
the 2013 ECORD research grant. This research used samples pro- sition of the predominant sulphide group. Next, the hypo-
vided by the ODP and IODP. The ODP was sponsored by the thetic whole rock concentration is calculated:
National Science Foundation (NSF) and participating countries
under management of Joint Oceanographic Institutions (JOI). EHWR ¼ Sulphw:f : ðF SA ESA þ F SB S SB þ Þ ðB:2Þ
The IODP was supported by NSF; Japan’s Ministry of Education,
Culture, Sports, Science, and Technology; the European Consor- with EHWR the hypothetical whole rock concentration of
tium for Ocean Research Drilling; the Australia-New Zealand the element E and ESA and ESB the element E concentration
IODP Consortium; and the People’s Republic of China Ministry
in the sulphide phase A and B respectively.
of Science and Technology.
B.2. Mass balance

APPENDIX A. LA-ICP-MS REFERENCE MATERIALS,
MONITORS AND MASS BALANCE RESULTS The mass balance of the extent to which the sulphide
minerals host the trace metals is determined by comparing
See electronic annexes A.1–A.4. the hypothetical whole rock concentration from Eq. (2)
with the actual one:
APPENDIX B. MASS BALANCE CALCULATION
E Sulph ¼ EHWR =EWR ðB:3Þ
The method used to carry out the mass balance is with E Sulph the fraction of the element E controlled by sul-
divided into two main steps: first, calculate a hypothetical phides and EWR the actual whole rock concentration of the
element E (Appendix A.3). The fraction of the element E (3) Sulphide weight fraction error:
controlled by silicate and oxide minerals is characterised
by: Errors associated with sulphide weight fraction can be
divided in two: the error associated with S whole rock anal-
E Sil; Ox ¼ 1 E Sulph ðB:4Þ yses which is quantifiable and the error associated with the
sulphide/anhydrite ratio in the sample which is not quantifi-
able. Sulphur whole rock error has been determined in
B.3. Error in mass balance calculation Patten et al. (2015; ESM1). Anhydrite is rare in the investi-
gated samples but its impact on the mass balance calculation
Mass balance calculations generally yield significant is unknown and the error associated could be significant.
errors. The main errors associated with this study mass
balance calculation are the whole rock data error (rWR), (4) Sulphide phase fraction error:
the in situ analyses data error (rISA), the sulphide weight
fraction error (rSWF) and the sulphide phase fraction error The sample assemblage observed from thin section
(rSPF). observation by reflected light microscopy enabled to esti-
mate the sulphide phase proportion in the different sulphide
(1) Whole rock analytical error: group. Because sulphide phase proportion is estimated from
a surface area but is assumed representative of sample vol-
Whole rock analytical errors are quantifiable and have been ume significant unquantifiable error is associated with this
determined in Patten et al. (2015; ESM1). They correspond assumption especially for small size phases (Godel, 2013).
to the error associated with the analyses by low Au tech- Errors for the mass balance calculation are propagated
nique for Au, HG-AFS for As, Sb, Se and Te and LA- as follows:
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ICP-MS for Cu, Zn and Pb (see Patten et al. (2015) for
DMass balance ¼ r2WR þ r2ISA þ r2SWF þ r2SPF ðB:5Þ
details).
with DMass balance the mass balance error, rWR the whole rock
(2) In situ analyses data error: analyses error, rISA the in-situ analyses error, rSWF the sul-
phide weight fraction error and rSPF the sulphide phase
Error associated with in situ analyses are determined fraction error. Minimum mass balance errors of the quan-
from the monitors in A.2 used to check the LA-IPC- tifiable errors are propagated:
Mass balance minimum errors

Error type Cu Zn As Se Sb Te Au Pb
Whole rock analyses r1 % 4.6 2.3 5.5 7.6 4.2 48.6 1.1 10.9
In situ analyses r2 % n.d. n.d. 4.1 13.5 32.8 n.d. 25.4 n.d.
Whole rock S r3 % 10.1 10.1 10.1 10.1 10.1 10.1 10.1 10.1
Anhydrite fraction r4 % n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d.
Sulphide phase fraction r5 % n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d.
pffiffi 2
Mass balance minimum error ðr1 þ r22 þ r32 r42 þ r52 Þ % 11.1 10.4 12.2 18.5 34.6 49.6 27.4 14.8
r 1 corresponds to the errors of the whole rock elemental analyses from Patten et al. (2015; ESM 1); r 2 corresponds to LA-ICP-MS monitor
errors in A.2; r 3 corresponds to whole rock S error from Patten et al. (2015; ESM 1).
MS calibration. Unfortunately no appropriate reference Because several and significant errors are associated with the
materials were available for monitoring 65Cu, 66Zn, mass balance calculation presented in this study the mass
125
Te, and 208Pb. Despite this lack of monitoring the balance results are considered to be semi-quantitative. The
implications on the mass balance calculation are minim. implication is that the mass balance result of a mineral con-
Mass balance calculation suggests that Cu is strongly trol over a metal is not absolute but informative (Appendix
controlled by chalcopyrite in all the sulphide groups in A.3). Thus, a metal is either strongly controlled by a mineral
which case the EPMA data are used. Similarly Zn is phase if the mass balance result is >50%, partly controlled if
strongly controlled by silicate except when sphalerite is the mass balance result is between 10 and 50 % or poorly
present in which case EPMA data are also used. Mass controlled is the result is <10%.
balance results for Te and Pb are not as straightforward
as Cu and Zn and the unknown error from the lack of APPENDIX C. SUPPLEMENTARY DATA
proper monitoring can be significant. This error, how-
ever, is likely to be less than the errors associated with Supplementary data associated with this article can be
the sulphide weight and sulphide phase fractions (see found, in the online version, at http://dx.doi.org/10.1016/
below). j.gca.2016.08.009.
REFERENCES Carmichael I. S. (1966) The iron-titanium oxides of salic volcanic

rocks and their associated ferromagnesian silicates. Contr. Min.
Alt J. C. (1995) Subseafloor processes in mid-ocean ridge Pet. 14, 36–64.
hydrothermal systems. Seafloor Hydrothermal Syst. Phys. Coogan L. A. and Dosso S. (2012) An internally consistent,
Chem. Biol. Geol. Interact., 85–114. probabilistic, determination of ridge-axis hydrothermal fluxes
Alt J. C. and Anderson T. F. (1991) The mineralogy and isotopic from basalt-hosted systems. Earth Planet. Sci. Lett. 323, 92–101.
composition of sulfur in Layer 3 gabbros from the Indian Coumou D., Driesner T. and Heinrich C. (2008) The structure and
Ocean, OOP Hole 735B. Proc. Ocean Drill. Progr. Sci. Results dynamics of mid-ocean ridge hydrothermal systems. Science
118, 113–125. 321, 1825–1828.
Alt J. C. and Chaussidon M. (1989) Ion microprobe analyses of the Craig J. and Kullerud G. (1969) Phase relations in the Cu–Fe–Ni–S
sulfur isotopic composition of sulfides in hydrothermally system and their application to magmatic ore deposits. Econ.
altered rocks, DSDP/ODP Hole 504B1. Proc. Ocean Drill. Geol. Monogr. 4, 344–358.
Progr. Sci. Results 111, 41–45. Craig J. R. (1973) Pyrite–pentlandite assemblages and other low
Alt J. C. and Shanks W. C. (2011) Microbial sulfate reduction and temperature relations in the Fe–Ni–S system. Am. J. Sci. 273,
the sulfur budget for a complete section of altered oceanic 496–510.
basalts, IODP Hole 1256D (eastern Pacific). Earth Planet. Sci. Crerar D. A., Susak N., Borcsik M. and Schwartz S. (1978)
Lett. 310, 73–83. Solubility of the buffer assemblage pyrite + pyrrhotite + mag-
Alt J. C. and Teagle D. A. (2000) Hydrothermal alteration and netite in NaCl solutions from 200 to 350 °C. Geochim.
fluid fluxes in ophiolites and oceanic crust. Spec. Papers Geol. Cosmochim. Acta 42, 1427–1437.
Soc. Am., 273–282. Czamanske G. K. and Moore J. G. (1977) Composition and phase
Alt J. C., Honnorez J., Laverne C. and Emmermann R. (1986) chemistry of sulfide globules in basalt from the Mid-Atlantic
Hydrothermal alteration of a 1 km section through the upper Ridge rift valley near 37°N lat. Geol. Soc. Am. Bull. 88, 587–
oceanic crust, Deep Sea Drilling Project Hole 504B: mineral- 599.
ogy, chemistry and evolution of seawater basalt interactions. J. Dare S., Barnes S.-J. and Prichard H. (2010) The distribution of
Geophys. Res. Solid Earth 91, 10309–10335. platinum group elements (PGE) and other chalcophile elements
Alt J. C., Anderson T. F. and Bonnell L. (1989) The geochemistry among sulfides from the Creighton Ni–Cu–PGE sulfide deposit,
of sulfur in a 1.3 km section of hydrothermally altered oceanic Sudbury, Canada, and the origin of palladium in pentlandite.
crust, DSDP Hole 504B. Geochim. Cosmochim. Acta 53, 1011– Min. Dep. 45, 765–793.
1023. Dare S. A., Barnes S.-J. and Beaudoin G. (2012) Variation in trace
Alt J. C., Laverne C., Vanko D. A., Tartarotti P., Teagle D. A., element content of magnetite crystallized from a fractionating
Bach W., Zuleger E., Erzinger J., Honnorez I. and Pezard P. A. sulfide liquid, Sudbury, Canada: implications for provenance
(1996) Hydrothermal alteration of a section of upper oceanic discrimination. Geochim. Cosmochim. Acta 88, 27–50.
crust in the eastern equatorial Pacific: a synthesis of results from Dare S. A., Barnes S.-J., Beaudoin G., Méric J., Boutroy E. and
Site 504 (DSDP Legs 69, 70, and 83, and ODP Legs 111, 137, Potvin-Doucet C. (2014) Trace elements in magnetite as
140, and 148). Proc. Ocean Drill. Progr. Sci. Results 148, 417– petrogenetic indicators. Min. Dep. 49, 785–796.
434. Djon M. and Barnes S.-J. (2012) Changes in sulfides and platinum-
Alt J. C., Teagle D. A. H., Brewer T., Shanks W. C. and Halliday group minerals with the degree of alteration in the Roby,
A. (1998) Alteration and mineralization of an oceanic forearc Twilight, and High Grade Zones of the Lac des Iles Complex,
and the ophiolite-ocean crust analogy. J. Geophys. Res. Solid Ontario, Canada. Min. Dep. 47, 1–22.
Earth 103, 12365–12380. Doe B. R. (1994) Zinc, copper, and lead in mid-ocean ridge basalts
Alt J. C., Laverne C., Coggon R. M., Teagle D. A. H., Banerjee N. and the source rock control on Zn/Pb in ocean-ridge
R., Morgan S., Smith-Duque C. E., Harris M. and Galli L. hydrothermal deposits. Geochim. Cosmochim. Acta 58, 2215–
(2010) Subsurface structure of a submarine hydrothermal 2223.
system in ocean crust formed at the East Pacific Rise, ODP/ Duran C., Barnes S.-J. and Corkery J. (2015) Chalcophile and
IODP Site 1256. Geochem. Geophys. Geosyst. 11. platinum-group element distribution in pyrites from the sulfide-
Andrews A. J. (1979) On the effect of low-temperature seawater- rich pods of the Lac des Iles Pd deposits, Western Ontario,
basalt interaction on the distribution of sulfur in oceanic crust, Canada: implications for post-cumulus re-equilibration of the
layer 2. Earth Planet. Sci. Lett. 46, 68–80. ore and the use of pyrite compositions in exploration. J.
Bach W., Peucker-Ehrenbrink B., Hart S. R. and Blusztajn J. S. Geochem. Expl. 158, 223–242.
(2003) Geochemistry of hydrothermally altered oceanic crust: Feely R., Gendron J., Baker E. and Lebon G. (1994a) Hydrother-
DSDP/ODP Hole 504B—Implications for seawater-crust mal plumes along the East Pacific Rise, 8 400 to 11 500 N:
exchange budgets and Sr-and Pb-isotopic evolution of the particle distribution and composition. Earth Planet. Sci. Lett.
mantle. Geochem. Geophys. Geosyst. 4. 128, 19–36.
Barnes S. J. and Lightfoot P. C. (2005) Formation of magmatic Feely R. A., Massoth G. J., Trefry J. H., Baker E. T., Paulson A. J.
nickel-sulfide ore deposits and processes affecting their copper and Lebon G. T. (1994b) Composition and sedimentation of
and platinum-group element contents. Econ. Geol. 100th Anniv. hydrothermal plume particles from North Cleft segment,
Vol., 179–213. Juan de Fuca Ridge. J. Geophys. Res. Solid Earth 99,
Barnes S. J., Cox R. A. and Zientek M. L. (2006) Platinum-group 4985–5006.
element, gold, silver and base metal distribution in composi- Fouquet Y., Auclair G., Cambon P. and Etoubleau J. (1988)
tionally zoned sulfide droplets from the Medvezky Creek Mine, Geological setting and mineralogical and geochemical investi-
Noril’sk, Russia. Contrib. Miner. Pet. 152, 187–200. gations on sulfide deposits near 13 N on the East Pacific Rise.
Barnes S. J., Wells M. A. and Verrall M. R. (2009) Effects of Mar. Geol. 84, 145–178.
magmatic processes, serpentinization, and talc-carbonate alter- Francis R. D. (1990) Sulfide globules in mid-ocean ridge basalts
ation on sulfide mineralogy and ore textures in the Black Swan (MORB), and the effect of oxygen abundance in Fe–S–O liquids
disseminated nickel sulfide deposit, Yilgarn Craton. Econ. Geol. on the ability of those liquids to partition metals from MORB
104, 539–562. and komatiite magmas. Chem. Geol. 85, 199–213.
Gao Y., Vils F., Cooper K., Banerjee N., Harris M., Hoefs J., studies in rock-buffered systems; I, Iron-copper-zinc-lead sulfide
Teagle D., Casey J., Elliott T. and Laverne C. (2012) Downhole solubility relations. Econ. Geol. 87, 1–22.
variation of lithium and oxygen isotopic compositions of Herzig P. M. and Hannington M. D. (1995) Polymetallic massive
oceanic crust at East Pacific Rise, ODP Site 1256. Geochem. sulfides at the modern seafloor – a review. Ore Geol. Rev. 10,
Geophys. Geosyst. 13. 95–115.
Genna D. and Gaboury D. (2015) Deciphering the hydrothermal Honnorez J. A., Jeffrey C., Honnorez-Guerstein B. M., Laverne C.,
evolution of a VMS system by LA-ICP-MS using trace elements Muehlenbachs K. and Saltzman E. (1985) Stockwork-like
in pyrite: an example from the Bracemac-McLeod Deposits, sulfide mineralization in young oceanic crust; Deep Sea Drilling
Abitibi, Canada, and implications for exploration. Econ. Geol. Project Hole 504B. Init. Rep. Deep Sea Drilling Proj. 83,
110, 2087–2108. 263–282.
Gibert F., Pascal M.-L. and Pichavant M. (1998) Gold solubility Jenner F. E. and O’Neill H. S. (2012) Analysis of 60 elements in 616
and speciation in hydrothermal solutions: Experimental study ocean floor basaltic glasses. Geochem. Geophys. Geosyst. 13.
of the stability of hydrosulphide complex of gold (AuHS) at 350 Jowitt S. M., Jenkin G. R. T., Coogan L. A. and Naden J. (2012)
to 450 °C and 500 bars. Geochim. Cosmochim. Acta 62, Quantifying the release of base metals from source rocks for
2931–2947. volcanogenic massive sulfide deposits: Effects of protolith
Gilbert S., Danyushevsky L., Robinson P., Wohlgemuth-Ueber- composition and alteration mineralogy. J. Geochem. Expl.
wasser C., Pearson N., Savard D., Norman M. and Hanley J. 118, 47–59.
(2013) A comparative study of five reference materials and the Jochum K. and Verma S. (1996) Extreme enrichment of Sb, Tl and
lombard meteorite for the determination of the Platinum- other trace elements in altered MORB. Chem. Geol. 130, 289–
Group Elements and Gold by LA-ICP-MS. Geostat. Geoanal. 299.
Res. 37, 51–64. Jochum K. P., Willbold M., Raczek I., Stoll B. and Herwig K.
Gillis K. M., Muehlenbachs K., Stewart M., Gleeson T. and (2005) Chemical characterisation of the USGS reference glasses
Karson J. (2001) Fluid flow patterns in fast spreading East GSA-1G, GSC-1G, GSD-1G, GSE-1G, BCR-2G, BHVO-2G
Pacific Rise crust exposed at Hess Deep. J. Geophys. Res. Solid and BIR-1G Using EPMA, ID-TIMS, ID-ICP-MS and LA-
Earth 106, 26311–26329. ICP-MS. Geostat. Geoanal. Res. 29, 285–302.
Godel B. (2013) High-resolution X-ray computed tomography and Jochum K. P., Weis U., Stoll B., Kuzmin D., Yang Q., Raczek I.,
its application to ore deposits: From data acquisition to Jacob D. E., Stracke A., Birbaum K. and Frick D. A. (2011)
quantitative three-dimensional measurements with case studies Determination of reference values for NIST SRM 610–617
from Ni–Cu–PGE deposits. Econ. Geol. 108, 2005–2019. glasses following ISO guidelines. Geostat. Geoanal. Res. 35,
Gurney J. and Ahrens L. (1973) The zinc content of some 397–429.
ultramafic and basic rocks. Trans. Geol. Soc. S. Afr. 73, 301– Jochum K. P. and Nohl U. (2008) Reference materials in
307. geochemistry and environmental research and the GeoReM
Han D. S., Batchelor B. and Abdel-Wahab A. (2012) Sorption of database. Chem. Geol. 253, 50–53.
selenium (IV) and selenium (VI) onto synthetic pyrite (FeS 2): Jochum K. P., Stoll B., Weis U., Jacob D. E., Mertz-Kraus R. and
spectroscopic and microscopic analyses. J. Col Inter. Sci. 368, Andreae M. O. (2014) Non-matrix-matched calibration for the
496–504. multi-element analysis of geological and environmental samples
Hannington M. D. (2013) The role of black smokers in the Cu mass using 200 nm femtosecond LA-ICP-MS: a comparison with
balance of the oceanic crust. Earth Planet. Sci. Lett. 374, nanosecond lasers. Geostat. Geoanal. Res. 38, 265–292.
215–226. Keays R. R. (1987) Principles of mobilization (dissolution) of
Hannington M. D. (2014) 13.18 – Volcanogenic massive sulfide metals in mafic and ultramafic rocks—the role of immiscible
deposits. In Treatise on Geochemistry (eds. Heinrich D. Holland magmatic sulphides in the generation of hydrothermal gold and
and K. K. Turekian), 2nd ed. Elsevier, Oxford, pp. 463–488. volcanogenic massive sulphide deposits. Ore Geol. Rev. 2, 47–
Hannington M. D. and Scott S. D. (1989) Sulfidation equilibria as 63.
guides to gold mineralization in volcanogenic massive sulfides; Keith M., Haase K. M., Schwarz-Schampera U., Klemd R.,
evidence from sulfide mineralogy and the composition of Petersen S. and Bach W. (2014) Effects of temperature, sulfur,
sphalerite. Econ. Geol. 84, 1978–1995. and oxygen fugacity on the composition of sphalerite from
Hannington M. D., Galley A. G., Herzig P. M. and Petersen S. submarine hydrothermal vents. Geology 42, 699–702.
(1998) 28. Comparison of the TAG mound and stockwork Keith M., Haase K. M., Klemd R., Krumm S. and Strauss H.
complex with Cyprus-type massive sulfide deposits. Proc. Ocean (2016) Systematic variations of trace element and sulfur isotope
Drill. Progr. Sci. Results, 389. compositions in pyrite with stratigraphic depth in the Skouri-
Harris M., Coggon R. M., Smith-Duque C. E., Cooper M. J., otissa volcanic-hosted massive sulfide deposit, Troodos ophio-
Milton J. A. and Teagle D. A. H. (2015) Channelling of lite, Cyprus. Chem. Geol. 423, 7–18.
hydrothermal fluids during the accretion and evolution of the Kelly D. and Vaughan D. (1983) Pyrrhotine–pentlandite ore
upper oceanic crust: Sr isotope evidence from ODP Hole textures: a mechanistic approach. Miner. Mag. 47, 453–463.
1256D. Earth Planet. Sci. Lett. 416, 56–66. Kempton P. D., Autio L. K., Rhodes J. M., Holdaway M. J.,
Heft K. L., Gillis K. M., Pollock M. A., Karson J. A. and Klein E. Dungan M. A. and Johnson P. (1985) Petrology of basalts from
M. (2008) Role of upwelling hydrothermal fluids in the hole-504B, deep-sea drilling project, LEG-83. Init. Rep. Deep
development of alteration patterns at fast spreading ridges: Sea Drilling Proj. 83, 129–164.
evidence from the sheeted dike complex at Pito Deep. Geochem. Kent D. V., Honnorez B., Opdyke N. D. and Fox P. J. (1978)
Geophys. Geosyst. 9. Magnetic properties of dredged oceanic gabbros and the source
Hellstrom J., Paton C., Woodhead J. and Hergt J. (2008) Iolite: of marine magnetic anomalies. Geophys. J. Int. 55, 513–537.
software for spatially resolved LA-(quad and MC) ICPMS Koepke J., Christie D., Dziony W., Holtz F., Lattard D.,
analysis. Mineral. Ass. Can. Short Course Series 40, Maclennan J., Park S., Scheibner B., Yamasaki T. and
343–348. Yamazaki S. (2008) Petrography of the dike-gabbro transition
Hemley J., Cygan G., Fein J., Robinson G. and d’Angelo W. at IODP Site 1256 (equatorial Pacific): the evolution of the
(1992) Hydrothermal ore-forming processes in the light of granoblastic dikes. Geochem. Geophys. Geosyst. 9.
Laverne C., Grauby O., Alt J. C. and Bohn M. (2006) Hydroschor- Schiffman P., Bettison L. and Smith B. (1987) Mineralogy and
lomite in altered basalts from Hole 1256D, ODP Leg 206: the geochemistry of epidosites from the Solea graben, Troodos
transition from low-temperature to hydrothermal alteration. ophiolite, Cyprus. Ophiolites: Oceanic Crustal Analogues:
Geochem. Geophys. Geosyst. 7. Proceedings of the Symposium Troodos, 673–683.
Lepage L. D. (2003) ILMAT: an Excel worksheet for ilmenite– Seewald J. S. and Seyfried, Jr., W. E. (1990) The effect of
magnetite geothermometry and geobarometry. Comput. Geosci. temperature on metal mobility in subseafloor hydrothermal
29, 673–678. systems: constraints from basalt alteration experiments. Earth
Li C. and Ripley E. (2005) Empirical equations to predict the sulfur Planet. Sci. Lett. 101, 388–403.
content of mafic magmas at sulfide saturation and applications Shanks W. C. and Seyfried W. E. (1987) Stable isotope studies of
to magmatic sulfide deposits. Min. Dep. 40, 218–230. vent fluids and chimney minerals, southern Juan de Fuca Ridge:
MacLean W. H. (1977) Sulfides in Leg 37 drill core from the Mid- Sodium metasomatism and seawater sulfate reduction. J.
Atlantic Ridge. Can. J. Earth Sci. 14, 674–683. Geophys. Res. Solid Earth 92, 11387–11399.
Mathez E. A. (1979) Sulfide relations in Hole 418A flows and sulfur Smedley P. and Kinniburgh D. (2002) A review of the source,
contents of glasses. Init. Rep. Deep Sea Drilling Proj. 51, 1069– behaviour and distribution of arsenic in natural waters. Appl.
1085. Geochem. 17, 517–568.
Metz S. and Trefry J. H. (2000) Chemical and mineralogical Teagle D., Alt J., Umino S., Miyashita S., Banerjee N. and Wilson
influences on concentrations of trace metals in hydrothermal D.the Expedition 309/312 Scientists (2006) Superfast spreading
fluids. Geochim. Cosmochim. Acta 64, 2267–2279. rate crust 2 and 3. Proc. Integr. Ocean Drill. Program 309, 312–
Monecke T., Petersen S., Hannington M. D., Grant H. and 350.
Samson I. (2016) The minor element endowment of modern Tolstoy M., Waldhauser F., Bohnenstiehl D., Weekly R. and Kim
sea-floor massive sulfide deposits and comparison with deposits W.-Y. (2008) Seismic identification of along-axis hydrothermal
hosted in ancient volcanic successions. Rev. Econ. Geol.. flow on the East Pacific Rise. Nature 451, 181–184.
Moore J. G. (1975) Mechanism of formation of pillow Lava: pillow Violay M., Pezard P. A., Ildefonse B., Célérier B. and Deleau A.
lava, produced as fluid lava cools underwater, is the most (2012) Structure of the hydrothermal root zone of the sheeted
abundant volcanic rock on earth, but only recently have divers dikes in fast-spread oceanic crust: a core-log integration study
observed it forming. Am. Sci. 63, 269–277. of ODP hole 1256D, Eastern Equatorial Pacific. Ofioliti 37, 1–
Morse J. and Luther G. (1999) Chemical influences on trace metal- 11.
sulfide interactions in anoxic sediments. Geochim. Cosmochim. Wilson S. A., Ridley W. I. and Koenig A. E. (2002) Development
Acta 63, 3373–3378. of sulfide calibration standards for the laser ablation induc-
Murowchick J. B. and Barnes H. (1986) Marcasite precipitation tively-coupled plasma mass spectrometry technique. J. Anal.
from hydrothermal solutions. Geochim. Cosmochim. Acta 50, Atom. Spec. 17, 406–409.
2615–2629. Wilson D. S. et al. (2003) Leg 206 summary. Proc. Ocean Drill.
Nadoll P., Angerer T., Mauk J. L., French D. and Walshe J. (2014) Progr. Sci. Results, 206.
The chemistry of hydrothermal magnetite: a review. Ore Geol. Wilson D. S. et al. (2006) Drilling to gabbro in intact ocean crust.
Rev. 61, 1–32. Science 312, 1016–1020.
Nesbitt B. E., St. Louis R. M. and Muehlenbachs K. (1987) Wohlgemuth-Ueberwasser C. C., Viljoen F. and McClung C. R.
Distribution of gold in altered basalts of DSDP hole 504B. Can. (2014) Metamorphic alteration of the massive sulfide horizon
J. Earth Sci. 24, 201–209. from the Salt River VMS deposit (South Africa). Ore Geol. Rev.
Paton C., Hellstrom J., Paul B., Woodhead J. and Hergt J. (2011) 56, 45–52.
Iolite: freeware for the visualisation and processing of mass Wohlgemuth-Ueberwasser C. C., Viljoen F., Petersen S. and
spectrometric data. J. Anal. Atom. Spec. 26, 2508–2518. Vorster C. (2015) Distribution and solubility limits of trace
Patten C., Barnes S.-J. and Mathez E. A. (2012) Textural variations elements in hydrothermal black smoker sulfides: an in-situ LA-
in MORB sulfide droplets due to differences in crystallization ICP-MS study. Geochim. Cosmochim. Acta 159, 16–41.
history. Can. Min. 50, 675–692. Wolthers M., Van der Linde P. and Van der Weijden C. (1998)
Patten C., Barnes S.-J., Mathez E. A. and Jenner F. E. (2013) Selenium and arsenic in sedimentary pyrites. Min. Mag. A 62,
Partition coefficients of chalcophile elements between sulfide 1660–1661.
and silicate melts and the early crystallization history of sulfide
liquid: LA-ICP-MS analysis of MORB sulfide droplets. Chem. FURTHER READING
Geol. 358, 170–188.
Patten C. G. C., Pitcairn I. K., Teagle D. A. and Harris M. (2015) Ji-Hai Y., Xiu-Chun Z., Chen-Zi F., Ling-Hao Z., Dong-Yang S.,
Mobility of Au and related elements during the hydrothermal Ze-Rong J., Ming-Yue H. and Li-Jun K. (2012) Quantitative
alteration of the oceanic crust: implications for the sources of analysis of sulfide minerals by laser ablation-inductively cou-
metals in VMS deposits. Min. Dep., 1–22. pled plasma-mass spectrometry using glass reference materials
Peach C. L., Mathez E. A. and Keays R. R. (1990) Sulfide melt with matrix normalization plus sulfur internal standardization
silicate melt distribution coefficients for noble-metals and other calibration. Chin. J. of Anal. Chem. 40, 201–207.
chalcophile elements as deduced from MORB – implications for Wohlgemuth-Ueberwasser C. C., Ballhaus C., Berndt J., Née
partial melting. Geochim. Cosmochim. Acta 54, 3379–3389. Paliulionyte V. S. and Meisel T. (2007) Synthesis of PGE sulfide
Richardson C. J., Cann J. R., Richards H. G. and Cowan J. G. standards for laser ablation inductively coupled plasma mass
(1987) Metal-depleted root zones of the Troodos ore-forming spectrometry (LA-ICP-MS). Cont. Miner. Pet. 154, 607–617.
hydrothermal systems, Cyprus. Earth Planet. Sci. Lett. 84, 243–
253.
Associate editor: Wolfgang Bach
Riley J. F. (1977) The pentlandite group (Fe, Ni, Co) 9S8: new data
and an appraisal of structure-composition relationships. Min.
Mag. 41, 345–349.

Patten 2016

Uploaded by

Copyright:

Available Formats

Patten 2016

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Patten 2016

Uploaded by

Copyright:

Available Formats

Available online at www.sciencedirect.

Sulphide mineral evolution and metal mobility during

1. INTRODUCTION Patten et al., 2015). These metals are depleted at diﬀerent

Sheet flow Massive flow Pillows Dykes Gabbro Granoblastic dykes

signal rather than due to 49Ti16O+ interference, as the

4.1. Sulphide classiﬁcation

LA-ICP-MS LA-ICP-MS LA-ICP-MS

LA-ICP-MS LA-ICP-MS LA-ICP-MS

using assemblage, texture, grain size and distribution. Three

temperature ﬂuid circulation, and (3) low temperature ﬂuid

4.1.1. Magmatic sulphides

Magmatic sulphides consist of pyrrhotite, chalcopyrite

magmatic sulphides (Fig. 3A and B; Czamanske and

Moore, 1977; Mathez, 1979; Peach et al., 1990; Barnes

in two samples from the volcanic section and the plutonic

complex at 445 mbsf and 1493 mbsf respectively (Fig. 2)

and represent only a minor proportion of the Hole 1256D

sulphide population. In the volcanic section, magmatic

sulphides occur as 10–50 micron-wide blebs composed of

pyrrhotite and chalcopyrite intergrowths representative of

quenched monosulphide solid solution (MSS) and interme-

diate solid solution (ISS; Fig. 3A; e.g. MacLean, 1977;

Czamanske and Moore, 1977; Patten et al., 2012). Similar

(Czamanske and Moore, 1977; Mathez, 1979; Francis,

tonic complex magmatic sulphides do not have systematic

s.d. = standard deviation; D.L. = detection limit; N.D. = not determined.

chalcopyrite and pentlandite (Fig. 3B). The latter occurs as

a lamellae exsolution within the pyrrhotite, which is

characteristic of slowly cooled magmatic sulphides that

Vaughan, 1983; Alt and Anderson, 1991; Barnes et al.,

2006; Dare et al., 2010; Djon and Barnes, 2012). Mineral

proportions by visual estimation are 90% pyrrhotite with

trace exsolution pentlandite and 10% chalcopyrite.

Quenched sulphide blebs from the volcanic section are a

mixture of MSS and ISS and have diﬀerent composition

from the magmatic sulphides in the plutonic complex

the Fe and S composition of pyrrhotite but show high and

variable Cu, Zn, Pb and trace metals concentrations due to

contamination from small chalcopyrite inclusions (Table 1).

Large pyrrhotite grains from the plutonic complex contain

detection limits. Chalcopyrite from the plutonic complex

has higher concentrations of Zn (1169 ppm, 2–5000 ppm),

C.G.C. Patten et al. / Geochimica et Cosmochimica Acta 193 (2016) 132–159

s.d. = standard deviation; D.L. = detection limit; N.D. = not determined.

C.G.C. Patten et al. / Geochimica et Cosmochimica Acta 193 (2016) 132–159

s.d. = standard deviation; D.L. = detection limit; N.D. = not determined.

s.d. = standard deviation; D.L. = detection limit; N.D. = not determined.

cracks (Fig. 5D–F). These pyrite fronts occur on the border

precipitation. Diﬀerences in composition between high tem- 4.1.5. Patchy sulphides

(A) Magmatic assemblage (B) Metasomatism

(C) Partial sulphide leaching (D) Extensive sulphide leaching

105 102 102

103 10-1 10-1

10-2 10-3 100

Magmatic sulphide group Metasomatised sulphide group

High temperature sulphide group Low temperature sulphide group

Patchy sulphide group

Silicates and oxides

ΔCu ΔZn ΔAs ΔSe