Ore forming process of magmatic nickel-copper-cobalt sulfide deposits in China

Abstract

The magmatic Ni-Cu-Co deposits are the main source of critical mineral resources in China, such as nickel, cobalt, and platinum group elements, which are formed generally in a continental rift setting or gravitational collapse at the end of an orogeny. Different from the deposits formed during the Archaean to Early Proterozoic in other countries, the magmatic Ni-Cu-Co deposits in China are of the Neoproterozoic to Paleozoic in age. The typical ones are represented by the world-class Neoproterozoic Jinchuan and the super-large Paleozoic Xiarihamu deposits, as well as many Early Permian to Early Triassic deposits in Tianshan-Beishan and Altai areas. As a product of mantle plume activity, the primitive magma (ca. 827 Ma) at Jinchuan was formed by the high degree of deep partial melting of asthenospheric mantle, resulting in the immiscibility between sulfide liquid phase and silicate melt. The extensive distributions of magmatic Ni-Cu-Co deposits in Phanerozoic orogens have also been discovered in China, and are classified as the “Tethys” and “Central Asian” types. The Palaeozoic (ca. 411 Ma) Xiarihamu deposit, one of the most representative Tethys type deposits, is associated with a mantle plume in a rifting tectonic setting, which is distal to the subduction and closure of the Proto-Tethys Ocean during the Devonian. The pyroxenite mantle source may be a characteristic of its mineralization. The Central Asian type is represented by many Early Permian deposits distributed in the Tianshan-Beishan and Altai areas of the Central Asian Orogen. The majority of them are related to the subduction of the Ancient Asian Ocean in the Carboniferous and the action of mantle plumes in the Tarim Large Igneous Province.

1. Introduction

Whether material sources of magmatic Ni-Cu-Co sulfides are derived from the mantle has interested the geological community for a long time [1]. The deposits are associated with continental rifting under discrete tectonic and metallogenic events [2,3]. It is further considered that the deposits are related to a mantle plume originating from the boundary between the Earth’s core and mantle [4], [5], [6]. It is believed that most magmatic Ni-Cu-Co sulfides sourced from komatiite formed during upwelling of mantle plumes and subsequent plate breakup [7,8]. With in-depth studies of oceanic plate subduction followed by continental collisions and development of orogenic belts, the distribution of magmatic Ni-Cu-Co deposits has received increasing interest from geologists and mineral exploration companies [9], [10], [11]. This is especially true after the discovery of the Xiarihamu super-large magmatic Ni-Cu-Co deposit in the East Kunlun Orogen of China, but mineralisation in a plate convergence setting has also received more interest [10], [11], [12], [13], [14], [15].

Mafic-ultramafic rocks with magmatic Ni-Cu-Co sulfide formed in orogens often exhibit island arc geochemistry, which is depleted in the incompatible Nd and Ta elements. This challenges the traditional idea that such deposits formed in discrete continental environments. It is not established if the major control on the deposition of Ni-Cu-Co sulfides is the tectonic setting or a partially melted mantle.

Recent research shows that ultramafic and mafic rocks in orogens are derived from the dehydration of a metasomatised subducted lithosphere mantle wedge [16]. Given that at least some of the ultramafic and mafic bodies can be preserved in orogens, it is not known which of these rock types are possible sources of economic magmatic Ni-Cu-Co mineralisation. A step towards this understanding is to better decipher the degree of partially melted mantle and the heat dynamics involved during sulfide deposition.

Current studies of sulphide mineralisation indicate that although the size of deposits may be relatively restricted in area, their genesis involves the presence of a large-scale partially melted mantle and a large and deep magmatic chamber [17,18]. In such magma chambers, there is sufficient immiscibility between sulfide and silicate melts, resulting in the formation of sulfide-rich magma that can accumulate in structural traps in the upper part of crust. An anomalously high heat source is also required resulting in the melting of the mantle, such as above the head of a mantle plume or upwelling mantle [5,10,19,20]. This would, in turn, drive the subduction of oceanic crust, dehydration, and metasomatism reducing the melting point of the lithospheric mantle. If enough silicate melts are produced, over 50% of the sulfide-rich magma can travel through conduits to shallow levels of the crust to form deposits.

In short, the genesis of magmatic Ni-Cu-Co deposits involves the supply of enough magma in a deep magma chamber. The magmatic Ni-Cu-Co deposits can be formed at continental rift or extensional tectonic setting of orogens, as long as there is a favourable source of the magma, sufficient partial melting, structural conduits, and sites of trapping.

2. Metallogenesis and broad characteristics

2.1. Discovery of Ni-Cu-Co deposits in China

Over 30 magmatic Ni-Cu-Co sulfide deposits have been discovered in China (Fig. 1), two of which are the world-class Jinchuan deposit and the super-large Xiarihamu deposit, with nickel metal resources exceeding one million tonnes representing ∼80% of China’s nickel and cobalt resources [21]. China's earliest discovered magmatic Ni-Cu-Co mineralisation is the Limahe deposit at Huili in Sichuan Province. The deposit was discovered in 1957 and mining began in the same year and closed at the end of the 1990s [22]. The first super-large magmatic deposit in China was discovered in 1959 at the Jinchuan Ni-Cu-Co deposit in Gansu Province [23]. The deposit has a resource of 6 Mt Ni, 3.5 Mt Cu, 0.16 Mt Co, and 200 t PGE. The Ni resource places the deposit in the world's third richest, after the Noril'sk deposit in Russia and the Sudbury deposit in Canada.

Fig. 1.

Geological distribution of magmatic copper-nickel-cobalt sulfide deposits in China (modified after Li, 2022[21]; Map of China outline according to the Ministry of Natural Resources, PRC, GS (2016) 1552).

The Jinchuan deposit is in the Longshou Terrane, which is a nappe composed of Proterozoic migmatite, gneisses, marble, and Palaeozoic granitic rocks along the southwestern edge of the Alxa Massif in the western North China Block (Fig. 2). The deposit is classified as “a super-large deposit formed from small intrusive rocks” [23]. The host rocks are ultramafic and consist of dunite, lherzolite, and a smaller amount of olivine pyroxenite. The discovery of the deposit resulted in a decade of Ni exploration focusing on mafic and ultramafic rocks with the discovery of the small-sized Zhubu deposit and the large Lengshuijing and Jinbaoshan deposits along the western edge of the Yangtze Block, and the large to middle-sized deposits at Chibaisong and Hongqiling in Jilin Province of northeast China [22].

Fig. 2.

Simplified geological map of the Longshoushan Terrane (modified after Tang et al., 1995[23]).

After the 1980s, significant large to middle-sized Ni-Cu-Co deposits were discovered at northern Xinjiang, East Huangshan, Tulaergen, and Kalatongke in the East Tianshan-Beishan and Altai orogens. This resulted in exploration of magmatic Ni-Cu-Co deposits in orogens.

The discovery of the super-large Xiarihamu magmatic Ni-Cu-Co sulfide deposit in 2011 opened the third wave of exploration for magmatic Ni-Cu-Co sulfide deposits in China. The deposit is the first explored super-large deposit in the global Tethys orogenic system, discovered in East Kunlun Orogen. The deposit is Early Devonian in Late Paleozoic in age, making it the first discovered of that age, and is thought to be the result of the fragmentation of the Paleo-Tethys plate [19,24].

2.2. Age and types of magmatic Ni-Cu-Co deposits in China

Most of the magmatic Ni-Cu-Co deposits in China are massif- and orogen-related. The massif-related deposits are in the southwestern and northern margin of the North China Block and Alxa Massif, the northern margin of the Tarim Block and Qaidam Massif, and the west margin of the Yangtze Block (Fig. 1; Table 1). The orogen-related type includes the East Tianshan-Beishan, Altai, and Jihei in CAO, and East Kunlun Paleo-Tethys [21].

Table 1.

Schedule of mineralization characteristics and types of magmatic Ni-Cu-Co sulfide deposits in China.

Mineralization background		Typical deposit	Main rock types	Ore-forming elements	Deposit scale	Chronology and mineralization age *
Continental rift setting	Continental margin rift	Jinchuan	Lherzolite, dunite	Ni, Cu, Co, PGE	Super-large	SHRIMP zircon U-Pb, 827 ± 8 Ma
		Xingdi	Gabbro, websterite, lherzolite	Ni, Cu	Small	Zircon U-Pb, 760 ± 6 Ma
		Dapoling	Meta-pyrolite, metapyroxenite, gabbro-diabase	Ni, Cu, Co, PGE	Small	SHRIMP zircon U-Pb, 828 ± 7 Ma
		Zhou'an	Lherzolite, olivine pyroxenolite	Ni, Cu, PGE	Large	Zircon U-Pb, 642 ± 4 Ma
		Taoke	Meta-olivine gabbronorite, metagabbronorite	Ni, Cu, PGE	Small	Zircon U-Pb, 2,715 ± 16 Ma
		Tongtongzi	Altered gabbro-diabase	Ni, Cu	Small	Pre-Cambrian
		Chibaisong	Gabbro-diabase, lherzolite	Ni, Cu, Co, PGE	Small	SHRIMP zircon U-Pb, 134 ± 7 Ma
	Large igneous province related to mantle plume	Limahe	Clinopyroxene peridotite, gabbro-diorite	Ni, Cu	Small	SHRIMP zircon U-Pb, 263 ± 3 Ma
		Baimazhai	Peridotite, olivine pyroxenolite, pyroxenolite, gabbro	Ni, Cu, Co, PGE	Small	SHRIMP zircon U-Pb, 259 ± 4 Ma
		Jinbaoshan	Altered clinopyroxene peridotite, gabbro-diabase	PGE, Ni, Cu, Co	Large	SHRIMP zircon U-Pb, 261 ± 4 Ma
		Yangliuping	Lherzolite, pyroxenolite, gabbro	PGE, Ni, Cu	Large	Late Permian
Extensional environment in the Orogenic belt	Tethys orogenic belt	Xiarihamu	Websterite, gabbro, peridotite	Ni, Cu, Co	Super-large	Zircon U-Pb, 412 ± 2 Ma
		Lashuixia	Altered peridotite	Ni, Cu, Co, PGE	Small	–
		Jianchaling	Serpentinite, talcized magnesitite, magnesitite	Ni, Co	Large	Sulfide Re-Os isochron, 878 ± 27 Ma
	Central Asian Orogenic Belt	Huangshandong	Hornblende-olivine gabbro, gabbro-peridotite	Ni, Cu, Co, PGE	Large	SHRIMP zircon U-Pb, 274 ± 3 Ma
		Tulaergen	Hornblende peridotite, pyroxenolite, gabbro	Ni, Cu, Co, PGE	Large	SHRIMP zircon U-Pb, 301 ± 3 Ma
		Poyi	Bojite, olivine pyroxenolite	Ni, Cu	Large	TIMS zircon U-Pb, 274 ± 4 Ma
		Jingbulake	Diorite, olivine-gabbro, peridotite	Ni, Cu	Small	SHRIMP zircon U-Pb, 434 ± 6 Ma
		Heishan	Plagio-cortlandtite, gabbro	Ni, Cu	Large	SHRIMP zircon U-Pb, 356 ± 1 Ma
		Xiaonanshan	Altered gabbro	Ni, Cu, Co, PGE	Small	Zircon U-Pb, 273 ± 3 Ma
		Karatongke	Harzburgite, gabbro-norite	Ni, Cu, PGE	Large	SHRIMP zircon U-Pb, 287 ± 5 Ma
		Hongqiling	Orthopyroxenite, peridotite, gabbro	Ni, Cu, Co	Large	SHRIMP zircon U-Pb, 240 ± 3 Ma
		Wuxing	Gabbro, pyroxenolite, olivine pyroxenolite	Ni, Cu, Co, PGE	Small	SHRIMP zircon U-Pb, 38 ± 1 Ma

^⁎Note: The data of ages is referenced after Li, 2022 [21].

The age of China’s magmatic Ni-Cu-Co deposits is Neoproterozoic, Late Silurian to Early Devonian, and Early Permian to Early Triassic. The Neoproterozoic deposits include the Jinchuan, Lengshuijing, Chibaisong, and Zhou'an that formed in a continental rift tectonic setting. It is proposed that the genesis of the Jinchuan deposit is related to the breakup of the Neoproterozoic Rodina Supercontinental caused by a super mantle plume [17,25]. The age of the deposit is based on the evidence of samples from Jinchuan. The whole rock and olivine Sm-Nd isochron ages of ca. 1,508 ± 31 Ma [23], the Re-Os isochron ages of 1,126 ± 96 and 840 ± 79 Ma yielded for the sulfide samples [26], and the U-Pb zircon age of ca. 827 Ma [27]. The small ca. 642 Ma Zhou'an deposit at the north margin of Yangtze Block in Henan Province has a Ni resource of 0.3 Mt [28].

The Early Permian to Early Triassic deposits includes: East Huangshan and Tulaergen in the CAO; Poyi in the Tarim Block; ca. 280 Ma Kalatongke deposit in the Altai Orogen [5,9,10]; ca. 273 Ma Xiaonanshan deposit in Inner Mongolia [29]; ca. 240 Ma Hongqiling deposit in Jihei Orogen [30]; and the ca. 260 Ma Zhubu, Limahe and Jinbaoshan deposits along the west edge of South China and Sanjiang tectonic belts [31], [32], [33]. These deposits formed in the post-orogenic aulacogen setting related to the Emei Large Igneous Province (LIP) [8,31,32].

The genesis of the East Tianshan-Beishan magmatic Cu-Ni-Co sulfide deposits may be related to an Early Permian mantle plume located in the Tarim Block following subduction of the ancient Asian Ocean, orogenesis, and post-orogenic gravitational collapse [5,10,18,34]. The known deposits in Inner Mongolia and Jihei gradually decrease in age from west to east with some deposits ranging from ca.156 to 38 Ma [35,36]. This probably relates to collision and post-orogenic extension of the CAO and subduction of the Pacific Plate. For the convenience of narration, the deposits in CAO are referred to as the Central Asian type. The deposits along the western margin of the Yangtze Block are thought to be related to the evolution of the Emei LIP linked to a mantle plume [32,37].

3. Mineralisation process

Magmatic Ni–Cu-Co sulfide deposits are commonly associated with olivine cumulates and form due to segregation and concentration of liquid sulfide droplets from mafic or ultramafic magmas and the partitioning of chalcophile elements into the silicate melt [7,38]. The sulfides accumulate in restricted localities, commonly at the base of magma conduits, giving rise to economic magmatic Ni–Cu-Co deposits [7]. The extent to which sulfur dissolves in silicate melts saturated in an immiscible sulfide phase before, during and after the emplacement of magma plays a primary role in the generation of magmatic Ni-Cu-Co deposits [1].

3.1. Jinchuan super-large Ni-Cu-Co deposit

During the early exploration of the Jinchuan deposit, it was recognised that the mineralisation followed the northern faulted boundary of the Longshoushan Terrane and that the fault was deep seated (Fig. 2) [39]. The fault was subsequently tested with the magnetotellurics geophysical method, which uses natural time variations of the Earth’s magnetic and electric fields to measure the electrical resistivity of the sub-surface. It was found that the fault extends to a depth of ∼10 km and the Jinchuan parental magmatic chamber was at least 10 km deep [23]. Given that the fault overprinted the deposit, it is not the conduit for the Jinchuan mineralisation, but is related to the south to north thrusting of the Longshoushan Terrane during the Mesozoic. The fault was reactivated as a tensional structure during the Cenozoic (Fig. 3) [19,[21], [22], [23],34].

Fig. 3.

Deep immiscibility-injection mineralization and displacement mode diagram of the Jinchuan super-large magmatic copper-nickel-cobalt sulfide deposit (modified after Li, 1996[22]).

Jinchuan deposit is a typical example of “forming big ore deposits in small intrusive rocks” type (mentioned earlier). The presence of “small intrusive rocks” means the Jinchuan deposit cannot be sourced in-situ from its host rocks [23]. However, systematic study of the deposit has led to the interpretation that the parental magma developed in the mantle and contained sulfide-rich melts that have migrated upwards through the crust to sites of deposition along structural conduits. The findings of the investigation in the last two decades have had a widespread impact on academia [40,41], and mineral exploration [12,42].

The average sulfur content of the host rock at Jinchuan is 6%. According to the calculation of sulfur solubility equilibrium, the volume proportion of Jinchuan mineralized magma accounts for ∼3% of the parental magma, indicating the original magma chamber was large. The calculation (mass ratio of silicate magma to sulfide (“R” factor) is 150–1,000, which also supports this point [17]. The high degree of partial melting of the upper mantle is crucial for the generation of a large parental magma. The Jinchuan parental magma is estimated to be the product of 20%−30% partial melting of the upper mantle, allowing sulfur and Ni, Cu, Co, and PGEs to enter the magma [23]. In addition, chromium spinel in the deposit has a relatively high Fe³⁺ content, reflecting the high oxygen fugacity in the Jinchuan magma [23]. An elevated oxygen fugacity is critical for the depletion of sulfides in the upper mantle, which brings chalcophile elements into the magma and provides favourable conditions for the formation of sulfide droplets within the parental magma.

The development of a large parental magma implies that significant partial melting of the asthenosphere mantle can be accommodated by the presence of a mantle plume. As mentioned earlier, the idea that a small volume and high-grade mineralisation can travel upward in a conduit against gravity in an extensional environment seems up likely [38]. Alternatively, similar mineralisation formed in a compressional environment. Lateral emplacement of the mineralisation from a magma, which would require a deep crustal site of deposition, is here considered a likely mechanism for the deposition of a small volume and rich magmatic sulfide deposit. Such a deposit would have been brought up the crust in a nappe structure during a compressional event during the Mesozoic to Cenozoic. This is consistent with what we see at the Jinchuan deposit.

3.2. Tectonic evolution and deep partial melting

The super-large Xiarihamu magmatic Ni-Cu-Co sulfide deposit is a typical example of a deposit formed in an orogen; the formation of which represents unique metallogenic events. It has been found recently that the total amount of Ni-Cu-Co sulfide ore at Xiarihamu is not insignificantly lower than at the Jinchuan [43]. The mass ratio of silicate magma to sulfide (“R”) factor ranges from 100 to 1,000 at Xiarihamu is like that of Jinchuan with a value of 150–1,000; and the Xiarihamu’s MgO content ranges from 9.8% to 12.5% [11,13,15,44], which is also similar to Jinchuan’s value of 11.8%–12.9%. These all show evidence of a higher degree of partial melting, estimated at Xiarihamu to be 15%−25%, and Jinchuan at 20%−30% [21]. However, the (⁸⁷Sr/⁸⁶Sr)_i value of the ore at Xiarihamu is lower than that at Jinchuan, and the εNd(t) value of Xiarihamu’s ore is higher than that at Jinchuan (Fig. 4), indicating that the Xiarihamu’s parental mantle magma is less enriched than that at Jinchuan.

Fig. 4.

Comparison of the Sr-Nd isotopes of ore-bearing mafic-ultramafic intrusions from the Xiarihamu deposit (modified after Zhang et al., 2021[44]).

Because both Cu and Pd are strongly incompatible elements and tend to be retained in the residual melt during magmatic differentiation. The distribution coefficients of PGE in sulfide melt/silicate melt are much higher than that of Cu. Once sulfur saturation and sulfide segregation occur, Pd will preferentially enter the sulfides more than Cu, leading to a significant increase in the Cu/Pd ratio in the residual magma after sulfide segregation [45], [46], [47]. Thus, the Cu/Pd ratio indirectly reflects the degree of deep-seated sulfide segregation. The Cu/Pd value for Xiarihamu’s host rocks averages 18.2 × 10⁴, higher than Jinchuan’s average value of 17.3 × 10⁴ [5,11,17,48]. Thus, the Xiarihamu deposit perhaps exhibits a higher degree of sulfide segregation than Jinchuan before emplacement. Jinchuan is a small intrusive rocks mineralization, and its deep part has experienced a long-time crystallization differentiation process. The Xiarihamu deposit perhaps exhibits a higher degree of sulfide segregation than Jinchuan, indicating the Xiarihamu deposit has undergone relatively prolonged and stable crystal differentiation. The prolonged duration of the mineralising process can be attributed to the presence of a mantle plume in a rifting tectonic setting, which is distal to the subduction and closure of the Proto-Tethys Ocean (Fig. 5), rather than being a product of true island-arc environment [15,24].

Fig. 5.

“Small intrusion forming large deposit” mineralization model of the Xiarihamu ore deposit.

It is envisaged that a subducting oceanic plate would produce different compositions of fluids at different depths and degrees of dehydration. Peridotite or pyroxenite in a mantle wedge that forms during the subduction of oceanic crust will be hydrated by partial melting of the subducted crust. The temperature at which this happens is lower than the melting point of the peridotite or pyroxenite, unless there are increased temperatures from a mantle plume, decrease in pressure, or the development of doubly thickened crust in an orogen. Under these conditions, the peridotite or pyroxenite might partially melt forming magnesian-ferroan melts that are the source of mafic and ultramafic in an orogen [16].

The distribution of mafic and ultramafic rocks in the Kunlun Orogen date between ca. 443 and 378 Ma, which covers the emplacement of mafic and ultramafic bodies during different tectonic events. The first mafic to ultramafic magmatism is pre-420 Ma, poorly mineralised, and associated with a Proto-Tethys collisional event. The second magmatism is post-420 Ma, associated with a Paleo-Tethys continental rifting event [19,24]. Mafic and ultramafic bodies generated in an island-arc setting are not endowered in economic concentrations of Ni-Cu-Co sulfide deposits. Similarly, extensional zones within active orogens are marked with limited magmatism emplaced at low temperatures and do not host economic concentrations of Ni-Cu-Co sulfides.

The immense geodynamic forces associated with mantle plumes, and the asthenosphere near gravitationally collapsed orogens is significantly modified generating large volumes of magma. This process is thought to form large magmatic Ni-Cu-Co sulfide deposits, but not during subduction or collisional orogenic events.

4. Discussion

4.1. Source of the magma in the Phanerozoic orogens of China

The East Tianshan-Beishan area of CAO is an important source for magmatic Ni-Cu-Co sulfide deposits [34]. The terrains in the area are: (1) the Jueluotage trench-arc-basin belts containing the Huangshan, East Huangshan, Xiangshan, Hulu, Tulaergen, Baixintan, and Lubei deposits; (2) the northern margin of Central Tianshan containing the Tianyu and Baishiquan deposits; and (3) the Beishan rift containing the Poyi, Bijiashan, Luodong and Xuanwoling deposits (Fig. 6).

Fig. 6.

Distribution of ore-bearing mafic-ultramafic intrusions in the East Tianshan and Beishan (modified after Xiao et al., 2004[49]and Su et al., 2011[51]).

Most of the deposits are ca. 280 Ma in age. Obviously, deformation events younger than Triassic have no bearing on their initial emplacement in the CAO, and there is considerable debate on the understanding of the tectonic evolution of the CAO; including the evolution and closure of the ancient Asian Ocean. One focal point of the debate is the timing of the closure of the ocean [49,50], and another is whether the ocean is related to the Proto-Tethyan Ocean that existed between the Laurasia and Gondwana after fragmentation of the Rodinian Supercontinent [3,19,51]. The current interpretation is that the Cathaysia paleocontinent separated the Proto-Tethys Ocean to its south and the Paleo-Asian Ocean to its north [52]. In addition, the evolution of the Tethys Ocean can be divided into: (i) the Proto-Tethys Ocean in the Early Paleozoic; (ii) Paleo-Tethys Ocean in the Late Paleozoic to Early Mesozoic; and (iii) the Neo-Tethys Ocean in the Mesozoic to Cenozoic [19,52,53].

Following the identification of separate Early Palaeozoic and Late Palaeozoic serpentinite bodies in Russia, it has been proposed that CAO was a continent during the Devonian [50,54]. In this period, the Early Permian magmatic Ni-Cu-Co sulfide deposits in the East Tianshan-Beishan area probably have an asthenospheric source following the effects of the proposed Tarim mantle plume [19].

The newly discovered Baixintan and Lubei deposits at south Shalong in the East Tianshan area show that some geochemical characteristics are similar to enriched mid-ocean ridge basalt (E-MORB) and the Sr-Nd isotopes are as same to oceanic island basalt (OIB). Therefore, it was proposed that the magmatic Ni-Cu-Co sulfide deposits at the East Tianshan area involved a mantle plume [20]. In addition, the trace element geochemistry shows that the source of the East Tianshan-Beishan deposits is the same as or like those in the Tethys Orogen, and the addition of fluid to a mantle wedge from subduction slab results in the depletion of Nb, Ta, and Ti [20].

The mafic-ultramafic samples from Limahe, Jinbaoshan deposits and Tarim LIP plot in the OIB field of the MORB-ORB array on the Nb/Yb-Th/Yb discrimination diagram, consistenting with a mantle plume origin [20,32,55]. Similarly, the mafic-ultramafic samples from the East Tianshan-Beishan area plot in the E-MORB field, indicating the characteristic of a mantle plume to mid-ocean ridge origin [5,20]. Therefore, it is not accurate to simply interpret the mantle source region of the East Tianshan-Beishan magmatic Ni-Cu-Co sulfide deposits as being subduction-related and hydrothermally altered mantle wedge. Further, the original temperature calculated by chromite and olivine ranges from 1,360 to 1,411 °C, which is significantly higher than the asthenospheric mantle temperature in the range of 1,280–1,350 °C, but slightly lower than the temperature of the mantle plume [56,57]. In addition, the (⁸⁷Sr/⁸⁶Sr)_i and ε_Nd(t) values for the samples plot in the OIB field, reflecting a possible association with a mantle plume (Fig. 7). A more direct evidence of mantle plume is the discovery of picrite or the evidence of the existence of picrite magma. A picrite, a rock that most likely derived from the hottest part of mantle plume centre, was found in the Bachu area (western part of Tarim Basin) [58]. Although picrite has not been found so far along the northeastern margin of Tarim, a picritic magma is suggested according to the inferred composition of olivine in Early Permian mafic-ultramafic rocks [59,60], implying that the northeastern margin of Tarim including the Beishan area was likely far from the center of the Tarim mantle plume [5]. However, based on the fact that the Beishan rift exists, Qin et al. (2011) hold the view that the centre of the mantle plume is in the Beishan area (Northeastern part of Tarim) rather than in the Bachu region (Western part of Tarim) [10]. Thus, there is still controversy over the location of the center of the Tarim mantle plume. The controversy may be due to that the magmatic Ni-Cu-Co sulfide deposits in the CAO is metasomatized by fluids derived from a subducting oceanic plate, which leads to the complexity in geochemical signals.

Fig. 7.

Comparison of the Sr-Nd isotopes of ore-bearing mafic-ultramafic intrusions from the east Tianshan and Beishan (modified after Zhou et al., 2008[32]and You, 2022[20]).

4.2. Fractionation and enrichment of nickel and cobalt

The average content of cobalt in the mantle is 102 ppm [61], in the continental crust is 27 ppm, in lower crust is 38 ppm, and in the upper crust is ∼70 ppm [62], and in the oceanic crust is ∼44 ppm [63]. The distribution coefficients of Co in olivine/basaltic melt, spinel/basaltic melt, clinopyroxene/basaltic melt, and orthopyroxene/basaltic melt are 3.63–6.3, 4.7–8.3, 1–1.7, and 2.06, respectively [64], [65], [66]. Therefore, during the mantle partial melting and the crystallization differentiation process of the melt, Co tends to enter minerals rather than melt. One method for cobalt not to enter silicate minerals is to achieve sulfide immiscibility before silicate crystallization due to the distribution coefficients of Co in sulfide/basaltic melt reach to 61–80 [67]. The distribution coefficients of Ni in olivine/basaltic melt, spinel/basaltic melt, clinopyroxene/basaltic melt, and orthopyroxene/basaltic melt are 32–47, 10, 1.1–10, and 11.3, respectively [64,67,68]. The priority order of Co entering silicate minerals is: spinel > olivine > orthopyroxene > clinopyroxene, while the order for Ni is: olivine > orthopyroxene > spinel > clinopyroxene. Thus, different degrees of partial melting in the mantle source region or different crystal differentiation sequences can lead to decoupling of Co and Ni. As the crystal differentiation sequences are nearly same in East Tianshan-Beishan and Jinchuan, the lower partial-melting degree of mantle in the East Tianshan-Beishan area perhaps is the reason why the cobalt content in the East Tianshan-Beishan magmatic Ni-Cu-Co sulfide deposits is lower than that in the Jinchuan deposit.

Based on the characteristics of high Ni/Cu ratio and high Ni content of olivine in the Xiarihamu deposit magma, Song et al. speculated that the mantle source of the Xiarihamu deposit is pyroxenite mantle [11]. The parent magma of the Xiarihamu deposit is believed to be high-Mg basaltic magma [11,15]. The partial melting of pyroxenite at 1.5 GPa/1,240 °C or 0.5 GPa/1,050 °C can indeed produce high-Mg basaltic melts, which supports the pyroxenite mantle source viewpoint [69]. The Co enrichment in the Xiarihamu deposit perhaps can also be explained by the pyroxenite mantle source. As Ni and Co are more compatible with olivine compared to pyroxene. In the pyroxenite mantle without olivine, Co will also be more likely to enter the melt during partial melting, achieving the initial enrichment of Co.

Another factor that affects the differential enrichment Co is the oxygen fugacity. The formation of magmatic sulfide deposits usually requires low oxygen fugacity, and the upper limit of oxygen fugacity must at least ensure that sulfide (S²⁻) can exist in the melt. At 0.5 GPa/1,050 °C, the oxygen fugacity value for the coexistence of S²⁻ and S⁶⁺ is about FMQ + 2.21 (lg fo₂ = −7.83, FMQ = fayalite–magnetite–quartz redox buffer, in lgfo₂) and for the coexistence of S²⁻ and S⁰ is FMQ–0.52 (lg fo₂ = −10.57) [69]. Thus, the oxygen fugacity of the magma lower than FMQ–0.52 probably is a necessary stage for the formation of magmatic Cu-Ni-Co-PGE deposits. The formation of the Xairihamu deposit follows this rule. The Xiarihamu Ni-Cu-Co deposit is considered to reach sulfide saturation at FMQ-0.6 during the crystallization differentiation progress [70]. The partition coefficient of cobalt (D^Co_SL/SM) in the sulfide liquid (SL) phase and silicate melt (SM) reaches the highest value at about FMQ-1 in the oxygen fugacity range from FMQ-3 to FMQ + 1 [71]. The oxygen fugacity of the mineralization in the Xiarihamu deposit happens to be near FMQ-1, which may also be a factor for the Co enrichment.

5. Conclusion

In this paper, we review the spatial and temporal distributions, tectonic models, and classifications of the magmatic Ni-Cu-Co deposits in China, which are represented by the world-class Neoproterozoic Jinchuan deposit, super-large Late Silurian to Early Devonian Xiarihamu deposit, and many Early Permian to Early Triassic deposits in Tianshan-Beishan and Altai areas. The main conclusions are summarised as follows.

(1) The world-class Jinchuan deposit, a typical Neoproterozoic example, is the product of mantle plume activity and continental rifting and fragmentation of the Rodinian Supercontinent. The genesis of Jinchuan is partial melting of the asthenospheric mantle resulting in the generation of high-Mg basaltic magma. The mineralised ultramafic bodies at Jinchuan are the products of the intrusion of mineralised magma emplaced at a depth of 10 km in the crust as a horizontal “sheet” during ca. 827 Ma. The sheet was uplifted to the surface and steepened in a nappe structure during the Mesozoic to Cenozoic.

(2) The super-large Late Silurian to Early Devonian Xiarihamu magmatic deposit is an example of a Devonian (ca. 411 Ma) deposit in the East Kunlun Orogen, which is classified as a Tethys-type magmatic Ni-Co sulfide deposit. The pyroxenite mantle source can explain the enrichment of Ni and Co. The Proto-Tethys oceanic crust subducted into asthenosphere during the Devonian, resulting in a large-scale partial melting of the asthenospheric mantle associated with the action of a mantle plume and the generation of mafic and ultramafic magma. Low oxygen fugacity is a favourable factor for the Xiarihamu mineralization.

(3) The mineralisation age of the magmatic Ni-Cu-Co deposits in the East Tianshan-Beishan region of the CAO ranges from Early Permian to Early Triassic, which are called the Central Asian type magmatic sulfide deposits. Many of them were formed during the subduction of Ancient Asian Ocean in the Carboniferous, and are related to the presence of a mantle-plume in the Tarim LIP. The center and influence range of the Tarim mantle plume are still controversial. The source mantle of these deposits is probably metasomatized by the subducted oceanic plate.

Declaration of competing interest

The authors declare that they have no conflicts of interest in this work.

Biographies

Wenyuan Li is a senior researcher and Ph.D. advisor at the Xi’an center of geological survey, China Geological Survey. He received the bachelor’s and master’s degrees in Chang’an University, and the doctorate degree in Northwest University. His current research interests include magmatic ore deposits and regional metallogenetics, etc. He is a director of the key laboratory for the study of focused magmatism and giant ore deposits, Ministry of Natural Resources and council member of China Geological Society.

Yongbao Gao is a researcher of the Xi’an center of mineral resources survey, China Geological Survey. He received the master’s and doctor’s degrees in Chang’an University. His current research interests include ore deposits and regional metallogenetics in northwestern China, etc. He is a secretary general of the key laboratory for the study of focused magmatism and giant ore deposits, Ministry of Natural Resources and vice chairman of the Youth Work Committee of the Chinese Geological Society.