Mineralization Age and Hydrothermal Evolution of the Fukeshan Cu (Mo) Deposit in the Northern Great Xing’an Range, Northeast China: Evidence from Fluid Inclusions, H–O–S–Pb Isotopes, and Re–Os Geochronology

: The Fukeshan Cu (Mo) deposit is a newfound porphyry deposit in the northern Great Xing’an Range (GXR), northeast China. In this paper, we present results of chalcopyrite Re–Os geochronology, microthermometry of the ﬂuid inclusions (FIs), and isotopic (H–O–S–Pb) compositions of the Fukeshan Cu (Mo) deposit. Its ore-forming process can be divided into sulﬁde-barren quartz veins (A vein; stage I), quartz + chalcopyrite + pyrite veins (B vein; stage II), quartz + polymetallic sulﬁde veins (D vein; stage III), and barren quartz + carbonate ± pyrite veins (E vein; stage IV), with Cu mineralization mainly occurred in stage II. Three types of FIs are identiﬁed in this deposit: liquid-rich two-phase (L-type) FIs, vapor-rich two-phase (V-type) FIs, daughter mineral-bearing three-phase (S-type) FIs. The homogenization temperatures of primary FIs hosted in quartz of stages I–IV are 381–494 ◦ C, 282–398 ◦ C, 233–340 ◦ C, and 144–239 ◦ C, with salinities of 7.2–58.6, 4.8–9.9, 1.4–7.9, and 0.9–3.9 wt. % NaCl equivalent, respectively. FIs microthermometry and H–O isotope data suggest that the ore-forming ﬂuids were magmatic in origin and were gradually mixed with meteoric water from stages II to IV. Sulfur and lead isotope results indicate that the ore-forming materials of the Fukeshan Cu (Mo) deposit were likely to have originated from Late Jurassic intrusive rocks. The available data suggest that ﬂuid cooling and incursions of meteoric water into the magmatic ﬂuids were two important factors for Cu precipitation in the Fukeshan Cu (Mo) deposit. Chalcopyrite Re–Os dating yielded an isochron age of 144.7 ± 5.4 Ma, which is similar to the zircon U–Pb age of the quartz diorite porphyry, indicating that Late Jurassic quartz diorite porphyry and Cu mineralization occurred contemporaneously. on detailed core microthermometry of the ﬂuid inclusions, isotopic (H–O–S–Pb) compositions, and chalcopyrite Re–Os geochronology. These results allowed us to further discuss the evolution of the ore-forming ﬂuids and the possible metallogenic mechanism. This information can provide further prospecting direction of Late Mesozoic porphyry deposits in the northern GXR.

The newfound Fukeshan porphyry Cu (Mo) deposit is located in the northern GXR. Previous researches on the Fukeshan deposit were mainly concentrated on the geological characteristics [20,21], and the geochronology and geochemistry of intrusions [20,22]. These studies have reached a consensus that the Fukeshan Cu (Mo) deposit is a porphyry deposit. However, the source of the ore-forming materials, and the origin and evolution of the ore-forming fluids have not been constrained.
To address these issues, based on detailed core logging, this paper presents microthermometry of the fluid inclusions, isotopic (H-O-S-Pb) compositions, and chalcopyrite Re-Os geochronology. These results allowed us to further discuss the evolution of the ore-forming fluids and the possible metallogenic mechanism. This information can provide further prospecting direction of Late Mesozoic porphyry deposits in the northern GXR.   [22]), showing the distribution of major ore deposits.

Ore Deposit Geology
The Fukeshan Cu (Mo) deposit is located~50 km southwest of the Mohe city, Heilongjiang Province ( Figure 1C). NW-trending faults are developed in the deposit (Figure 2A). Multiphase intrusive rocks are observed in the Fukeshan mining area, including Early Jurassic medium-coarse-grained monzogranite (ca. 192 Ma), Late Jurassic medium-grained granodiorite (ca. 149 Ma) and quartz diorite porphyry (ca. 148 Ma), as well as Early Cretaceous intermediate and acidic dykes (Figure 2A,B) [22]. The quartz diorite porphyry intrudes the medium-coarse-grained monzogranite and medium-grained granodiorite rocks, Cu (Mo) mineralization is mainly developed in quartz diorite porphyry and its surrounding rock (Figure 2A,B), therefore the quartz diorite porphyry is considered to be the main host rock of the Cu (Mo) mineralization and closely related to hydrothermal alterations of the deposit. The quartz diorite porphyry is dark gray in color and has a massive structure with a porphyritic texture, it consists of 30-35% phenocrysts and 65-70% cryptocrystalline matrix, and phenocrysts are dominantly composed of plagioclase (20-25%), quartz (5-8%), and biotite (5-8%). A number of ore bodies have been found in the Fukeshan Cu (Mo) deposit, with Cu grades of 0.20-1.16% and Mo grades of 0.03-0.32% [22]. This deposit currently contains inferred ore reserves of >200,000 t, the reserves will increase with ongoing exploration.
The mineralization and alteration characteristics of the Fukeshan Cu (Mo) deposit have been discussed based on the short wavelength infrared (SWIR) analysis [21], in this study, based on detailed field observations, mineral assemblages, and previous studies [21], alteration in the Fukeshan Cu (Mo) deposit can be divided, from early to late, into potassic, silicic, chlorite-illite/sericite, phyllic, and carbonate alterations. However, the alteration zonation is not obvious due to superposition of multiple post-metallogenic magmatism. Potassic and silicic alterations are mainly distributed in the quartz diorite porphyry and its surrounding rocks ( Figure 3A-C). Potassic alteration are characterized by the mineral assemblage of secondary K-feldspar and biotite, irregular secondary K-feldspar was intensively developed in the quartz diorite porphyry, with fuzzy boundary between the crystals ( Figure 3A). Metallic minerals that coexist with potassic and silicic alterations are mainly composed of magnetite, chalcopyrite, and minor molybdenite ( Figures 3B and 4A-C). Chlorite-illite/sericite alteration contains chlorite, illite/sericite, and minor epidote, it only locally overprints the potassic and silicic alteration ( Figure 3D,E). The main Cu mineralization is closely associated with potassic, silicic, and chlorite-illite/sericite alterations. Phyllic alteration is characterized by the mineral assemblage of quartz and sericite. Metallic minerals that coexist with phyllic alteration are mainly composed of pyrite, chalcopyrite, molybdenite, as well as minor sphalerite and galena. Phyllic alteration overprints the preexisting potassic, silicic, and chlorite-illite/sericite alterations. Sericite can completely/partially replace feldspars ( Figure 3F). Minor veined and disseminated Cu (Mo) mineralization develop in the phyllic alteration zone. The mineralization and alteration characteristics of the Fukeshan Cu (Mo) deposit have been discussed based on the short wavelength infrared (SWIR) analysis [21], in this study, based on detailed field observations, mineral assemblages, and previous studies [21], alteration in the Fukeshan Cu (Mo) deposit can be divided, from early to late, into potassic, silicic, chlorite-illite/sericite, phyllic, and carbonate alterations. However, the alteration zonation is not obvious due to superposition of multiple post-metallogenic magmatism. Potassic and silicic alterations are mainly distributed in the quartz diorite porphyry and its surrounding rocks ( Figure  3A-C). Potassic alteration are characterized by the mineral assemblage of secondary K-feldspar and biotite, irregular secondary K-feldspar was intensively developed in the quartz diorite porphyry, with fuzzy boundary between the crystals ( Figure 3A). Metallic minerals that coexist with potassic and silicic alterations are mainly composed of magnetite, chalcopyrite, and minor molybdenite (Figures 3B and 4A-C). Chlorite-illite/sericite alteration contains chlorite, illite/sericite, and minor epidote, it only locally overprints the potassic and silicic alteration ( Figure 3D,E). The main Cu mineralization is closely associated with potassic, silicic, and chlorite-illite/sericite alterations. Phyllic alteration is characterized by the mineral assemblage of quartz and sericite. Metallic minerals that coexist with phyllic alteration are mainly composed of pyrite, chalcopyrite, molybdenite, as well as minor sphalerite and galena. Phyllic alteration overprints the preexisting potassic, silicic, and chlorite-illite/sericite alterations. Sericite can completely/partially replace feldspars ( Figure 3F). Minor veined and disseminated Cu (Mo) mineralization develop in the phyllic alteration zone.       Based on mineral assemblages and crosscutting relationships between different veins in the Fukeshan Cu (Mo) deposit (Figure 4), we have identified four types of veins (i.e., A, B, D, and E veins) [35]. The vein sequence corresponds to four mineralization stages, from early to late: sulfide-barren quartz veins (A vein; stage I), quartz + chalcopyrite + pyrite veins (B vein; stage II), quartz + polymetallic sulfide veins (D vein; stage III), and barren quartz + carbonate ± pyrite veins (E vein; stage IV) ( Figure 5). Their characteristics are described as follows.

Fluid Inclusion Measurements
Fluid inclusions (FIs) petrography, laser Raman spectra, and microthermometry analyses were performed at the Key Laboratory of Mineral Resources Evaluation in Northeast Asia, Ministry of Land and Resources, Changchun, China. A total of 36 quartz samples collected from drilling operations were prepared as two-sided, 0.2 mm-thick polished sections. FIs were examined via petrographic observation under a binocular microscope, the liquid phase and the gaseous phase components of the representative FIs were analyzed using laser Raman spectroscopy by an RM-2000 laser Raman microprobe (Renishaw, New Mills, UK). The laser source is an Ar ion laser, with a wavelength of 514 nm and a laser power of 20 Mw. The laser beam width was 1 μm and the spectral resolution was 0.14 cm −1 . Then, the representative primary FIs were selected for microthermometric analyses. Secondary FIs were not analyzed [36]. Microthermometric analyses were conducted using a Linkam THMS-600 (Linkam Scientific Instruments Ltd., Epsom, UK) heating-freezing stage. The heating rates were generally 0.2-5.0 °C /min but were reduced to 0.2 °C /min when phase transitions were approached, and 1 °C /min when approaching FIs homogenization temperatures. The estimated precision of the measurements is ±0.1 °C for final ice melting temperatures and ±2 °C for the homogenization temperatures.

Fluid Inclusion Measurements
Fluid inclusions (FIs) petrography, laser Raman spectra, and microthermometry analyses were performed at the Key Laboratory of Mineral Resources Evaluation in Northeast Asia, Ministry of Land and Resources, Changchun, China. A total of 36 quartz samples collected from drilling operations were prepared as two-sided, 0.2 mm-thick polished sections. FIs were examined via petrographic observation under a binocular microscope, the liquid phase and the gaseous phase components of the representative FIs were analyzed using laser Raman spectroscopy by an RM-2000 laser Raman microprobe (Renishaw, New Mills, UK). The laser source is an Ar ion laser, with a wavelength of 514 nm and a laser power of 20 Mw. The laser beam width was 1 µm and the spectral resolution was 0.14 cm −1 . Then, the representative primary FIs were selected for microthermometric analyses. Secondary FIs were not analyzed [36]. Microthermometric analyses were conducted using a Linkam THMS-600 (Linkam Scientific Instruments Ltd., Epsom, UK) heating-freezing stage. The heating rates were generally 0.2-5.0 • C/min but were reduced to 0.2 • C/min when phase transitions were approached, and 1 • C/min when approaching FIs homogenization temperatures. The estimated precision of the measurements is ±0.1 • C for final ice melting temperatures and ±2 • C for the homogenization temperatures.

H-O Isotope Analyses
Four representative quartz samples were collected from quartz + K-feldspar + pyrite vein (stage I), quartz + chalcopyrite vein (stage II), quartz + molybdenite + pyrite vein (stage III), as well as quartz + calcite vein (stage IV). The H isotope compositions of FIs hosted in quartz samples and the O isotope compositions of quartz samples were analyzed at the Center of Analytical Laboratory in the Beijing Research Institute of Uranium Geology, China National Nuclear Corporation (CNNC). Quartz veins from different mineralization stages were crushed to quartz grains (each 0.25-0.35 mm in size). Then quartz grains were carefully handpicked under a binocular microscope to remove visual impurities and ensure that the purity was not less than 95% and washed with double distilled water. H and O isotope compositions were analyzed using a Finnigan MAT253-EM mass spectrometer. The O isotope compositions of quartz were analyzed using the conventional BrF 5 method [37]. The H isotope compositions of FIs hosted in quartz were analyzed using the Zn reduction method [38]. The H and O isotope values were normalized with Vienna Standard Mean Ocean Water (V-SMOW) standards, with an analytical precision greater than 2% for δD and greater than 0.2% for δ 18 O.

S Isotope Analyses
Nine sulfide samples, which were collected from quartz + K-feldspar + pyrite veins (stage I), quartz + chalcopyrite ± pyrite veins (stage II), as well as quartz + molybdenite + pyrite and quartz + chalcopyrite + sphalerite + galena veins (stage III), were selected for S isotope analysis. The S isotopes of the samples were analyzed at CNNC. The individual metal sulfide mineral (purity > 99%) was mixed with cuprous oxide and powdered to 200 mesh (0.076 mm). After the samples were mixed evenly, they were heated at 980 • C under a vacuum pressure of 2 × 10 −2 Pa to generate SO 2 gas. The SO 2 gas was collected by freezing in a vacuum. Then, S isotopes were analyzed using a Delta V Plus mass spectrometer. Sulfur isotope ratios are reported in the standard notation as per mil (% ) deviations from the sulfur isotope composition of the Vienna Cañon Diablo Troilite (V-CDT) standard, with an analytical precision greater than 0.2% . The sulfide reference materials were the GBW-04414 and GBW-04415 Ag sulfide standards, with δ 34 S values determined in this study of −0.07 ± 0.13% and 22.15 ± 0.14% , respectively. And the accuracy with respect to the sulfide reference materials was greater than 99%.

Pb Isotope Analyses
Four sulfide and one quartz diorite porphyry samples were selected for Pb isotope analysis, also analyzed at CNNC. Four sulfide samples were collected from quartz + K-feldspar + pyrite vein (stage I), quartz + chalcopyrite vein (stage II), as well as quartz + molybdenite + pyrite and quartz + chalcopyrite + galena veins (stage III). Approximately 30-100 mg of powdered sample was dissolved in Teflon bombs with an ultrapure HF + HNO 3 mixture. The samples were dried after complete dissolution. The residue was then dissolved in an HBr + HNO 3 mixture and loaded into a column of AG 1-X8 anionic resin. The extracted Pb was then purified in a second column. About 100 ng of Pb was loaded onto single rhenium filaments using the silica gel technique. The Pb isotopes were analyzed with an ISOPROBE-T thermal ionization mass spectrometer, with an analytical precision greater than 0.09%. The measured Pb isotope ratios were corrected for instrumental mass fractionation of 0.11% per atomic mass unit by comparisons with repeated analyses of the standard sample (NBS-981).

Chalcopyrite Re-Os Dating
Six chalcopyrite samples, which were collected from quartz + chalcopyrite veins (stage II), were collected for Re-Os dating. Chalcopyrite grains (each 0.25-0.35 mm in size) were magnetically separated and handpicked under a binocular microscope to ensure a purity higher than 99%. The Re-Os isotope analyses were conducted at the Re-Os Isotope Laboratory, National Research Center of Geoanalysis, Chinese Academy of Geological Sciences (CAGS). The detailed procedures used here were described in [39]. The chalcopyrite standard GBW04477/JCBY [40] was used in this study to control reproducibility and instrument stability. Isoplot/Ex ver. 3.0 [41] was used to calculate the Re-Os isochron age.

Petrography and Types of Fluid Inclusion
Petrographic studies indicated that primary fluid inclusions (FIs) in four mineralization stages were distributed regularly or randomly along growth bands in quartz crystals, and occasionally existed as isolations, showing characteristic of primary FIs [36]. Secondary FIs are usually smaller than primary FIs, and generally occur along fractures or grain boundaries in clusters and linear arrays. All FIs data collected in this study were from primary FIs. At room temperature (25 • C), three types of FIs were determined by examining their phases, filling degree, and assemblage relations. The specific characteristics of the FIs are as follows.
Liquid-rich two-phase (L-type) FIs consist of vapor and liquid water (as indicated by Raman spectroscopy) with filling degrees of 70-95 vol % ( Figure 6A,C-E), in other words, the vapor phase mainly occupies 5-30 vol % of the inclusion volume. L-type FIs from stages II and III show various vapor phase proportion but display relatively constant vapor phase proportion in stages I and IV. L-type FIs generally display relatively larger vapor phase proportion (15-30 vol %) in stage I, with a few up to 45 vol % of vapor phase proportion ( Figure 6A). However, L-type FIs from stage IV show relatively lower vapor phase proportion (5-15 vol %) ( Figure 6E). L-type FIs range in size from 5 to 15 µm and exhibit native-crystal, irregular or round shapes. L-type FIs commonly homogenize to liquid during heating. L-type FIs are the most common type in all mineralization stages ( Figure 6A,C-E).

Petrography and Types of Fluid Inclusion
Petrographic studies indicated that primary fluid inclusions (FIs) in four mineralization stages were distributed regularly or randomly along growth bands in quartz crystals, and occasionally existed as isolations, showing characteristic of primary FIs [36]. Secondary FIs are usually smaller than primary FIs, and generally occur along fractures or grain boundaries in clusters and linear arrays. All FIs data collected in this study were from primary FIs. At room temperature (25 °C), three types of FIs were determined by examining their phases, filling degree, and assemblage relations. The specific characteristics of the FIs are as follows.
Liquid-rich two-phase (L-type) FIs consist of vapor and liquid water (as indicated by Raman spectroscopy) with filling degrees of 70-95 vol % ( Figure 6A,C-E), in other words, the vapor phase mainly occupies 5-30 vol % of the inclusion volume. L-type FIs from stages II and III show various vapor phase proportion but display relatively constant vapor phase proportion in stages I and IV. L-type FIs generally display relatively larger vapor phase proportion (15-30 vol %) in stage I, with a few up to 45 vol % of vapor phase proportion ( Figure 6A). However, L-type FIs from stage IV show relatively lower vapor phase proportion (5-15 vol %) ( Figure 6E). L-type FIs range in size from 5 to 15 μm and exhibit native-crystal, irregular or round shapes. L-type FIs commonly homogenize to liquid during heating. L-type FIs are the most common type in all mineralization stages ( Figure  6A,C-E).  Vapor-rich two-phase (V-type) FIs are identified in stages I, II, and III. They consist of vapor and liquid water (as indicated by Raman spectroscopy) with filling degrees of 10-30 vol %, in other words, the vapor phase generally occupies 70-90 vol % ( Figure 6A,C,D). They are normally round or oval in shape and 8-16 µm in size. V-type FIs commonly homogenize to vapor during heating. V-type FIs in different stages show various vapor phase proportion ( Figure 6A,C,D).
Daughter mineral-bearing three-phase (S-type) FIs are only observed in stage I, and consist of liquid, a vapor bubble, and one or more solid daughter minerals ( Figure 6B). They mainly occur in irregular or oval shapes and vary from 5 to 15 µm in size, with a few up to 20 µm. The daughter minerals observed are halite and unidentified metallic minerals ( Figure 6B). The halite daughter minerals are generally transparent and angular cube shapes ( Figure 6B), and metallic minerals are opaque. During heating, the halite daughter minerals generally dissolved after the vapor bubble disappeared. Table 1 lists the microthermometric data for primary FIs from four different mineralization stages, which are also shown in Figure 7. The liquid phase and the gaseous phase components of S-type FIs from stage I, as well as the gaseous phase components of L-and V-type FIs from four different mineralization stages, were analyzed using laser Raman spectroscopy. The spectra indicate that the liquid and vapor phases are almost entirely H 2 O (Figure 8). On freezing/warming, eutectic first melting temperatures of the FIs from stages I to III ranged from −27.9 to −23.7 • C, indicating the fluid is mainly composed of Na-chloride solutions with low concentrations of K and Mg cations [42]. However, the liquid phase of FIs from stage IV are too small to observe the eutectic temperature. According to the analysis of the eutectic temperatures, we can assume the fluids as approximating the simple H 2 O-NaCl system, so we used the program HOKIEFLINCS_H 2 O-NaCl to estimate the salinities of L-and V-type FIs [43]. As most S-type FIs showed homogenization by halite dissolution in the Fukeshan Cu (Mo) deposit, a direct application of halite dissolution temperature can lead to uncertainties [44,45]. Therefore, we calculated the salinities of S-type FIs according to the equations of [46].
We observed L-, and V-type FIs in stage III quartz grains ( Figure 6D). During heating, all L-type FIs homogenized to the aqueous liquid phase at 233-340 • C (n = 50), peaking at 260-280 • C ( Figure 7E). These L-type FIs yielded T m-ice ranging from −5.0 to −0.8 • C, corresponding to salinities of 1.4-7.9 wt. % NaCl equivalent (n = 50), concentrated between 4 and 6 wt. % NaCl equivalent ( Figure 7F). The V-type FIs homogenized to the vapor phase at 261-334 • C (n = 19) ( Figure 7E). They yielded T m-ice varying from −4.0 to −1.5 • C, with calculated salinities of 2.6 to 6.4 wt. % NaCl equivalent (n = 19) ( Figure 7F). Tm-eu = eutectic temperature (first ice melting temperature); Tm-ice = temperature of final ice melting; Th-s = dissolution temperature of halite; Th-v = homogenization temperature of V-type or L-type FIs, vapor disappearance temperature of S-type FIs; and Qz = quartz.  Tm-eu = eutectic temperature (first ice melting temperature); Tm-ice = temperature of final ice melting; Th-s = dissolution temperature of halite; Th-v = homogenization temperature of V-type or L-type FIs, vapor disappearance temperature of S-type FIs; and Qz = quartz.   Tm-eu = eutectic temperature (first ice melting temperature); Tm-ice = temperature of final ice melting; Th-s = dissolution temperature of halite; Th-v = homogenization temperature of V-type or L-type FIs, vapor disappearance temperature of S-type FIs; and Qz = quartz.

H-O Isotopes
The H and O analytical results are listed in Table 2 and shown in Figure 9. The measured δ 18 O V-SMOW values of four quartz samples from four mineralization stages range from 6.2 to 9.0% . The quartz-water equilibrium equation was used to calculate the δ 18 O H2O values (1000lnα quartz-water = 3.38 × 10 6 /T 2 − 3.4) [47]. The calculated δ 18 O H2O values from stages I to IV are 5.6% , 1.2% , −1.0% , and −6.2% , respectively ( Table 2). The measured δD values of FIs in quartz from stages I to IV are -109.9% , −125.7% , −140.9% , and −152.3% , respectively ( Table 2).    Table 3 lists the S isotope compositions of nine sulfide samples (including pyrite, chalcopyrite, molybdenite, sphalerite, and galena) from mineralization stages I to III in the Fukeshan deposit, which are shown in Figure 10. The δ 34 SV-CDT values of the sulfides from mineralization stages I to III range from 3.3‰ to 3.4‰, −2.3‰ to 1.7‰, and 0.3‰ to 1.1‰, respectively (Table 3). Table 3. S isotope data of sulfide minerals from the Fukeshan Cu (Mo) deposit.  Table 3 lists the S isotope compositions of nine sulfide samples (including pyrite, chalcopyrite, molybdenite, sphalerite, and galena) from mineralization stages I to III in the Fukeshan deposit, which are shown in Figure 10. The δ 34 S V-CDT values of the sulfides from mineralization stages I to III range from 3.3% to 3.4% , −2.3% to 1.7% , and 0.3% to 1.1% , respectively (Table 3).   Table 4 lists the Pb isotope compositions of four sulfide samples (including pyrite, chalcopyrite, molybdenite, and galena) from mineralization stages I to III in the Fukeshan deposit, which are plotted in Figure 11. The Pb isotope compositions of the sulfides have limited variation in their 206 Table 4 lists the Pb isotope compositions of four sulfide samples (including pyrite, chalcopyrite, molybdenite, and galena) from mineralization stages I to III in the Fukeshan deposit, which are plotted in Figure 11. The Pb isotope compositions of the sulfides have limited variation in their 206 [55]. Pb isotope data for Late Jurassic-Early Cretaceous ore in the GXR are from [51,52,56]. Pb isotope data for Mesozoic igneous rocks in the GXR are from [57].  Table 5 presents the Re-Os isotope compositions of six chalcopyrite samples from mineralization stage II in the Fukeshan deposit. The concentrations of Re and common Os in the samples range from 0.4177 to 3.7997 ng/g and from 0.0021 to 0.0515 ng/g, respectively. Their 187 Re/ 188 Os values range from 130 to 5997, and their 187 Os/ 188 Os ratios range from 0.738 to 15.1. The data for the six chalcopyrite samples yielded an isochron age of 144.7 ± 5.4 Ma (MSWD = 3.8, initial 187 Os/ 188 Os = 0.45 ± 0.05; Figure 12), which represents the Cu mineralization age of the Fukeshan Cu (Mo) deposit.  Table 5 presents the Re-Os isotope compositions of six chalcopyrite samples from mineralization stage II in the Fukeshan deposit. The concentrations of Re and common Os in the samples range from 0.4177 to 3.7997 ng/g and from 0.0021 to 0.0515 ng/g, respectively. Their 187 Re/ 188 Os values range from 130 to 5997, and their 187 Os/ 188 Os ratios range from 0.738 to 15.1. The data for the six chalcopyrite samples yielded an isochron age of 144.7 ± 5.4 Ma (MSWD = 3.8, initial 187 Os/ 188 Os = 0.45 ± 0.05; Figure 12), which represents the Cu mineralization age of the Fukeshan Cu (Mo) deposit.  6. Discussion

Timing of Cu (Mo) Mineralization
The ore bodies in the Fukeshan deposit are mainly hosted within the quartz diorite porphyries ( Figure 2B). Geochronology data indicate that chalcopyrite Re-Os isochron age (144.7 ± 5.4 Ma) corresponds well with the zircon U-Pb weighted mean age of the Fukeshan quartz diorite porphyry (148.7 ± 0.8 Ma) [22], indicating that the quartz diorite porphyry and Cu mineralization in the Fukeshan Cu (Mo) deposit occurred contemporaneously. In recent years, the Late Jurassic porphyry Cu (Mo) mineralization events have been increasingly reported in the northern GXR, such as the Chalukou porphyry Mo deposit (148 ± 1 Ma) [17], the Xiaokele porphyry Cu (Mo) deposit (150.0 ± 1.6) [19], and the Daheishan (147 ± 2 Ma) porphyry Cu (Mo) deposit [18]; these results indicate that Late Jurassic was a crucial time for porphyry Cu (Mo) mineralization in the northern GXR.

Timing of Cu (Mo) Mineralization
The ore bodies in the Fukeshan deposit are mainly hosted within the quartz diorite porphyries ( Figure 2B). Geochronology data indicate that chalcopyrite Re-Os isochron age (144.7 ± 5.4 Ma) corresponds well with the zircon U-Pb weighted mean age of the Fukeshan quartz diorite porphyry (148.7 ± 0.8 Ma) [22], indicating that the quartz diorite porphyry and Cu mineralization in the Fukeshan Cu (Mo) deposit occurred contemporaneously. In recent years, the Late Jurassic porphyry Cu (Mo) mineralization events have been increasingly reported in the northern GXR, such as the Chalukou porphyry Mo deposit (148 ± 1 Ma) [17], the Xiaokele porphyry Cu (Mo) deposit (150.0 ± 1.6) [19], and the Daheishan (147 ± 2 Ma) porphyry Cu (Mo) deposit [18]; these results indicate that Late Jurassic was a crucial time for porphyry Cu (Mo) mineralization in the northern GXR.

Source of the Ore-Forming Materials
The δ 34 S values of sulfides from mineralization stages I to III in the Fukeshan Cu (Mo) deposit have a narrow range of −2.3% to 3.4% (Table 3; Figure 10), reflecting a remarkably homogeneous sulfur source. There are no sulphate minerals in the Fukeshan Cu (Mo) deposit, which indicates weak sulfur fractionation in various valence states during different mineralization stages. The δ 34 S values are near 0% , indicating a magmatic source and are interpreted to reflect uniformity of the ore-forming fluids (−5% to 5% ) [58]. The δ 34 S values of the sulfides in the Fukeshan deposit overlap with those of typical porphyry deposits in the world ( Figure 10) [59], such as Butte porphyry Cu-Mo deposit in USA (−0.1% to 4.7% ) [50], Wunugetushan (0.76% to 3.20% ) [53], and Badaguan (−2.4% to 3.5% ) [54]. In addition, the δ 34 S values of the sulfides in the Fukeshan deposit match the sulfur isotope compositions of Late Jurassic porphyry deposits in the northern GXR, such as Chalukou Mo (−1.9% to 3.6% ) [52], Xiaokele Cu (Mo) (−1.2% to 2.4% ; our unpublished data), and Daheishan Cu (Mo) (0.4% to 2.3% ) [51] (Figure 10). All of these deposits have sulfur origins from Late Jurassic intermediate-acid porphyries (150-147 Ma). Considering the intrusive magmatic activities and the chalcopyrite Re-Os isochron age (144.7 ± 5.4 Ma) in the Fukeshan deposit, we suggest that the sulfur in the ores was most likely derived from Late Jurassic intrusive rocks.
In order to further evaluate the metal source, we made use of previously published Pb isotope data, including Pb isotope data for Mesozoic ore in the GXR [51,52,56], and Pb isotope data for Mesozoic igneous rocks in the GXR [57] (Figure 11). The Pb isotope ratios of Mesozoic igneous rocks in the GXR are mainly distributed between upper crust and mantle values, implying a mixed source [58] ( Figure 11). Significantly, the Pb isotope ratios of Mesozoic ore in the GXR plot in the lead region of Mesozoic igneous rocks, suggesting that the source of ore lead was most likely related to Mesozoic igneous rocks. Likewise, Pb isotope data from the Fukeshan deposit all plot in the lead region of an orogenic belt between upper crust and mantle sources, indicating that Pb was most likely derived from a deep-seated magma that involved some crustal materials [54]. Pb isotope ratios of the sulfides from different mineralization stages in the Fukeshan deposit are uniform, suggesting the hydrothermal fluids were derived from a well-mixed and homogeneous magma [59]. We have obtained only one set of Pb isotope data for the quartz diorite porphyry from the Fukeshan deposit, but Pb isotope compositions of the quartz diorite porphyry do not overlap with those of sulfides well ( Figure 11). Even so, in the Fukeshan deposit, the Pb isotope ratios of the sulfides and the quartz diorite porphyry plot in the lead region of Mesozoic ore and igneous rocks in the GXR (Figure 11), suggesting that the Pb isotope compositions of the sulfides and the quartz diorite porphyry may have shared a common Pb source with the Mesozoic igneous rocks in the GXR. Considering Late Jurassic ore-bearing intrusive rocks and mineralization age (chalcopyrite Re-Os isochron age of 144.7 ± 5.4 Ma) in the Fukeshan deposit, we suggest that the lead and ore-forming metals were most likely derived from Late Jurassic intrusive rocks.

Fluid Immiscibility and Pressure Estimates
In the Fukeshan deposit, fluid immiscibility most likely happened in stage I, supported by the following evidence (1) S-type and V-type FIs coexisting in the same quartz crystal show close genetic relationship, and (2) S-type and V-type FIs display similar homogenization temperatures but contrasting salinities ( Figure 7A,B). Trapping pressure can be accurately estimated when the fluid immiscibility was occurring in the fluid system at the capture time [60]. Considering the evidence of fluid immiscibility in stage I, based on the simple H 2 O-NaCl system, trapping pressures in stage I are estimated to range from~200 to~500 bar ( Figure 13) using the isobar equations reported in [61].
L-type and V-type FIs assemblage from stages II to IV are not the real immiscibility assemblage in porphyry systems and they are not suitable to estimate the accurate trapping pressure conditions.
Pressures obtained for such assemblage can only represent minimum trapping values [62]. Based on the simple H 2 O-NaCl system, using the isobar equations reported in [61], we estimate the minimum trapping pressures in stages II and III range from~70 to~300 bar and~30 to~150 bar, respectively ( Figure 13). The minimum trapping pressures in stage IV occurred at pressures <50 bar ( Figure 13). L-type and V-type FIs assemblage from stages II to IV are not the real immiscibility assemblage in porphyry systems and they are not suitable to estimate the accurate trapping pressure conditions. Pressures obtained for such assemblage can only represent minimum trapping values [62]. Based on the simple H2O-NaCl system, using the isobar equations reported in [61], we estimate the minimum trapping pressures in stages II and III range from ~70 to ~300 bar and ~30 to ~150 bar, respectively ( Figure 13). The minimum trapping pressures in stage IV occurred at pressures <50 bar ( Figure 13).

Origin and Evolution of the Ore-Forming Fluids
FIs provide a record of ore-forming fluid systems, and investigations of FIs can yield certain indications as to the nature and genesis of the original fluids [36,63]. The δ 18 OH2O values for stage I veins (5.6‰) are similar to magmatic water values [51], suggesting that the ore-forming fluids from the early stage derived from a magmatic source, but the δD values (−109.9‰) of the ore-forming fluids in stage I are obviously lower than typical magmatic water values (Figure 9) [64]. Previous studies have suggested that the low δD values in most porphyry deposits may be due to magma degassing and fluid boiling [65][66][67]. During the late crystallization period, continuous degassing of parent magma in an open system would reduce the δD values of the residual water with little influence on the δ 18 OH2O values [68,69]. Furthermore, fluid boiling occurred in stage I, vapor separation caused by fluid boiling would also decrease the δD values in the remaining ore-forming fluids [70]. Therefore, the low δD values in stage I are likely a consequence of magma degassing and fluid boiling. In addition, previous studies have also shown that such depleted δD isotope values are widely recorded by ore-forming fluids in the early stage of other porphyry deposits in the GXR [7], such as the Hashitu porphyry Mo deposit [71], Wulandele porphyry Mo deposit [72], and Chalukou porphyry Mo deposit [55], which likely have a predominantly magmatic origin. The samples from stages II to IV display relatively lower δD and δ 18 OH2O values than those from stage I, plot in the region between the meteoric water line and the magmatic water field (Figure 9), this indicates the involvement of meteoric water. In addition, ore-forming fluids from stages II to IV have characteristics of middle-low salinities and are different from typical exsolved magmatic fluids, which are characterized by high temperature and high salinity [73]. This also implies that meteoric water may be involved in the ore-forming fluids, especially in the late stage. In addition, the bulk fluid inclusion decrepitation method, which was employed to provide the water for the δD analyses, inevitably led to sampling a mixture of primary and secondary fluid inclusions, resulting in the low δD values from stages I to IV.
The development of the L-, V-and S-type FIs in stage I hydrothermal quartz suggests that the

Origin and Evolution of the Ore-Forming Fluids
FIs provide a record of ore-forming fluid systems, and investigations of FIs can yield certain indications as to the nature and genesis of the original fluids [36,63]. The δ 18 O H2O values for stage I veins (5.6% ) are similar to magmatic water values [51], suggesting that the ore-forming fluids from the early stage derived from a magmatic source, but the δD values (−109.9% ) of the ore-forming fluids in stage I are obviously lower than typical magmatic water values (Figure 9) [64]. Previous studies have suggested that the low δD values in most porphyry deposits may be due to magma degassing and fluid boiling [65][66][67]. During the late crystallization period, continuous degassing of parent magma in an open system would reduce the δD values of the residual water with little influence on the δ 18 O H2O values [68,69]. Furthermore, fluid boiling occurred in stage I, vapor separation caused by fluid boiling would also decrease the δD values in the remaining ore-forming fluids [70]. Therefore, the low δD values in stage I are likely a consequence of magma degassing and fluid boiling. In addition, previous studies have also shown that such depleted δD isotope values are widely recorded by ore-forming fluids in the early stage of other porphyry deposits in the GXR [7], such as the Hashitu porphyry Mo deposit [71], Wulandele porphyry Mo deposit [72], and Chalukou porphyry Mo deposit [55], which likely have a predominantly magmatic origin. The samples from stages II to IV display relatively lower δD and δ 18 O H2O values than those from stage I, plot in the region between the meteoric water line and the magmatic water field (Figure 9), this indicates the involvement of meteoric water. In addition, ore-forming fluids from stages II to IV have characteristics of middle-low salinities and are different from typical exsolved magmatic fluids, which are characterized by high temperature and high salinity [73]. This also implies that meteoric water may be involved in the ore-forming fluids, especially in the late stage. In addition, the bulk fluid inclusion decrepitation method, which was employed to provide the water for the δD analyses, inevitably led to sampling a mixture of primary and secondary fluid inclusions, resulting in the low δD values from stages I to IV.
The development of the L-, V-and S-type FIs in stage I hydrothermal quartz suggests that the initial hydrothermal fluids are characterized by high homogenization temperatures and high salinities ( Figure 7A,B). Observations of mineral association, such as hematite and magnetite ( Figure 4A,B), but rare sulfides in stage I veinlets, indicate highly oxidized initial hydrothermal fluids [74], typical features of porphyry-type deposits [75,76]. Because of the high oxygen fugacity, the sulfur within the magma was mostly present as sulfate [53,77]. Sulfide mineralization, however, was inevitably constrained as sulfur in ore minerals, mainly existing in the reduced form of S 2− , although fluid boiling may have occurred in stage I. L-and V-type FIs were observed in quartz crystals from stages II and III ( Figure 6C,D). Compared with stage I, hematite and magnetite were not observed in stages II and III, indicating that the oxygen fugacity of the fluids was distinctly lower than in stage I. Most of the stage IV quartz crystals contain single L-type FIs (Figure 6E), where the ore-forming fluids belonged to an homogeneous H 2 O-NaCl system with medium-low temperature (144-239 • C) and low salinity (0.9-3.9 wt. % NaCl equivalent). This implies that the fluids in stage IV were more diluted and cooled due to a large amount of meteoric water flow during stage IV. This conclusion is consistent with results of H−O isotope compositions in this study.

Mechanism for Cu Transportation and Precipitation
Most researchers have proposed that the transportation of metals mainly occurs as complexes in aqueous liquids [64,[78][79][80]. Cu transportation process is still a controversial issue although much research has been done on it. Most studies believe that Cu exists as chlorine complex and migrates in the brine phase [78,80]. However, many LA-ICP-MS (laser ablation-inductively coupled plasma-mass spectrometry) analysis and simulation experiments also supported that Cu is preferentially partitioned into the vapor phase rather than brine by diffusion [81][82][83]. In this study, most sulfide minerals are present in the S-type FIs rather than high vapor-liquid ratio FIs during different mineralization stages, indicating that Cu chloride complexes are the predominant species in the brine phase.
As described above, the initial hydrothermal fluids in stage I have high oxygen fugacity in the Fukeshan deposit. In contrast, stage II is the main Cu mineralization stage in the Fukeshan deposit, which is demonstrated to be relatively less oxidizing. A decrease in the oxygen fugacity could have resulted from either CO 2 -escape that was effectively caused by fluid boiling [84][85][86] and/or magnetite crystallization [87][88][89]. Our results show that the fluid system is not a CO 2 -rich system. Thus, the unmixing between water and CO 2 may not have been an important process in the Fukeshan deposit. As a result, magnetite precipitation in stage I could have led to the reduction of S 6+ to S 2− (12[FeO] + H 2 SO 4 = 4Fe 3 O 4 + H 2 S) [77,89], providing suitable condition for significant sulfide mineralization [89]. In this study, two main mechanisms have been proposed as the effective agents in the Cu precipitation in the Fukeshan deposit. The first factor is that Cu solubility (as chloride complex) decreases dramatically with decline of temperatures [90][91][92]. FIs studies show that homogenization temperatures decrease from 381-494 • C in stage I to 282-398 • C in stage II in the Fukeshan deposit (Table 1). In addition, Pb-Zn mainly precipitate in stage III with homogenization temperatures vary from 233 to 340 • C (Table 1), but only a small amount of Cu precipitate in this temperature interval. The homogenization temperatures range in the main Cu mineralization stage (stage II) are consistent with the most favorable temperature interval (~400-300 • C) for Cu precipitation in porphyry systems [90][91][92][93][94]. These temperature conditions are also in agreement with the thermodynamic calculations of the chloride complex in the Sungun porphyry Cu deposit (northwestern Iran) [93]. Furthermore, previous research has found that the Cu solubility in aqueous fluids was approximately 50,000 ppm at higher temperatures (~450 • C) and dropped sharply to approximately 50 ppm at about 360 • C [93]. Therefore, fluid cooling may play a very important role in precipitation of Cu precipitation in the Fukeshan deposit. In addition, decrease in temperature has a major effect on the disproportionation of SO 2 to H 2 S and H 2 SO 4 (4SO 2 + 4H 2 O → 3H 2 SO 4 + H 2 S) [95,96]. The disproportionation reaction is only effective below 400 • C [97]. This reaction resulted in an increase in the reduced S species (H 2 S) in the fluids, which can decrease the solubility of chalcopyrite and trigger Cu precipitation [93,[98][99][100]. The second factor is fluid mixing between magmatic fluids and meteoric water, which has long been thought to effectively lead to the deposition of metals from ore solutions [101]. Combined with the H-O isotope data in this study, it can be concluded that the mixing between magmatic fluids and meteoric water has occurred since stage II. In addition, from stages II to IV, the salinities of the ore-forming solutions decrease with the decreasing of temperature, indicating the input of meteoric water with low-salinities and low-temperatures [102]. Fluid mixing would decrease chloride ion concentration of the ore-forming fluid, thus destabilizing the chloride complex of Cu and depositing Cu [93]. Incursions of cooler meteoric water into the magmatic fluids may have also assisted in the temperature decrease and further promoted Cu precipitation, as discussed already.

Conclusions
(1) The Fukeshan Cu (Mo) deposit is a typical porphyry deposit in the northern GXR. Its ore-forming process can be divided into sulfide-barren quartz veins (A vein; stage I), quartz + chalcopyrite + pyrite veins (B vein; stage II), quartz + polymetallic sulfide veins (D vein; stage III), and barren quartz + carbonate ± pyrite veins (E vein; stage IV), with Cu mineralization mainly occurred in stage II.
(2) Microthermometry of the FIs and H-O isotope data suggest that the ore-forming fluids were magmatic in origin and were gradually mixed with meteoric water from stages II to IV. Sulfur and lead isotopic results indicate that the ore-forming materials of the Fukeshan Cu (Mo) deposit were likely to originate from Late Jurassic intrusive rocks.
(3) Fluid cooling and incursions of meteoric water into the magmatic fluids were two important factors for Cu precipitation in the Fukeshan Cu (Mo) deposit.
(4) Late Jurassic quartz diorite porphyry and Cu mineralization occurred contemporaneously in the Fukeshan Cu (Mo) deposit.