2.1 Geological setting

The study area is located ~30 km north of the Bangong–Nujiang Suture Zone (BGNS) within the southern margin of the South Qiangtang Terrane. The Tiegelongnan, Duobuza, Bolong, and Naruo large-superlarge porphyry copper (gold) deposits as well as other mineralization spots comprise the Duolong metallogenic district. The stratigraphy of the Duolong district includes the Upper Triassic Riganpeicuo Formation (T₃r), Lower Jurassic Quse Formation (J₁q), Middle–Lower Jurassic Sewa Formation (J_1–2s), Lower Cretaceous Meiriqiecuo Formation (K₁m), and Upper Cretaceous Abushan Formation (K₂a), all of which are partly overlain by Quaternary sediments. The Riganpeicuo Formation is dominated by marble and limestone. The Quse Formation contains interbedded sandstones and siltstones as well as chert and pillow units. The Sewa Formation mainly contains interbedded sandstones and siltstones. Both of the formations have undergone weak metamorphism, with siltstones partly metamorphosed to slates. The Meiriqiecuo Formation is mainly composed of volcanic rocks, such as basalts, basaltic andesites, and andesites. The basalt yields a zircon U–Pb age of 110.4 ± 1.7 Ma (Liu et al., 2017); the ⁴⁰Ar/³⁹Ar geochronology of the basaltic andesites illustrates two episodes of igneous rock activity at approximately 118 and 106 Ma (Li et al., 2011), and the andesite was emplaced at 118–111 Ma (Li et al., 2011, 2016). The Abushan Formation is comprised of conglomerates and sandstone. Regionally, tectonism is dominated by NE–SW- and NW–SE-trending fault sets (Hu, Xu, Wan, Wu, & Yi, 2018; Zhu et al., 2019). Magmatic activity occurred along these faults and mainly formed intermediate-acidic intrusions in the Duolong area.

According to the porphyry copper deposit classification system (McMillan & Panteleyev, 1980), deposits in Duolong district are of the typical volcanic type, which was formed by Bangonghu–Nujiang Ocean (BNO) subduction and accompanied by the occurrence of a magmatic arc. The BNO commenced subduction towards the north in the Middle to Late Jurassic, possibly within 175–145 Ma (Kapp et al., 2003; Liu et al., 2014; Xu et al., 2017). While the closure time of the BNO is controversial, some studies suggest that the BNO may have closed during the Late Jurassic–Early Cretaceous through arc–arc ‘soft’ collision, most likely within 140–130 Ma (Zhu et al., 2016), and then entered the post-collisional stage of the orogeny at approximately 110 Ma (Qu, Wang, Xin, Jiang, & Chen, 2012; Song et al., 2018). However, some studies have recommended that the BNO was still subducting northward during the Early Cretaceous, possibly within 106–125 Ma (Geng et al., 2016; Li et al., 2014; Sun et al., 2017; Xu et al., 2017).

2.2 Deposit geology

The Tiegelongnan deposit is a representative high-sulfidation porphyry mineralization type that has been newly recognized in recent years in the western part of the Bangonghu–Nujiang metallogenic belt, central Tibet, China. The main ore body presents as a thick slab-like form, of which the width, length, and depth are approximately 1,400, 1,800, and 960 m, respectively. The total reserves are >11 million tons of copper and ~120 tons of gold, with average grades of 0.53% and 0.12 g/t, respectively (He et al., 2018; Song et al., 2018). Mineralization mainly occurs in the Sewa Formation and the granodiorite porphyry zone (Figure 1a). The alteration zone is expressed as successive developments of advanced argillization, phyllic alteration, propylitization, and K-silicate alteration from the centre to the periphery and from the top to the bottom of the deposit. K-silicate is weakly developed in the deeper part of the system. Propylitic alteration has a widespread distribution mainly located around the ore-forming system. The typical cross-sections in the central part of the deposit are shown in Figure 1b,c. Its ores are pervasively enveloped under advanced argillic and phyllic alteration, and mineralization appears in most drill holes at approximately 100–200 m below the surface. The ore body is preserved below the Meiriqiecuo Formation with a zircon U–Pb age of 111.9 Ma (Li et al., 2011), which is key to the preservation of this deposit from intense erosion with rapid lifting due to the continuous collision of the Qiangtang and Lhasa terranes (Song et al., 2018). Advanced argillic alteration is developed in the upper levels of the deposit and yields variable associations consisting of alunite, kaolinite, APS, pyrophyllite, and Cu–Fe–S–As minerals. The phyllic zone is superimposed by an advanced argillic alteration event and typically contains sericite, carbonate (±chlorite, ±kaolinite, etc.), and Cu–Fe–S minerals. According to a previous study, the Ar–Ar dating of hydrothermal sericite and alunite reports ages of 120.8 ± 0.7 Ma and 117.9 ± 1.6 Ma, respectively (Lin, Chen, et al., 2017; Lin, Tang, et al., 2017); therefore, it is possible that the advanced argillic alteration postdates the phyllic alteration. However, the life spans present here are dramatically longer than the theoretically modelled time required for the consolidation of individual porphyry intrusions (<40,000 years; Cathles, 1977; Cathles, Erendi, & Barrie, 1997), the formation of copper ores in porphyry deposits (<100,000 years; McInnes et al., 2005), or dominant potassic alteration (<20,000 years; Cathles & Shannon, 2007; Shinohara & Hedenquist, 1997), which has been summarized by Sillitoe (2010).

3.1 Sample collecting

A total of 230 samples collected from drill holes zk0804, zk1604, zk2404, zk2420, zk2428, zk3205, zk3220, zk4004, zk4804, and zk5604 were identified by optical microscopy, among which zk0804, zk1604, zk2404, zk4004, zk4804, and zk5604 along the east–west profile cross-cut the Tiegelongnan mineralization system (Figure 1a). These selected samples occur as disseminated and veinlet ores, which are the primary types present in the ore body. Based on the detailed optical microscopic study, 23 samples were selected for analysis by scanning electron microscope–energy dispersive spectrometer (SEM–EDS). Sulphides from the chalcopyrite–sericite–dolomite and bornite–sericite assemblages presenting typical phyllic alteration found at relatively deep levels; tennantite–alunite assemblages, which are typical of high-sulfidation alteration at shallow levels; and covellite–sericite–kaolinite assemblages, which are intensively developed in phyllic to high-sulfidation transition zones were studied by LA–ICPMS (laser ablation inductively coupled plasma mass spectrometry).

3.2 Scanning electron microscopy–energy dispersive spectrometry

Micromineral inclusions, mineral textures, and semiquantitative geochemical compositions were investigated in the back-scattered electron (BSE) mode using a ZEISS Ultra Plus SEM equipped with Oxford Inca 305 energy-dispersive X-ray spectroscopy (EDS) at the Institute of Mineral Resources, Chinese Academy of Geological Sciences, Beijing, China. The EDS spectra were collected at an accelerating voltage of 15 kV, and a beam diameter of 1 μm was used for optimized quantification. However, the analytical EDS data are problematic for evaluating the quality of these elemental compositions, which is probably true at % of weight percent for the elements shown in Data S1.

3.3 LA–ICPMS analyses

LA–ICPMS spot analyses were performed on four representative polished thin sections (i.e., samples zk1604-263, zk2420-938, zk2412-786, and zk4804-1,206) using a Teledyne Cetac Technologies Analyte Excite laser ablation system (Bozeman, MT) coupled to an Agilent Technologies 7700x quadrupole ICP–MS (Hachioji, Tokyo, Japan) at Nanjing FocuMS Technology Co. Ltd., China. The specifics of the instrument include a 40 μm spot diameter, a fluence of 6.06 J/cm², and a repetition rate of 6 Hz. Each analysis comprised a 15 s background measurement followed by 40 s of sample ablation. Two 40 μm-diameter analyses were performed on a USGS polymetal sulphide pressed MASS-1 pellet and synthetic basaltic GSE-1 glasses after every six analyses of unknowns. ICPMSD-Data-Cal software was used to carry out data calculations. The trace element data and the corresponding average minimum detection limits in each sample are provided in Data S1.

LA–ICPMS element mapping was performed using the same instrument consisting of ablating sets of parallel line rasters in a grid across the sample surface. A square laser beam profile was used throughout the sample surface with spot sizes varying between 15 and 40 μm to provide an adequate spatial resolution for each map. Similarly, the laser bean scan speed was set to 15–40 μm/s, where a 15 μm spot size was used. A laser repetition of 10 Hz was selected at a constant energy output of 50 mJ, resulting in an energy density of ~3 J/cm² at the target. A 30 s background acquisition was set at the start of scanning. To allow for cell wash-out, gas stabilization, and computer processing, a delay of 20 s was used after ablation. At the start and end of each mapping reference material piece, a USGS polymetal sulphide pressed MASS-1pellet and synthetic basaltic GSE-1 glasses were analysed for data calibration. Images were compiled and processed using the program LIMS (Wang et al., 2017; Xiao et al., 2018). For each raster and every element, the average background was subtracted from its corresponding raster, and the rasters were compiled into a 2D image displaying the combined background/drift corrected intensity of each element (Wang et al., 2017).

4.1 Mineralogy

The spatial zoning of ore mineral assemblages is a distinctive feature of the Tiegelongnan deposit. In the ore body, chalcopyrite produced in the phyllic alteration zone gives way to tennantite, enargite, digenite, anilite, etc., in the advanced argillic alteration zone. Tennantite-bearing veinlets with various ratios of alunite, ±APS, ±kaolinite, pyrite, quartz, etc., regularly cross cut each other within a sample, similar to chalcopyrite-bearing veinlets with different proportions of quartz, ±sericite, ±dolomite, and pyrite. Bornite and covellite occur quite commonly in both alteration zones. Covellite frequently replaces bornite and chalcopyrite, while quartz-bornite veins are also observed cross-cutting quartz-covellite veins locally. To identify these various gangue minerals around the chief sulphides, optical microscopy and SEM-EDS are used in this study. Sericite, alunite, and kaolinite are mainly identified by optical microscopy, while pyrophyllite, APS (woodhouseite, svanbergite), carbonate (siderite, magnesite, dolomite), chlorite, and trace Au–Ag-bearing minerals are mainly recognized by SEM-EDS. The obtained EDS data are shown in Data S1.

Tennantite, mainly found coexisting with alunite ± kaolinite ± APS (Figure 2a,b-1,b-2), is one of the dominant ore minerals in the AAZ, whereas chalcopyrite occurs within a very limited extent and occasionally presents an extremely low proportion in these sulphide assemblages (Table 1). Tennantite can be found in abundance at ~180 to >550 m in the AAZ, where well crystalline tennantite, that is, up to 3,000-μm sized grains, would occur. Tennantite could either present as individual grains (Figure 2b-1,b-2) or coexist with other sulphides, such as enargite and bornite, which are the most preferred accompaniments. In addition, micron-scale calaverite is also contained in tennantite, similar to that reported by He et al. (2018). Triangular enargite is enclosed in tennantite, and a mutual boundary texture occurs between the two minerals, indicating that successive or simultaneous precipitation could both exist (Figure 2b-3,b-4). In most samples, bornite (±anilite) is rimmed by tennantite; however, the opposite relations are occasionally observed (Figure 2c-1). At relatively deep levels of the AAZ, a certain number of bornite aggregates occur within the alunite matrix without tennantite (Figure 2c-3,c-4). Generally, towards deeper areas of the AAZ, a relatively higher proportion of bornite can be observed in the tennantite–bornite (±anilite) association. Specifically, where the mineral forming space is sufficient, the interspace area of euhedral alunite (±APS, ±kaolinite) is infilled with tennantite (±enargite, ±bornite, and ±anilite), suggesting that the sulphides could continuously be deposited after these gangue minerals.

Bornite, as one of the most widespread sulphides, is commonly associated with various sulphides. Such variations, however, apparently correlate with the different alteration zones (Table 1). In the AAZ, except for the tennantite–bornite assemblage mentioned above, bornite can form mineral assemblages with anilite, covellite, digenite, chalcocite, yarrowite, etc. These sulphides generally coexist with euhedral–subhedral alunite, kaolinite, svanbergite, woodhouseite, etc. (Figure 2d-1,d-2). With increasing depths in the AAZ, bornite is locally observed as being intergrown with pyrophyllite coinciding with the occurrence of kaolinite (Figure 2e-1–e-3). In the PZ, it occurs as bornite (±chalcopyrite) and sericite intergrowths (Figure 2f). Lamellae, patches, or worms, such as those consisting of chalcopyrite, regularly form exsolution textures in bornite, which obviously originate from simultaneous precipitation (Durazzo & Taylor, 1982). At a great depth (~938 m) in the PZ, bornite and pyrite contain the same-sized sericite and share a straight contact boundary (Figure 2g-1), suggesting the great possibility of the simultaneous deposition of solid Cu–Fe–S and K–Si–Al solutions. In the same sample, a nanoscale native gold grain occurs within bornite (Figures 2g-2 and 3). In the contact areas between the phyllic and propylitic zones, bornite with minor chalcopyrite exsolutions is occasionally observed in Mg–Fe carbonate (i.e., siderite and magnesite) veins, and in phyllic alteration rocks, bornite with/without chalcopyrite could also occur to some extent as dispersive grains with the presence of Mg–Fe carbonates (Figure 2h). In short, bornite from the PZ presents the character of having a contemporaneous deposition texture with chalcopyrite, pyrite, sericite, and Mg–Fe carbonate minerals, while in the AAZ, it can either be individually present as aggregates within the interspace of the fine-grained matrix of alunite, APS, and pyrophyllite or partly replaced with covellite, anilite, enargite, etc., from the inside grain voids or fringes (Figure 2i).

Covellite, which is the most dominant Cu–S mineral, broadly occurs in both the AAZ and PZ (Table 1). In the AAZ, covellite rarely occurs as a monomineralic assemblage but is rather closely related to anilite, digenite, and enargite in the presence of kaolinite (±alunite). These sulphides usually replace bornite along its boundaries or within its voids (Figure 2i-1,i-2). In the phyllic to advanced argillic zone, which is located on the southwest side of the ore-body centre, notable covellite aggregates occur with varying amounts of sericite and kaolinite and occasionally APS (Figure 2j-1–j-4). In the phyllic to prophylitic transitional zone at relatively deep levels, covellite (±anilite) tends to present with variable amounts of dolomite, Fe–Mg carbonate, chlorite, and apatite (Figure 2k-1–k-4). Under hypogenic conditions, the cores of chalcopyrite are partially altered to covellite, with the presence of chlorite, sericite, and kaolinite (Figure 2l-1,l-2).

Chalcopyrite mainly occurs as monomineralic grains and associates with idiomorphic sericite and dolomite at a depth of approximately >900–1,200 m in the PZ (Figure 2m and Table 1). It is worth noting that the coprecipitation of chalcopyrite and dolomite, occurring as clustered grains or veinlets, is well developed at this depth, while the S-bearing gangue minerals are not as pervasive, which indicates that the CO₂-bearing fluids are likely involved in early copper precipitation events. Within this depth range, the transitional zone between phyllic and propylitic alteration is characterized by the presence of minor chalcopyrite with chlorite, sericite, and dolomite (Figure 2n). In addition, the spatial coexistence of sulphide–gangue mineral associations from the AAZ and the PZ is locally observed. Typically, the representative acid-resistant mineral assemblages, consisting of alunite and kaolinite plus minor enargite with/without bornite and covellite, replace large chalcopyrite grains at depth (~985 m), which contributes to the formation of a wave-like shape on the contact boundary (Figure 2o). Because the chalcopyrite-dominated sulphide assemblage is not a type of AAZ, we suggest that the photomicrograph, that is, Figure 2o, is promising geological evidence that indicates that phyllic alteration had been penetrated by these sluggish-formed acid-resistant minerals from the AAZ.

4.2 LA–ICPMS analyses of the main ore sulphides

Element maps of tennantite–alunite, bornite–sericite, covellite–sericite–kaolinite, and chalcopyrite–sericite are shown in Figures 4-7. Representative spot analyses of these copper sulphides are presented in Figure 8.

5.1 Alteration and mineralization process

The primary ore minerals, namely, chalcopyrite, bornite, covellite, and tennantite, produced in the Tiegelongnan deposit are also present in some other well-known porphyry deposits, such as Butte, Mantos Blancos, Río Blanco, etc. (Crespo, Reich, Barra, Verdugo, & Martínez, 2018; Reich, Palacios, Barra, & Chryssoulis, 2013; Sales & Meyer, 1949). Despite the pervasive coexistence of sulphides and gangue minerals, there are still uncertainties as to the stability of the sulphide–gangue mineral assemblages, and how they coexist, especially those present in transition zones. Thus, in this study, we aim to understand the ore-forming fluids related to typical sulphide–gangue mineral assemblages that are able to coexist under similar temperature and pH conditions, such as chalcopyrite–sericite (±dolomite) and tennantite–alunite (±kaolinite ±APS) in the phyllic and advanced argillic alteration zones, respectively.

First, the majority of tennantite aggregates are produced in the AAZ with abundant acid-resistant alunite crystals, whereas chalcopyrite grains are mainly associated with sericite and dolomite at depth, implying the distinguishable pH values of the fluids that deposited the two sulphides. Second, the AAZ present on the west side of the deposit at relatively shallow depths notably postdated the PZ on the east side in relatively deep-seated areas, which suggests that the causative hydrothermal fluids flared upwards from east to west. Third, covellite (±anilite ±enargite) spatially coexists with kaolinite in the AAZ, with abundant sericite–kaolinite (±APS) in the transition zone adjacent to the ore-body centre, or with chlorite, carbonate, and sericite minerals in the phyllic to propylitic transition zone. Bornite tends to coexist with sericite (±pyrite), pyrophyllite (±kaolinite ±APS) and alunite (±APS ± kaolinite) within the phyllic to advanced argillic alteration zone. These assemblages imply that covellite and bornite might be produced under a broad range of pH values and form stable or substable assemblages with various gangue minerals. Fourth, the frequent pulses of fluids could induce multiple precipitation episodes even for one copper mineral and result in the occurrence of bornite replaced by covellite and inversely quartz–bornite veins cross-cutting quartz–covellite veins. In total, chalcopyrite, bornite, and tennantite successively precipitated at deep, deep-intermediate, and shallow levels, while covellite was found to predominate in areas adjacent to the ore-body centre on the southwest side (Figure 9). Last, the pervasive SO₄²⁻ and CO₃²⁻ bearing gangue minerals with different coexisting sulphides from the AAZ and the PZ, respectively, indicate that the volatile components in the fluids changed considerably when migrating to shallow levels.

Experimental observations suggest that the initial hydrothermal alteration of granitic rocks generally occurs in plagioclase at temperatures of 340–400°C (Boulvais, Ruffet, Cornishet, & Mermet, 2007; Lee & Parsons, 1997; Morad et al., 2010; Yuguchi et al., 2019). In this temperature range, a certain amount of liquid H₂O would occur (Audétat, 2019), which is instrumental to prompting the hydrolysis of volatiles, such as CO₂, H₂S, SO₂, etc. This process would result in the appearance of extra H⁺ ions and other species, such as CO₃²⁻, HCO₃⁻, HS⁻, S₂⁻, etc., and finally form acidic conditions that would substantially accelerate the alteration process (Morad et al., 2010; Seyama et al., 2002). Under acidic conditions, experimental observations have illustrated that biotite chloritization and plagioclase alteration could be readily activated, which is needed to concentrate more Ca²⁺, Mg²⁺, and Fe²⁺/Fe³⁺ in the fluids and results in the formation of sericite, carbonate, chlorite, etc. (Morad et al., 2010). Meanwhile, in porphyry-type or some magmatic fluid–related deposits, ore precipitation temperatures are also in the range of ~350 to ~400°C (Audétat, 2019; Cao et al., 2021; Sillitoe, 2010), which approximately matches the foreside plagioclase alteration temperature. Herein, under slightly acidic and proper temperature conditions, chalcopyrite–sericite–dolomite would first be present in hypogene ores, followed by the precipitation of bornite–sericite (±Fe–Mg carbonate). Moreover, carbonates evolving from Ca–Mg carbonates (dolomite) to Fe–Mg carbonates (siderite, magnesite) are also documented in other porphyry deposits (i.e., Peschanka, Marushchenko et al., 2015).

In the near-surface environment, more H⁺, SO₄²⁻, and H₂S could be produced by SO₂ disproportionation (Simon, Pettke, Candela, Piccoli, & Heinrich, 2007), forming stronger acidic fluids at <300°C (Khashgerel, Kavalieris, & Hayashi, 2008). This important transformation is reflected by an increase in the number and content of Al species and the total Al in solution, which formed abundant Al–(K) silicates (mainly pyrophyllite and kaolinite) ± APS (mainly woodhouseite and svanbergite). These gangue minerals in the AAZ are generally adapted to acidic conditions. At shallower levels, sulphates (dominantly alunite) and Al-silicates (mainly kaolinite) are subsequently precipitated, which suggests the influence of a stronger acidic solution pH. In this episode, tennantite, enargite, bornite, minor anilite, covellite, etc., would precipitate with the decaying heat source (Brimhall & Ghiorso, 1983).

5.2 Trace element distribution

According to the LA–ICPMS element maps and individual spot analyses, Se, Bi, In, Ag, Pb, Au, Te, Ge, Hg, Zn, and Sn are comparatively abundant in the selected sulphides with respect to the surrounding gangue minerals. Most of these trace element concentrations rarely reach hundreds of ppm. As Figure 10 shows, Se concentrations recorded in tennantite (5.9–6.7 ppm) are approximately one order of magnitude lower than that of chalcopyrite (22.1–33.2 ppm), bornite (42.0–64.9 ppm), and covellite (31.5–53.5 ppm), and the lowest levels of Bi are measured in bornite (1.4–1.6 ppm). However, indium contents in the early formed chalcopyrite (4.1–8.0 ppm) present at least one order of magnitude higher than the other three sulphides, in which most measured concentrations are typically <0.6 ppm; silver is conspicuously relatively enriched in bornite (28.3–44.4 ppm) with respect to chalcopyrite (0.8–4.2 ppm), covellite (1.1–3.3 ppm), and tennantite (6.8–26.5 ppm); the measured Pb (647.1–1831.0 ppm), Te (7.7–58.8 ppm), Ge (4.8–19.2 ppm), Hg (2.3–7.9 ppm), and Au (0.5–2.0 ppm) concentrations in tennantite are highest compared to the other three sulphides. Excluding the highly variable Zn and Sn contents due to the inadvertently measured gangue minerals, the two elements are mainly concentrated in tennantite. Overall, although gangue minerals were inevitably measured, some trace elements consistently enriched in sulphides still present notable distribution characteristics in these selected assemblages. Their implications are illustrated below.

First, selenium is uniformly homogeneous in chalcopyrite, bornite, covellite, and tennantite, because this element commonly substitutes for S in most sulphide structures (George, Cook, & Ciobanu, 2017; Wohlgemuth-Ueberwasser, Viljoen, Petersen, & Vorster, 2015). In contrast to these aforementioned cations, this anion is generally present within sulphides at the time of initial deposition, which is not readily subject to subsequent modification as are cations (Owen, Ciobanu, Cook, Slattery, & Basak, 2018). Thus, the selenium contents in these selected sulphides are strongly related to the initial ore-forming fluids and offer the possibility to track a fluid source (Cook et al., 2011). Compositional data for chalcopyrite, bornite, covellite, and tennantite show that the first three sulphides corresponding to the PZ contain almost equivalent Se concentrations, which are remarkably more than those in tennantite from the AAZ, thus suggesting a closer relationship among the first three sulphides. In addition, the relative enrichment of indium in chalcopyrite might be caused by the high proportion of Fe⁺³ and probably Fe⁺² in chalcopyrite (Liu, Chen, & Yang, 2020; Pearce, Pattrick, Vaughan, Henderson, & van der Laan, 2006), which is readily replaced by In⁺³.

Second, although only up to 283.5 ppm, Sn is detected in bornite, which is mainly attributed to the presence of exotic minerals. In Figure 10, the median Sn content of bornite is 1.2 ppm, implying that the selected bornite grains contain low Sn concentrations. Excluding bornite, the highest Sn concentrations are present in tennantite (49.7 ppm), followed by covellite (30.4 ppm) and chalcopyrite (2.9 ppm). Compared to chalcopyrite, notable Sn contents are present in tennantite, which is probably due to the fact that Sn tends to be held and transported in reduced fluids and deposited in oxidized fluids (Eugster, 1986). In other words, instead of incorporating Sn into chalcopyrite, Sn tends to migrate upwards to shallower levels in relatively reduced fluids and then partitions into tennantite from relatively oxidized fluids in the AAZ. Additionally, coexisting alunite also tends to deposit under oxidized conditions (Hedenquist & Taran, 2013). Tellurium and Ge are also preferentially incorporated in tennantite. It is plausible to postulate that the charge imbalance resulting from divalent cations, such as Zn²⁺, Fe²⁺, Pb²⁺, and possibly Cu²⁺, and substitution into the Cu⁺ site leads to the incorporation of Sn⁴⁺, Te⁴⁺, and Ge⁴⁺ via coupled substitution; for instance, As⁵⁺ + Cu⁺ is substituted by Ge⁴⁺+Zn²⁺, as proposed by Bernstein (1985). In summary, the high valences of these metalloids, which precede Sb and As in substituting in tennantite (Johnson, Craig, & Rimstidt, 1986; Krismer, Vavtar, Tropper, Kaindl, & Sartory, 2011; Makovicky & Karup-Møller, 2017), also reflect the presence of a relatively oxidative fluid in the AAZ.

Third, two distribution forms of Au and Ag are summarized in the study. For one situation, Ag and Au are homogeneously distributed in tennantite and bornite with relatively high and consistent signals compared to those in chalcopyrite and covellite. The range of Au concentrations in tennantite and bornite is within 0.5–2.0 ppm and 0.5–1.8 ppm at approximately 310–330°C and 330–360°C, respectively, according to the microthermometry-derived results of a previous study (He et al., 2017). Compared with the experimental Au saturation contents of bornite, namely, 2–4 ppm at 400°C (Simon, Kesler, Essene, & Chryssoulis, 2000), the Au contents in solution likely reach a maximal value during bornite precipitation. The invisible Au and Ag components, which exhibit a homogeneous distribution, are likely incorporated into the Cu–(Fe)–sulphides via substitution with Cu atoms (Catchpole, Kouzmanov, & Fontbote, 2012; Tagirov et al., 2016). For the other situation, Au–Ag trace minerals are present in sulphides. Nano-sized native gold droplets are first found in bornite in the PZ (Figure 2g-2). At shallower levels, the Au–Ag-bearing minerals of micron-scale grain sizes grade to calaverite, hessite, and native gold within tennantite (Figure 2c-2; detailed Au–Ag trace mineral investigation referring to He et al., 2018). The presence of micron-sized Au–Ag tellurides in the AAZ can be attributed to the enrichment of tellurium, which prefers to maintain high concentrations in volatiles (Te: up to 100 ppm in fluid inclusions for forming advanced argillic alteration, Pudack, Heinrich, & Pettke, 2009) that highly accelerate the efficiency of Au precipitation (Cooke & McPhail, 2001). We thus suggest that the enrichment of Au and Ag is affected by Te to a certain degree and is prone to accumulation at relatively shallow depths.