Recognition of Late Triassic Cu-Mo Mineralization in the Northern Yidun Arc (S.E. Tibetan Plateau): Implications for Regional Exploration

: The Yidun arc, located in the southeastern Tibetan Plateau, was formed by the westward subduction of the Ganze-Litang Paleo-Tethys ocean in Late Triassic. It is well-known for the formation of numerous Mesozoic porphyry-skarn Cu-Mo-(Au) deposits in the arc. To date, more than 20 Cu-Mo-(Au) deposits ( > 10 million tonnes Cu resources) have been discovered in the southern Cu-Mo mineralization formed by similar metallogenic event and geodynamic setting as the Cu-Mo-(Au) mineralization in the south. In order to determine the metallogenic age and shed light on potential links between Cu-Mo mineralization and regional magmatic events, we present molybdenite Re-Os and zircon U-Pb ages to constrain the timing of two types of Cu-Mo mineralization in the northern Eastern Yidun arc (type I and type II). Molybdenite ICP-MS Re-Os dating results show that type I mineralization was formed at 217.7 ± 3.6 Ma, which is highly consistent with the formation ages of the host granite (218.1 ± 1.5 Ma, 2 σ , n = 15, MSWD = 0.92) and aplite dyke (217.3 ± 1.3 Ma, 2 σ , n = 16, MSWD = 0.50) within error. While the type II mineralization has a relatively younger formation age of 211.8 ± 4.7 Ma than the host granite (217.1 ± 1.5 Ma, 2 σ , n = 14, MSWD = 0.96) and type I Cu-Mo mineralization. These data indicate that the Cu-Mo mineralization in the northern Eastern Yidun arc was temporally and spatially related to the Late Triassic magmatism in the region. from type ranging from ppm 100 ppm), indicate that the ore-forming were from a Re in molybdenite from the type II mineralization, ranging from 7.983 to 10.40 ppm, that the ore-forming metals were from a mixed mantle and crustal source with a predominantly crustal component. study conﬁrms that the northern Eastern Yidun Late Cu-Mo metallogenesis, and thus much attention should be paid on this region to ﬁnd more Late Triassic Cu-Mo resources.


Introduction
The Yidun arc is one of the largest volcanic island arcs in the Sanjiang Tethyan Metallogenic Domain of the southeastern Tibetan Plateau (Figure 1a) [1,2]. The arc was developed through two phases of magmatic activity in Late Triassic and Late Cretaceous, respectively (Figure 1a), which contributed to the formation of numerous deposits and occurrences [2][3][4]. In the region, more than 20 porphyry, skarn, and quartz-vein type Cu-Mo-(Au-W) polymetallic deposits of various sizes have been discovered and explored (Figure 1b) [5][6][7][8][9][10][11]. The porphyry Cu-Mo-Au deposits include the giant Pulang Cu-Au deposit, the large Xuejiping Cu-Au deposit, and several medium to small Cu deposits (e.g., Lannitang, Chundu, Songnuo) (Figure 1b) [6,12,13]. The skarn Cu-Mo deposits include two large Cu-Mo deposits (Hongshan and Tongchanggou), and several medium to small Cu-Mo deposits (e.g., Langdu, Gaochiping) [8,9]. The quartz-vein type W-Mo deposits include the Relin and Xiuwacu W-Mo deposits with medium scale (Figure 1b) [5,14,15]. However, these deposits are exclusively clustered in the Southern Eastern Yidun Arc (SEYA). Recently, some porphyry Cu-(Mo-Au) deposits (e.g., Changdagou, Zhujiding; Figure 1a) have been successively discovered in the Northern Eastern Yidun Arc (NEYA) [16][17][18], indicating that the NEYA may also have potential to find Cu-Mo resources. However, the ore-forming ages of the Cu-Mo deposits or occurrences have not been well constrained, which hampers our understanding of the relationship between the Cu-Mo mineralization and regional magmatism. In addition, the lack of ore-forming age also obscures our prospecting target where should we put effort into finding Cu-Mo deposit, in late Triassic intrusions or in late Cretaceous intrusions?
Molybdenite Re-Os and zircon U-Pb isotopic systems have high closure temperature and have proven to be powerful tools to determine the precise metallogenic age [5,[19][20][21]. In this study, we present new molybdenite Re-Os and zircon LA-ICP-MS U-Pb ages to constrain the timing of Cu-Mo mineralization and their host granitic rocks, respectively. This data provides first constraints on the timing of Cu-Mo mineralization, the source of ore-forming metals, as well as the implications for the regional exploration.

Regional Geology
The Yidun arc, situated in the southeastern margin of the Tibetan Plateau (Figure 1a). To the west, it is bounded by the Jinshajiang suture which is considered to be a Late Paleozoic Paleo-Tethyan oceanic subduction zone dipping to the west (Figure 1a) [22]. To the east, it is bounded by the Ganze-Litang suture which is considered to be a westward-dipping Paleo-Tethyan oceanic subduction zone during the Middle-Late Triassic (Figure 1a) [22]. The Yidun arc was formed by the westward subduction of the Ganze-Litang Paleo-Tethys ocean in Late Triassic [22][23][24]. Tectonically, the arc can be divided into two principal geological units, the Western Yidun arc (WYA) and Eastern Yidun arc (EYA), by the NNW-trending Xiangcheng-Geza fault (Figure 1a). The WYA, also named as Zhongza massif, consists of Paleozoic shallow to deep marine carbonates and clastic rocks interlayered with volcanic rocks, comparable to the Paleozoic passive continental margin sedimentary sequences of the western Yangtze Block. Based on the similarity of Paleozoic successions and paleontological fossils between the Yidun arc and Yangtze Block, it was traditionally considered that the WYA was rifted from the Yangtze Block during Middle to Late Paleozoic due to the opening of the Ganze-Litang Paleo-Tethyan ocean [25,26]. During Middle to Late Triassic, the Paleozoic sedimentary rocks in the WYA have undergone collision-related greenschist to lower amphibolite facies metamorphism due to the closure of the Jinshajiang Paleo-Tethys ocean, which subsequently led to the collision of WYA with the Qiangtang terrane [1,22]. The EYA consists of sporadically exposed Precambrian metamorphic basement, and Paleozoic to Triassic sedimentary covers [23]. The basement rocks include schist, leptynite, quartzite, marble, and felsic volcanic rocks interlayers [27,28]. The Paleozoic strata are composed of clastic rocks, shallow to deep marine carbonates intercalated with mafic volcanic rocks [23].The Late Triassic strata, from the base upward, include the Qugasi Formation, Tumugou Formation, Lanashan Formation, and Lamaya Formation [23]. The Qugasi Formation is composed of sandstone, slate, phyllite, limestone and mafic arc volcanic rocks [23,29]. These rocks are intruded by~230 Ma (zircon U-Pb age) quartz diorite, indicating that the deposition time of the Qugasi Formation should be older than 230 Ma [29]. The Tumugou Formation, conformably overlying the Qugasi Formation, comprises conglomerate sandstone, slate, limestone, intermediate to felsic arc volcanic rocks (e.g., rhyolite, and andesite) [1,23,29,30]. LA-ICP-MS zircon U-Pb dating show that the volcanic rocks in the lower and upper Tumugou Formation were erupted at~230 Ma and~220 Ma, respectively [1,30], indicating that the Tumugou Formation was deposited at 230-220 Ma.
deposition time of the Qugasi Formation should be older than 230 Ma [29]. The Tumugou Formation, conformably overlying the Qugasi Formation, comprises conglomerate sandstone, slate, limestone, intermediate to felsic arc volcanic rocks (e.g., rhyolite, and andesite) [1,23,29,30]. LA-ICP-MS zircon U-Pb dating show that the volcanic rocks in the lower and upper Tumugou Formation were erupted at ~230 Ma and ~220 Ma, respectively [1,30], indicating that the Tumugou Formation was deposited at 230-220 Ma. The Lanashan Formation is in conformable contact with the Tumugou Formation and consists of sandstone, slate, limestone, mafic volcanic rocks, and conglomerate in the bottom [23,29]. Conformably overlying the Lanashan Formation, the Lamaya Formation is dominant by the dark slate and sandstone [23,29]. The arc volcanic rocks interlayered in the Late Triassic strata have been interpreted as the products of the westward subduction of the Ganze-Litang Paleo-Tethyan ocean [1,30]. The Late Triassic volcanic-sedimentary successions were intruded by voluminous intermediate to felsic intrusions (225-215 Ma; see below). In addition to Tertiary and Quaternary The Lanashan Formation is in conformable contact with the Tumugou Formation and consists of sandstone, slate, limestone, mafic volcanic rocks, and conglomerate in the bottom [23,29]. Conformably overlying the Lanashan Formation, the Lamaya Formation is dominant by the dark slate and sandstone [23,29]. The arc volcanic rocks interlayered in the Late Triassic strata have been interpreted as the products of the westward subduction of the Ganze-Litang Paleo-Tethyan ocean [1,30]. The Late Triassic volcanic-sedimentary successions were intruded by voluminous intermediate to felsic intrusions (225-215 Ma; see below). In addition to Tertiary and Quaternary sediments, other strata younger than Triassic (e.g., Jurassic and Cretaceous) are absent in the whole Yidun arc, though there is belt of Cretaceous granites in the EYA (88-80 Ma; see below). The EYA was collided with the Songpan-Ganze Fold Belt at the end of the Triassic owing to the closure of the Ganze-Litang Paleo-Tethyan ocean [29]. Following the collision of India with Asia during the Tertiary, the Yidun arc was incorporated into the modern Tibetan Plateau [6].
As stated above, magmatism in the EYA was principally emplaced during the Late Triassic and Late Cretaceous times [4,10]. The Late Triassic intrusions intruded into the Upper Triassic volcanic-sedimentary successions in the northern part of Yidun arc. They are composed of biotite monzogranite, granodiorite and quartz diorite. Previous studies show that the intrusions emplaced at ca. 225-215 Ma (zircon U-Pb age) [3,31,32]. These granitic rocks are metaluminous or slightly peraluminous and belong to high-K calc-alkaline I-type granitoid [3], with negative to positive ε Hf(t) values (−9.8 to 3.4) and negative ε Nd(t) values (−7.8 to −5.7) [3]. They have been interpreted to be partial melting products of Late Paleoproterozoic to Early Mesoproterozoic mafic-intermediate lower crust with minor involvement of mantle-derived materials [3]. In the southern segment of the EYA, the Late Triassic magmatic rocks are composed of granodiorite, monzonite, quartz diorite porphyry, quartz monzonite porphyry, andesite, formed at 221-211 Ma (zircon U-Pb ages) [33]. These intermediate-felsic plutonic rocks host numerous porphyry Cu-Mo-Au deposits (e.g., Pulang, Xuejiping, Chundu, Disuga), and skarn Cu-Mo deposits (e.g., Langdu; Figure 1b). The detailed relationship between the Late Triassic magmatism and regional Cu-Mo-Au mineralization has been reviewed by Li et al. [4].
The distribution of the Late Triassic and Late Cretaceous magmatic rocks and their associated deposits are controlled by the Late Triassic NW-trending and NE-trending faults [4]. These NWand NE-trending faults are regarded as the main channel of ore-forming fluid migration [4]. The NW-trending faults are reverse fault, with strike direction of~320-350 • and the dip direction of NE [23]. The NW-trending faults were cut the NE-trending faults (Figures 1b, 2 and 3). The NE-trending faults belong to normal faults. They have strike directions of 65-85 • or 30-50 • , with the dip direction of SE [23].

Sampling and Sample Descriptions
Field and hand sample observations show that the type I Cu-Mo mineralization (100 • 15 31.7"E, 28 • 43 33.8"N; Figure 2) is located within the contact surface between the granite and the granitic aplite (Figure 4a,c-e). The granite intrudes the strata of the Lamaya and Lanashan formations and is itself cut by granitic aplite dykes (Figures 2 and 4a). Molybdenite is the major sulfide in the type I Cu-Mo mineralization and occurs as aggregates or thin coating of the contact surfaces (Figure 4d,e,i). Chalcopyrite and pyrite are also observed in the type I Cu-Mo mineralization (Figure 4i). The type II Cu-Mo mineralization (99 • 54 25.5"E, 31 • 19 19.3"N; Figure 3) is distributed within the veins and cracks in granite (Figure 4f-h). The granite intrudes the strata of Lanashan and Qugasi formations ( Figure 3).

Sampling and Sample Descriptions
Field and hand sample observations show that the type I Cu-Mo mineralization (100°15′31.7″E, 28°43′33.8″N; Figure 2) is located within the contact surface between the granite and the granitic aplite ( Figure 4a,c-e). The granite intrudes the strata of the Lamaya and Lanashan formations and is itself cut by granitic aplite dykes (Figures 2 and 4a). Molybdenite is the major sulfide in the type I Cu-Mo mineralization and occurs as aggregates or thin coating of the contact surfaces (Figure 4d,e,i). Chalcopyrite and pyrite are also observed in the type I Cu-Mo mineralization (Figure 4i). The type II Cu-Mo mineralization (99°54′25.5″E, 31°19′19.3″N; Figure 3) is distributed within the veins and cracks in granite (Figure 4f-h). The granite intrudes the strata of Lanashan and Qugasi formations ( Figure 3).
The main ore mineral assemblages are molybdenite, pyrrhotite, and chalcopyrite. Molybdenite is disseminated or occurs as speckles within the fractures of the host granite (Figure 4j  Accessory minerals include apatite, titanite, and zircon. The aplitic dyke, crosscutting the host granite of type I Cu-Mo mineralization, is light-gray, 5-15 cm wide (Figure 4a,c), and exhibits fine-grained structure and massive texture. It has a sharp contact with the granite (Figure 4a,c).

LA-ICP-MS U-Pb Dating
Zircon grains for Laser Ablation Inductively Coupled Plasma Mass Spectrometry (LA-ICP-MS) U-Pb dating were separated using conventional magnetic and heavy liquid techniques and then handpicked under a binocular microscope. They were then mounted in epoxy resin on a 2 cm diameter disk, which would be polished to section the crystals in half for analyses. Prior to in-situ U-Pb isotopic analyses, all zircons were examined under transmitted and reflected light with an optical microscope at the State key Laboratory of Ore Deposit Geochemistry (SKLODG), Institute of Geochemistry, Chinese Academy of Sciences. Cathodoluminescence (CL) images were obtained using a JSM-7088F type thermal field scanning electron microscope equipped with a Gatan Mono CL4 detector at the SKLODG. Based on the transmitted and reflected light, and BSE observations, the inclusion-free domains were selected for the U-Pb isotopic analyses.
A 7900 ICP-MS (Agilent, Santa Clara, CA, USA) equipped with a GeoLas Pro 193 nm ArF excimer laser at SKLODG was used to measure the U-Pb ages of zircon. Helium was used as carrier gas mixed with argon via a T-connector before entering the spectrometer. A 32 µm laser spot size was selected during the ablation with a repetition rate of 5 Hz. Each analysis consists of 20 s background signal acquisition followed by 50 s ablation signal acquisition. Zircon 91500 was used as external standard to correct elemental fractionation and zircon GJ-1 and Plešovice were analyzed as quality controls. NIST SRM 610 glass was used as external standard to normalize U, Th, Pb contents, with zircon 29 Si concentrations used for internal standardization. The LA-ICP-MS zircon U-Pb dating results of standard zircons are listed in Table 1 Off-line raw data selection and integration of background and analytic signals, time-drift correction and quantitative calibration for U-Pb dating were performed by ICPMSDataCal program [40,41]. The age calculations, the plotting of concordia diagrams (Figure 6a,c,e) and weighted mean age diagrams (Figure 6b,d,f) were made using Isoplot Ver_3.0 [42]. The complete U-Pb dating results from three samples are listed in Table 2.

ICP-MS Re-Os Dating
Molybdenite samples were separated by hand-picking. Fresh and unoxidized molybdenite powders (<0.1 mm in size and purity > 99%) were used for Re-Os isotopic analyses. The Re-Os isotopic analyses were performed at the Re-Os Laboratory of the National Research Center of Geoanalysis, Chinese Academy of Geological Science. Detailed operated processes, including sample preparation, chemical separation and mass spectroscopy, were done according to Du et al. [43] and Shirey and Walker [44]. Re and Os concentrations were determined by TJA X-series ICP-MS. The procedural blanks for this analysis were 0.0010 (±0.0011) for Re and 0.0001 for Os, which are far less than the contents of Re and Os in the analyzed molybdenite samples. The molybdenite standard sample GBW04435(HLP) yielded a model age of 220.5 ± 3.0 Ma, which is in good agreement with the certified value (221.4 ± 5.6 Ma) within error. The molybdenite model age was calculated by the formula of t = [ln (1 + 187 Os/ 187 Re)]/λ, where λ is the 187 Re decay constant of 1.666 × 10 −11 per year [45]. The Re-Os isochron (Figure 7a,c) and weighted mean age (Figure 7b,d) were calculated and plotted by Isoplot Ver_3.0 [42].

Zircon U-Pb Ages
Zircon grains separated from granitic aplite (DC16-65; Figure 2) are mainly colorless, euhedral to subhedral shape, with lengths varying from 50 to 120 µm and the length to width ratios ranging from about 2:1 to 1:1. In the CL images, these zircons show obvious oscillatory zoning (Figure 5a), indicative of the igneous origin [46,47]. Sixteen U-Pb analyses on 16 zircon grains were obtained. The concentrations of Th and U of these zircon grains vary from 648 to 3111 ppm and 1109 to 8616 ppm, respectively, with Th/U ratios varying from 0.34 to 0.76 ( Table 2). All analyses are concordant within analytical errors and yield a concordia age of 217.3 ± 1.4 Ma (MSWD = 0.22, n = 16), with a weighted mean age of 217.3 ± 1.3 Ma (MSWD = 0.50, n = 16) (Figure 6a,b), representing the formation age of the granitic aplite. Zircon grains from sample DC16-68 ( Figure 2) are transparent, euhedral. They have grain sizes ranging from 70 to110 μm in length and from 50 to 80 μm in width, with length/width ratios of 2:1-1:1. CL images show that most of the zircon grains exhibit obvious oscillatory zoning (Figure 5b), indicating the igneous origin [46,47]. The Th/U ratios of zircons vary from 0.32 to 0.77, which further supports the igneous origin [48]. Thirty U-Pb ages were obtained on 30 zircon grains, of which two grains have 206 Pb/ 238 U ages of 505.7 ± 7.0 Ma and 513.0 ± 6.8 Ma, and thirteen zircons have 206 Pb/ 238 U ages ranging from 223.7 ± 5.7 Ma to 234.5 ± 2.9 Ma, and fifteen have 206 Pb/ 238 U ages of 222.2 ± 3.3 Ma to 213.3 ± 3 Ma ( Table 2). The older zircon grains (513.0 ± 6.8 Ma to 223.7 ± 5.7 Ma) are interpreted as inherited or xenocrystic grains that were captured during the ascent of magma. The remaining fifteen zircon grains yield a concordia age of 218.0 ± 1.5 Ma (MSWD = 1.5, n = 15), with a weighted mean 206 Pb/ 238 U age of 218.1 ± 1.5 Ma (MSWD = 0.92, n = 15) (Figure 6c,d), representing the formation Zircon grains from sample DC16-68 ( Figure 2) are transparent, euhedral. They have grain sizes ranging from 70 to110 µm in length and from 50 to 80 µm in width, with length/width ratios of 2:1-1:1. CL images show that most of the zircon grains exhibit obvious oscillatory zoning (Figure 5b), indicating the igneous origin [46,47]. The Th/U ratios of zircons vary from 0.32 to 0.77, which further supports the igneous origin [48]. Thirty U-Pb ages were obtained on 30 zircon grains, of which two grains have 206 Pb/ 238 U ages of 505.7 ± 7.0 Ma and 513.0 ± 6.8 Ma, and thirteen zircons have 206 Pb/ 238 U ages ranging from 223.7 ± 5.7 Ma to 234.5 ± 2.9 Ma, and fifteen have 206 Pb/ 238 U ages of 222.2 ± 3.3 Ma to 213.3 ± 3 Ma ( Table 2). The older zircon grains (513.0 ± 6.8 Ma to 223.7 ± 5.7 Ma) are interpreted as inherited or xenocrystic grains that were captured during the ascent of magma. The remaining fifteen zircon grains yield a concordia age of 218.0 ± 1.5 Ma (MSWD = 1.5, n = 15), with a weighted mean 206 Pb/ 238 U age of 218.1 ± 1.5 Ma (MSWD = 0.92, n = 15) (Figure 6c,d), representing the formation age of this sample. These zircon grains generally show clear oscillatory zoning (Figure 5c) and high Th/U ratios (0.38-0.87) ( Table 2), indicating the magmatic origin [46][47][48]. Twenty-six U-Pb ages were obtained on 26 zircon grains from this sample, and their 206 Pb/ 238 U ages vary from 296. 8   Zircon grains from sample GZ16-40 ( Figure 3) are colorless and transparent, with euhedral and prismatic morphology. They are mostly ranging from 80 to 120 µm in length and from 60 to 80 µm in width, with the length/width ratio of 2:1-1.2:1.

Molybdenite Re-Os Ages
The Re-Os analytical results for seven molybdenite samples are given in Table 3 Table 3). The Re-Os model ages range from 214.8 ± 3.3 Ma to 222.3 ± 3.9 Ma with an average mean age of 217.7 ± 3.6 Ma. Four molybdenite samples yield a 187 Re-187 Os isochron age of 220 ± 18 Ma, which is coherent well with the weight mean age of 217.7 ± 3.6 Ma (Figure 7a,b).  (Figure 7d). This age agrees well with the 187 Re-187 Os isochron age of 205 ± 11 Ma (MSWD = 2.8) within error (Figure 7c), and represents the crystallization age of molybdenite in this type of Cu-Mo mineralization. well with the weight mean age of 217.7 ± 3.6 Ma (Figure 7a,b).
The total Re, 187 Re and 187 Os concentrations of molybdenite collected from type II Cu-Mo mineralization vary from 7.983 to 10.40 ppm, 5.018 to 6.536 ppm, and 17.81 to 22.89 ppb, respectively (Table 3). Three molybdenite samples yield relatively tight Re-Os model age of 212.6 ± 3.4 Ma, 213.4 ± 3.6 Ma, and 209.8 ± 3.1 Ma, with an average mean age of 211.8 ± 4.7 Ma (Figure 7d). This age agrees well with the 187 Re-187 Os isochron age of 205 ± 11 Ma (MSWD = 2.8) within error (Figure 7c), and represents the crystallization age of molybdenite in this type of Cu-Mo mineralization.

Timing of Magmatism and Cu-Mo Mineralization
Zircon U-Pb and molybdenite Re-Os dating are commonly used to determine the formation age of ore-related intrusion and ore-forming age of hydrothermal deposit [5,6,8,11,34]. The zircon LA-ICP-MS U-Pb dating results show that the host rocks of granitic aplite and granite of type I Cu-Mo mineralization were formed at 217.3 ± 1.3 Ma and 218.1 ± 1.5 Ma, respectively (Figure 6b,d). And the host rock of granite of type II mineralization has a similar formation age of 217.1 ± 1.5 Ma (Figure 6f) to the host rocks of type I mineralization. These LA-ICP-MS zircon ages are highly consistent with previous dating results [2,3,32,49]. The ore-forming ages of two types of Cu-Mo mineralization have a relatively large age interval. Molybdenite separated from contact surface between granite and granitic aplite (i.e., type I Cu-Mo mineralization) occurred at 217.7 ± 3.6 Ma (Figure 7b), which is coeval with the emplacement age of granite (218.1 ± 1.5 Ma) and granitic aplite (217.3 ± 1.3 Ma). However, the molybdenite of type II Cu-Mo mineralization yield a younger average mean age of 211.8 ± 4.7 Ma (Figure 7d), which postdates the emplacement event of host granite (217.1 ± 1.5 Ma). The younger molybdenite Re-Os age of 211.8 ± 4.7 Ma may indicate that the type II Cu-Mo mineralization was formed in the end stage of magmatic evolution. This is supported by the geological observation that the type II mineralization occurs near granitic pegmatite (Figure 4b).

Source of Ore-Forming Metals
Re contents in molybdenite from different hydrothermal deposits vary greatly. Berzina et al. [50] suggested that Re contents of molybdenite may be related to the concentration of Re in ore-forming fluid, the composition of parent magma, physical-chemical conditions (e.g., temperature, pressure, and f O 2 ) of crystallization, and sources of ore-forming materials. Based on systematic and comprehensive investigations on different types of endogenous Mo deposit in China, Mao et al. [19] demonstrated that Re content in molybdenite decreases from a mantle source (>100 ppm), to a mixed mantle/crustal source (10-100 ppm), and to a crustal source (<10 ppm). Similarity, Stein et al. [21] proposed that molybdenite from deposits involved mantle metasomatism, underplating, or melting of mafic/ultramafic rocks generally have high Re contents. In addition, the study showed that deposits originating from intermediate crustal rocks or organic-poor sedimentary rocks are expected to have low Re content [21].
The Re contents in molybdenite from type I Cu-Mo mineralization show large variation ranging from 12.77 ppm to 111.1 ppm (Table 3). Three out of four molybdenites with model ages of 222.3-215.2 Ma have high Re contents, indicating the ore-forming materials were derived from a mantle source. One sample (DC16-67) with youngest model age of 214.8 ± 3.3 Ma has lowest Re content of 12.77 ppm, which probably indicative of crustal contamination of the mineralizing fluid (at 214.8 ± 3.3 Ma) after the peak of the magmatic pulse responsible for granite formation and emplacement (at 218.1 ± 1.5 Ma). The type II Cu-Mo mineralization has relatively constant and low Re contents (from 7.983 ppm to 10.40 ppm; Table 3), indicative of a mixed mantle and crustal source with a predominantly crustal component.
In this study, the type I and tye II Cu-Mo mineralization were dated at 217.7 ± 3.6 Ma and 211.8 ± 4.7 Ma, respectively, which coincides well with the metallogenic ages of Late Triassic Cu-Mo-(Au) mineralization in the SEYA (Table 4). Therefore, we consider that the Cu-Mo mineralization in the NEYA is likely related to the subduction of Ganze-Litang Paleo-Tethys ocean slab in Late Triassic. The subduction-related granitoid in the NEYA is widely exposed, with the outcrops more than 5200 km 2 [3]. Previous studies revealed that these rocks belong to high-K calc-alkaline I-type granitoid, which is favorable to the formation of Cu-Mo deposits [64,65]. Except for the two types Cu-Mo mineralization presented in this study, some porphyry Cu deposits (e.g., Changdagou, Zhujiding; Figure 1a) have also been documented in the NEYA [16][17][18]66]. Previous studies show that the ore-bearing porphyries of Changdagou porphyry Cu deposit were formed at 216-208 Ma [18,52], which were synchronously formed to the ore-related intrusions in the SEYA. In addition, the Changdagou and Zhujiding porphyry Cu deposit develop similar hydrothermal alteration (e.g., silification, propylitization, phyllic and potassic alteration), and ore fabric features (e.g., veinlet and disseminated ore structures) to those of porphyry Cu-Mo deposits in the SEYA [16][17][18]52,66]. Recognition of these porphyry Cu deposits and Cu-Mo mineralization indicate that the NEYA exists Late Triassic Cu-Mo metallogenesis. Therefore, a renewed exploration should be encouraged to find late Triassic Cu-Mo resources in the NEYA.   Table 4. The Late Triassic (221-213 Ma) mineralization (e.g., Pulang porphyry Cu-Mo-Au deposit) are genetically associated with Late Triassic subduction-related intermediate-felsic porphyritic intrusions [4,6,13,24]. The Late Cretaceous (88-80 Ma) mineralization, including the large Hongshan and Tongchang Cu-Mo deposits, have close temporal and spatial relationships with Late Cretaceous I-type granitoids [4,9,11,14,15]. The formation of the Late Triassic and Late Cretaceous Cu-Mo-(Au) mineralization has been attributed to the westward subduction of Ganze-Litang Paleo-Tethys ocean slab in Late Triassic, and crustal extension in Late Cretaceous, respectively [4,6,10,12,15].
In this study, the type I and tye II Cu-Mo mineralization were dated at 217.7 ± 3.6 Ma and 211.8 ± 4.7 Ma, respectively, which coincides well with the metallogenic ages of Late Triassic Cu-Mo-(Au) mineralization in the SEYA (Table 4). Therefore, we consider that the Cu-Mo mineralization in the NEYA is likely related to the subduction of Ganze-Litang Paleo-Tethys ocean slab in Late Triassic. The subduction-related granitoid in the NEYA is widely exposed, with the outcrops more than 5200 km 2 [3]. Previous studies revealed that these rocks belong to high-K calc-alkaline I-type granitoid, which is favorable to the formation of Cu-Mo deposits [64,65]. Except for the two types Cu-Mo mineralization presented in this study, some porphyry Cu deposits (e.g., Changdagou, Zhujiding; Figure 1a) have also been documented in the NEYA [16][17][18]66]. Previous studies show that the ore-bearing porphyries of Changdagou porphyry Cu deposit were formed at 216-208 Ma [18,52], which were synchronously formed to the ore-related intrusions in the SEYA. In addition, the Changdagou and Zhujiding porphyry Cu deposit develop similar hydrothermal alteration (e.g., silification, propylitization, phyllic and potassic alteration), and ore fabric features (e.g., veinlet and disseminated ore structures) to those of porphyry Cu-Mo deposits in the SEYA [16][17][18]52,66]. Recognition of these porphyry Cu deposits and Cu-Mo mineralization indicate that the NEYA exists Late Triassic Cu-Mo metallogenesis. Therefore, a renewed exploration should be encouraged to find late Triassic Cu-Mo resources in the NEYA.

Conclusions
Molybdenite Re-Os dating indicates that the type I and type II Cu-Mo mineralization occurred at 218 Ma and~212 Ma, respectively. Zircon LA-ICP-MS U-Pb dating shows that the granite and granitic aplite associated with type I Cu-Mo mineralization were formed at 218.1 ± 1.5 Ma and 217.3 ± 1.3 Ma, respectively. The host granite of type II Cu-Mo mineralization was formed at 217.1 ± 1.5 Ma. Re content in molybdenite suggests that the ore-forming materials of type I Cu-Mo mineralization were derived from a mantle source, while the type II Cu-Mo mineralization was sourced from a mixed mantle and crustal source. The relatively low Re contents and younger Re-Os ages of molybdenite in type II mineralization may indicate that type II Cu-Mo mineralization was formed in the late stage of magmatic evolution, accompanying with addition of crustal-derived materials. This study provides significant evidence to support that the NEYA hosts Late Triassic Cu-Mo mineralization. Recognition of Late Triassic porphyry Cu deposits and Cu-Mo mineralization in the NEYA should encourage renewed investigations and ore prospecting in the NEYA to find late Triassic Cu-Mo deposits.