Accepted Manuscript

Magmatic Cu-Ni sulfide mineralization of the Huangshannan mafic-untramafic

intrusion, Eastern Tianshan, China

Yun Zhao, Chunji Xue, Xiaobo Zhao, YongQiang Yang, Junjun Ke

PII: S1367-9120(15)00165-0

DOI: http://dx.doi.org/10.1016/j.jseaes.2015.03.031

Reference: JAES 2310

To appear in: Journal of Asian Earth Sciences

Received Date: 18 November 2014

Revised Date: 17 March 2015

Accepted Date: 18 March 2015

Please cite this article as: Zhao, Y., Xue, C., Zhao, X., Yang, Y., Ke, J., Magmatic Cu-Ni sulfide mineralization of

the Huangshannan mafic-untramafic intrusion, Eastern Tianshan, China, Journal of Asian Earth Sciences (2015),

doi: http://dx.doi.org/10.1016/j.jseaes.2015.03.031

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers

we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and

review of the resulting proof before it is published in its final form. Please note that during the production process

errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Magmatic Cu-Ni sulfide mineralization of the Huangshannan mafic-untramafic

intrusion, Eastern Tianshan, China

Yun Zhao, Chunji Xue *, Xiaobo Zhao, YongQiang Yang, Junjun Ke

State Key Laboratory of Geological Processes and Mineral Resources, China

University of Geosciences, Beijing 100083, China

*Corresponding author. Email: chunji.xue@cugb.edu.cn

Tel.: +86 10 82321895; Fax: +86 10 82322175

Abstract

The Huangshannan Ni-Cu (-PGE) sulfide deposit, a new discovery from geological

prospecting in Eastern Tianshan, is in a belt of magmatic Ni-Cu (-PGE) sulfide

deposits along the southern margin of the Central Asian Orogenic Belt. The host

intrusion of the Huangshannan deposit is composed of a layered ultramafic sequence

and a massive gabbro-diorite unit. The major sulfide orebodies occur mainly within

websterite and lherzolite in the layered ultramafic sequence. In-situ zircon U-Pb

dating analyses yielded a crystallization age of 282.5 ±1.4 Ma, similar to the ages of

the Permian Tarim mantle plume. Samples from the Huangshannan intrusion are

characterized by nearly flat rare earth elements patterns, negative Zr, Ti and Nb

anomalies, arc-like Th/Yb and Nb/Yb ratios, and significantly lower rare earth

element and immobile trace element contents than the Tarim basalts. These

characteristics suggest that the Huangshannan intrusion was not generated from the

Tarim mantle plume. The primary magma for the Huangshannan intrusion and its

associated sulfide mineralization were formed from different pulses of picritic magma

with different degrees of crustal contamination. The first pulse underwent an initial

removal of 0.016% sulfide in the deep magma chamber. The evolved magma reached

sulfide saturation again in the shallow magma chamber and formed sulfide ores in

lherzolite. The second pulse of magma reached a level of 0.022% sulfide segregation

at staging chamber before ascending up to the shallow magma chamber. In the

shallow conduit system, this sulfide-unsaturated magma mixed with the first pulse of

magma and with contamination from the country rocks, leading to the formation of

sulfide ores in websterite. The third magma pulse from the deep chamber formed the

unmineralized massive gabbro-diorite unit of the Huangshannan intrusion.

Keywords: Central Asian Orogenic Belt, Eastern Tianshan, Huangshannan

intrusion, Permian, magmatic Cu-Ni sulfide deposit

1. Introduction

Most world-class magmatic sulfide deposits are formed within cratons or on their

margins in association with intra-plate magmatism (Barnes and Lightfoot, 2005; Begg

et al., 2010). Such examples include Jinchuan on the western margin of the North

China Craton (Fig. 1a; Tang and Li, 1995; Tang et al., 2002; Lehmann et al., 2007),

the Bushveld igneous complex in the Kaapvaal Craton (Clarke et al., 2009), and

Noril'sk in the Siberian Craton (Naldrett, 1992; Maier et al., 2000). Thus the

magmatic evolution and mineralization processes forming magmatic Cu-Ni deposits

within cratons have been well documented (Barnes and Lightfoot, 2005; Naldrett,

2009; Begg et al., 2010). Conversely, the relatively small magmatic sulfide deposits

formed in orogenic belts that have undergone complex evolution histories have not

been well studied, resulting in a debatable understanding of their sulfide

mineralization and magmatic conduit systems (Gao et al., 2012, 2013; Su et al.,

2013).

Many important magmatic Ni-Cu (-PGE) sulfide deposits are distributed along

the southern margin of the Central Asian Orogenic Belt (CAOB), including

Kalatongke (Gao et al., 2012), Huangshandong (Sun et al., 2013), Erbutu (Peng et al.,

2013), Heishan (Xie et al., 2013) and Hongqiling (Wei et al., 2013) (Fig. 1a). The

Eastern Tianshan region is projected to become an important source of Ni and Cu

metal in the CAOB, especially from those deposits along the Huangshan-Kangguer

fault (Fig. 1b). Study of these Ni-Cu (-PGE) sulfide deposits provides us with the

possibility to understand how their magmatic conduit systems evolved and to examine

the relationship between their sulfide mineralization and the sequence of magmatic

emplacement in the Eastern Tianshan region. In addition, most mafic-ultramafic

intrusions in the Eastern Tianshan were intruded between 300 to 270 Ma, which is

similar to the eruption age of the ~280 Ma alkaline basalts of Tarim plume in

northwest China. Whether these mafic-ultramafic intrusions were related to the Tarim

mantle plume (e.g., Pirajno et al., 2008; Qin et al., 2011; Su et al., 2011; Tang et al.,

2011; Mao, et al., 2013) or to a post-collisional setting (e.g., Gao and Zhou 2013; Gao

et al., 2013;Sun et al., 2013) is still debatable.

The Huangshannan Ni-Cu (-PGE) sulfide deposit is a new discovery from

geological prospecting in Eastern Tianshan in recent years and has attracted the

interest of many Chinese geologists. The deposit contains about 30 million metric

tons (Mt) of sulfide ore with an average grade of 0.4 wt % Ni and 0.12 wt % Cu. The

size and grade of the Huangshannan deposit makes it a good choice for studying the

evolution of a magmatic mineralization system in an orogenic belt, given that there

has been almost no geological research to date on the topic.

In this paper, we describe the geology of the Huangshannan intrusion and the

varieties of sulfide ores found within its different host rocks. In-situ zircon U-Pb

dating is used to constrain the tectonic setting of emplacement for the Huangshannan

intrusion. We use whole-rock composition data to evaluate the source mantle

characteristics and the abundances of chalcophile elements, PGE, Cu and Ni to

constrain the nature of the parental magmas and sulfide mineralization processes in

the magmatic conduit system.

2. Geologic setting

The Huangshannan deposit is located in Eastern Tianshan in the southern area of

the CAOB (Fig. 1a). The CAOB is the largest Phanerozoic juvenile orogenic belt in

the world. It extends for more than 7000 km from the Pacific Ocean to the Eastern

European Craton, and it is bounded by the Tarim and North China Cratons to the

south and the Siberian Craton to the north (Sengör et al., 1993). The CAOB is

composed of fragments of Precambrian continental blocks and Paleozoic island arcs,

ophiolites, and volcanic rock assemblages (Sengör et al., 1993; Windley et al., 2002;

Jahn et al., 2004; Xiao et al., 2009). The Tianshan Orogenic Belt is the southernmost

part of the CAOB and it is separated from the Tarim Block by the North Tarim and

Xingxingxia Faults (Fig.1b). The North Tianshan and South Tianshan suture zones

divide the Tianshan Orogenic Belt in China into the North, Middle and South

Tianshan terranes from north to south (Fig.1b).

The Huangshan area includes the Middle and North Tianshan terranes that are

separated by the Shaquanzi fault (Fig. 2). The Middle Tianshan terrane consists

mainly of Precambrian basement complexes overlain by Paleozoic sedimentary and

volcanic strata. The North Tianshan terrane is composed of abundant Paleozoic

volcanic and sedimentary strata that include Lower Devonian to Lower Carboniferous

sandstone and pelitic slate interlayered with conglomerate, pyrite-bearing mudstone

and limestone, and Middle-Upper Carboniferous mafic to intermediate volcanic rocks

associated with abundant chert and limestone (Gao et al., 2013). Many Paleozoic

granitic plutons occur in this region.

The evolution of the Eastern Tianshan region is much debated. Some researchers

proposed that the Eastern Tianshan resulted from southward subduction of the

Junggar Ocean (Su et al., 2012b; Zhang et al., 2004), whereas others have suggested a

northward subduction (Li et al., 2003; Wang et al., 2006). The end of oceanic

subduction in the Eastern Tianshan region is also controversial with viewpoints

varying from Late Carboniferous (Mao et al., 2008; Su et al., 2012b;Wang et al.,

2006) to Late Permian or Early Triassic (Xiao et al., 2009, 2010). However, most

researchers have agreed that the mantle beneath the Eastern Tianshan region was

continuously modified by oceanic subduction prior to the Late Permian (Gao et al.,

2013; Mao et al., 2008; Su et al., 2012b; Zhou et al., 2004).

The Ni-Cu (-PGE) deposits in Huangshan area are hosted mainly in small mafic-

ultramafic intrusions that are distributed between the NE-trending Kangguer-

Huangshan and Shaquanzi faults. The Huangshannan intrusion is located along the

middle part of the Kangguer-Huangshan fault (Fig. 2).

3. Geology of the intrusion

The Huangshannan mafic–ultramafic intrusion is lens-shaped with an exposed

length of 5.2 km and width of 1.3 km with an area of about 4 km2 (Fig. 2). The pluton

is emplaced into the Carboniferous Gandun Formation that is composed of biotite

quartz schist, two-mica quartz schist and garnet-bearing biotite quartz schist.

The Huangshannan intrusion is composed of a lherzolite–websterite unit with a

layered sequence in the eastern part and a massive gabbro-diorite unit in the eastern

and western part (Fig. 3a). Interfingering occurs at the contact of the layered and

massive units, (Wang et al., 1987) and websterite fragments are found in the massive

unit (Fig. 3), indicating that the massive unit formed later than the layered unit. The

layered sequence is composed of harzburgite, lherzolite, olivine websterite, websterite

and hornblende websterite. Also, lherzolite xenoliths are enclosed by websterite (Fig.

3), which may indicate that the layered unit was formed from different magma pulses.

The lherzolite is composed of 50-60% olivine, 8-20% orthopyroxene, 5-15%

clinopyroxene, 1-3% hornblende with minor phlogopite and chrome spinel. Chrome

spinel is common in lherzolite and is usually enclosed in olivine and orthopyroxene

but may occur also as single interstitial grains with olivine and pyroxenes. Websterite

consists of ~10% olivine, ~35% orthopyroxene, 50-60% clinopyroxene and ~3%

hornblende with minor phlogopite.

The massive unit is composed dominantly of norite, gabbro, hornblende gabbro

and diorite (Fig. 3a). Norite is made up of 50-60% plagioclase, 20-30%

orthopyroxene, 10-20% clinopyroxene, <5% olivine, and minor hornblende (5-10%).

Hornblende gabbro is made up of ~60% plagioclase, ~25% clinopyroxene, 15-20%

hornblende, ~2% biotite and minor apatite. Gabbro is comprised of 40-50%

plagioclase, 30-40% clinopyroxene, 5-10% orthopyroxene and 5-10% biotite, and

minor hornblende. Diorite contains plagioclase (30-60%), clinopyroxene (5-10%),

hornblende (20-30%), biotite (5-15%) and quartz (5-10%). Through detailed optical

microscopy (e.g., Fig. 4), we propose the following crystallization sequence: olivine

→ orthopyroxene → clinopyroxene + plagioclase → hornblende + biotite.

In the ultramafic rocks, the olivine crystals are enclosed in clinopyroxene and

orthopyroxene (Fig. 4a, 4b) and are generally weakly altered, whereas those rocks that

are in direct contact with sulfides are more commonly altered to serpentine.

Remaining olivine cores are rare in the websterite. Most orthopyroxene and

clinopyroxene crystals have been partially or totally altered to tremolite and talc,

however the orthopyroxene is more strongly altered than the clinopyroxene.

Compared to the layered unit, the silicate rocks from the massive unit have

experienced weaker alteration. The euhedral-subhedral plagioclase laths vary in

length from 0.1 to 1 mm and form a latticework in which olivine, orthopyroxene and

clinopyroxene crystals infill the interstitial spaces. Some plagioclase crystals have

been partially altered to sericite. Hornblende and phlogopite crystals occur in isolated

interstitial spaces.

4. Ore geology

There are 18 identified orebodies in the Huangshannan intrusion between

exploration lines 27 to 40, including four relatively large orebodies. These orebodies

dip generally to the west with a length of 50 to 330 m, width of 20 to 70 m, and

thickness of 15 to 60 m.

The abundance of sulfides in these orebodies decreases upward and the orebodies

in the western part are more deeply buried (Fig. 3b). These lenticular and veined Ni–

Cu sulfide orebodies occur mainly within websterite and lherzolite at the base and in

the lower parts of the intrusion (Fig. 3b). The sulfide ores are generally weakly

disseminated and disseminated whereas net-textured, semi-massive and massive ores

are relatively rare. The massive and semi-massive sulfide accumulations are generally

small with a tabular, lenticular or pipe-like shape. The contacts between these

mineralization types and the disseminated sulfide ores and country rocks are sharp,

whereas the contacts with the weakly disseminated and disseminated sulfide orebodies

are generally gradational.

The sulfide assemblages are composed mainly of pyrrhotite, pentlandite and

chalcopyrite, and they occur mainly in the interstitial spaces of olivine + pyroxene

cumulates (Fig. 4a, 4b, 4c). Pentlandite occurs as flames or thin veinlets either within

pyrrhotite or as distinct crystals. Chalcopyrite is developed generally at the edge of

pyrrhotite and/or pentlandite crystals or as inclusions in them (Fig. 4e, 4f). The sulfide

assemblages occur as irregular isolated patches in the interstitial spaces of olivine,

orthopyroxene or clinopyroxene crystals (Fig. 4a, 4b).

5. Sampling and Analytical Methods

The sulfide-bearing samples for this study were collected from underground adits

in the eastern part of the ore deposit (Fig. 2) because only underground mining is

carried on between prospecting lines 27 and 40. Most of the lherzolite and websterite

samples were collected from the layered unit and gabbro, hornblende gabbro and

diorite samples were collected from the massive unit (Fig. 2).

5.1. Zircon U-Pb dating

Zircon grains were separated using standard density and magnetic separation

techniques. Zircon separates together with zircon standards TEMORA and 91500

were mounted in epoxy and then polished to section the crystals for analysis. All

zircons were photographed in transmitted, reflected and cathodoluminescence (CL)

light to reveal their internal structures. The mount was vacuum-coated with high-

purity gold prior to running the U-Pb isotopic analyses.

Measurements of U, Th and Pb isotopes were conducted using a Cameca IMS

1280 large-radius secondary ion mass spectrometer (SIMS) at the Institute of Geology

and Geophysics, Chinese Academy of Sciences, Beijing. Analytical procedures are the

same as those described by Li et al. (2009). The O2−primary ion beam was accelerated

to 13 kV with an intensity of ca. 8 nA. Positive secondary ions were extracted using a

10 kV potential. The ellipsoidal spot was approximately 20 ×30 μm in size. The

oxygen flooding technique was used to increase the O2 pressure to ca. 5 ×10-6

Torr in

the sample chamber, thereby enhancing the Pb+ sensitivity to a value of 24-28

cps/nA/ppm for zircon.

Precise mass calibration was maintained by using an automatic routine in the

Cameca CIPS software to scan over large peaks and to extrapolate the mass of the B-

field curves for peaks between these reference points (Whitehouse and Kamber,

2005). Correction of common lead was made using the measured 204

Pb. An average

Pb of present-day crustal composition (Stacey and Kramers, 1975) was used for the

common Pb correction, assuming that it is largely due to surface contamination

introduced during sample preparation. Uncertainties for the individual analyses in the

data tables are reported at the 1σ level; mean ages for pooled U/Pb analyses are

quoted at the 2σ or 95% confidence level. Data reduction was carried out using the

ISOPLOT 3.23 program (Ludwig, 2003).

5.2. Elements, S and PGE analyses

The collected samples were powdered in an agate mortar. Whole-rock major,

trace element, and PGE contents were analyzed at the National Research Center of

Geoanalysis in Beijing, China. Whole-rock major elements were determined by

mixing the samples with lithium metaborate and then fusing them at 1000°C. The

melts were poured into glass beakers with 15 ml of 5% aqua regia and then the

beakers were subjected to ultrasonic dissolution for 15 min. When the melts

dissolved, the solution was diluted to 25 ml with 5% aqua regia for determination by

inductively coupled plasma-atomic emission spectrometry (ICP-AES). The analytical

precision is better than 1.5%.

Whole-rock trace element, Cu and Ni compositions were determined by acid

digestion in steel-jacketed Teflon “bombs” followed by analysis by inductively

coupled plasma-mass spectrometry (ICP-MS). The accuracies are estimated to be

better than 2~10% RSD. Whole-rock S contents were measured using a high-

frequency infrared carbon sulfur analyzer. The detection limits are 0.005% and the

accuracies are estimated to be better than 10% RSD.

The concentrations of PGE were determined by nickel sulfide fire assay and Te

coprecipitation, followed by ICP-MS analysis. A detailed description of the method

for PGE analysis is presented by Asif and Parry (1991). Precision and accuracy as

demonstrated by analyzing reference materials such as UMT-1 and WPR-1 are better

than 10%.

6. Analytical Results

SIMS zircon U-Pb isotope data for the Huangshannan intrusion are listed in Table

1. Whole-rock major and trace element compositions, and chalcophile elements (PGE,

Cu, and Ni) are given in Appendix Table A1.

6.1. Zircon U–Pb dating

A gabbro sample from the massive unit in the eastern part of the Huangshannan

intrusion was selected for zircon U–Pb dating. Zircons from the sample vary from

euhedral to anhedral with most occurring as crystal fragments with rounded

terminations from initially equant to short or long prismatic crystals (Fig. 5). The

lengths of the crystals range from 80 to 200 μm with aspect ratios from 1:1 to 3:1.

Most crystals display oscillatory or patchy linear zoning with variable luminescence

in CL images (Fig. 5). U, Th and Pb contents vary from 836 to 2589 ppm, 134 to 2689

ppm, and 43 to 28.7 ppm, respectively. Th / U ratios are ca. 0.12–1.09. All analyses

have concordant U–Pb ages within analytical errors (Fig. 6), yielding a concordia age

of 282.5 ±1.4 Ma (MSWD = 0.25). This age is considered to be the crystallization

age, which is identical to those from other mineralized mafic-ultramafic intrusions in

Eastern Tianshan.

6.2. Major and trace elements

We compare the LOI (loss on ignition)- and sulfide-free compositions of gabbro,

diorite, hornblende gabbro, lherzolite, weakly disseminated sulfide ores (<1.2 wt % S)

and disseminated sulfide ores (up to 12.66 wt % S) (Fig. 7). The assumptions and

procedure used for calculating the bulk sulfide composition from whole rock Ni, Cu

and S analyses were first introduced by Naldrett and Duke (1980) and explained in

detail by Li et al. (2001). We assume that the magmatic sulfide melt only crystallizes

pyrrhotite, pentlandite and chalcopyrite and that the sulfide ores haven’t been affected

by alteration. The amounts of total sulfides in the samples were estimated from the

contents of S, Cu and Ni. We ignore the Ni and Cu in pyrrhotite and then use the

assay values of Ni and Cu in whole rock and ideal stoichiometric compositions to

calculate the percentage of pentlandite (Fe4.5Ni4.5S8) and chalcopyrite (CuFeS2) in a

given sample, respectively. We can then use the balance of S left over to calculate the

percentage of pyrrhotite (Fe7S8) in the sample. The normalized data show that the

SiO2, Al2O3, TiO2 and Na2O + K2O contents increase with decreasing MgO content

(Fig. 7a-e), whereas the FeO total content decreases with increasing MgO content

(Fig. 7f), indicating that different lithofacies result from the fractional crystallization

of the parental magma sourced from a coherent magma chamber.

The chondrite-normalized rare earth element (REE) and primitive mantle-

normalized immobile trace element patterns for the Huangshannan mafic-ultramafic

rocks and sulfide-bearing samples occurring in the ultramafic rocks are illustrated in

Figure 8. The average compositions from the Permian Tarim basalts and global

volcanic arc basalts are presented for comparison. The Huangshannan mafic rocks are

characterized by high REE abundances, the sulfide ores show moderate light REE

enrichments, and the lherzolite samples exhibit the low REE enrichments (Fig. 8a). In

addition, the immobile trace elements show similar enrichment characteristics as the

REE. In contrast with the Tarim basalts, the Huangshannan mafic-ultramafic rocks

and sulfide-mineralized samples have much lower REE abundances and show slighter

light REE enrichments relative to heavy REE. Unlike the Tarim basalts also, the

Huangshannan intrusion shows pronounced depletions of Zr and Ti, as well as

negative Nb anomalies relative to Th and Ta (Fig. 8b). These REE and immobile trace

element characteristics show that the Huangshannan intrusion did not result from

intrusive activity related to the Tarim mantle plume even though the Huangshannan

intrusion was emplaced at a similar age as the eruption of basalts from the Tarim

plume.

6.3. Ni, Cu, and PGE

All of the disseminated sulfide ores show higher PGE concentrations than the

weakly disseminated sulfide ores, and the mafic-ultramafic rocks retain the lowest

PGE contents (Fig. 9). All samples from the Huangshannan intrusion show significant

depletions of PGE relative to Cu and Ni. In addition, they show similar fractionated

PGE patterns, except for the variation in the Pt anomaly (Fig. 9). The sulfide-bearing

samples show an excellent positive correlation between Ir and Os, Ru and Rh, and a

poor correlation between Ir and Pt or Pd (Fig. 10). The Pt/Ir ratios from 0.71 to 414.29

for the sulfide ore samples have a much wider range than the Pd/Ir ratios from 12.86

to 201.43, and Pt exhibits a poor positive correlation with Pd (Fig. 10). The Cu/Pd

ratios of the Huangshannan intrusion range from 1.6 ×104 to 2.98 ×10

7 and there is no

obvious difference between the mafic-ultramafic rocks and the sulfide-bearing

samples (Fig. 11).

7. Discussion

7.1. Alteration effects

The Huangshannan intrusion has undergone variable alteration, especially the

websterite, which may have led to the loss of some mobile elements. Such alteration

may affect our following discussion, so it is important to assess our samples to see if

they are appropriate for modeling the ore-forming process.

Hydrothermal or metamorphic sulfide assemblages of magmatic Ni-Cu (-PGE)

deposits commonly have moderate or significantly variable Pd and extremely low Ir

contents (Keays, 1982; Molnar et al., 1997; 2001). Given that Pd and Ni are usually

relatively more mobile whereas Ir remains stable and that hydrothermal or

metamorphic sulfide assemblages commonly have extremely low Ir, the result is very

high Pd/Ir and Ni/Ir ratios. Conversely, sulfide assemblages without hydrothermal or

metamorphic alteration show a positive correlation between Pd versus Ir and Ni

versus Ir. Most of the samples from the Huangshannan intrusion have low Pd/Ir ratios

of <30 (Fig. 10) that are within the range of typical magmatic sulfide deposits (see

summary in Naldrett, 2011). Positive correlations of Pd versus Ir (Fig. 10) and Ni

versus Ir (Fig. 11) exist in these samples. The distributions of sulfides and PGE in the

intrusion are generally consistent with normal primary magmatic controls. The sulfide

ores show excellent positive correlations between Ir and Os, Ru or Rh, and poor

correlations between Ir and Pt or Pd (Fig. 10), which is consistent with the results of

the incomplete crystallization of monosulfide solid solution (mss).

High field-strength elements (HFSE) and REE may be mobilized during alteration

(Pearce et al., 1992) and mineralization. Zirconium has been considered to be one of

the least mobile elements and thus gives an alteration-independent index of

geochemical variation. The variation diagram shows a main axis of dispersion for all

elements that increases with rising Zr content and that is consistent with the

crystallization of a major phase (Fig. 12). The correlations of REE (represented by La,

Sm and Lu) and HFSE (such as Nd, Yb and Th) with Zr suggest that these elements

were immobile during the alteration and mineralization process in the Huangshannan

intrusion, which is consistent with the results of the PGE analyses.

7.2. Parental and primary magmas

Previous studies have shown that the Ir, Ru, Os and Ni are more compatible in the

mantle than the Rh, Pt, Pd and Cu during partial melting of the mantle (Barnes et al.,

1985; Barnes and Lightfoot, 2005). Thus, magma formed by a high degree of partial

melting of the mantle would result in relatively high Ni/Cu and low Pd/Ir ratios. Most

sulfide-mineralized samples from the Huangshannan intrusion have moderate Ni/Cu

ratios of 1.2−15.9 and Pd/Ir ratios of 2.3−29. In a plot of Pd/Ir versus Ni/Cu, most of

the samples from the Huangshannan intrusion plotted within the field of high Mg

basalts except for one in the komatiites field (Fig. 13), indicating that the primary

magmas of the intrusion were likely high Mg basaltic in composition.

The mantle beneath the Eastern Tianshan region was modified by slab-derived

melts and fluids due to oceanic subduction before the Late Permian (Gao et al., 2013;

Mao et al., 2008; Su et al., 2012b). The parental magma of the Huangshannan

intrusion with its high Mg content may indicate a metasomatized and hydrous mantle

source after modification by the subducted slab (e.g., Zhou et al., 2004; Zhang et al.,

2011; Gao and Zhou, 2013), which is consistent with the magma being emplaced in a

post-collisional setting at 282.5 ±1.4 Ma.

In the diagram of Th/Yb versus Nb/Yb, the Huangshannan intrusive rocks differ

from those of Permian mafic-ultramafic volcanic rocks and associated ultramafic

dikes in the Tarim Craton, and most plot within the field of global volcanic arc basalts

(Fig. 14). This result can be explained by having the parental magma of

Huangshannan intrusion being sourced from a depleted mantle-derived magma plus

crustal materials rather than from mantle plume-related activity.

Compared to the important Ni-Cu (-PGE) deposits in the Eurasian plate, including

in the Siberian large igneous provinces and the CAOB (Fig. 15), the sulfide ores of

the Huangshannan intrusion (recalculated to 100% sulfide) have high Ni (average =

17.8 wt %) and PGE contents (average of 334.79 ppb) and thus high Ni/Cu ratios

(average of 5.04), indicating high Ni and PGE contents and low Cu content in the

parental magma. The Ni content and Ni/Cu ratio will decrease during fractional

crystallization of olivine and pyroxenes from a basaltic magma because Ni is

compatible whereas Cu is relatively incompatible in this process (Wei et al., 2013;

Barnes and Lightfoot, 2005). Thus the Ni tenor (in 100% sulfides) of the ores from

the Huangshannan intrusion would be higher without olivine and pyroxene fractional

crystallization at depth. Therefore, we propose that the primary magma of the

Huangshannan intrusion was sourced from a Ni-rich magma with a high degree of

mantle partial melting.

7.3. Possible factors controlling the formation of PGE-depleted sulfide ores

A key factor in forming magmatic Ni-Cu (-PGE) sulfide deposits is that the

primary magmas from the mantle must be efficiently transported to surface with a

minimum of olivine fractionation and sulfide segregation at depth (Barnes and

Lightfoot, 2005). In practice, primary magmas approaching the crust can be

considered to be in equilibrium (batch) melting (Naldrett, 2010). The minimum

degree of partial melting to entirely exhaust the sulfide present in the mantle source

by dissolving it into the silicate magma is thought to be about 11% (Naldrett, 2010) or

20 to 40 % (Barnes et al., 1985; Keays, 1995), depending on the sulfur contents of the

mantle and the magmas from partial melting (Naldrett, 2010; Lightfoot et al., 2012).

Given that olivine makes up a much greater proportion of the mantle than sulfide,

the main phase controlling Ni is olivine rather than sulfide. High degrees of partial

melting result in high Ni contents in the magma. From the discussion in the previous

section, the Huangshannan intrusion originated from a Ni-rich magma. Therefore, the

degree of partial melting required to produce high Mg basalt with so high Ni content

can be speculated to be higher than 11%. In contrast to Ni, Cu and PGE do not prefer

to be partitioned into olivine. Although the phases controlling the PGE content remain

open to debate (Barnes and Lightfoot, 2005), the sulfide content is clearly the key

factor controlling the Cu and PGE concentrations in the primary magma.

The mantle under the Eastern Tianshan has been modified by slab-derived melts

and fluids (Su et al., 2012). The addition of melts and fluids would result in a high

oxygen fugacity in the primary magma from this mantle source and would have high

sulfur solubility (cf. Parkinson and Arculus, 1999; Gao et al., 2012). Therefore, the

primary magma of the Huangshannan intrusion should not be PGE depleted relative to

Cu and Ni.

All of the silicate rocks and sulfide ores from the Huangshannan intrusion are

strongly depleted in Os, Ir, Ru and Rh relative to Cu and Ni (Fig. 9). There may be

three possible factors that account for the PGE depletion in the parental magma of the

intrusion: (1) a PGE-depleted source mantle; (2) a low degree of partial melting of

PGE-undepleted mantle; or (3) previous sulfide segregation prior to the parental

magma emplacement. From the discussion above, the Huangshannan intrusion was a

Ni-rich magma that resulted from a high degree of mantle partial melting. Thus, the

possibility of a low degree of partial melting of a PGE-undepleted mantle can be ruled

out. There are two types of sulfides in the mantle, i.e. Cu-rich intermediate solid

solution (iss) and Fe-rich mss (Barnes et al. 2001; Brockrath et al., 2004; Peregoedova

et al., 2004). Osmium, Ir, Ru and Pd are strongly partitioned into mss and iss,

respectively, whereas Pt is assumed into the Fe-Pt alloys. In a low degree of partial

melting, the Pt-alloy and mss retain Os, Ir, Ru and Pt in the mantle. However, the Cu-

Pd sulfide droplets go preferentially into the silicate melt and are then carried into the

crust. This magma will have depleted Os, Ir and Ru and enriched Pd. In a high degree

of partial melting, all of the sulfur is dissolved into the silicate magma and thus the

primary magma would entrain all of the PGEs. This type of silicate magma would

have a nearly flat, primitive mantle-normalized chalcophile element pattern. In

addition, the available analyses for the oxide-rich sulfide ores from the

Huangshandong deposit show no obvious PGE depletion (Gao et al., 2013).

Therefore, the possibility of a PGE-depleted source mantle can be abandoned. To sum

up, the parental magma forming the Huangshannan intrusion has undergone sulfide

removal at staging chamber in its early stage of magma evolution.

7.4. Variable amounts of early sulfide removal

During sulfide segregation from silicate magmas, Cu and Pd behave completely

differently because Pd has a much higher partition coefficient into sulfide than Cu.

Magmas with Cu/Pd ratios higher than the mantle value of about 7000 (Barnes and

Maier, 1999) may reflect the removal of sulfide (Gao et al., 2012; Song et al., 2011).

Both the host rock and sulfide ores of the Huangshannan intrusion have much higher

Cu/Pd ratios (16705−384348) than the mantle value (Fig. 11), implying the possible

removal of sulfide from its magma before intrusion emplacement.

The primary magma of the Huangshannan intrusion was a high Mg basaltic

magma. We suggest that this magma would have a PGE-undepleted composition with

10 ppb Pd and 600 ppm Ni, which are within the range of data from the typical

undepleted picritic basalts of Qeqertarssuaq, West Greenland (80−1400 ppm Ni and

4.2−12.9 ppb Pd; Keays and Lightfoot, 2007). Partition coefficients of Pd and Ni

between a sulfide liquid and a silicate magma for typical mafic magmas are set at

20000 and 500 (e.g., Francis, 1990; Fleet et al., 1991, 1999), respectively. Using the

mass-balance equation of Campbell and Naldrett (1979), Pd and Ni tenors of the

sulfide ores found in websterite from the Huangshannan intrusion can be modeled

using a <0.016% removal line and an earlier 0.022% sulfide removal line with mass

ratio for silicate- to sulfide-melts (R factor; Campbell and Naldrett, 1979) of about

500. However, the sulfide ores found in lherzolite are plotted on the range of a

0.008% removal line and an earlier 0.016% sulfide removal line with an R factor of

about 1000 (Fig. 16). The modeling calculations indicate that the PGE depletion

relative to Ni and Cu can be attributed to the previous removal of small amounts of

sulfide liquid at staging chamber. To be more specific, the sulfide ores found in

websterite have undergone a greater degree of early sulfide segregation at relatively

low R factors than the sulfide ores that occur in lherzolite at relatively high R factors

(Fig. 16).

After earlier sulfide removal from the parental magma at staging chamber, the

PGE become depleted due to the high partition coefficient between sulfide liquid and

silicate magma, and then the parental magma ascends into a shallow conduit system.

The evolved magma enters the shallow conduit system and reaches sulfide saturation

due to crustal contamination and/or magma mixing as described later in section 7.7,

resulting in sulfide liquid segregation.

7.5. Crustal contamination and modification of the mantle source

The primitive mantle is characterized by an Nb/Th ratio of about 8 (Sun and

McDonough, 1989). By comparison, Archean and younger continental crust has

fractionated LREE with negative Nb, P and Ti anomalies (Taylor and McLennan,

1985; Rudnick and Gao, 2003). The Huangshannan intrusion has LREE enrichment

(Fig. 8a) and Nb, P and Ti depletion (Fig. 8b), which may indicate continental crustal

contamination.

The values of (Nb/Th) PM and (Th/Yb) PM indicate the magnitude of the Nb

anomaly and the crustal contamination, respectively (Wei et al., 2013). The (Nb/Th)

PM ratios of the Huangshannan intrusion range from 0.093 to 0.437, which are much

lower than the average value of N-MORB (2.3). In the plot of (Nb/Th) PM vs. (Th/Yb)

PM, the samples are very close to the mixing line between the fields for the upper crust

and for a depleted mantle-derived magma (Fig. 17). It is worth noting that single

samples of gabbro, diorite and sulfide ore show much more intense upper crustal

contamination with (Th/Yb) PM values from 15% to 20% and that there are no samples

that plot within the range of 10%-15%. This suggests that the parental magmas have

undergone different degrees of crustal contamination and thus may originate from

different magma pulses.

From the Th/Yb versus Nb/Yb diagram, the parental magma of the Huangshannan

intrusion may be generated from a depleted mantle-derived magma that was

influenced by crustal materials (see details in Fig. 14). However, this result cannot be

explained just by upper crustal contamination from relatively low Nb/Yb and high

Th/Yb ratios. Cerium and Pb as well as Ba and Nb have similar geochemical

characteristics during partial melting of a mantle source (Sun and McDonough, 1989).

However, these elements fractionate differently during subduction because Ce is less

mobile than Pb, and Nb is essentially immobile during slab dehydration (McCulloch

and Gamble, 1991). Therefore, the low Ce/Pb (0.32-3.6) and high Ba/Nb (3.95-

104.17) ratios of the Huangshannan intrusion are diagnostic of a metasomatized

mantle source after modification by subducted slab. Similar trace element

characteristics exist in the sulfide-bearing mafic-ultramafic intrusions of Eastern

Tianshan such as the Huangshandong (Sun et al., 2013), Huangshanxi (Zhang et al.,

2011) and Tianyu intrusions (Tang et al., 2011). The mixing between a subduction-

metasomatized mantle-derived magma and coeval A-type granites, plus upper crustal

materials, may be the best explanation for this phenomenon (e.g. Sun et al., 2013; Wei

et al., 2013; Tang et al., 2011).

7.6. The magma mixing of multiple magma pulse

Palladium has a much higher partition coefficient than either Cu or Ni between

sulfide and silicate magmas, and Cu has a slightly lower partition coefficient than Ni

in mafic melts. Thus, the removal of sulfide melts and fractional crystallization would

increase both the Ni/Cu and Cu/Pd ratios of the residual silicate magmas (Gao et al.,

2012). On the other hand, during mss fractionation from sulfide melts, Pd has a lower

partition coefficient than Cu and Cu is lower than Ni. Therefore this process would

decrease the Ni/Cu and Cu/Pd ratios in the residual sulfide melt (cf. Naldrett et al.,

1996; Gao et al., 2013). Thus, if the segregation and fractionation of the sulfide melts

took place in the same magma pulse, there would be a positive correlation between

the Ni/Cu and Cu/Pd ratios. However, the sulfide ores and silicate rocks from the

Huangshannan intrusion show a negative correlation between the Ni/Cu and Cu/Pd

ratios (Fig. 18). Sulfide mineralization with a high Cu tenor has relatively high Cu/Pd

ratios and Os, Ir and Ru contents, whereas the high Ni sulfide ores have relatively low

Cu/Pd ratios and Os, Ir and Ru concentrations, indicating the mixing of different

silicate magma pulses.

7.7. Genetic model for the Huangshannan deposit

The Huangshannan intrusion was formed in a post-collisional extensional

environment, and the mantle beneath this area was metasomatized by subducted slabs

(Zhou et al., 2004; Zhang et al., 2011; Gao and Zhou, 2013). This environment

generated a series of mantle-derived magmatic events and Cu-Ni sulfide deposits in

Eastern Tianshan.

At Huangshannan, the websterite is found in lherzolite (Fig. 3b), implying that the

websterite formed earlier than lherzolite. In addition, the disseminated sulfide ores

that occur in websterite and lherzolite show different PGE (Fig. 16) and trace element

features (Fig. 8 and Fig. 17). These differences cannot be the result of alteration and

fractional crystallization in-situ or at depth (see details in Sections 7.1, 7.4 and 7.6),

suggesting two different histories of magma evolution. Thus, we propose that the

Huangshannan magmatic sulfide deposit was generated from three different pulses of

magma.

The first pulse of magma underwent crustal contamination and fractional

crystallization at staging chamber, causing the magma to become sulfide-saturated

with 0.016% sulfide removal. The PGE became depleted relative to Cu and Ni in this

magma because the PGE have a much higher partition coefficient than Cu and Ni

between the sulfide liquid and silicate magma. The PGE-depleted magma then

ascended and entered the shallow conduit system and reached sulfide saturation again

because of the addition of crustal sulfide from the country rocks, resulting in large

amount of segregation in the sulfide melts to form the sulfide ores in lherzolite. The

second pulse of magma experienced 0.022% sulfide segregation at staging chamber

and then ascended up into the shallow magma chamber. After entering the conduit

system, the sulfide unsaturated magma was contaminated by the country rocks,

leading to the formation of sulfide ores in the websterite. The third pulse of magma

that emplaced in the eastern and western part of the Huangshannan intrusion and

formed the massive gabbro-diorite unit of the Huangshannan intrusion with no

obvious mineralization.

8. Conclusions

(1) Zircon SIMS U-Pb age dating of the Huangshannan intrusion yielded a weighted

206Pb/

238U age of 282.5 ± 1.4 Ma (N=9, MSWD=0.25), which is coeval with the

ages of other mineralized mafic-ultramafic intrusions in the Eastern Tianshan

region. Major and trace element geochemistry of the Huangshannan intrusion

showed that its primary magmas were not the result of mantle plume activity.

(2) Geological conditions and our PGE modeling imply that the Huangshannan

deposit was formed from three pulses of magma. The first magmatic pulse

underwent the removal of 0.016% sulfide in the deep magma chamber and then

ascended into the shallow conduit system to form the sulfide ores in lherzolite.

The second magmatic pulse, which had experienced 0.022% sulfide segregation at

depth, formed the sulfide ores in websterite. The third pulse of magma formed the

massive gabbro-diorite unit with no obvious mineralization. Also the different

pulses of magma underwent different degrees of crustal contamination.

(3) The higher PGE contents in the sulfide ores than in the whole rocks indicate that

the sulfide liquids were the primary collectors of the PGE in the Huangshannan

deposit.

Acknowledgement

We thank JiaquanYang of the Hami Great Wall Industry Co., Ltd for sharing his

viewpoint about the local geology and his assistance in our field work. The

manuscript benefitted from the enthusiastic help of Benxun Su, Bo Wei, Stephen

Barnes, Runmin Peng, Wei Xie and Liemeng Chen. We especially thank Jianfeng

Gao of the University of Hong Kong for a helpful scientific review of an earlier

version of this manuscript. The authors extend their profound gratitude to Prof.

Xianhua Li, Xiaoxiao Ling and Jiao Li for their assistance in doing the zircon age

analyses at the Institute of Geology and Geophysics, Chinese Academy of Sciences.

Constructive comments from Prof. David Symons, J.G. Liou and an anonymous

reviewer are greatly appreciated. This study was financially supported by the National

Natural Science Foundation of China (U1303292), the National Science and

Technology Support Program of China (No. 2011BAB06B02), the Chinese

Geological Survey Program (No.121211220926) and a research grant from the China

Geological Survey (12120113089400).

References

Asif , M., Parry, S.J., 1991. Study of the digestion of chromite during nickel sulphide

fire assay for the platinum group elements and gold. Analyst 116, 1071 –1073.

Begg, G.C., Hronsky, J.A.M., Arndt, N.T., Griffin, W.L., O’Reilly, S.Y., Hayward,

N., 2010. Lithospheric, cratonic and geodynamic setting of Ni−Cu−PGE sulfide

deposits. Economic Geology 105, 1057−1070.

Barnes, S.J., and Lightfoot, P.C., 2005. Formation of magmatic nickel sulfide ore

deposits and processes affecting their copper and platinum group element contents.

Economic Geology, 100th Anniversary Volume, 179−213.

Barnes, S.J., and Maier, W.D., 1999. The fractionation of Ni, Cu and the noble metals

in silicate and sulfide liquids: Geological Association of Canada Short Course

Notes 13, 69−106.

Barnes, S.J., Acterberg, E., Makovicky, E., Li, C., 2001. Proton probe results for

partitioning of platinum group elements between monosulphide solid solution and

sulphide liquid. South African Journal of Geology 104, 337–351.

Brockrath, C., Ballhaus, C., Holzheid, A., 2004. Fractionation of platinum group

elements during partial melting of the mantle. Science 305, 1951–1953.

Clarke, B., Uken, R., Reinhardt, J., 2009. Structural and compositional constraints on

the emplacement of the Bushveld Complex, South Africa. Lithos 111, 21–36.

Fleet, M.E., Stone, W.E., Crocket, J.H., 1991. Partitioning of palladium, iridium and

platinum between sulfide liquid and basalt melt: Effects of melt composition,

concentration and oxygen fugacity. Geochimica et Cosmochimica Acta 55,

2545−2554.

Fleet, M.E., Crocket, J.H., Liu, M.H., Stone, W.E., 1999. Laboratory partitioning of

platinum−group elements (PGE) and gold with application to magmatic

sulfide−PGE deposits. Lithos 47, 127−142.

Francis, R.D., 1990. Sulfide globules in mid−ocean ridge basalts (MORB), and the

effect of oxygen abundance in Fe−S−O liquids on the ability of those liquids to

partition metals from MORB and komatiite magmas. Chemical Geology 85,

199−213.

Gao, J.F., and Zhou, M.F., 2013, Generation and evolution of siliceous high-

magnesium basaltic magmas in the formation of the Permian Huangshandong

intrusion (Xinjiang, NW China). Lithos 162−163, 128−139.

Gao, J.F., Zhou, M.F., Lightfoot, P.C., Wang, C.Y., Qi, L., Sun, M., 2013. Sulfide

saturation and magma emplacement in the formation of the Permian

Huangshandong Ni−Cu sulfide deposit, Xinjiang, NW China. Economic Geology

108, 1833–1848.

Gao, J.F., Zhou, M.F., Lightfoot, P.C., Wang, C.Y., Qi, L., 2012. Origin of PGE−poor

and Cu−rich magmatic sulfides from the Kalatongke deposit, Xinjiang, Northwest

China. Economic Geology 107, 481−506.

Jahn, B.M., Windley, B., Natal'in, B., Dobretsov, N., 2004. Phanerozoic continental

growth in Central Asia. Journal of Asian Earth Sciences 23, 599–603.

Keays, R.R., 1982. Palladium and iridium in komatiites and associated rocks:

Application to petrogenetic problems, in Arndt N.T., and Nisbet E.G., eds.,

Komatiites: London, George Allen and Unwin, 435−455.

Keays, R.R., 1995. The role of komatiitic and picritic magmatism and S−saturation in

the formation of the ore deposits. Lithos 34, 1–18.

Lehmann, J., Arndt, N., Windley, B., Zhou, M.F., Wang, C.Y., Harris, C., 2007. Field

relationships and geochemical constraints on the emplacement of the Jinchuan

intrusion and its Ni−Cu−PGE sulfide deposit, Gansu, China. Economic Geology

102, 75−94.

Li, C., Maier, W.D., and Waal, S.A., 2001. Magmatic Ni−Cu versus PGE deposits:

Contrasting genetic controls and exploration implication. South Africa Journal of

Geology 104, 205−214.

Li, J.Y., Wang, K.Z., Li, W.Q., Guo, H.C., Song, B., Wang, Y., Mo, S.G., Zhao, Z.R.,

Zhu, Z.X., Pan, C.Z., 2003. Tectonic evlution since the late Paleozoic and mineral

prospecting in eastern Tianshan Mountains, NW China. Xinjiang Geology 20,

295– 301 (in Chinese with English abstract).

Li, X.H., Liu, Y., Li, Q.L., Guo, C.H., Chamberlain, K.R., 2009. Precise

determination of Phanerozoic zircon Pb/Pb age by multi−collector SIMS without

external standardization. Geochemistry, Geophysics, Geosystems 10, Q04010.

http://dx.doi.org/10.1029/2009GC002400.

Ludwig, K.R., 2003. ISOPLOT 3: A geochronological toolkit for Microsoft Excel.

Berkeley Geochronology Centre Special Publication no.4, pp1-74.

Maier, W.D., Arndt, N.T., Curl, E.A., 2000. Progressive crustal contamination of the

Bushveld Complex: evidence from Nd isotopic analysis of the cumulate rocks.

Contributions to Mineralogy and Petrology 140, 316–327.

Mao, J., Yang, J., Qu, W., Du, A., Wang, Z., Han, C., 2003. Re−Os age of Cu−Ni

ores from the Huangshandong Cu−Ni sulfide deposit in the East Tianshan

Mountains and its implication for geodynamic processes. Acta Geologica Sinica

77, 220–226 (in Chinese with English abstract).

Mao, J.W., Pirajno, F., Zhang, Z.H., Chai, F.M., Wu, H., Chen, S.P., Cheng, S.L.,

Yang, J.M., Zhang, C.Q., 2008. A review of the Cu−Ni sulfide deposits in the

Chinese Tianshan and Altay orogens (Xinjiang Autonomous Region, NW China):

principal characteristics and ore−forming processes. Journal of Asian Earth

Sciences 32, 184−203.

McCulloch, M.T., Gamble, A.J., 1991. Geoch emical and geodynamical constraints

on subduction zone magmatism. Earth and Planetary Science Letters 102, 358–

374.

Molnar F., Watkinson D.H., Jones P.C., 1997. Fluid inclusion evidence for

hydrothermal enrichment of magmatic ore at the contact zone of the

Ni−Cu−Platinum−group element deposit, Lindsley mine, Sudbury, Canada.

Economic Geology 92, 674−685.

Molnar F., Watkinson D.H., Jones P.C., 2001. Multiple hydrothermal processes in

footwall units of the North Range, Sudbury igneous complex, Canada, and

implications for the genesis of vein−type Cu−Ni−PGE deposits. Economic

Geology 96, 1645−1670.

Malaviarachchi, S.P.K., 2011. U–Pb ages and Hf−O isotopes of zircons from Late

Paleozoic mafic–ultramafic units in the southern Central Asian Orogenic Belt:

Tectonic implications and evidence for an Early−Permian mantle plume.

Gondwana Research 20, 516−531.

McDonough, W.F., Sun, S.S., 1995. The composition of the Earth: Chemical Geology

120, 223−253.

Naldrett, A.J., 2009, Fundamentals of magmatic sulfide deposits, in Li, C., and

Ripley, E.M., eds., New Developments in Magmatic Ni−Cu and PGE Deposits.

Geological Publishing House, Beijing, pp. 1−26.

Naldrett, A.J., Duke, J.M, 1980. Platinum metals in magmatic sulfide ores. Science

208, 1417−1428.

Naldrett, A.J., Lightfoot, P.C., Fedorenko, V., Doherty, W., Gorbachev, N.S., 1992.

Geology and geochemistry of intrusions and flood basalts of the Noril'sk region,

USSR, with implications for the origin of the Ni–Cu ores. Economic Geology 87,

975−1004.

Naldrett, A.J., 2010. From the mantle to the bank: the life of a Ni−Cu−(PGE) sulfide

deposit. South African Journal of Geology 113, 1−32.

Naldrett, A.J., Fedorenko, V.A., Asif, M., Lin, S., Kunilov, V.E., Stekhin, A.I.,

Lightfoot, P.C., Gorbachev, N.S., 1996. Controls on the composition of Ni−Cu

sulfide deposits as illustrated by those at Noril’sk, Siberia. Economic Geology 91,

751−773.

Naldrett, A.J., 2011. Fundamentals of magmatic sulfide deposits: Society of

Economic Geology Special Publication 17, 1−26.

Parkinson, I.J., Arculus, R.J., 1999. The redox state of subduction zones: Insights

from arc−peridotites. Chemical Geology 160, 409−423.

Pearce, J.A., Thirlwall, M.F., Ingram G., Murton, B.J., Arculus, R.J., Vanderlaan,

S.R., 1992. Isotopic evidence for the origin of boninites and related rocks drilled in

the Izu−Bonin (Ogasawara) forearc, Leg 125. In: Fryer P, Pearce J A and Stokking

L (eds.). Proceedings of the Ocean Drilling Program Scientific Results 125,

237−261.

Peregoedova, A., Barnes, S. J., Baker, D.R., 2004. The formation of the Pt−Ir alloys

and Cu−Pd rich sulfide melts by partial desulfurization of Fe−Ni−Cu sulfides:

Results of experiments and implications for natural systems. Chemical Geology

208, 247–264.

Pirajno, F., Mao, J.W., Zhang, Z.C., Zhang, Z.H., Chai, F.M., 2008. The association

of mafic–ultramafic intrusions and A−type magmatism in the Tianshan and Altay

orogens, NW China: Implications for geodynamic evolution and potential for the

discovery of new ore deposits. Journal of Asian Earth Sciences 32, 165–183.

Qin, K.Z., Su, B.X., Sakyi, P.A., Tang, D.M., Li, X.H., Sun, H., Xiao, Q.H., Liu, P.P.,

2011. SIMS zircon U–Pb geochronology and Sr–Nd isotopes of Ni–Cu−bearing

mafic–ultramafic intrusions in Eastern Thanshan and Beishan in correlation with

flood basalts in the Tarim Basin (NW China): Constraints on a ca. 280 Ma mantle

plume. American Journal of Science 311, 237−260.

Rudnick, R.L., and Gao, S., 2003. Composition of the continental crust. Treatise on

Geochemistry 3, 1–64.

Sengör, A.M.C., Natal'in, B.A., Burtman, V.S., 1993. Evolution of the Altaid tectonic

collage and Paleozoic crustal growth in Eurasia. Nature 364, 299–307.

Stacey, J.S., Kramers, J.D., 1975. Approximation of terrestrial lead isotope evolution

by a two−stage model. Earth and Planetary Science Letters 26, 207–221.

Su, B.X., Qin, K.Z., Sun, H., Tang, D.M., Sakyi, P.A., Chu, Z.Y., Liu, P.P., Xiao,

Q.H., 2012. Subduction−induced mantle heterogeneity beneath Eastern Tianshan

and Beishan: Insights from Nd–Sr–Hf–O isotopic mapping of Late Paleozoic

mafic–ultramafic complexes. Lithos 134–135, 41–51.

Su, B.X, Qin K.Z, Tang D.M, Sakyi, P.A, Liu P.P, Sun H, Xiao Q.H., 2013. Late

Paleozoic mafic–ultramafic intrusions in southern Central Asian Orogenic Belt

(NW China): Insight into magmatic Ni–Cu sulfide mineralization in orogenic

setting. Ore Geology Reviews 51, 57−73.

Sun, S. S., McDonough, W.F., 1989. Chemical and isotopic systematics in ocean

basalt: Implication for mantle composition and processes. Geological Society

Special Publication, pp. 313–345.

Sun T., Qian Z. Z., Deng Y. F., Li C. S., Song X. Y., Tang Q. Y., 2013. PGE and

Isotope (Hf−Sr−Nd−Pb) Constraints on the origin of the Huangshandong

magmatic Ni−Cu sulfide deposit in the Central Asian Orogenic Belt, Northwestern

China. Economic Geology 108, 1849−1864.

Song, X.Y., Keays, R.R., Zhou, M.F., Qi, L., Ihlenfeld, C., Xiao, J.F., 2009.

Siderophile and chalcophile elemental constraints on the origin of the Jinchuan

Ni−Cu−(PGE) sulfide deposit, NW China. Geochimica et Cosmochimica Acta 73,

404−424.

Song, X.Y., Wang Y.S, Chen L.M., 2011. Magmatic Ni−Cu−(PGE) deposits in

magma plumbing systems: Features, formation and exploration. Geoscience

Frontiers 2, 375−384.

Taylor, S.R., McLennan., S.M., 1985. The Continental Crust: Its Composition and

Evolution. Blackwell Scientific Publications, Oxford, pp. 1−312.

Tang, Z.L., and Li, W.Y., 1995. Mineralization model and geology of the Jinchuan

deposit bearing PGE. Geological Publishing House, Beijing, pp. 1−208 (in

Chinese).

Tang, Z.L., Bai, Y.L., Li, Z.L., 2002. Geotectonic setting of large and super large

mineral deposits on the southwestern margin of the north China Plate. Acta

Geologica Sinica 76, 367–377 (in Chinese with English abstract).

Tang, D.M., Qin, K.Z., Li, C.S., Qi, L., Su, B.X., Qu, W.J., 2011. Zircon dating,

Hf−Sr−Nd−Os isotopes and PGE geochemistry of the Tianyu sulfide−bearing

mafic−ultramafic intrusion in the Central Asian orogenic belt, NW China. Lithos

126, 84–98.

Tian, W., Campbell, I.H., Allen, C.M., Guan, P., Pan, W., Chen, M., Yu, H., Zhu, W.,

2010. The Tarim picrite−basalt−rhyolite suite, a Permian flood basalt from

northwest China with contrasting rhyolites produced by fractional crystallization

and anatexis. Contributions to Mineralogy and Petrology 160, 407–425.

Wang, R.M., Liu, D.Q., Yin, D.T., 1987. The conditions of controlling metallogny of

Cu, Ni sulphide ore deposits and the orientation of finding ore Hami, Xinjiang,

China. Minerals and Rocks 1, 115–121 (in Chinese).

Wang, Y.W., Wang, J.B., Wang, L.J., 2006. Comparison of host rocks between two

vanadic titianomagnetite deposit types from the eastern Tian−shan mountains.

Acta Petrologica Sinica 22, 1425−1436 (in Chinese with English abstract).

Wei B., Wang, C.Y., Li C.S., Sun Y.L., 2013. Origin of PGE−depleted Ni−Cu sulfide

mineralization in the Triassic Hongqiling No. 7 orthopyroxenite intrusion, Central

Asian Orogenic Belt, Northeastern China. Economic Geology 108, 1813−1831.

Windley, B.F., Kroner, A., Guo, J.H., Qu, G.S., Li, Y.Y., Zhang, C., 2002.

Neoproterozoic to Paleozoic geology of the Altai orogen, NW China: New zircon

age data and tectonic evolution. Journal of Geology 110, 719−737.

Whitehouse, M.J., Kamber, B.S., 2005. Assigning dates to thin gneissic veins in high

grade metamorphic terranes: a cautionary tale from Akilia, Southwest Greenland.

Journal of Petrology 46, 291–318.

Xiao, Q.H., Qin, K.Z., Tang, D.M., Sun, B.X., Sun, H., San, J.Z., Cao, M.J., Hui,

W.D., 2010. Xiangshanxi composite Cu–Ni–Ti–Fe deposit belongs to comagmatic

evolution product: evidence from ore microscopy, zircon U–Pb chronology and

petrological geochemistry, Hami, Xinjiang, NW China. Acta Petrologica Sinica

26, 503–522 (in Chinese with English abstract).

Xiao, W.J., Windley, B.F., Yuan, C., Sun, M., Han, C.M., Lin, S.F., Chen, H.L., Yan,

Q.R., Liu, D.Y., Qin, K.Z., Li, J.L., Sun, S., 2009. Paleozoic multiple subduction–

accretion processes of the southern Altaids. American Journal of Science 309,

221–270.

Xie W., Song X.Y., Chen L.M., Deng Y.F., Zheng W.Q., Wang Y.S., Ba D.H., Yin

M.H., Luan Y., 2013. Geochemistry insights on the genesis of the

subduction−related Heishan magmatic Ni−Cu−(PGE) deposit in Gansu, NW

China, at the southern margin of the Central Asian Orogenic Belt. Economic

Geology 109, 1563−1583.

Yuan, F., Zhou, T.F., Zhang, D., Fan, Y., Jowitt. S.J., Keays, R.R., Liu, S., Fan, Y.,

2012. Siderophile and chalcophile metal variations in basalts: Implications for the

sulfide saturation history and Ni−Cu−PGE mineralization potential of the Tarim

continental flood basalt province, Xinjiang Province, China. Ore Geology Reviews

45, 5–15.

Zhang, L.C., Qin, K.Z., Ying, J.F., Xia, B., Shu, J.S., 2004. The relationship between

oreforming processes and adakitic rock in Tuwu−Yandong porphyry copper

metallogenic belt, eastern Tianshan mountains. Acta Petrologica Sinica 20, 259–

268 (in Chinese with English abstract).

Zhang, M.J., Li, C.S., Fu, P. A.E., Hu, P. Q., Ripley, E.M., 2011. The Permian

Huangshanxi Cu−Ni deposit in western China: Intrusive−extrusive association, ore

genesis, and exploration implications. Mineralium Deposita 46, 153−170.

Zhou, M.F., Lesher, C.M., Yang, Z.X., Li, J.W., Sun, M., 2004. Geochemistry and

petrogenesis of 270 Ma Ni−Cu−(PGE) sulfide−bearing mafic intrusions in the

Huangshan district, eastern Xinjiang, northwest China: Implications for the

tectonic evolution of the Central Asian Orogenic Belt. Chemical Geology 209,

233−257.

Zhou, M.F., Zhao, J.H., Jiang, C.Y., Gao, J.F., Wang, W., Yang, S.H., 2009.

OIB−like, heterogeneous mantle sources of Permian basaltic magmatism in the

western Tarim basin, NW China: Implications for a possible Permian large

igneous province. Lithos 113, 583–594.

Figure Captions

Fig. 1. (a). Simplified tectonic units of Asia (modified from Jahn, 2004). (b). The

distribution of magmatic Cu-Ni sulfide deposits associated with mafic-ultramafic

intrusions in northern Xinjiang (modified from Sun, 2013).

Fig. 2. Simplified geological map of Huangshan area and the distribution of magmatic

Ni-Cu (-PGE) deposits. (Modified from Wang et al., 2006 and Gao et al., 2013).

Fig. 3. (a). Simplified geological map of the Huangshannan sulfide-bearing ultramafic

intrusion. (b). A-A’ prospecting line of the Huangshannan deposit (modified after

Inner Mongolia Mineral Experimental Research Institute and Wang et al., 1987).

Fig. 4. Microphotographs of mineral assemblages and sulfide textures in the

Huangshannan intrusion. (a). Sulfide assemblages occurring in the interstitial spaces

of olivine + pyroxene cumulates and olivine crystals enclosed in large orthopyroxene

crystals in lherzolite. (b). Sulfide assemblages occurring in the interstitial spaces of

orthopyroxene + clinopyroxene cumulates and olivine crystals enclosed in large

pyroxene crystals in websterite. (c). Intergranular texture of gabbro in the layered

sequence. (d). Intergranular texture of hornblende gabbro in the massive unit. (e). The

relationship of sulfide assemblages in the interstitial sulfides. (f). The relationship of

sulfide assemblages in the interstitial sulfides. Abbreviations: Amp = amphibole, Ol =

olivine, Opx = orthopyroxene, Cpx= clinopyroxene, Pl = plagioclase, Bt=biotite, Cpy

= chalcopyrite, Po = pyrrhotite, Pn = pentlandite, Sulf = sulfide

Fig. 5. Cathodoluminescence images of representative zircons from gabbros in the

Huangshannan mafic-ultramafic complex. The ellipses indicate the analyzed spots for

U–Pb isotopic analyses.

Fig. 6. U–Pb concordia plot (1σ error) for zircons from gabbro of the Huangshannan

mafic–ultramafic complex.

Fig. 7. Plots of SiO2 (a), Al2O3 (b), CaO (c), TiO2 (d), (Na2O + K2O) (e), and FeO

total (f) vs. MgO in LOI- and sulfide-free whole rocks in the Huangshannan intrusion.

Fig. 8. Chondrite-normalized REE (a) and mantle-normalized immobile trace element

patterns (b) for the Huangshannan intrusion. The normalization values are from Sun

and McDonough (1989). Data for the Permian Tarim basalts (PTB) are from Zhou et

al. (2009) and Yuan et al. (2012).

Fig. 9. Primitive mantle-normalized chalcophile element patterns for the silicate rocks

and sulfide ores from the Huangshannan intrusion. PGE are represented in whole-rock

compositions. Normalization values are from McDonough and Sun (1995) and Barnes

and Maier (1999).

Fig. 10. Plots of Pt vs. Pd, and Ir vs. Pd, Os, Ru and Rh for the Huangshannan

intrusion.

Fig. 11. Plot of Ni vs. Ir and Cu vs. Pd for the Huangshannan intrusion.

Fig. 12. Plots of Zr vs. La, Sm, Lu, Nd, Yb and Th to test for the effects of alteration

and mineralization in the Huangshannan intrusion.

Fig. 13. Plot of Pd/Ir vs. Ni/Cu for sulfide mineralization from the Huangshannan

intrusion. The fields were defined by Barnes and Lightfoot (2005)

Fig. 14. Plot of whole-rock Th/Yb vs. Nb/Yb for the Huangshannan mafic-ultramafic

intrusion. The data for the upper, middle, and lower crusts (UC, MC, LC) are from

Rudnick and Gao (2003); the primitive mantle (PM) values are from McDonough and

Sun(1995); the data of OIB, N-MORB, and E-MORB are from Sun and McDonough

(1989); the data for the Permian Tarim basalts (PTB) and associated ultramafic dikes

(PTUD) are from Zhou et al. (2009), Yuan et al. (2012) and Tian et al. (2010); and,

the data for global volcanic arc basalts (VAB) are from a public database

(http://www.petdb.org).

Fig. 15. The comparison of primitive mantle-normalized chalcophile element patterns

between the Huangshannan deposit and other important Ni-Cu (-PGE) deposits. The

abundances have been recalculated to 100% sulfide. Sources of data: Noril’sk,

Naldrett (2011); Jinchuan, Song et al. (2009); Huangshanxi, Zhang et al. (2010);

Kalatongke, Gao et al. (2012); Hongqiling, Wei et al. (2013); Huangshandong, Gao et

al. (2013); and Tianyu, Tang (2009). The normalization values for Ni and Cu are from

McDonough and Sun (1995) and for PGE are from Barnes and Maier (1999).

Fig. 16. Modeling of R factors with Ni and Pd tenors of sulfide mineralization. Model

1: Segregation of the sulfide melts from picritic basalt magma with 10 ppb Pd and 600

ppm Ni under various R values from 100 to 100000. Model 2: Segregation of sulfide

melts from magma with 2.02 ppb Pd and 576.52 ppm Ni under various R values.

Model 3: Sulfide liquids produced by magma with 0.41 ppb Pd and 553.95 ppm Ni

(for 0.016% sulfide removal from the magma of model 1 at different R factors).

Model 4: Sulfide liquids produced by magma with 0.12 ppb Pd and 537.61ppm Ni

(for 0.022% sulfide removal from the magma of model 1 at different R factors).

Fig. 17. Plots of (Nb/Th) PM vs. (Th/Yb) PM ratios of the Huangshannan intrusion.

Sources of data: N-MORB and OIB, Sun and McDonough (1989); the lower, middle

and upper crust, Rudnick and Gao (2004).

Fig. 18. Ni/Cu ratios vs. Cu/Pd ratios of sulfide mineralization and hosted silicate

rocks of the Huangshannan intrusion.

Table Captions

Table 1. SIMS U–Pb ages for zircons separated from gabbro in the Huangshannan

intrusion.

Appendix Captions

Table A1. Major and trace element compositions and chalcophile element contents

for samples from the Huangshannan intrusion.

Fig. 1.

Fig. 2.

Fig. 3.

Fig. 4.

Fig. 5.

Fig. 6.

Fig. 7.

Fig. 8.

Fig. 9.

Fig.10.

Fig. 11.

Fig. 12.

Fig. 13.

Fig. 14.

Fig. 15.

Fig. 16

Fig. 17.

Fig. 18.

Table 1.

Spot U(ppm) Th(ppm) Pb(ppm) Th/U 207Pb/235U ±σ% 206Pb/238U ±σ% 207Pb/206Pb ±σ% t207/206 ±σ% t207/235 ±σ% t206/238 ±σ%

Hsn-32@1 836 321 43 0.384 0.31800 1.91 0.0435 1.65 0.05300 0.96 328.6 21.6 280.4 4.7 274.6 4.4

Hsn-32@2 1217 724 66 0.595 0.31134 1.83 0.0436 1.63 0.05177 0.82 275.4 18.8 275.2 4.4 275.2 4.4

Hsn-32@3 1138 134 55 0.118 0.32263 2.19 0.0445 1.50 0.05254 1.59 309.1 35.8 283.9 5.4 280.9 4.1

Hsn-32@4 2589 2284 153 0.882 0.31368 5.65 0.0448 1.64 0.05083 5.40 233.3 120.1 277.0 13.8 282.2 4.5

Hsn-32@5 864 249 44 0.288 0.31317 2.56 0.0449 1.52 0.05057 2.05 221.1 46.8 276.6 6.2 283.2 4.2

Hsn-32@6 1169 750 66 0.642 0.32323 1.77 0.0451 1.53 0.05202 0.89 286.2 20.3 284.4 4.4 284.2 4.3

Hsn-32@7 1553 1051 89 0.677 0.32011 1.77 0.0451 1.51 0.05143 0.93 260.2 21.2 282.0 4.4 284.6 4.2

Hsn-32@8 1520 211 75 0.139 0.32396 2.23 0.0454 1.51 0.05174 1.64 273.9 37.1 284.9 5.6 286.3 4.2

Hsn-32@9 2467 2682 158 1.087 0.33098 1.65 0.0460 1.55 0.05214 0.57 291.4 13.0 290.3 4.2 290.2 4.4

Table

Table 1.

Spot U(ppm) Th(ppm) Pb(ppm)

Hsn-32@1 836 321 43

Hsn-32@2 1217 724 66

Hsn-32@3 1138 134 55

Hsn-32@4 2589 2284 153

Hsn-32@5 864 249 44

Hsn-32@6 1169 750 66

Hsn-32@7 1553 1051 89

Hsn-32@8 1520 211 75

Hsn-32@9 2467 2682 158

Table A1.

Rock type Diorite Gabbro Gabbro Diorite Hornblende

gabbro Iherzolite

Sample no. hsn-1 hsn-2 hsn-17 hsn32 2hsn-3 2hsn-3-1 hsn-12 hsn-6

Major oxides(wt%)

SiO2 56.06 52.49 51.75 58.52 53.9 54.13 37.36 45.63

Al2O3 18.29 17.56 19.84 14.28 16.81 17.23 2.13 3.56

CaO 6.72 7.04 10.2 7.07 8.99 9.32 0.78 1.8

Fe2O3 1.15 2.32 0.52 0.32 0.58 0.41 5.1 0.88

FeO 3.88 4.93 4.64 5.87 5.23 5.15 5.43 8.2

K2O 0.64 1.18 0.51 1.28 0.4 0.43 <0.01 0.05

MgO 5.68 6.76 7.41 8.6 8.98 8.66 36.6 33.45

MnO 0.06 0.06 0.08 0.12 0.12 0.11 0.13 0.14

Na2O 5.51 4.27 3.8 2.87 3.38 3.43 0.04 0.1

P2O5 <0.01 <0.01 <0.01 0.03 0.06 0.04 0.02 0.03

TiO2 0.42 0.36 0.84 0.52 0.52 0.49 0.13 0.16

LOI 1.2 1.9 0.79 0.86 0.53 0.46 11.25 5.08

Total 99.61 98.87 100.38 100.34 99.5 99.86 98.97 99.08

TFeo 4.915 7.018 5.108 6.158 5.752 5.519 10.02 8.992

Trace elements(ppm)

Li 28.9 43.1 43.3 59.5 6.04 10.7 6.8 5.59

Be 2.26 2.14 1.13 1.51 0.29 0.48 0.34 0.17

Cr 61.9 56.4 100 458 240 344 3714 3252

Mn 460 547 742 956 662 900 1076 1189

Co 121 159 39.3 41 29.5 41.9 147 127

Ni 6782 9210 510 244 128 182 3130 2741

Cu 2843 3579 342 139 14.7 18.6 198 340

Zn 35.1 37.7 46.4 64.5 37.5 49.4 86.2 67.1

Ga 15.7 15.2 17.6 15.6 10.4 15.2 3.23 4.28

Rb 29.4 63.7 24.5 41.6 5.12 7.42 0.53 0.82

Sr 809 661 801 340 285 414 24.4 23.6

Mo 0.38 0.37 0.24 0.55 0.16 0.18 0.48 0.34

Cs 2.85 4.81 6.39 4.08 0.71 0.68 0.26 2.41

Ba 140 236 92 244 67.6 100 2.46 3.48

Pb 59.5 56.4 37.6 8.15 1.62 1.93 2.17 2.3

Th 4.8 4.2 1.96 3.18 0.54 0.67 0.52 0.41

U 1.27 0.99 0.8 1.08 0.37 0.35 0.17 0.13

Nb 4.17 3.28 2.39 4.54 0.71 0.96 0.6 0.51

Ta 0.59 0.46 0.3 0.54 0.09 0.14 0.24 0.26

Zr 65.1 45.7 50.8 65.2 25.3 33.5 13.6 14.4

Hf 2.16 1.61 1.5 2.2 0.83 1.01 0.4 0.39

Ti 2831 2328 5616 3495 2428 3360 907 1035

W 6.24 3.52 1.67 0.06 <0.05 0.19 3.36 2.31

V 77.8 85.5 153 148 102 144 40.1 56.3

La 11.3 8.22 5.68 9.8 2.24 3.03 1.34 1.22

Ce 25.1 18.1 13.2 22.4 5.18 6.95 3.07 2.95

Pr 3.19 2.37 1.9 2.98 0.78 1.05 0.41 0.36

Nd 13.1 9.4 8.68 12.3 3.76 4.87 1.71 1.7

Sm 3.12 2.23 2.29 3.12 1.16 1.42 0.37 0.42

Eu 0.95 0.89 0.95 0.88 0.56 0.83 0.18 0.14

Gd 3.04 2.13 2.81 3.43 1.47 1.9 0.38 0.47

Tb 0.54 0.39 0.45 0.62 0.24 0.34 0.07 0.07

Dy 3.22 2.3 2.85 4.03 1.51 2.09 0.45 0.53

Ho 0.63 0.51 0.56 0.81 0.32 0.46 0.09 0.1

Er 1.91 1.53 1.66 2.55 0.93 1.39 0.33 0.34

Tm 0.25 0.2 0.21 0.32 0.14 0.18 <0.05 0.05

Yb 1.68 1.32 1.43 2.24 0.89 1.17 0.3 0.33

Lu 0.25 0.21 0.21 0.31 0.13 0.17 <0.05 0.06

Sc 17.7 19.2 26.2 26.2 18 25.4 7.11 13.8

Y 18 13.8 14.7 21.9 8.54 11.8 2.57 2.99

PGE(ppb)

Pt 1.16 9.36 2.67 0.95 1.07 1.35 29.8 6.91

Pd 9.52 12.7 1.72 0.57 0.88 0.67 11.8 4.63

Rh 0.52 1.19 0.07 <0.02 0.04 0.04 0.87 0.3

Ru 0.22 1.39 0.25 0.05 0.08 0.1 1.67 0.72

Ir 0.11 0.64 0.12 <0.02 0.03 0.04 0.73 0.36

Os 0.21 1.26 0.37 0.07 0.13 0.16 2.57 0.99

S(%) 1.12 1.99 0.058 0.077 0.006 0.005 0.18 0.21

Table A1. (Cont.)

Rock type Weakly desseminated sulfide ores Desseminated sulfide ores in Iherzolite

Sample no. hsn-5 hsn-11 hsn-8 hsn-9 hsn-3 hsn-4 hsn-7 hsn-10

Major oxides(wt%)

SiO2 50.43 52.91 48.22 41.32 37.93 38.12 37.84 39.21

Al2O3 3.61 7.72 8.05 4.15 4.08 4.1 3.4 3.76

CaO 3.8 5.46 10.04 2.3 2.32 2.31 2.15 2.56

Fe2O3 1.26 0.94 0.56 2.32 4.45 4.51 4.39 5.41

FeO 9.45 7.32 9.52 9.77 10.37 10.74 11.1 8.85

K2O 0.05 0.81 0.68 0.07 0.06 0.07 0.07 0.08

MgO 25.72 19.89 17.74 32.65 29.8 30.21 30.59 30.58

MnO 0.19 0.14 0.15 0.17 0.15 0.13 0.14 0.15

Na2O 0.2 1.39 1.24 0.17 0.15 0.15 0.13 0.14

P2O5 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01

TiO2 0.32 0.48 0.56 0.36 0.38 0.49 0.24 0.4

LOI 3.58 2.88 2.11 5.8 6.2 6.04 7.09 6.4

Total 98.61 99.94 98.87 99.08 95.89 96.87 97.14 97.54

TFeo 10.584 8.166 10.024 11.858 14.375 14.799 15.051 13.719

Trace elements(ppm)

Li 8.53 56.6 47.8 5.14 4.33 4.75 4.22 5.4

Be 0.2 0.68 1.68 0.26 0.25 0.29 0.21 0.2

Cr 2525 1675 1016 2680 2826 2800 3142 3367

Mn 1484 1140 1263 1203 1119 1099 1108 1107

Co 122 104 102 165 364 353 333 323

Ni 4476 2434 4591 5730 21380 19080 16490 17090

Cu 1166 1016 2773 2082 5371 3377 5100 7256

Zn 92.7 58.7 68 98.4 86.4 87.4 96.2 97.1

Ga 6.25 9.23 11.3 5.4 5.62 5.84 4.74 5.4

Rb 0.36 30.5 21.6 0.72 0.92 0.92 0.7 0.8

Sr 28.1 145 114 38.2 35 32.9 38.7 38.5

Mo 0.38 0.51 0.23 0.25 0.53 0.58 0.4 0.37

Cs 0.72 4.65 2.13 1.41 1.87 2.93 0.38 2.26

Ba 2.92 104 139 8.01 6.54 6.86 7.79 8.58

Pb 4.23 6.52 10.7 4.75 12.5 6.68 9.46 8.57

Th 0.59 1.44 1.53 0.51 0.61 0.54 0.68 0.27

U 0.16 0.56 0.79 0.16 0.19 0.2 0.18 0.08

Nb 0.74 1.3 3.82 1.29 1.18 1.41 0.96 0.99

Ta 0.16 0.22 0.59 0.29 0.2 0.25 0.2 0.19

Zr 18.8 37.8 53.5 20.6 21.2 17.5 21.1 15.5

Hf 0.63 1.16 1.97 0.77 0.79 0.72 0.73 0.59

Ti 2168 3204 3732 2381 2528 3225 1573 2575

W 0.52 1.56 1 1.58 1.82 1.4 1.21 0.91

V 150 165 189 63.2 70.9 76.6 62.3 79.9

La 1.66 2.9 9.24 1.9 2.07 1.75 2.04 1.27

Ce 4.1 7.55 22.2 5.27 5.3 5.04 4.82 3.87

Pr 0.66 1.15 2.84 0.85 0.76 0.84 0.8 0.66

Nd 3.13 5.49 12.1 4.22 3.97 4.09 3.64 3.52

Sm 1.05 1.64 3.1 1.16 1.13 1.26 0.82 1.2

Eu 0.29 0.51 1.01 0.37 0.31 0.38 0.26 0.28

Gd 1.21 2.03 3.58 1.38 1.45 1.62 0.99 1.25

Tb 0.22 0.32 0.59 0.22 0.22 0.26 0.16 0.22

Dy 1.28 2.14 3.73 1.41 1.33 1.51 0.94 1.28

Ho 0.29 0.46 0.75 0.3 0.28 0.33 0.21 0.26

Er 0.86 1.34 2.27 0.8 0.87 1.02 0.6 0.87

Tm 0.13 0.2 0.3 0.11 0.1 0.12 0.08 0.11

Yb 0.83 1.19 1.95 0.71 0.65 0.86 0.59 0.66

Lu 0.13 0.18 0.3 0.1 0.1 0.12 0.07 0.1

Sc 27.1 29.1 43.4 9.5 11.1 11.3 10 11.3

Y 7.26 12.3 19.8 7.52 7.35 8.41 5.52 7.34

PGE(ppb)

Pt 13.3 6.24 58 42.2 151 36.6 187 261

Pd 15.5 6.78 28.2 15.4 99.7 56.7 37.7 57.3

Rh 0.44 0.3 0.16 0.52 5.16 2.57 1.6 1.93

Ru 0.67 0.72 0.21 1.03 10.3 4.71 3 3.35

Ir 0.45 0.41 0.14 0.56 5.6 3 1.98 2.29

Os 1.12 1.41 1.28 1.62 15.1 7.04 3.37 4.59

S(%) 1.03 0.65 0.74 1.13 3.58 3.3 3.27 3.44

Table A1. (Cont.)

Rock type

Desseminated sulfide ores in Iherzolite Desseminated sulfide ores in pyroxenite

Sample no. hsn-13 hsn-14 hsn-20-1 hsn-20-2 hsn-16 hsn-21 2hsn-4-2 2hsn-4-3 2hsn-4-4

Major oxides(wt%)

SiO2 40.13 40.31 40.23 47.94 45.64 39.73 29.02

Al2O3 3.51 3.66 4.27 3.24 4.08 4.16 3.18

CaO 2.22 2.77 2.37 3.47 2.32 2.29 5.09

Fe2O3 2.7 4.99 3.19 4.36 3.12 4.06 16.3

FeO 10.47 9.26 10.05 8.21 9.55 9.39 8.65

K2O 0.08 0.07 0.07 0.06 0.07 0.07 0.03

MgO 31.93 30.77 30.5 25.72 27.87 30.67 17.8

MnO 0.15 0.13 0.17 0.15 0.15 0.14 0.07

Na2O 0.15 0.14 0.15 0.2 0.13 0.14 <0.05

P2O5 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01

TiO2 0.23 0.37 0.41 0.27 0.41 0.42 0.2

LOI 6.61 5.67 5.78 4.25 5.21 6.08 7.93

Total 98.18 98.14 97.19 97.87 98.55 97.15 88.27

TFeo 12.9 13.751 12.921 12.134 12.358 13.044 23.32

Trace elements(ppm)

Li 5.8 4.91 4.46 7.58 7.83 5.55 8.55

Be 0.3 0.29 0.28 0.37 0.2 0.28 0.17

Cr 3313 2421 2853 2487 2842 3182 1711

Mn 1188 980 1199 1207 1180 1208 576

Co 206 340 273 266 256 294 920

Ni 8700 17020 14740 13630 13770 15830 39420 38220 40020

Cu 7059 3576 3222 5373 3796 4767 2486

Zn 103 80.9 85.2 71.2 76.1 92.9 36

Ga 5.13 5.06 5.69 5.35 5.79 5.71 5.2

Rb 0.91 1.25 0.7 0.27 1.57 0.74 0.3

Sr 39.2 38.7 35.3 25.7 26.7 34.6 39.6

Mo 0.24 0.47 0.46 0.72 0.38 0.44 0.38

Cs 2.65 1.69 1.01 0.26 5.13 2.74 0.19

Ba 7.79 8.3 6.6 2.11 5.77 5.33 4.13

Pb 10.3 8.43 9.69 9.85 8.64 14.9 10.9

Th 0.44 0.38 0.65 0.66 0.46 0.61 1.62

U 0.15 0.11 0.19 0.2 0.17 0.23 0.35

Nb 0.97 0.97 1.31 0.82 0.94 1.14 1.43

Ta 0.2 0.2 0.24 0.15 0.18 0.21 0.22

Zr 17.7 17.6 21.2 16 26.9 27.5 33.3

Hf 0.62 0.66 0.64 0.57 0.8 0.9 0.96

Ti 1539 2393 2639 1776 2746 2499 1340

W 1.73 0.51 1.81 0.36 0.83 2.59 0.25

V 60.4 84.2 75.6 141 94.2 87.9 40

La 1.8 1.52 2.28 1.91 1.38 2.17 5.06

Ce 4.6 4.22 5.64 4.66 3.77 5.32 11

Pr 0.7 0.69 0.83 0.7 0.62 0.82 1.43

Nd 3.21 3.5 3.94 3.2 3.07 3.86 5.81

Sm 0.86 1.06 1.08 0.96 0.85 0.98 1.2

Eu 0.26 0.29 0.34 0.24 0.24 0.33 0.24

Gd 0.94 1.24 1.3 1.15 1.22 1.37 1.22

Tb 0.15 0.22 0.2 0.19 0.21 0.23 0.19

Dy 0.99 1.43 1.37 1.37 1.26 1.33 1.02

Ho 0.23 0.29 0.28 0.27 0.24 0.28 0.21

Er 0.66 0.82 0.85 0.83 0.85 0.86 0.61

Tm 0.09 0.11 0.12 0.12 0.11 0.11 0.09

Yb 0.52 0.73 0.73 0.78 0.74 0.73 0.58

Lu 0.09 0.1 0.1 0.11 0.11 0.11 0.08

Sc 9.68 13.3 12.5 27.2 16.7 12.3 6.67

Y 5.59 7.28 7.43 6.83 7.01 7.49 5.53

PGE(ppb)

Pt 16.3 31.9 92.5 44.2 14.7 44.4 1.52 1.02 1.57

Pd 24.8 61.3 67.4 44.3 45.3 65.8 28.5 32 29.4

Rh 0.84 2.9 3.35 1.66 1.48 3.01 1.95 2.21 2.05

Ru 1.85 5.02 6.95 3 2.59 5 1.82 2.69 1.91

Ir 1.15 3 3.34 1.87 1.73 2.75 2.14 2.01 2.02

Os 3.17 7.02 10.2 4.87 3.83 7.61 3.86 3.00 3.11

S(%) 2.07 3.23 2.33 3.16 2.28 2.61 12.66 10.25 13.42

Highlights

1. The Huangshannan mafic-untramafic intrusion was emplaced in 282.5 ±1.4 Ma and

not related to mantle plume activity.

2. The Huangshannan deposit was formed in three pulses of magma.

3. The multiple magmas may have undergone different evolution histories.