Volatile variations in magmas related to porphyry Cu-Au deposits: Insights from amphibole geochemistry, Duolong district, central Tibet

Accepted Manuscript Volatile variations in magmas related to porphyry Cu-Au deposits: Insights from amphibole geochemistry, Duolong district, central Tibet Jin-Xiang Li, Ke-Zhang Qin, Guang-Ming Li, Noreen J. Evans, Jun-Xing Zhao, Ya-Hui Yue, Jing Xie PII: DOI: Reference:

S0169-1368(17)30820-X https://doi.org/10.1016/j.oregeorev.2018.03.019 OREGEO 2535

To appear in:

Ore Geology Reviews

Received Date: Revised Date: Accepted Date:

2 November 2017 14 March 2018 21 March 2018

Please cite this article as: J-X. Li, K-Z. Qin, G-M. Li, N.J. Evans, J-X. Zhao, Y-H. Yue, J. Xie, Volatile variations in magmas related to porphyry Cu-Au deposits: Insights from amphibole geochemistry, Duolong district, central Tibet, Ore Geology Reviews (2018), doi: https://doi.org/10.1016/j.oregeorev.2018.03.019

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Volatile variations in magmas related to porphyry Cu-Au deposits: Insights from amphibole geochemistry, Duolong district, central Tibet

Jin-Xiang Li a, b*, Ke-Zhang Qin b, c, Guang-Ming Li b, c, Noreen J. Evans d, Jun-Xing Zhao c, Ya-Hui Yue a, Jing Xie a

a

Key Laboratory of Continental Collision and Plateau Uplift, Institute of Tibetan

Plateau Research, Chinese Academy of Sciences, Beijing 100101, China b

CAS Center for Excellence in Tibetan Plateau Earth Sciences, Beijing 100101,

China c

Key Laboratory of Mineral Resources, Institute of Geology and Geophysics, Chinese

Academy of Sciences, Beijing 100029, China d

John de Laeter Center, TIGeR, Applied Geology, Curtin University, Perth, WA 6945,

Australia

*Corresponding author. Present address: Key Laboratory of Continental Collision and Plateau Uplift, Institute of Tibetan Plateau Research, Chinese Academy of Sciences, Beijing 100101, China. Tel.: +86 10 84097034; fax: +86 10 84097079; E-mail: [email protected]. 1

ABSTRACT Ore-forming fluid exsolution in a shallow magma chamber is a critical step in the formation of porphyry Cu±Mo±Au deposits but one for which direct evidence is rarely found. Here, Cl abundance, major-trace element content and H isotope analysis of amphibole in diorite and barren granodiorite porphyry samples from the Duolong porphyry-epithermal Cu-Au district are presented in order to reveal processes associated with fluid exsolution and magma evolution. Low-Al Mg-hornblende formed in ore-bearing diorite at ~ 750-860 °C and ~ 80-200 MPa. On primitive mantle-normalized diagrams, these low-Al amphibole grains have slightly convex upward REE patterns with distinctly negative Eu anomalies and negative anomalies in Pb, Sr, Eu, Zr, Hf, and Ti anomalies, suggesting crystallization from the same arc magma after plagioclase and magnetite crystallization. A large variation in estimated melt H2O (~ 6 to 3 wt. %) and Cl content (0.09-0.38 wt. %), as well as low δD values (-103 to -113‰) indicate that the magma underwent large-scale fluid exsolution, contributing to the formation of Duolong Cu-Au mineralization. Additionally, amphibole from a barren granodiorite porphyry shows two distinct populations, distinguished by their Al content. This intrusion has low-Al amphibole (Mg-hornblende) which formed at similar conditions to low-Al amphiboles in the diorite (~ 790-870 °C and ~ 100-230 MPa), whereas high-Al amphibole (tschermakite) formed at ~ 880-970 °C and ~ 210-400 MPa. Some high-Al amphibole phenocrysts have slightly convex REE patterns with no negative Eu anomalies, a depletion in Nb, Zr, and Hf, and positive Sr, Ba, and Pb anomalies, likely consistent with amphibole crystallization from more mafic basaltic-andesitic melts. Low-Al amphibole in the granodiorite porphyry shows different compositional trends compared to those in the diorite, suggesting that they crystallized from different magmas. Considering the

2

evidence for mafic magma replenishment suggested by the positive correlation between AlIV and Mg# values in high-Al amphibole, the low-Al amphibole in the barren granodiorite porphyry likely crystallized from a new hybrid magma with mafic to intermediate-felsic magma compositions. Meanwhile, smaller variations in Cl content (0.08-0.24 wt. %) in low-Al amphibole was dominantly controlled by magma evolution (shown by variable Mg#) rather than fluid exsolution. Thus, the low δD values (-102 to -122‰) were likely inherited from evolved dioritic magmas that underwent fluid exsolution. Importantly, the Cl compositional variation of amphibole has a potential application as a powerful tool to identify ore-bearing and barren intrusions within porphyry Cu±Mo±Au deposits.

Keywords: Amphibole; H isotope; Fluid exsolution; Duolong; Tibet

1. Introduction It is well known that magmatic volatiles (e.g., chlorine, water, and sulfur) play a major role in the formation of porphyry-epithermal Cu ± Mo ± Au deposits (e.g., Shinohara and Hedenquist, 1997; Sillitoe, 2010; Richards, 2011). Chlorine (Cl) is incompatible in the melt and is partitioned strongly into an exsolved fluid phase that can effectively transport ore metals (e.g, Cu) to form mineralization (Heinrich et al., 1999; Webster, 2004). Generally, the volatiles directly exsolved from the underlying

3

magma chamber sequester metals (e.g, Cu, Au) to form the ore-forming fluids, which ascend and give rise to porphyry-epithermal ore deposits surrounding shallow porphyries (e.g., Hedenquist and Lowenstern, 1994; Sillitoe, 2010; Richards, 2011). Most studies suggesting that ore-forming fluids can exsolve from magmas have relied on thermodynamic modeling, fluid inclusions in shallow mineralized veins, or experimental investigations (e.g., Candela and Holland, 1986; Cline and Bodnar, 1991; Shinohara and Hedenquist, 1997; Heinrich et al., 1999; Harris et al., 2003; Webster, 2004; Williams-Jones and Heinrich, 2005; Zajacz et al., 2012a). However, direct evidence of fluid exsolution in the underlying magma chamber is rarely observed and, therefore, the role of magmatic volatiles in melt evolution and fluid exsolution are poorly understood (Sillitoe, 2010; Richards, 2011). Moreover, multi-phase porphyries generally occur in porphyry-epithermal Cu ± Mo ± Au deposits, but only one or two phase porphyries (ore-bearing) are genetically related with mineralization (Sillitoe, 2010; Richards, 2011). The factors controlling the formation of ore-bearing and barren porphyries are unclear. Amphibole, a common mineral in magmatic rocks, can be used to track variations in magmatic volatile contents and to establish magma conditions at the time of crystallization because its composition responds to redox state, temperature, pressure, Cl and H2O content, and melt composition (e.g., Sato et al., 2005; Humphreys et al., 2009; Ridolfi et al., 2010; Krawczynski et al., 2012; Nandedkar et al., 2016; Bao et al., 2016; Duan and Jiang, 2017; Chelle-Michou and Chiaradia, 2017). Moreover, hydrogen (H) isotope analysis of OH-bearing amphibole can shed light on magma degassing and fluid exsolution processes, as these cause depleted H isotopic compositions compared to normal magmas (Taylor et al., 1983; Underwood et al., 2013; Chambefort et al., 2013). In porphyry-related hydrothermal systems,

4

amphibole is rarely preserved in extensively altered ore-bearing porphyritic rocks, but fresh amphibole in cogenetic ore-bearing plutons provides a good substitute (e.g., Dilles, 1987; Zhang et al., 2017). This study focuses on the texture and composition of amphibole in diorite (ore-bearing) and granodiorite porphyry (barren) from the giant Duolong porphyry-high sulfidation epithermal Cu-Au district, central Tibet (Fig. 1). Major and trace element analysis, as well as H isotope analysis of amphibole is presented in order to decipher the evolution of magmatic volatiles and reveal magma conditions during fluid exsolution. The results will help elucidate the key factors controlling the formation of ore-bearing and barren porphyries. 2. Geological setting The Tibetan plateau consists of five terranes (from north to south: Qaidam, Songpan–Ganze, Qiangtang, Lhasa, and India), separated by four Mesozoic and Cenozoic sutures (Ayimaquin–Kunlun, Jinshajiang, Bangong–Nujiang, and Indus–Yarlung, respectively; Fig. 1a, b; Yin and Harrison, 2000; Kapp et al., 2005; Zhang et al., 2012; Zhu et al., 2013). The Qiangtang terrane is divided by the Mesozoic Longmu-Shuanghu suture (LSS) into the northern and southern Qiangtang terrane (Fig. 1b; Zhai et al., 2011; Zhang et al., 2012; Zhu et al., 2013). The Bangong–Nujiang suture zone is characterized by a >1200 km-long east-west belt of mainly Jurassic–Cretaceous flysch, mélange, and ophiolitic fragments, and represents the remnants of a Bangong-Nujiang ocean basin (Pan et al., 2012; Shi et al., 2008; Yin and Harrison, 2000). The ages of ophiolites and radiolarians from this zone indicate that the Bangong–Nujiang ocean existed at least from the Carboniferous to the Late Cretaceous (Pan et al., 2012; Shi et al., 2008; Huang et al., 2015a; Fan et al., 2015; Wang et al., 2016). During northward subduction of the Bangong-Nujiang oceanic lithosphere, Jurassic-Cretaceous arc magmatism, including intermediate–felsic

5

intrusions and volcanic rocks, was extensive in the southern Qiangtang terrane (Guynn et al., 2006; Zhang et al., 2012; Hao et al., 2016; Huang et al., 2015b; Li et al., 2014a, 2014b, 2016c; Liu et al., 2014; Geng et al., 2016; Wu et al., 2016; Zhu et al., 2016). The giant Duolong porphyry–high sulfidation epithermal Cu–Au district, the focus of this study, is closely related to Early Cretaceous arc magmatism (~ 118 Ma; Li et al., 2013, 2016a, 2016b, 2017; Zhu et al., 2015; Hou and Zhang, 2015; Sun et al., 2017). The Duolong Cu–Au district (prospective metal resources of ~ 25 Mt Cu and ~ 400 t Au; Tang et al., 2016; Li et al., 2016b) covers more than 120 km2, and includes the Duolong (Duobuza and Bolong) and Naruo porphyry deposits and the Rongna and Nadun porphyry-high sulfidation epithermal deposits (Fig. 1c). The stratigraphy mainly comprises the Late Triassic Riganpeicuo Group, Middle Jurassic Quse Group, Late Cretaceous Meiriqie Group, and Neogene Kangtuo Group (Fig. 1c). The Late Triassic Riganpeicuo Group is predominantly comprised of thick-bedded carbonate deposits whereas the Middle Jurassic Quse Group is a clastic-interbedded volcanic sequence of littoral facies composed of arkosic sandstone and siltstone-interbeded siliceous rock. The Late Cretaceous Meiriqie Group contains basaltic andesite, dacite, volcanic-clastic rocks, andesite porphyry and andesite. The Neogene Kangtuo Group is composed of brown-red clay and sandy gravel (Fig. 1c). The intrusive rocks mainly comprise diorite and ore-bearing and barren granodiorite porphyries, which intruded into the Middle Jurassic Quse Group as stocks and dykes (Fig. 1c). Ore-bearing and barren granodiorite porphyries (~ 60-64 wt.% SiO2 ; Li et al., 2013, 2016a, 2017; Sun et al., 2017) show porphyritic texture and are mineralogically similar consisting of amphibole (10-15 vol.%), biotite (5-10 vol.%), plagioclase (45-50 vol.%), quartz (20-30 vol.%), and K-feldspar (5-10 vol.%),

6

with accessory zircon, apatite, titanite, and Fe-Ti oxides (each <1 vol.%). Ore-bearing granodiorite porphyry shows extensive potassic and argillic alteration, including secondary K-feldspar, biotite, chlorite, and clay minerals. Fresh diorites (~ 55-62 wt.% SiO2 ; Li et al., 2014a, 2017) have medium-grained equigranular textures and contain the same minerals as granodiorite porphyries but with a higher proportion of mafic minerals (20-25 vol.% amphibole and 10-15 vol.% biotite). Precise zircon U–Pb dating results (Li et al., 2011a, 2013, 2017; Sun et al., 2017; Zhu et al., 2015) indicated that these intrusions were synchronously emplaced at ~118 ± 1.5 Ma, consistent within error with the molybdenite Re-Os age (~119 ± 2.0 Ma; Sun et al., 2017; Zhu et al., 2015). This suggests there is a genetic relationship between the intrusions and the Cu-Au mineralization. Additionally, pre-ore Early Cretaceous volcanic basalt (40Ar/39Ar plateau age of ~142 Ma; Li et al., 2013) has alkaline characteristics with an enrichment of high field strength elements (HFSE: Nb and Ta) and large ion lithophile elements (LILE: Rb and Ba; Li et al., 2016a), consistent with OIB-type magmas. Cretaceous basalt, basaltic andesite and andesite from the Late Cretaceous Meiriqie Group are syn- and post-ore volcanic rocks with zircon U–Pb ages of ~105-118 Ma (Li et al., 2011a, 2016a). Syn- and post-ore volcanic rocks, diorite, and ore-bearing/barren granodiorite porphyry samples show similar geochemical features with high-K calc-alkaline series classification, enrichment in large ion lithophile elements (LILE: such as Rb and Ba), and depletion of high field strength elements (HFSE: such as Nb and Ta; Li et al., 2013, 2016a), consistent with arc magmas. 3. Sampling and analytical methods 3.1. Sampling and petrography Amphiboles were identified from six diorite (including drillhole GCZK01) and

7

five barren granodiorite porphyry samples (Fig. 1c) in the Duolong ore district. Amphibole in fresh diorite is euhedral and/or subhedral and commonly associated with magnetite and apatite (Fig. 2a, b). Most amphibole from the fresh barren granodiorite porphyry is euhedral with some subhedral or anhedral grains, and occasionally inclusions of magnetite, apatite, and plagioclase were noted (Fig. 2c, d, h). One amphibole phenocryst displayed a homogeneous core grading out to a weakly oscillatory zoned rim (Fig. 2f). Additionally, some amphibole phenocrysts in the barren granodiorite porphyry samples have a breakdown rim that commonly consists of fine-grained plagioclase and Fe-Ti oxides (Fig. 2e). In another breakdown texture, plagioclase and Fe-Ti oxides occur as patches and along cleavages in amphibole (Fig. 2c, g). 3.2. Analytical methods Amphibole was analyzed for major and trace elements both in thin sections and in epoxy grain mounts. Major elements were measured at the Institute of Tibetan Plateau Research, Chinese Academy of Sciences (ITPCAS), using a JXA-8230 Electron Microprobe Analyzer (EMPA) with accelerating voltage of 15 kV, a beam current of 20 nA and a spot size of 5 μm. Kα lines for each element are analyzed. Data were corrected on-line using a modified ZAF (atomic number, absorption, fluorescence) correction procedure. The peak counting time was 10 s and background counting time was 5 s on the high- and low-energy background positions. The following standards were used: Jadeite (Na, Al), Olivine (Mg), Diopside (Si, Ca), Orthoclase (K), Rutile (Ti), Rhodonite (Mn), Hematite (Fe), Fluorite (F), and NaCl (Cl). Analytical precision for Cl is ± 0.01 wt. % and 0.01-0.2 wt. % for other elements. Trace elements were determined using an Agilent 7500 inductively coupled

8

plasma mass spectrometer (ICP-MS) in combination with a NewWave 193 nm ArF excimer laser ablation system at ITPCAS. A laser repetition rate of 6~8 Hz with ~10 J/cm2 fluence and spot size of 40-60 μm were used. The ICP-MS was tuned to obtain < 0.5 % oxide production as monitored by ThO/Th. Every 5 sample analyses were followed by analysis of NIST 610 and NIST 612 standard glasses. Quantitative concentrations were calculated using GLITTER4.0 (Macquarie University) with NIST 610 as the primary standard and Ca (determined by EMPA) as the internal reference element. NIST 612 was treated as an unknown throughout the analysis and elemental abundances were within 5-10 % of recommended values. Major and trace amphibole data are provided Supplementary Data Table 1. Hydrogen isotope analyses of nine amphibole separates (diorite and barren granodiorite porphyry) were performed on the MAT 253 mass spectrometer using the closed tube technique (Vennemann and O’Neil, 1993) at the Analytical Laboratory Beijing Research Institute of Uranium Geology (ALBRUG), China. The NBS-30 biotite standard yielded a mean δD value of -66.2 ± 1.8 ‰ (2σ, n=3), which is within analytical uncertainty of the published δD value of -65.7 ‰ (Brand et al., 2014). All data are reported in per mil relative to V-SMOW, and accuracy is better than 5 ‰. 4. Results 4.1. Amphibole compositions According to the classification of amphibole, all Duolong intrusions contain Ca amphibole whereas amphibole in the diorite is Mg hornblende (Si > 6.5; Leake et al., 1997; Hawthorne and Oberti, 2007). The barren granodiorite porphyry samples also contain primarily Mg hornblende with some minor tschermakite (Si < 6.5; Fig. 3a). In this study, amphibole grains roughly fall into two groups; low- (Al2O3 < ~ 10 wt. % and Si/AlT > 4) and high-Al (Al2O3 > ~ 10 wt. % and Si/AlT < 4; Fig. 3b). The diorite

9

contains low-Al amphibole and the barren granodiorite porphyry contains low-Al amphibole and high-Al tschermakite (Fig. 3a, b). Based on the composition of major elements, low-Al amphibole can be further divided into two groups (Fig. 4). Low-Al amphibole in diorite has lower Al2O3, MnO, and higher Na2O, TiO2 and Cl than amphibole in barren granodiorite porphyry rocks (Fig. 4c, e, f, h). Importantly, Cl contents in low-Al amphibole from the diorite (0.09-0.38 wt. %) extend to higher values than those from the barren granodiorite porphyry samples (0.08-0.24 wt. %), and show a positive correlation with Al2O3 content. Cl contents in high-Al amphibole from the barren granodiorite porphyry samples (0.02-0.21 wt. %) extend to the lowest values of all studied rocks and correlate negatively with Al2O3 content (Fig. 4h). Indeed, high-Al amphibole in the barren granodiorite porphyry samples has the highest Al2O3, Na2O, and lowest Cl, MnO contents (Fig. 4c, f, h). Moreover, amphibole AlIV content shows a positive correlation with Ti, (Na+K) A, and AlVI, suggesting compositional control by crystallization temperature- and pressure-sensitive Ti-tschermark, edenite, and Al-tschermark substitutions (Fig. 5b, c, d; Rutherford and Devine, 2003; Kiss et al., 2014). Low-Al amphibole from diorite and barren granodiorite porphyry samples have relatively consistent REE patterns that are slightly convex upward with Ce-Pr-Nd-Sm enrichment and distinctive negative Eu anomalies (Fig. 6a, b). On primitive mantle-normalized diagrams, these amphibole grains show negative Pb, Sr, Eu, Zr, Hf, and Ti anomalies (Fig. 6c, d). However, trace elements in high-Al amphibole from the barren granodiorite porphyry samples can be separated into two groups (Fig. 6b, d). One group shows similar REE patterns and trace element characteristics to the low-Al amphibole, but lower total REE abundances. Another group has the lowest REE

10

abundances, slightly convex REE patterns with MREE-enrichment and the absence of negative Eu anomalies (Fig. 6b), negative Nb, Zr, and Hf anomalies and positive Sr, Ba, and Pb anomalies (Fig. 6d). 4.2. Hydrogen isotopic compositions of amphibole Amphibole (n=4) in diorite shows H isotopic values (δD) ranging from -103 to -113‰, while barren granodiorite porphyry samples overlap these and extend to lower values (n=5; δD = -102 to -122‰) (Fig. 7a; Table 1). 5. Discussion 5.1. Temperature, pressure, oxygen fugacity Empirical and experimental studies indicate that amphibole composition can be effectively used to quantify magma conditions during crystallization (e.g., Ridolfi et al., 2010; Krawczynski et al., 2012). Here, pressure, temperature, melt H2O concentration, and oxygen fugacity of the Duolong magmas are determined using the empirical amphibole formulation (Ridolfi et al., 2010). Low-Al amphibole in diorite yield pressures between ~ 200 and 80 MPa (~ 8-3 km depth) and temperatures from ~ 860 to 750 °C, which overlap conditions of low-Al amphibole (~ 230-100 MPa and ~ 870-790 °C) in barren granodiorite porphyry samples. High-Al amphibole yield higher pressure and temperature conditions of formation (~ 400-210 MPa corresponding to 15-8 km and ~ 970-880 °C) in barren granodiorite porphyry samples (Fig. 8a, b). Moreover, the equilibrium melts with low-Al amphibole in diorite yield the same oxygen fugacities (ƒO2) (from NNO + 1.1 to NNO + 2.3) and extend to lower H2O contents (3-5.5 wt. %) than the equilibrium melts with low-Al amphibole in the barren granodiorite porphyry samples (NNO + 0.6 to NNO + 2.1 and 4.2-6.0 wt. %). The high-Al amphiboles yield relatively consistent ƒO2 (~ NNO + 1.3) and melt H2O

11

contents of ~ 5.4-6.4 wt. % (Fig. 8a, b). The equilibrium melts with low/high-Al amphibole have high oxygen fugacities (NNO + 0.6 to NNO + 2.3), consistent with ore-bearing magmas at porphyry Cu±Mo±Au deposits (NNO + 0.5 to NNO + 2) worldwide (Sun et al., 2015) and with the pervasive magnetite crystallization observed at Duolong (Fig. 2). In addition, experimental studies show that H2O solubility in melts is dominantly controlled by pressure (Baker and Alletti, 2012; Botcharnikov et al., 2015). If the H2O content in the melt is higher than its solubility, the melt will undergo fluid exsolution, either by decompression (decreases H2O solubility) and/or isobaric fractional crystallization of anhydrous mineral (increases H2O content; Candela, 1997; Webster, 2004). Based on H2O solubility in andesitic melts (Botcharnikov et al., 2015), the wide range of H2O contents (~ 3-6 wt. %) in magmas in equilibrium with low-Al amphibole from the diorite may suggest that fluid exsolution occurred below a pressure of ~200MPa (~7-8km depth) and a temperature of ~850 °C (Fig. 8b). This conclusion is consistent with previous studies on fluid inclusions hosted in quartz phenocrysts from the Duolong ore-bearing granodiorite porphyry, which suggested that primary fluid exsolution may have occurred in the magma chamber at ~900 °C and ~ 200MPa (Li et al., 2011b). Similarly, equilibrium magmas with low-Al amphibole from the barren granodiorite porphyries (except for some amphiboles with similar compositions with those in diorite; Fig. 4) have a range of H2Omelt contents (~ 4.2-6.0 wt. %; Fig. 8b), suggesting they also likely underwent fluid exsolution during amphibole crystallization. Exsolution during formation of the low-Al amphibole in the barren granodiorite porphyry is further supported by the presence of breakdown rims (> 200 μm; Fig. 2e) on high-Al amphibole phenocrysts, resulting from a concomitant decrease in H2Omelt content with decreasing pressure (e.g., Rutherford and Hill, 1993).

12

Evidence for exsolution in both the diorite and the barren granodiorite porphyry during low-Al amphibole growth is found in elevated oxygen fugacities (up to NNO + 2.3) of equilibrium melts (Bell and Simon, 2011; Dilles et al., 2015; Richards, 2015). Equilibrium melts with high-Al amphibole yield relatively consistent H2O contents of ~ 5.4-6.4 wt. % (Fig. 8b) that are mostly lower than H2O solubility at 200-400 MPa (~ 6-9 wt. %; Baker and Alletti, 2012), suggesting that these amphibole grains formed in an H2O-unsaturated melt. 5.2. Magmatic evolution Low-Al amphibole in diorite and barren granodiorite porphyry samples are typically characterized by low AlIV (0.8-1.5; Fig. 5) and low crystallization temperature (~ 870 to 750 °C; Fig. 8a) suggesting that they were derived from a cold silicic magma (< 900 °C andesitic-dacitic melts; Ridolfi et al., 2010; Chambefort et al., 2013; Kiss et al., 2014). However, the amphibole studied here shows two compositional trends on Al2O3 versus CaO, MnO, TiO2, Na2O, Cl (Fig. 4b, c, e, f, h), and Sr diagrams (Fig. 9a), likely suggesting crystallization from different magmas. Moreover, the low-Al amphibole in diorite and barren granodiorite porphyry samples show a positive correlation between Sr content and Eu/Eu* ratios, and between Ti and V content (Fig. 9b, c), indicating that it formed after plagioclase and magnetite crystallization (Figs. 2, 10; Krawczynski et al., 2012; Coint et al., 2013; Chambefort et al., 2013; Kiss et al., 2014). This is supported by trace element characteristics including negative Eu anomalies (Fig. 6a, b) and depletion in Sr and Ti (Fig. 6c, d). In addition, a few low-Al amphibole in barren granodiorite porphyry samples have a similar composition to those from the diorite (Fig. 4), possibly suggesting that the former were derived from dioritic magmas and captured by the late-stage magmas that formed the barren granodiorite porphyry samples. Moreover, based on partition

13

coefficients between amphibole and melt (Nandedkar et al., 2016), the calculated equilibrium melts show a similar REE composition to the host diorite and barren granodiorite porphyry (Fig. 6a, b; Li et al., 2013, 2016a, 2017). High-Al amphibole (AlIV=1.8-1.5; Fig. 5) from the barren granodiorite porphyry is characterized by elevated crystallization temperature (~ 880 to 970 °C; Fig. 8a), suggesting crystallization from a hotter, more mafic magma (Ridolfi et al., 2010). Experimental studies show that amphibole crystallized from more felsic magmas at low temperature and pressure has low AlIV and Mg# (Mg/Mg+FeT). In contrast, amphibole crystallized from more mafic magmas at high temperature and pressure has higher AlIV and Mg# (e.g., Ridolfi et al., 2010; Krawczynski et al., 2012; Kiss et al., 2014). For high-Al amphibole from the Duolong barren granodiorite porphyries, positive correlation between AlIV and Mg# values (Fig. 5a) likely indicates mafic magma replenishment (Sato et al., 2005; Ridolfi et al., 2010; Kiss et al., 2014) that caused a decrease in the Mg/(Mg+Fe) ratio of the mixed magma. Moreover, some high-Al amphiboles have the lowest REE content of all amphibole studied, and a complete absence of negative Eu anomaly (Figs. 6b, 9b), as well as enrichment in Sr (> 200 ppm; Fig. 9a), no depletion in Ti (Fig. 6d), and high V (> ~ 500 ppm; Fig. 9c) and Cr contents (Fig. 9d). The trace element compositions generally suggest that the high-Al amphibole crystallized from a high temperature mafic magma before plagioclase and magnetite crystallization (Fig. 10; Krawczynski et al., 2012; Chambefort et al., 2013; Kiss et al., 2014; Ribeiro et al., 2016). This conclusion is also supported by the relatively consistent REE compositions between the calculated equilibrium melts and syn-ore basaltic andesite (Fig. 6b; Li et al., 2013, 2016a). Other high-Al amphiboles show similar REE patterns and trace element characteristics to low-Al amphibole, but lower REE abundances and weakly negative Eu anomalies

14

(Fig. 6b). Combined with the positive correlation between AlIV and Mg# values (Fig. 5a), these high-Al amphibole with sharply decreasing V and Cr contents (Fig. 9c, d) in barren granodiorite porphyry samples likely record magma mixing between mafic and felsic melts below ~ 900 °C and 300 MPa where plagioclase and amphibole are stable (Fig. 10; Ridolfi et al., 2010; Krawczynski et al., 2012). 5.3. Volatile variations during magma evolution and fluid exsolution Most studies have revealed that Cl is preferentially incorporated into Fe-rich amphibole (governed by the preference of Cl for Cl-Fe2+ over Cl-Mg bonds; Oberti et al., 1993). Further experimental data has shown that the distribution coefficient (KD = (Cl/OH)Amp/(Cl/OH)melt) for Cl and OH between amphibole and melt depends on the Mg# of the amphibole and temperature. KD dominantly decreases as amphibole Mg# increases (Sato et al., 2005; Giesting and Filiberto, 2014). Thus, the Cl content and Mg# of amphibole can be estimated assuming a constant Cl/OH ratio in the melt (Fig. 11; Sato et al., 2005; Humphreys et al., 2009). For low-Al amphibole from the Duolong ore district, Cl content is negatively correlated with Mg#, consistent with experimental results (Oberti et al., 1993; Sato et al., 2005; Giesting and Filiberto, 2014). However, the Cl content of low-Al amphibole in diorite fits within the wide range of Cl/OH ratios in equilibrium melts (0.03-0.085; Fig. 11), which cannot be solely explained by the dependence of the Cl/OH partition coefficient on Mg# in amphibole. Previous studies indicated that the Cl/OH ratio of melt in H2O-unsaturated magmas remains constant or slightly increases with progressive fractional crystallization (Webster, 2004). Conversely, Cl/OH ratios (0.085-0.03) in equilibrium melts decrease during magma evolution (Figs. 8 and 11a). Therefore, the large variation in the Cl content of low-Al amphibole may be caused by an influx of Cl-poor mafic magmas or fluid exsolution (Sato et al., 2005; Humphreys et al., 2009;

15

Chambefort et al., 2013). The negative correlation between AlIV and Mg# values (Fig. 5a) indicates an absence of Cl-poor mafic magma replenishment (Sato et al., 2005; Ridolfi et al., 2010; Kiss et al., 2014). Combined with the H2Omelt content decreasing from ~ 6 to 3 wt. % (Fig. 8b), the Cl variation in low-Al amphibole from diorite could be dominantly caused by fluid exsolution from ore-bearing magmas. Moreover, most studies suggest that fluid exsolution/magma degassing can reduce the δD values of residue melts (Nabelek et al., 1983; Taylor et al., 1983; Underwood et al., 2013; Chambefort et al., 2013). Generally, H isotopic fractionation for “melt-fluid” and “amphibole-fluid” are approximately the same (i.e. there is no fractionation between melt and amphibole). Thus, the H isotopic composition of amphibole could represent the isotopic composition of its equilibrium melt (Graham et al., 1984; Dobson et al., 1989). The low-Al amphibole in Duolong diorites has an obviously lower δD value (-103 to -113 ‰; Fig. 7a) than the amphibole from undegassed arc magmas (-40 to -60 ‰; Underwood et al., 2013; Chambefort et al., 2013). The H isotopic composition of the melt can be evaluated by modeling both closed- and open-system fluid exsolution (Fig. 7b). The formula for closed-system fluid exsolution is δDmelti = δDmelt0 –(1-F)1000lnαH2O-melt and for open-system is δDmelti = δDmelt0 –(1000+δDmelt0)(1-FαH2O-melt), where δDmelt0 is the initial H isotopic value of -50‰ consistent with undegassed arc magmas (e.g., Underwood et al., 2013) and F is the fraction of H2O remaining in the melt (initial H2O content is 6 wt.%). The H isotopic fractionation coefficient (αH2O-melt) is calculated by the temperature-dependent equation at 800 °C (Dalou et al., 2015). Modeling results indicate that the low H isotopic compositions are consistent with ~ 40-50 % H2O loss during open system fluid exsolution (Fig. 7b). For low-Al amphibole in barren granodiorite porphyry samples, Cl contents

16

plotted against Mg/Mg+FeT generally fit the dashed line defined by Cl/OHmelt of 0.03 (Fig. 11) suggesting that these amphibole grains likely crystallized from a melt with a relatively uniform Cl/OH ratio. Thus, the Cl compositional variation in low-Al amphibole from barren granodiorite porphyry samples was dominantly controlled by amphibole composition rather than by fluid exsolution (Mg#; Sato et al., 2005; Humphreys et al., 2009). The latter likely had a minor impact on the Cl contents of these amphibole grains, consistent with weak alteration and a lack of mineralization in the host porphyries. On the other hand, the Cl content of high-Al amphibole from the barren granodiorite porphyry rocks fit the modeled dashed line defined by Cl/OHmelt of 0.01-0.03 on the Cl versus Mg/Mg+FeT plot (Fig. 11). Combined with the positive correlation between AlIV and Mg# values (Fig. 5a), influx of Cl-poor mafic magmas could account for the Cl compositional variations observed in high-Al amphibole (Fig. 11). Moreover, amphibole grains from the barren granodiorite porphyry samples have low δD values (-102 to -122‰), similar to those of amphibole from the diorite (Fig. 7a). Based on consistent Cl/OH ratio of equilibrium magmas with low-Al amphibole (Fig. 11), low δD values caused by fluid exsolution in the barren granodiorite porphyry has been ruled out. Thus, amphibole grains with lower δD values might have crystallized from hybrid magmas that had a composition between mafic and intermediate-felsic (Figs. 4, 5a, 6b, 9a). The intermediate-felsic magmas would have evolved from early-stage dioritic magmas that underwent large scale fluid exsolution (Fig. 12a, b). Additionally, the thickness of breakdown rims on amphibole phenocrysts (e.g., Fig. 2e) depends on magma ascent rate, with a concomitant decrease in H2Omelt content with decreasing pressure (Rutherford and Hill, 1993; Rutherford and Devine, 2003). High-Al amphibole phenocrysts from the barren granodiorite porphyry rocks have thick breakdown rims (> 200 μm; Fig. 2e),

17

suggesting that the magma ascended slowly (more than 100 days residence; Rutherford and Devine, 2003). Based on experimental data on diffusivity of H in amphibole (Ingrin and Blanchard, 2006), a 1mm long amphibole at ~ 800°C completely reaches H isotopic exchange equilibrium with the surrounding melt in ~ 30 days (Chambefort et al., 2013). Thus, the slow magma ascent rate and long residence time could insure effective H isotope exchange (Nabelek et al., 1983; Rutherford and Devine, 2003; Chambefort et al., 2013) between high-Al amphibole and degassed dioritic magmas with low δD values during mafic magmas injection. 5.4. A scenario for fluid exsolution and amphibole crystallization in the Duolong evolved magma chamber Previous studies showed that the Duolong ore-bearing diorite and granodiorite porphyry have relatively uniform zircon U-Pb ages, similar trace element characteristics and whole rock Sr-Nd and zircon Hf-O isotopic compositions, suggesting they crystallized from a series of cogenetic melts with various degrees of fractional crystallization (e.g, Li et al., 2013, 2017; Sun et al., 2017). Generally, amphibole grains from ore-bearing porphyries in porphyry Cu-Au deposits are not well preserved given extensive alteration. However, a few low-Al remnant amphibole grains in the Duolong ore-bearing granodiorite porphyries (Sun et al., 2017) have similar compositions to amphibole from diorite (Figs, 4, 5, 11a), suggesting they were derived from the same magma chamber under large-scale fluid exsolution conditions (Fig, 12a). Also, it is well known that formation of porphyry Cu ± Mo ± Au deposits commonly requires a large magma chamber (up to 2000 km3) in the shallow crust (e.g., Richards, 2011; Zhang et al., 2017), such as the exposed ~ 1300 km3 ore-bearing plutons at the Bingham deposit, USA (Dilles, 1987). Thus, the diorite in the Duolong ore district should represent the ore-bearing plutonic intrusion connected to a shallow

18

ore-bearing granodiorite porphyry and Cu-Au mineralization (Fig, 12a), consistent with the petrogenetic model for porphyry Cu ± Mo ± Au deposits (e.g., Sillitoe, 2010; Richards, 2011). Based on amphibole compositions in this study and previous petrogeochemistry studies (Li et al., 2013, 2016a, 2017; Sun et al., 2017), the magmatic evolution and fluid exsolution model for the Duolong Cu-Au district is presented in Fig. 12. At ~ 118 Ma, oxidized and sulfur-rich, Cl-rich intermediate melts derived from the MASH zone (melting, assimilation, storage and homogenization; Hildreth and Moorbath, 1988) ascended rapidly to shallow crustal levels (3-8 km) and formed a magma chamber. Meanwhile, intermediate melts underwent fractional crystallization (low-Al amphibole, plagioclase, and magnetite) and large-scale ore-forming fluid exsolution (Fig, 12a). The fluid exsolution led to decreasing Cl/OHmelt (Fig, 11a) and depleted H isotopic compositions (δD = -103 to -113 ‰; Fig. 7a) in residual melts. The differentiated melts rose to form ore-bearing granodiorite porphyries and a channel of ore-forming fluids led to forming the Doulong Cu-Au mineralization. Subsequently, deep-sourced basaltic-andesitic arc melts (Cl-poor) with high-Al amphibole (~ 400-300 MPa, ~ 970-940 °C, and lowest REE; Figs, 6d, 8) were injected into the differentiated intermediate-felsic magma chamber, which underwent fluid exsolution and crystallized early-stage low-Al amphiboles. The new hybrid melts also crystallized late-stage high-Al (~ 210-260 MPa, ~ 880-900 °C, and higher REE; Figs, 6d, 8) and low-Al amphibole (~ 790-870 °C and ~ 100-230 MPa; Fig, 8) and rose to form barren granodiorite porphyries (Fig, 12b). 5.5. Implications for metallogeny Fluid exsolution is a critical step in the formation of porphyry Cu ± Mo ± Au deposits (e.g., Cline and Bodnar, 1991; Shinohara and Hedenquist, 1997; Webster, 2004; Sillitoe, 2010; Richards, 2011). Fluids exsolved early from magmas can

19

sequestrate more ore-forming elements (e.g., Cu and Au) than fluids exsolved later (Candela and Holland, 1986; Shinohara and Hedenquist, 1997; Webster, 2004), because fractional crystallization of minerals (Candela and Holland, 1986; Cline and Bodnar, 1991; Liu et al., 2015) and/or sulfide saturation (Jenner et al., 2010) reduce the absolute contents of available Cu and Au in melts. A recent study showed that interaction between exsolved fluids and early sulfide melts in a magma chamber might be a new mechanism for producing ore-forming fluids (Nodeau et al., 2010; Park et al., 2015). At Duolong, elemental composition and H isotopic characteristics in low-Al amphibole from ore-bearing diorite indicate that large-scale fluid exsolution from the magma may have begun at a pressure of ~200MPa (~7-8km depth) and a temperature of ~850 °C (Fig. 8b). Compositions of Cl- and S-rich apatite in the Duolong ore-bearing porphyry (Li et al., 2012) and amphibole in the ore-bearing diorite (Figs, 8a, 11) indicate that the ore-bearing magmas were Cl-rich, S-rich and oxidized. Based on the estimated Cl/OHmelt of ~0.085 in ore-bearing dioritic melts (Fig. 11) and partitioning Cl, S strongly into the fluids (e.g., Heinrich et al., 1999, 2004; Ulrich et al., 1999; Harris et al., 2003; Williams-Jones and Heinrich, 2005; Zajacz et al., 2012a), the early exsolved, high temperature Cl- and S-rich fluids could effectively sequestrate metal elements from the magma and/or sulfide in the magma chamber to form the giant Duolong Cu-Au deposits. In contrast, small amounts of Cl-poor fluids also exsolved from melts (lower and consistent Cl/OHmelt of ~0.03; Fig. 11) in equilibrium with low-Al amphibole in barren porphyries. These fluids may be similar to other low-salinity fluids in barren intrusions (New Mexico, USA; Audétat and Pettke, 2003). Most studies show that magma degassing (fluid exsolution) commonly leads to lower Cl content and lower δD values in residual melts (e.g., Nabelek et al., 1983;

20

Taylor et al., 1983; Taran et al., 1997; Sato et al., 2005; Chambefort et al., 2013). For formation of porphyry Cu deposits, the ore-bearing magmas must undergo large-scale fluid exsolution to effectively sequester metals in the magma chamber (e.g., Shinohara and Hedenquist, 1997; Webster, 2004; Richards, 2011). Therefore, the ore-bearing magmas associated with fluid exsolution should show decreasing Cl contents, as shown by the Cl variations of low-Al amphibole in the Doulong ore-bearing diorites (Fig. 11). Conversely, the barren magmas with small-scale fluid exsolution should have limited variations of Cl contents and relatively consistent Cl/OHmelt ratios, as seen in the low-Al amphibole from Doulong barren granodiorite porphyries (Fig. 11). Additionally, the ore-bearing magmas should have depleted H isotopic compositions due to large-scale fluid exsolution, as demonstrated by low δD values of amphibole in ore-bearing diorite (Fig. 7a) and consistent with the degassed arc magmas (e.g., Taylor et al., 1983; Taran et al., 1997). In contrast, barren magmas with small-scale fluid exsolution should have normal H isotopic compositions (δD = -50 to -80‰; Taylor et al., 1983; Taran et al., 1997; Chambefort et al., 2013; Underwood et al., 2013), in the absence of long-lived interaction with degassed melts. As seen at Doulong, Cl compositional variation of amphibole could become a powerful tool for identifying ore-bearing and barren intrusions within porphyry Cu ± Mo ± Au deposits. Speculatively, H isotope signature of amphibole also might be an effective indicator for discriminating ore-bearing and barren intrusions in regional exploration, because barren magmas apart from ore deposit could not undergo long-lived interaction with degassed magmas. 6. Conclusions (1) Low-Al amphibole grains from ore-bearing diorite were formed in a shallow magma chamber (~ 750-860 °C and ~ 80-200 MPa). Distinctive negative Eu

21

anomalies and negative anomalies in Sr, Eu, and Ti suggest that these amphibole grains crystallized from magmas that previously crystallized plagioclase and magnetite. Large Cl content variations (0.09-0.38 wt. %) and low δD values (-103 to -113‰) in low-Al amphibole was likely caused by large scale fluid exsolution that formed the Duolong Cu-Au mineralization. (2) Low-Al amphibole in barren granodiorite porphyry samples have distinctly different compositions from those of the diorite, suggesting they were crystallized from a seperate magma, which experienced mixing between more mafic and intermediate-felsic magmas prior to formation. (3) High-Al amphibole crystallized at high temperature and pressure (880-970 °C and 210-400 MPa). They show an absence of negative Eu anomalies and high Sr, Ba, Ti contents, likely consistent with amphibole crystallization from more mafic (basaltic-andesitic) melts. (4) Positive correlation between AlIV and Mg# values in high-Al amphiboles indicates mafic magma replenishment. Meanwhile, small variations of Cl content (0.08-0.24 wt. %) in low-Al amphibole in the same barren granodiorite porphyry samples correlate negatively with Mg# under relatively constant melt Cl/OH ratios, suggesting compositions were not strongly controlled by fluid exsolution. (5) Low δD values (-102 to -122‰) in amphibole from barren granodiorite porphyry samples likely inherited the isotopic characteristics of dioritic magmas that underwent fluid exsolution. (6) Cl compositional variation in amphibole could be used as a powerful tool to distinguish between ore-bearing and barren intrusions within porphyry Cu ± Mo ± Au deposits worldwide.

22

Acknowledgments This article was funded by the Natural Science Foundation Project (Grant No. 41672091; 41490613), and the Ministry of Science and Technology of China (2016YFC0600303). We obtained support and help from senior geologists Hong-Qi Chen and Yu-Bin Li at the No.5 Geological Team, Tibet Bureau of Geology and Exploration. This manuscript benefited from constructive comments by the editor Franco Pirajno and Harald G. Dill, and an anonymous reviewer and Dr. Ryan Mathur.

References Audétat, A., Pettke, T., 2003. The magmatic-hydrothermal evolution of two barren granites: A melt and fluid inclusion study of the Rito del Medio and Canada Pinabete plutons in northern New Mexico (USA). Geochimica et Cosmochimica Acta 67, 97-121. Baker, D. R., Alletti, M., 2012. Fluid saturation and volatile partitioning between melts and hydrous fluids in crustal magmatic systems: The contribution of experimental measurements and solubility models. Earth Science Reviews 114, 298-324. Bao, B., Webster, J. D., Zhang, D. H., Goldoff, B. A., Zhang, R. Z., 2016. Compositions of biotite, amphibole, apatite and silicate melt inclusions from the Tongchang mine, Dexing porphyry deposit, SE China: Implications for the behavior of halogens in mineralized porphyry systems. Ore Geology Reviews 79, 443-462. Bell, A. S., Simon, A., 2011. Experimental evidence for the alteration of the Fe3+/ ΣFe of silicate melt caused by the degassing of chlorine-bearing aqueous volatiles. Geology 39, 499-502. 23

Botcharnikov, R. E., Holtz, F., Behrens, H., 2015. Solubility and fluid–melt partitioning of H2O and Cl in andesitic magmas as a function of pressure between 50 and 500MPa. Chemical Geology 418, 117-131. Brand, W.A., Coplen, T.B., Vogl, J., Rosner, M., Prohaska, T., 2014. Assessment of international reference materials for isotope-ratio analysis (IUPAC Technical Report). Pure and Applied Chemistry 86, 425-467. Candela, P.A., 1997. A review of shallow, ore-related granites: Textures, volatiles, and ore metals. Journal of Petrology 38, 1619-1633. Candela, P.A., Holland, H.D., 1986. A Mass-Transfer Model for Copper and Molybdenum in Magmatic Hydrothermal Systems - the Origin of Porphyry-Type Ore-Deposits. Economic Geology 81, 1-19. Chambefort, I., Dilles, J. H. Longo, A. A., 2013. Amphibole geochemistry of the Yanacocha volcanics, Peru: evidence for diverse sources of magmatic volatiles related to gold ores. Journal of Petrology 54, 1017-1046. Chelle-Michou, C., Chiaradia, M., 2017. Amphibole and apatite insights into the evolution and mass balance of Cl and S in magmas associated with porphyry copper deposits. Contributions to Mineralogy and Petrology 172(11-12), 105. Cline, J.S., Bodnar, R.J., 1991. Can economic porphyry copper mineralization be generated by a typical calc-alkaline melt? Journal of Geophysical Research 96, 8113-8126. Coint, N., Barnes, C. G., Yoshinobu, A. S., Barnes, M. A., Buck, S., 2013. Use of trace element abundances in augite and hornblende to determine the size, connectivity, timing, and evolution of magma batches in a tilted batholith. Geosphere 9, 1747-1765. Dalou, C., Le Losq, C., Mysen, B. O., 2015. In situ study of the fractionation of

24

hydrogen isotopes between aluminosilicate melts and coexisting aqueous fluids at high pressure and high temperature – Implications for the δD in magmatic processes. Earth and Planetary Science Letters 426, 158-166. Dilles, J.H., Kent, A.J.R., Wooden, J.L., Tosdal R.M., Koleszar, A., Lee, R.G., Farmer, L.P., 2015. Zircon compositional evidence for sulfur-degassing from ore-forming arc magmas. Economic Geology 110, 241-251. Dilles, J.H., 1987. Petrology of the Yerington Batholith, Nevada: Evidence for Evolution of Porphyry Copper Ore Fluids. Economic Geology 82, 1750-1789. Dobson, P.F., Epstein, S., Stolper, E.M., 1989. Hydrogen isotope fractionation between coexisting vapor and silicate-glasses and melts at low-pressure. Geochimica Et Cosmochimica Acta 53, 2723-2730. Duan, D.F., Jiang, S.Y., 2017. In situ major and trace element analysis of amphiboles in quartz monzodiorite porphyry from the Tonglvshan Cu–Fe (Au) deposit, Hubei Province, China: insights into magma evolution and related mineralization. Contributions to Mineralogy and Petrology 172(5): 36. Fan, J.J., Li, C., Xie, C.M., Wang, M., Chen, J.W., 2015. The evolution of the Bangong–Nujiang Neo-Tethys ocean: Evidence from zircon U–Pb and Lu–Hf isotopic analyses of Early Cretaceous oceanic islands and ophiolites. Tectonophysics 655, 27-40. Geng, Q.R., Zhang, Z., Peng, Z.M., Guan, J.L., Zhu, X.P., Mao, X.C., 2016. Jurassic–Cretaceous granitoids and related tectono-metallogenesis in the Zapug–Duobuza arc, western Tibet. Ore Geology Reviews 77, 163-175. Giesting, P.A., Filiberto, J., 2014. Quantitative models linking igneous amphibole composition with magma Cl and OH content. American Mineralogist 99, 852-865.

25

Graham, C.M., Harmon, R.S., Sheppard, S.M.F., 1984. Experimental hydrogen isotope studies: hydrogen isotope exchange between amphibole and water. American Mineralogist 69, 128-138. Guynn, J.H., Kapp, P., Pullen, A., Heizier, M., Gehrels, G., Ding, L., 2006. Tibetan basement rocks near Amdo reveal "missing" Mesozoic tectonism along the Bangong suture, central Tibet. Geology 34, 505-508. Hao, L.L., Wang, Q., Wyman, D.A., Ou, Q., Dan, W., Jiang, Z. Q., Wu, F.Y., Yang, J.H., Long, X.P., Li, J., 2016. Underplating of basaltic magmas and crustal growth in a continental arc: Evidence from Late Mesozoic intermediate–felsic intrusive rocks in southern Qiangtang, central Tibet. Lithos 245, 223-242. Harris, A.C., Kamenetsky, V.S., White, N.C., van Achterbergh, E., Ryan, C.G., 2003. Melt inclusions in veins: linking magmas and porphyry Cu deposits. Science 302, 2109-2111. Hawthorne, F.C., Oberti, R., 2007. Classification of the Amphiboles. Reviews in Mineralogy and Geochemistry 67, 55-88. Hedenquist, J.W., Lowenstern, J.B., 1994. The role of magmas in the formation of hydrothermal ore deposits. Nature 370, 519-527. Heinrich, C.A., Driesner, T., Stefansson, A., Seward, T.M., 2004. Magmatic vapor contraction and the transport of gold from the porphyry environment to epithermal ore deposits. Geology 32, 761-764. Heinrich, C.A., Gunther, D., Audetat, A., Ulrich, T., Frischknecht, R., 1999. Metal fractionation between magmatic brine and vapor, determined by microanalysis of fluid inclusions. Geology 27, 755-758. Hildreth, W., Moorbath, S., 1988. Crustal contributions to arc magmatism in the andes of Central Chile. Contributions to Mineralogy and Petrology 98, 455-489.

26

Hou, Z.Q., Zhang, H.R., 2015. Geodynamics and metallogeny of the eastern Tethyan metallogenic domain. Ore Geology Reviews 70, 346-384. Hou, Z.Q., Gao, Y.F., Qu, X.M., Rui, Z.Y., Mo, X.X., 2004. Origin of adakitic intrusives generated during mid-Miocene east-west extension in southern Tibet. Earth and Planetary Science Letters 220, 139-155. Huang, Q.S., Shi, R.D., O’Reilly, S.Y., Griffin, W.L., Zhang, M., Liu, D.L., Zhang, X. R., 2015a. Re-Os isotopic constraints on the evolution of the Bangong-Nujiang Tethyan oceanic mantle, Central Tibet. Lithos 224-225, 32-45. Huang, Q.T., Cai, Z.R., Xia, B., Li, J.F., Xia, L.Z., Liu, H.C., 2015b. Geochronology, geochemistry, and Sr–Nd–Pb isotopes of Cretaceous granitoids from western Tibet: petrogenesis and tectonic implications for the evolution of the Bangong Meso-Tethys. International Geology Review 58, 95-111. Humphreys, M.C.S., Edmonds, M., Christopher, T., Hards, V., 2009. Chlorine variations in the magma of Soufrière Hills Volcano, Montserrat: Insights from Cl in hornblende and melt inclusions. Geochimica et Cosmochimica Acta 73, 5693-5708. Ingrin, J., Blanchard, M., 2006. Diffusion of hydrogen in minerals. Reviews in Mineralogy and Geochemistry 62, 291-320. Jenner, F.E., O'Neill, H.S.C., Arculus, R.J., Mavrogenes, J.A., 2010. The magnetite crisis in the evolution of arc-related magmas and the initial concentration of Au, Ag and Cu. Journal of Petrology 51, 2445-2464. Kapp, P., Yin, A., Harrison, T.M., Ding, L., 2005. Cretaceous-Tertiary shortening, basin development, and volcanism in central Tibet. Geological Society of America Bulletin 117, 865-878. Kiss, B., Harangi, S., Ntaflos, T., Mason, P. R. D., Pál-Molnár, E., 2014. Amphibole

27

perspective to unravel pre-eruptive processes and conditions in volcanic plumbing systems beneath intermediate arc volcanoes: a case study from Ciomadul volcano (SE Carpathians). Contributions to Mineralogy and Petrology 167, 986. Krawczynski, M. J., Grove, T.L., Behrens, H., 2012. Amphibole stability in primitive arc magmas: effects of temperature, H2O content, and oxygen fugacity. Contributions to Mineralogy and Petrology 164, 317-339. Leake, B. E., Woolley, A. R., Arps, C.E.S., Birch, W.D., Gilbert, M.C., Grice, J.D., Hawthorne, F.C., Kato, A., Kisch, H.J., Krivovichev, V.G., Linthout, K., Laird, J., Mandarino, J.A., Maresch, W.V., Nickel, E.H., Rock, N.M.S., Schumacher, J.C., Smith, D.C., Stephenson, N.C.N., Ungaretti, L., Whittaker, E.J.W., Guo, Y.Z., 1997. Nomenclature of amphiboles: Report of the subcommittee on amphiboles of the International Mineralogical Association, Commission on New Minerals and Mineral Names. Canadian Mineralogist 35, 219-246. Li, J. X., Qin, K. Z., Li, G. M., Richards, J. P., Zhao, J. X., Cao, M. J., 2014a. Geochronology, geochemistry, and zircon Hf isotopic compositions of Mesozoic intermediate-felsic intrusions in central Tibet: Petrogenetic and tectonic implications. Lithos 198, 77-91. Li, J. X., Qin, K. Z., Li, G. M., Xiao, B., Zhao, J. X., Cao, M. J., Chen, L., 2013. Petrogenesis of ore-bearing porphyries from the Duolong porphyry Cu-Au deposit, central Tibet: Evidence from U-Pb geochronology, petrochemistry and Sr-Nd-Hf-O isotope characteristics. Lithos 160, 216-227. Li, J. X., Qin, K. Z., Li, G. M., Xiao, B., Zhao, J. X., Chen, L., 2016a. Petrogenesis of Cretaceous igneous rocks from the Duolong porphyry Cu-Au deposit, central Tibet: evidence from zircon U-Pb geochronology, petrochemistry and

28

Sr-Nd-Pb-Hf isotope characteristics. Geological Journal 51, 285-307. Li, J. X., Qin, K.Z., Li, G.M., Xiao, B., Zhao, J.X., Chen, L., 2011a. Magmatic-hydrothermal evolution of the Cretaceous Duolong gold-rich porphyry copper deposit in the Bangongco metallogenic belt, Tibet: Evidence from U-Pb and 40Ar/39Ar geochronology. Journal of Asian Earth Sciences 41, 525-536. Li, J. X., Qin, K. Z., Li, G. M., Evans, N. J., Zhao, J. X., Cao, M. J., Huang, F., 2016b. The Nadun Cu–Au mineralization, central Tibet: Root of a high sulfidation epithermal deposit. Ore Geology Reviews 78, 371-387. Li, J.X., Li, G.M., Qin, K.Z., Xiao, B., 2011b. High-temperature magmatic fluid exsolved from magma at the Duobuza porphyry copper-gold deposit, Northern Tibet. Geofluids 11, 134-143. Li, J.X., Li, G.M., Qin, K.Z., Xiao, B., Chen, L., Zhao, J.X., 2012. Mineralogy and Mineral Chemistry of the Cretaceous Duolong Gold-Rich Porphyry Copper Deposit in the Bangongco Arc, Northern Tibet. Resource Geology 62, 19-41. Li, S. M., Zhu, D. C., Wang, Q., Zhao, Z. D., Sui, Q. L., Liu, S. A., Liu, D., Mo, X. X., 2014b. Northward subduction of Bangong–Nujiang Tethys: Insight from Late Jurassic intrusive rocks from Bangong Tso in western Tibet. Lithos 205, 284-297. Li, S.M., Zhu, D.C., Wang, Q., Zhao, Z., Zhang, L.L., Liu, S.A., Chang, Q.S., Lu, Y.H., Dai, J.G., Zheng, Y.C., 2016c. Slab-derived adakites and subslab asthenosphere-derived OIB-type rocks at 156±2Ma from the north of Gerze, central Tibet: Records of the Bangong–Nujiang oceanic ridge subduction during the Late Jurassic. Lithos 262, 456-469. Li, X.K., Li, C., Sun, Z.M., Wang, M., 2017. Origin and tectonic setting of the giant Duolong Cu–Au deposit, South Qiangtang Terrane, Tibet: Evidence from

29

geochronology and geochemistry of Early Cretaceous intrusive rocks. Ore Geology Reviews 80, 61-78. Liu, D.L., Huang, Q.S., Fan, S.Q., Zhang, L.Y., Shi, R.D., Ding, L., 2014. Subduction of the Bangong-Nujiang Ocean: constraints from granites in the Bangong Co area, Tibet. Geological Journal 49, 188-206. Liu, X., Xiong, X., Audétat, A., Li, Y., 2015. Partitioning of Cu between mafic minerals, Fe–Ti oxides and intermediate to felsic melts. Geochimica et Cosmochimica Acta 151, 86-102. Nabelek, P. I., Oneil, J. R., Papike, J. J., 1983. Vapor Phase Exsolution as a Controlling Factor in Hydrogen Isotope Variation in Granitic Rocks - the Notch Peak Granitic Stock, Utah. Earth and Planetary Science Letters 66, 137-150. Nadeau, O., Williams-Jones, A.E., Stix, J., 2010. Sulphide magma as a source of metals in arc-related magmatic hydrothermal ore fluids. Nat. Geosci. 3, 501-505. Nandedkar, R.H., Hürlimann, N., Ulmer, P., Müntener, O., 2016. Amphibole–melt trace element partitioning of fractionating calc-alkaline magmas in the lower crust: an experimental study. Contributions to Mineralogy and Petrology 171. Oberti, R., Ungaretti, L., Cannillo, E., Hawthorne, F.C., 1993. The mechanism of Cl incorporation in amphibole. American Mineralogist 78, 746-752. Pan, G.T., Wang, L.Q., Li, R.S., Yuan, S.H., Ji, W.H., Yin, F.G., Zhang, W.P., Wang, B.D., 2012. Tectonic evolution of the Qinghai-Tibet Plateau. Journal of Asian Earth Sciences 53, 3-14. Park, J.W., Campbell, I.H., Kim, J., Moon, J.W., 2015. The Role of Late Sulfide Saturation in the Formation of a Cu- and Au-rich Magma: Insights from the Platinum Group Element Geochemistry of Niuatahi-Motutahi Lavas, Tonga Rear Arc. Journal of Petrology 56, 59-81.

30

Ribeiro, J.M., Maury, R.C., Grégoire, M., 2016. Are adakites slab melts or high-pressure fractionated mantle melts? Journal of Petrology 57, 839-862. Richards, J. P., 2011. Magmatic to hydrothermal metal fluxes in convergent and collided margins. Ore Geology Reviews 40, 1-26. Richards, J. P., 2015. The oxidation state, and sulfur and Cu contents of arc magmas: implications for metallogeny. Lithos 233, 27-45. Ridolfi, F., Renzulli, A., Puerini, M., 2010. Stability and chemical equilibrium of amphibole in calc-alkaline magmas: an overview, new thermobarometric formulations and application to subduction-related volcanoes. Contributions to Mineralogy and Petrology 160, 45-66. Rutherford, M.J., Devine, J.D., 2003. Magmatic conditions and magma ascent as indicated by hornblende phase equilibria and reactions in the 1995-2002 Soufriere Hills magma. Journal of Petrology 44, 1433-1454. Rutherford, M.J., Hill, P.M., 1993. Magma ascent rates from amphibole breakdown: an experimental study applied to the 1980-1986 Mount St-Helens eruptions. Journal of Geophysical Research-Solid Earth 98, 19667-19685. Sato, H., 2005. Experimental Petrology of the 1991-1995 Unzen Dacite, Japan. Part II: Cl/OH Partitioning between Hornblende and Melt and its Implications for the Origin of Oscillatory Zoning of Hornblende Phenocrysts. Journal of Petrology 46, 339-354. Shi, R. D., Yang, J. S., Xu, Z. Q., Qi, X. X., 2008. The Bangong Lake ophiolite (NW Tibet) and its bearing on the tectonic evolution of the Bangong-Nujiang suture zone. Journal of Asian Earth Sciences 32, 438-457. Shinohara. H., Hedenquist, J. W., 1997. Constraints on magma degassing beneath the far southeast porphyry Cu-Au deposit, Philippines. Journal of Petrology 38,

31

1741-1752. Sillitoe, R.H., 2010. Porphyry copper systems. Economic Geology 105, 3-41. Sun, J., Mao, J.W., Beaudoin, G., Duan, X.Z., Yao, F.J., Ouyang, H., Wu, Y., Li, Y.B., Meng, X.Y., 2017. Geochronology and geochemistry of porphyritic intrusions in the Duolong porphyry and epithermal Cu-Au district, central Tibet: Implications for the genesis and exploration of porphyry copper deposits. Ore Geology Reviews 80, 1004-1019. Sun, W.D., Huang, R.F., Li, H., Hu, Y.B., Zhang, C.C., Sun, S.J., Zhang, L.P., Ding, X., Li, C.Y., Zartman, R.E., Ling, M.X., 2015. Porphyry deposits and oxidized magmas. Ore Geology Reviews 65, 97-131. Sun, S. S., McDonough, W. F., 1989. Chemical and isotopic systematics of oceanic basalts: implications for mantle composition and processes. Geological Society, London, Special Publications 42, 313-345. Tang, J. X., Song, Y., Wang, Q., Lin, B., Yang, C., Guo, N., Fang, X., Yang, H. H., Wang, Y. Y., Gao, K., Ding, S., Zhang, Z., Duan, J. L., Chen, H. Q., Su, D. K., Feng, J., Liu, Z. B., Wei, S. G., He, W., Song, J. L., Li, Y. B., Wei, L. J., 2016.Geological characteristics and exploration model of the Tiegelongnan Cu (Au-Ag) deposit: The first ten million tons metal resources of a porphyry-epithermal deposit in Tibet. Acta Geoscientica Sinica 37, 663-690 (in Chinese with English abstract). Taran, Y.A., Pokrovsky, B.G., Volynets, O.N., 1997. Hydrogen isotopes in hornblendes and biotites from Quaternary volcanic rocks of the Kamchatka-Kurile arc. Geochemical Journal 31, 203-221. Taylor, B. E., Eichelberger, J. C., Westrich, H. R., 1983. Hydrogen isotopic evidence of rhyolitic magma degassing during shallow intrusion and eruption. Nature 306,

32

541-545. Ulrich, T., Gunther, D., Heinrich, C.A., 1999. Gold concentrations of magmatic brines and the metal budget of porphyry copper deposits. Nature 399, 676-679. Underwood, S.J., Feeley, T.C., Clynne, M.A., 2013. Hydrogen isotope investigation of amphibole and glass in dacite magmas erupted in 1980-1986 and 2005 at Mount St. Helens, Washington. Journal of Petrology 54, 1047-1070. Vennemann, T.W., O’Neil, J.R., 1993. A simple and inexpensive method of hydrogen isotope and water analyses of minerals and rocks based on zinc reagent. Chemical Geology 103, 227-234. Wang, B.D., Wang, L.Q., Chung, S.L., Chen, J.L., Yin, F.G., Liu, H., Li, X.B., Chen, L.K., 2016. Evolution of the Bangong–Nujiang Tethyan ocean: Insights from the geochronology and geochemistry of mafic rocks within ophiolites. Lithos 245, 18-33. Webster, J. D., 2004. The exsolution of magmatic hydrosaline chloride liquids. Chemical Geology 210, 33-48. Williams-Jones, A.E., Heinrich, C.A., 2005. 100th Anniversary special paper: Vapor transport of metals and the formation of magmatic-hydrothermal ore deposits. Economic Geology 100, 1287-1312. Wones, D. R., 1989. Significance of the assemblage titanite + magnetite + quartz in granitic rocks. American Mineralogist 74, 744-749. Wu, H., Xie, C.M., Li, C., Wang, M., Fan, J.J, Xu, W.L., 2016. Tectonic shortening and crustal thickening in subduction zones: Evidence from Middle–Late Jurassic magmatism in Southern Qiangtang, China. Gondwana Research 39, 1-13. Yin, A., Harrison, T. M., 2000. Geologic evolution of the Himalayan-Tibetan orogen. Annual Review of Earth and Planetary Sciences 28, 211-280.

33

Zajacz, Z., Candela, P.A., Piccoli, P.M., Sanchez-Valle, C., 2012a. The partitioning of sulfur and chlorine between andesite melts and magmatic volatiles and the exchange coefficients of major cations. Geochimica et Cosmochimica Acta 89, 81-101. Zajacz, Z., Candela, P.A., Piccoli, P.M., Wälle, M., Sanchez-Valle, C., 2012b. Gold and copper in volatile saturated mafic to intermediate magmas: Solubilities, partitioning, and implications for ore deposit formation. Geochimica et Cosmochimica Acta 91, 140-159. Zhai, Q.G., Jahn, B.M., Zhang, R.Y., Wang, J., Su, L., 2011. Triassic Subduction of the Paleo-Tethys in northern Tibet, China: Evidence from the geochemical and isotopic characteristics of eclogites and blueschists of the Qiangtang Block. Journal of Asian Earth Sciences 42, 1356-1370. Zhang, D., Audétat, A., 2017. What caused the formation of the giant Bingham Canyon porphyry Cu-Mo-Au deposit? Insights from melt Inclusions and magmatic Sulfides. Economic Geology 112, 221-244. Zhang, K.J., Zhang, Y.X., Tang, X.C., Xia, B., 2012. Late Mesozoic tectonic evolution and growth of the Tibetan plateau prior to the Indo-Asian collision. Earth-Science Reviews 114, 236-249. Zhu, D.C., Li, S.M., Cawood, P.A., Wang, Q., Zhao, Z.D., Liu, S.A, Wang, L.Q., 2016. Assembly of the Lhasa and Qiangtang terranes in central Tibet by divergent double subduction. Lithos 245, 7-17. Zhu, D.C., Zhao, Z.D., Niu, Y.L., Dilek, Y., Hou, Z.Q., Mo, X.X., 2013. The origin and pre-Cenozoic evolution of the Tibetan Plateau. Gondwana Research 23, 1429-1454. Zhu, X.P., Li, G.M., Chen, H.A., Ma, D.F., Huang, H.X., 2015. Zircon U-Pb,

34

molybdenite Re-Os and K-feldspar 40Ar/39Ar dating of the Bolong porphyry Cu-Au deposit, Tibet, China. Resource Geology 65, 122-135.

Figure Captions

Fig. 1. (a) Geographic map, (b) sketch tectonic map (Hou et al., 2004; Zhang et al., 2012) and generalized geologic map (c) of the Duolong Cu-Au ore district (Li et al., 2017).

Fig. 2. Photomicrographs and back-scattered electron images of Duolong amphibole from diorite (a, b) and barren granodiorite porphyry (c, d, e, f, g, h). (a, b) Subhedral low-Al amphibole in diorite. (c) Low-Al amphibole with magnetite (Mt) inclusion in barren granodiorite porphyry. (d) Low-Al amphibole with apatite (Ap) inclusion in barren granodiorite porphyry. (e) Resorbed high-Al amphibole (Amp) with the rim replaced by plagioclase and Fe-Ti oxide in barren granodiorite porphyry. (f) Decomposed Low-Al amphibole with weak oscillatory zoning and magnetite and apatite inclusions in barren granodiorite porphyry. (g) Low-Al amphibole with magnetite and plagioclase inclusions in barren granodiorite porphyry. (h) Low-Al amphibole with magnetite, plagioclase and apatite inclusions in barren granodiorite porphyry.

Fig. 3. Classification diagrams of Si vs. Mg/(Mg+Fe2+) (a, Leake et al., 1997) and Si/AlT vs. Mg/(Mg+Fe2+) (b) for Duolong amphibole from diorite and barren granodiorite porphyry. Published amphibole data in ore-bearing and barren granodiorite porphyry samples are from Sun et al. (2017).

35

Fig. 4. Diagrams of MgO (a), CaO (b), MnO (c), FeOT (d), TiO2 (e), Na2O (f), K2O (g), and Cl (h) vs. Al2O3 of Duolong amphibole from diorite and barren granodiorite porphyry. Published amphibole data in ore-bearing and barren granodiorite porphyry samples are from Sun et al. (2017).

Fig. 5. Diagrams of AlIV vs. Mg# (a), Ti (b), (Na+K) A (c), and AlVI (d) illustrating the substitution mechanisms for Duolong amphibole. Positive correlation between AlIV and Mg# values possibly causes by decreased Mg# in mixed magma, suggesting mafic magma replenishment during magma ascent. Published amphibole data in ore-bearing and barren granodiorite porphyry samples are from Sun et al. (2017).

Fig. 6. Chondrite-normalized REE patterns (a, b) and primitive mantle-normalized trace element variation diagrams (c, d) of Duolong amphibole. The normalizing values for REE and trace elements are from Sun and McDonough (1989). The compositions of equilibrium melts are calculated by partition coefficients between amphibole and melt (Nandedkar et al., 2016). Mean compositions of diorite, barren granodiorite porphyry, and basaltic andesite are from Li et al. (2013, 2016a, 2017) and Sun et al. (2017).

Fig. 7. H isotope compositions (a) of amphibole in Duolong diorite and barren porphyries showing fluid exsolution. (b) Modeling of H isotope fractionation for open- and close-system degassing, showing that Duolong amphibole is in equilibrium with degassed melt. Initial H2O content and δD value in melt is 6 wt.% and -50‰, respectively. H isotopic fractionation coefficient is calculated by the equation (Dalou et al., 2015) at 800 °C. The percentage values at curves in (b) represent fraction of 36

H2O lost from the magma.

Fig. 8. Temperature, oxygen fugacity (a), and H2O content (b) for melt equilibrium with Duolong amphibole. Ni-NiO (NNO) and hematite-magnetite (HM) buffers are from Wones (1989). H2O solubility in Cl-rich andesitic melt is from Botcharnikov et al. (2015). Published amphibole data in ore-bearing and barren granodiorite porphyry samples are from Sun et al. (2017). Pressure, temperature, melt H2O concentration, and oxygen fugacity of the Duolong magmas are determined using the empirical amphibole formulation (Ridolfi et al., 2010). Error bars represent relatively consistent model uncertainties of temperature, oxygen fugacity, and H2O content and increased pressure uncertainties from 100 MPa to 400 MPa (Supplementary Data Table 1; Ridolfi et al., 2010).

Fig. 9. Variations of amphibole composition as function of plagioclase and magnetite crystallization, and mafic magma replenishment. (a) Sr (ppm) vs. Al2O3 (wt.%); (b) Sr (ppm) vs. Eu/Eu*; (c) Ti (ppm) vs. V (ppm); (d) AlIV vs. Cr (ppm).

Fig. 10. Experimental phase diagram (Krawczynski et al., 2012) of oxidized arc magmas showing stability of Duolong amphibole. At ~930-960 °C and 300-400 MPa, high-Al amphibole in barren porphyry are stable and show high Sr contents and absence of Eu anomalies. At ~800-900 °C and 100-300 MPa, low-Al amphibole in diorite and barren porphyries show low Sr contents and negative Eu anomalies, caused by plagioclase crystallization.

Fig. 11. Cl content (wt. %) vs. Mg/Mg+FeT (Mg#) in amphibole from the Duolong Cu-Au deposit. Error bar represents Cl uncertainties (2σ). Lines indicate amphibole 37

compositions calculated by equation in Sato et al. (2005), assuming constant Cl/OH ratios in the melt. Published amphibole data in ore-bearing and barren granodiorite porphyry samples are from Sun et al. (2017).

Fig. 12. Simplified schematic evolution of the Cretaceous Duolong magma chamber. (a) Large-scale ore-forming fluid exsolved and early-stage low-Al amphibole crystallized from dioritic magma chamber (~3-8 km). The exsolved fluid ascended and caused hydrothermal alteration, and Cu-Au mineralization surrounding ore-bearing granodiorite porphyry. (b) During crystallization of magma, a crystalline carapace is formed at the edge of the magma chamber. Newly deep-sourced mafic magma with high-Al amphibole was injected into degassed dioritic magma. Subsequently, the newly formed hybrid melt crystallized late-stage low-Al amphibole, and rose to form barren granodiorite porphyry.

Table Captions

Table 1 Hydrogen isotope compositions of amphibole in diorite and barren granodiorite porphyry samples from the Duolong ore district

38

90°

80° 70° Kazakhstan 40°

M

Kyrghyzstan

go

78°

lia

BC

B

angt

Fig.1c

N

India

ang t

ngtan

erran

g terra

ne

an

-G

34°

an

ze

BNS

Lhasa terrane Suture

gp

YCB

e

Coqen

(b)

98°

on

Te r

ra

ne

Changdu

30°

GCB Lhasa

Zhongba Xigaze

ep

Strike-slip fault

al

Thrust belt uta

n

30°

500km 90°

Burm

a

Bh

80°

LSS

N IYS

Fig.1b

32 ° 54´

S.Qi

Gerze

40°

94° AKS S

100°

Vietnam

India

Detachment fault

IYS Indus-Yarlung suture BNS Bangong-Nujiang suture LSS Longmu-Shuanghu suture JSS Jinshajiang suture AKS Ayimaquin-Kunlun suture

0

26°

200km

YCB Yulong copper belt BCB Bangonghu copper belt GCB Gangdese copper belt Duolong ore district

83 ° 13´

83 ° 45´

(c)

2km

Na

ru

o

Sena

N Rongna

E

W

32 ° 54´

gh

ta n

China

90°

N.Qia

30°

20°

86°

Gangar

P a k is

Af

82°

JSS

an

ist

an

Tajikistan

on

100° 50°

S Gaerqing Duobuza

Tiegelong

Bolong

Nadun

Neogene Kangtuo Group Cretaceous Meiriqie Group

Nating

32 ° 43´

GCZK01

83 ° 13´

Ore-bearing granodiorite porphyry Barren granodiorite porphyry

Jurassic Quse Group

Porphyry Cu-Au

Triassic Riganpeicou Group

Epithermal Cu-Au

Early Cretaceous basalt

Drill hole

Diorite

Sample position

83 ° 45´

32 ° 43´

Quaternary

39

1.0 Magnesiohornblende

0.8

0.8

0.7

0.7

Low-Al amphibole

Amphibole in barren porphyry Tsch (high-Al)

0.6

0.6

Mg-Hbl (literature)

High-Al amphibole

Amphibole in diorite

6.2

6.4

6.6

Si

6.8

7.0

7.2

7.4

High-Al amphibole

0.5

Mg-Hbl (low-Al)

Mg-Hbl in ore-bearing porphyry (literature)

6.0

Low-Al amphibole

Mg-Hbl (low-Al)

Tsch (literature)

0.5

Mg/(Mg+Fe 2+)

(b)

0.9

Tschermakite

0.9

1.0

(a)

3

4

5

6

Si/Al T

7

8

9

12.5

Amphibole in barren porphyry

18

Tsch (high-Al) Tsch (literature)

Mg-Hbl (low-Al)

(b)

12.0

(a)

Mg-Hbl (literature)

Amphibole in diorite

11.5

CaO

MgO

10.5

14

6

8

10

12

14

2

18

4

10.0

12 1.0

2

4

6

8

10

12

14

4

6

8

10

12

14

4

6

8

10

12

14

4

6

8

10

12

14

(d)

14

FeO T

8

0.2

10

12

0.6

0.8

16

(c)

0.4

MnO

11.0

16

Mg-Hbl (low-Al)

Mg-Hbl in ore-bearing porphyry (literature)

6

8

10

12

2

14

(f) 2.0

Na 2O

1.5

2.0

(e)

4

6

8

10

12

14

(h)

Cl

0.6

0.1

0.4

0.2

0.8

0.3

1.0

(g)

2

4

6

8

Al 2O 3

10

12

14

0.0

0.2

K 2O

2

0.4

1.2

2

0.5

0.5

1.0

1.0

TiO 2

4

1.5

2

2

Al 2O 3

0.25 0.20

n

e

Ti

a

le

m

0.10

Amphibole in barren porphyry Mg-Hbl (low-Al)

0.55

Tsch (high-Al)

(b)

0.15

0.80 0.65

m

m ag

p re

h is

nt

0.60

Mg#

0.70

0.75

(a)

sc -T Ti

Tsch (literature) Mg-Hbl (literature)

Amphibole in diorite

h

m er

ak

1.0

0.05 1.2

1.4

1.6

0.8

1.8

0.25 0.8

1.0

en

1.2

1.4

Al

IV

1.4

1.6

1.8

h

1.4

1.6

1.8

ak

0.10

A

sc l-T

m er

ak

0.05

c Ts

m

1.2

(d)

0.15

0.20

Ed

ite

r he

1.0

0.00

0.1

0.2

0.3

Al VI

0.5 0.4

(c)

0.0

(Na+K) A

0.6

0.8

0.30

0.50

Mg-Hbl (low-Al)

Mg-Hbl in ore-bearing porphyry (literature)

1.6

1.8

0.8

1.0

1.2

Al

IV

1000

1000

(a)

(b)

Amphibole in diorite

Amphibole in barren porphyry

Sample/ REE chondrite

Mg-Hbl (low-Al)

Tsch (high-Al)

100

100

Basaltic andesite Diorite

10

10

Equilibrium melt with low-Al amphibole Equilibrium melt

Equilibrium melt with high-Al amphibole

Barren granodiorite porphyry

1

1 La Ce Pr Nd Sm Eu Gd Tb Dy

La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu

Ho Er Tm Yb Lu 1000

1000

(d)

(c) Sample/ Primitive Mantle

Mg-Hbl (Low-Al)

100

100

10

10

1

1

0.1

0.1 Rb Ba Nb La Ce Pb Pr Sr Nd Zr Hf SmEu Ti Gd Dy Y Er Yb Lu

Rb Ba Nb La Ce Pb Pr Sr Nd Zr Hf SmEu Ti Gd Dy Y Er Yb Lu

-20

-40

(a)

-40

-100

Closed-system degassing

Arc granite

δD (‰)

-80

-60

Magma degassing

δD (‰)

-60

(b)

Arc magmatic water

20%

30%

-80 40%

20%

30%

50%

-100

40%

Amphibole in equilibrium with degassed melt

-120

-120

50%

Open-system degassing

-140

-140 Amphibole in diorite

Amphibole in barren granodiorite porphyry

2

3

4

H 2O (wt.%)

5

6

O

N

N

O

(b)

+2

M +1

-12

N

rs

ol

u

in

C

h

a

te

Fluid exsolution

O

4

N

W

e at

ty

ic

si

5

O

H 2O melt

-11

N

N

li bi

l-r

e nd

Amphibole in barren porphyry Tsch (high-Al)

-13

Tsch (literature)

Mg-Hbl (low-Al)

Mg-Hbl (literature)

3

Amphibole in diorite Mg-Hbl (low-Al)

Mg-Hbl in ore-bearing porphyry (literature)

-14

log(ƒO 2)

H

N

+3

6

N

-10

-9

(a)

700

750

800

850

900

Temperature (°C)

950

1000

0

100

200

P (MPa)

300

400

1.4

30

1.0

n

0.8

na

ta

io

ag

se

F

c ra

tio

n

c al

ry

st

liz al

at

io

f no

pl

a

o gi

cla

se

50

0.4

0.6

tio

ys

at

pl

la

Eu/Eu*

c ra

r lc

z lli

of

c io

1.2

250 200 150

F

100

Sr (ppm)

(b)

(a)

Amphibole in barren porphyry Mg-Hbl (low-Al)

Tsch (high-Al)

Amphibole in diorite

0.2 12

al

l

of

200

300

250

(d)

20 0

100 2000

150

Sr (ppm)

100

120

tite

100

40

on

tal

on

ne

50

80

500 400

Fr

ti ac

s cry

ti iza

g ma

Cr (ppm)

Mafic magma replenishment

(c)

200

300

0

14

60

10

600

700

8

Al 2O 3 (wt%)

Mafic magma replenishment

0

6

4

V (ppm)

Mg-Hbl (low-Al)

4000

6000

8000

Ti (ppm)

10000

12000

0.6

0.8

1.0

1.2

1.4

Al

IV

1.6

1.8

2.0

+3 p in N

NO

O

am

in N N amp

700

pla

in barren porphyry

opx

300

in

Pressure (MPa)

cpx in

Tsch (high-Al)

o lv in

g in

500

100 Mg-Hbl (low-Al) in diorite and barren porphyry

800

900

1000

1100

Temperature (°C)

1200

0.5

Cl (wt%)

0.4

(C

(C

l/O

Amphibole in barren porphyry

l/O

H

Tsch (high-Al)

)m

=

Mg-Hbl (low-Al)

Tsch (literature)

0.

08

Mg-Hbl (literature)

5

Amphibole in diorite Mg-Hbl (low-Al)

H)

m

=0

.0

Mg-Hbl in ore-bearing porphyry (literature)

55

0.3

0.2

2σ

0.1 (C l/ O H ) m =

(Cl /OH ) m = 0.0 1

0.0 0.50

0.55

0.60

0.65

0.70

Mg/Mg+Fe T

0.75

0 .0 3

0.80

(a)

0

(b)

Current surface

Duolong porphyry-epithermal Cu-Au deposit

Current surface Barren granodiorite porphyry

2

Duolong Cu-Au deposit Ore-forming granodiorite porphyry

Crystalline carapace Low-Al amphibole

4 6 8

Small-scale fluid exsolution from magma

Ore-forming fluid exsolution from magma

Degassed magma Low-Al amphibole

Basalitic magma input

10 12 14

km

Low-Al amphibole

Oxidized and sulfur-rich magma input

High-Al amphibole

Oxidized and sulfur-rich magma input

Table 1

Sample No. Dbz-2 Dbz-1 DL-28 Nd-10 Dd-8 SJ-2 SJ-1 GCZK001-461 GCZK001-466

Lithology Barren granodiorite porphyry Barren granodiorite porphyry Barren granodiorite porphyry Barren granodiorite porphyry Barren granodiorite porphyry Diorite Diorite Diorite Diorite

40

Mineral Amphibole Amphibole Amphibole Amphibole Amphibole Amphibole Amphibole Amphibole Amphibole

δD (‰) -116 -122 -102 -109 -116 -105 -113 -103 -111

(a)

0

(b)

Current surface

Duolong porphyry-epithermal Cu-Au deposit

Current surface Barren granodiorite porphyry

2

Duolong Cu-Au deposit Ore-forming granodiorite porphyry

Crystalline carapace Low-Al amphibole

4 6 8

Small-scale fluid exsolution from magma

Ore-forming fluid exsolution from magma

Degassed magma Low-Al amphibole

Basalitic magma input

10 12 14

km

Low-Al amphibole

Oxidized and sulfur-rich magma input

High-Al amphibole

Oxidized and sulfur-rich magma input

Highlights > Low-Al Mg-hornblende in ore-bearing diorite and barren porphyry formed at ~ 750-870 °C and ~ 80-200 MPa. > Large Cl variations and low δD values in amphibole indicate that the dioritic magma underwent fluid exsolution. > Small Cl variations in amphibole from barren porphyry were controlled by Mg# rather than fluid exsolution. > Cl content variation of amphibole may be a powerful tool to identify ore-bearing and barren intrusions.

41