Petrogenesis and Metallogenesis of Late Cretaceous Adakites in the Nuri Large Cu-W-Mo Deposit, Tibet, China: Constraints from Geochronology, Geochemistry, and Hf Isotopes

Abstract

The Gangdese metallogenic belt in Tibet is an important polymetallic metallogenic belt formed during the subduction of the Neo-Tethys Ocean and subsequent India–Asia collision. Adakitic rocks are widely distributed in this belt and are considered to be closely related to porphyry–skarn Cu-Mo polymetallic mineralization. However, the petrogenesis and geodynamic setting of the Late Cretaceous adakites in the Gangdese belt remain controversial. In this study, we focus on the quartz diorite in the Nuri Cu-W-Mo deposit along the southern margin of the eastern Gangdese belt. LA-ICP-MS zircon U-Pb dating yields a Late Cretaceous age of 93.6 ± 0.4 Ma for the quartz diorite. Whole-rock geochemistry shows that the quartz diorite possesses typical adakitic signatures, with high SiO₂, Al₂O_3, and Sr contents, but low Y and Yb contents. The relatively low K₂O content and high MgO, Cr, and Ni contents, as well as the positive zircon εHf(t) values (+6.58 to +14.52), suggest that the adakites were derived from the partial melting of the subducted Neo-Tethys oceanic slab, with subsequent interaction with the overlying mantle wedge. The Late Cretaceous magmatic flare-up and coeval high-temperature granulite-facies metamorphism in the Gangdese belt were likely triggered by Neo-Tethys mid-ocean ridge subduction. The widespread occurrence of Late Cretaceous adakitic intrusions and associated Cu mineralization in the Nuri ore district indicate a strong tectono-magmatic-metallogenic event related to the Neo-Tethys subduction during this period. This study provides new insights into the petrogenesis and geodynamic setting of the Late Cretaceous adakites in the Gangdese belt, and has important implications for Cu polymetallic deposit exploration in this region.

1. Introduction

The term “adakite” was first introduced by Defant and Drummond [1] to specifically refer to intermediate to felsic magmas formed by the partial melting of subducted young oceanic crust. These magmas are characterized by high silica content (SiO₂ ≥ 56 wt.%), high aluminum oxide content (Al₂O₃ ≥ 15 wt.%), elevated strontium (Sr ≥ 400 ppm), a depletion in yttrium (Y ≤ 18 ppm), and heavy rare earth elements (HREEs, Yb ≤ 1.9 ppm). However, subsequent research revealed that igneous rocks with adakitic geochemical compositions similar to adakites are also distributed outside oceanic subduction environments, including continental collision and intracontinental extension settings [2,3,4]. The petrogenesis of adakites remains controversial, mainly focusing on the following: (1) the partial melting of subducted oceanic crust [1,5,6]; (2) the partial melting of thickened mafic lower crust [2,7,8,9]; (3) the partial melting of delaminated crust [3,4]; (4) the partial melting of subducted continental crust [10]; (5) the crustal contamination and fractional crystallization of basaltic parental magmas [11,12]; (6) and magma mixing between mafic and felsic magmas [13,14].

Following the subduction and eventual closure of the Mesozoic Neo-Tethys Ocean, the Gangdese area in Tibet successively underwent accretionary orogenesis and collisional orogenesis, forming a gigantic tectono-magmatic belt [15,16,17,18]. Long-term and in-depth studies over the past few decades have shown that this belt is widely distributed with rocks possessing adakitic geochemical characteristics, spanning a long time period, from the Jurassic to the Miocene [2,7,19,20,21,22,23,24,25]. These Cretaceous adakites are mainly distributed in the Nimu–Lhasa area, Zhanang–Sangri area, and Langxian–Milin area in the eastern part of the Gangdese belt [23,24,25,26,27,28,29]. The rock types of the Cretaceous adakites in the Gangdese belt are diverse, including volcanic rocks like andesite and dacite, and intrusive rocks mainly comprising quartz diorite, quartz monzonite, granodiorite, quartz porphyry, tonalite, and trondhjemite. However, the petrogenesis of adakites in this stage, especially in the Late Cretaceous, is quite controversial, mainly including three genetic models: (1) the partial melting of the subducted oceanic slab [6,24,25,30,31,32]; (2) the partial melting of thickened lower crust [25]; (3) and the fractional crystallization of basaltic magmas [27]. At the same time, the geodynamic mechanism producing the Late Cretaceous adakitic magmatism also remains disputed, including (1) the mid-ocean ridge subduction model [25,28,33]; (2) the low-angle or flat subduction model [23,29]; (3) and the slab rollback model [30,31,32,34].

The Gangdese belt is a key metallogenic region within the Qinghai–Tibet Plateau collisional orogenic belt, where both continental collision porphyry Cu-Mo and magmatic arc porphyry Cu-Au deposits have been discovered, with their formation considered to be closely related to adakites [2,7,15,16,21,35,36,37,38]. In terms of the metallogenic time scale, typical large to giant porphyry deposits within the Gangdese belt mainly formed during the Cenozoic collisional orogenic stage. Notably, during the northward subduction of the Mesozoic Neo-Tethys oceanic plate, significant magmatism created the 1500 km long Gangdese batholith [29,39,40,41]. Unlike the Jurassic magmatic arc where the Xiongcun giant porphyry Cu-Au deposit has been discovered [21,42], no large porphyry deposits related to the Cretaceous magmatic arc have been found so far. According to the minerogenetic series of mineral deposits [43,44,45,46], there should exist a series of metallogenic responses to the northward subduction of the Yarlung Zangbo Tethys Ocean leading to the collision between the Indian and Asian continents in the Gangdese belt [36,42,47,48,49,50,51,52,53]. Therefore, clarifying the petrogenesis of the Cretaceous adakites in the Gangdese metallogenic belt and constraining their corresponding geodynamic settings are crucial for the further exploration of the relationship between the Cretaceous magmatic arc and porphyry mineralization.

Extensive Late Cretaceous adakites are exposed along the southern margin of the Gangdese belt, making it an ideal site to study Late Cretaceous adakitic magmatism. Given the varied interpretations of the petrogenesis and geodynamic settings of these adakites by previous researchers, this study focuses on the quartz diorite dykes in the southern Nuri region along the eastern Gangdese belt. Through LA-ICP-MS zircon U-Pb geochronology and whole-rock geochemistry, this research aims to investigate the source and petrogenesis of the Late Cretaceous adakitic magmatism in the Gangdese metallogenic belt, further constrains their formation geodynamic background, and helps to expand the prospecting direction in this area.

2. Geological Setting

The Gangdese metallogenic belt spans from the Bangong Lake–Nujiang suture zone in the north to the Indus–Yarlung Zangbo suture zone in the south (Figure 1). Its formation is intricately linked to the geological evolution of the Paleo-Tethys and Neo-Tethys oceans, undergoing a multifaceted tectono-magmatic process [54]. As a result, it has gained global recognition as a significant Cu-Mo-Pb-Zn-Au-Ag polymetallic metallogenic belt [41,55,56]. The eastern region, among the most mineral-rich sections of the Gangdese belt, hosts numerous porphyry systems, which originated from the Mesozoic northwards subduction of the Yarlung Zangbo Neo-Tethys Ocean and continued through the Cenozoic collision between the Indian and Asian continents, followed by a subsequent extensional phase [57] (Figure 1).

Based on the types of metallogenic combinations and their spatial distribution, the central and eastern parts of the Gangdese belt can be broadly categorized into three metallogenic subzones: the northern, central, and southern subzones [41]. The northern subzone exhibits skarn–porphyry Pb-Zn-Ag polymetallic deposits, such as the Sharang porphyry Mo, Yaguila skarn–porphyry Pb-Zn-Mo, Mongyaa skarn Pb-Zn, and Dongzhongsongduo skarn Pb-Zn deposits. The central subzone is the most substantial concentration of mineralization in the Qinghai–Tibet Plateau orogenic belt, comprising a porphyry–skarn Cu-Mo polymetallic metallogenic subzone with ultra-large deposits, exemplified by the Jiama and Qulong, and some medium-large deposits (e.g., the Chongjiang, Zhunuo, and Tinggong deposits). The southern subzone is characterized by porphyry–skarn Cu-W-Mo polymetallic mineralization, hosting notable deposits such as the Nuri Cu-W-Mo, Mingze Cu-Mo, Chengba Mo-Cu, Chenba Cu-Au, Chongmuda Cu-Au, and Kelu Cu-Au deposits [34,59].

This study focuses on the Nuri region, the southern subzone of the eastern Gangdese metallogenic belt (Figure 1). Tectonically, it is situated along the southern boundary of the eastern Gangdese volcanic magmatic arc belt and at the northern edge of the Yarlung Zangbo suture zone. Based on different metal combinations and mineralization epochs, the deposit can be divided into two distinct mineralization sections: northern and southern. The northern section comprises skarn-type Cu-Fe deposits, while the southern section consists of skarn-type Cu-W-Mo deposits [49] (Figure 2).

The exploration level of the skarn-type Cu-Fe deposits in the northern section is relatively low. Currently, the exposed Cu-Fe ore bodies are mainly controlled by a single flat lens, with dimensions of approximately 200 m in length and 5–10 m in width. These ore bodies are predominantly hosted within medium-grained garnet-rich skarns and are closely associated with a suite of Late Cretaceous adakitic quartz diorites (zircon U-Pb dating yields 93.0 ± 1.4 Ma) [60]. The skarn Cu-W-Mo deposits in the southern section consist of Cu, W, and Mo co-existing/mineralizing bodies. These minerals are spatially distributed as both single-mineral independent bodies and multi-mineral coexisting bodies. They primarily occur in the fourth segment (K₁b⁴) of the Lower Cretaceous Bima Formation, generally situated within the interlayer detachment fault zone (F₂) between carbonate rocks and clastic rocks, displaying distinct layer-like to pseudo-layer-like characteristics. Due to oblique slip along the F₃ fault, the ore layers extend in an “S” or reverse “S” shape on the plane, trending overall in a northwest direction, with the dip angle gradually decreasing from south to north. The estimated and inferred metal resources are approximately 553,800 tons of Cu, averaging 0.70% in grade; 197,200 tons of WO₃, averaging 0.22% in grade; and 32,400 tons of Mo, averaging 0.067% in grade. Both Cu and WO₃ resources are classified as large-scale, while Mo is considered medium-scale. The Nuri deposit stands out as the only polymetallic deposit with WO₃ resources reaching the large-scale classification, rendering it immensely valuable for exploration and research purposes.

The magmatic activity in the Nuri ore district is intense, with a large time span of rock formation, from the Late Cretaceous to the Miocene [34,60,61]. The rock types are complex, mainly including andesite, quartz diorite, quartz porphyry, quartz monzonite, granodiorite, monzogranite, and minor granite porphyry. Quartz diorite is mainly distributed in the northern and southern parts of this region. Quartz–sulfide veins are often embedded in the quartz diorite of the southern ore segment, especially in the deep part of drill holes, where disseminated chalcopyrite is commonly developed, with local continuous mineralization reaching 10.69 m and an average Cu grade of 0.37 wt.% (according to the internal company report). The quartz diorite samples in this study were all collected from drill hole No. ZK4501, and the sampling locations are shown in Figure 3. The samples are generally gray in color, with a hypidiomorphic granular texture and massive structure (Figure 4). The rock composition is mainly plagioclase and hornblende, followed by quartz, and minor K-feldspar and biotite. Plagioclase accounts for more than 60%, occurring as hypidiomorphic tabular, with a grain size of 2.0–5.0 mm, developing polysynthetic twinning, and showing zoning structure, randomly distributed. Meanwhile, uneven sericitization, epidotization, and carbonation can be observed. Hornblende accounts for about 15%, occurring as hypidiomorphic–xenomorphic columnar grains, 0.2–3.0 mm in size, commonly actinolitized, and minorly epidotized and biotitized, with sulfides precipitated from it. Quartz accounts for about 10%, occurring as xenomorphic granular grains, 0.2–3.0 mm in size, showing undulatory extinction, interstitially or randomly distributed. K-feldspar accounts for about 5%, occurring as xenomorphic granular grains, 0.2–2.0 mm in size, developing cross-hatched twinning, kaolinized, and sporadically distributed in the rock. Biotite accounts for less than 5%, occurring as flaky, locally chloritized, with sulfides precipitated from it, and sporadically distributed in the rock.

3. Analytical Methods

Samples for geochemical analyses were ground to pass through a 200-mesh sieve and further ground and homogenized in an agate mortar under alcohol. The major and trace elements in this study were conducted at the Analysis and Testing Center of the Beijing Research Institute of Uranium Geology, China. The major elements were analyzed using a Riken Primus II X-ray fluorescence spectrometer (XRF) from Japan, and the analytical precision was better than 1%. The trace elements were determined by the through acid digestion of samples in Teflon screw-cap capsules, followed by analysis using a Thermo Scientific X Series II ICPMS instrument (Germany). The uncertainty of the trace elements examined here was within 5%.

Zircon separation, mounting, transmitted light, reflected light, cathodoluminescence imaging, U-Pb isotope dating, and trace element content testing were all completed at Tuoyan Testing Technology Co., Ltd., Guangzhou, China. Zircons were separated using heavy-liquid and magnetic methods, followed by hand-picking purification under a binocular microscope. Cathodoluminescence (CL) images were obtained using an environmental scanning electron microscope equipped with a Gatan CL3 detector, with operating conditions set at 15 kV and 20 nA.

Zircon U-Pb dating was utilized using a New Wave Research 193 nm ArF excimer laser ablation system (Bozeman, Montana, USA) coupled with a Thermo Scientific iCap-RQ quadrupole inductively coupled plasma mass spectrometer (ICP-MS) (Hachioji, Tokyo, Japan). The laser ablation parameters included a 30 μm spot diameter, 6 Hz frequency, and 3.5 J/cm² energy density. Helium served as the carrier gas, and argon as the compensation gas during laser ablation. Each time-resolved analysis data included approximately 45 s of a blank signal and 40 s of a sample signal. Eight unknown samples were tested between every two sets of standard sample tests. Instrumental mass discrimination and elemental fractionation were corrected using the zircon standard 91,500 as the external standard. NIST SRM 610 glass was used as the external standard, with Si as the internal standard element to calibrate the Pb content in the zircon, and Zr as the internal standard element to calibrate the remaining trace element content in the zircon. The standard zircons Plešovice and Tan-Z were used as monitoring samples to evaluate the quality of the U-Pb dating data. In this experiment, the ages of the monitoring zircons Plešovice and Tan-Z were 337.37 ± 1.03 Ma and 565.16 ± 1.97 Ma, consistent with the recommended values within error [62,63]. The original test data were processed offline using iolite4 software (including the selection of samples and blank signals, correction of instrument sensitivity drift, and the calculation of element content, U-Pb isotope ratios, and ages). IsoplotR was used for age calculation and concordia diagram plotting. For more detailed analysis details, please refer to [64].

Zircon Hf isotope analysis was conducted by a Neptune multi-collector inductively coupled plasma mass spectrometer (MC-ICP-MS) (Wrexham, Wales, UK) and its matching laser ablation system (Bozeman, Montana, USA) based on the same spot or equivalent location of U-Pb dating, with the analysis spots shown in Figure 5. During the test, the laser spot diameter was 40 μm, the ablation time was about 26 s, the laser pulse rate was 6 Hz, and the pulse energy was 100 mJ/cm. The zircon international standard 91,500 was used as the external standard during the experiment. Detailed operating and analytical procedures are the same as [64]. The calculation formulas and related parameters for εHf(t) values and Hf model ages are as follows [65,66,67,68]:

ε_Hf(t) = 1000{[(¹⁷⁶Hf/¹⁷⁷Hf)_sample − (¹⁷⁶Lu/¹⁷⁷Hf)_sample × (e^λt − 1)]/[(¹⁷⁶Hf/¹⁷⁷Hf)_CHUR,0 − (¹⁷⁶Lu/¹⁷⁷Hf)_CHUR × (e^λt − 1) − 1]};

T_DM = 1/λ × ln{1 + [(¹⁷⁶Hf/¹⁷⁷Hf) _sample − (¹⁷⁶Hf/¹⁷⁷Hf)_DM]/[(¹⁷⁶Lu/¹⁷⁷Hf) _sample − (¹⁷⁶Lu/¹⁷⁷Hf) _DM]};

T_DM^C = 1/λ × ln{1 + [(¹⁷⁶Hf/¹⁷⁷Hf) _sample,t − (¹⁷⁶Hf/¹⁷⁷Hf)_DM,t]/[(¹⁷⁶Lu/¹⁷⁷Hf)_C − (¹⁷⁶Lu/¹⁷⁷Hf) _DM]} + t;

f_Lu/Hf = (¹⁷⁶Lu/¹⁷⁷Hf)_sample/(¹⁷⁶Lu/¹⁷⁷Hf)_CHUR − 1;

(¹⁷⁶Lu/¹⁷⁷Hf)_CHUR = 0.0332, (¹⁷⁶Hf/¹⁷⁷Hf)_CHUR,0 = 0.282772;

(¹⁷⁶Lu/¹⁷⁷Hf) _DM = 0.0384, (¹⁷⁶Hf/¹⁷⁷Hf)_DM = 0.28325;

λ = 1.867 × 10⁻¹¹year⁻¹;

(¹⁷⁶Lu/¹⁷⁷Hf)_C = 0.015;

(¹⁷⁶Hf/¹⁷⁷Hf)_sample and (¹⁷⁶Lu/¹⁷⁷Hf)_sample are the experimental measured values of the samples, and t is the corresponding U-Pb age of the zircon grain.

4. Results

4.1. Zircon U-Pb Ages

CL images of zircons from the quartz diorite in the drilling hole (ZK4501-450) show that they are grayish-white in color, and most of them are well-formed, euhedral–subhedral, and mainly long prismatic, with a few short prismatic (Figure 5). Their long axes range from 50 to 130 μm, and short axes from 40 to 80 μm, with the length-to-width ratios mostly between 1:1 and 2:1.

Although a few grains of the CL images show some core–rim textures, the LA-ICP-MS spots were carefully placed on the clean, oscillatory-zoned rims to avoid any potential inheritance. The highly concordant U-Pb ages obtained indicate the analyzed domains reflect the primary magmatic growth phase rather than mixed ages (Figure 5 and

Table 1. LA-ICP-MS zircon U-Pb analytical results of quartz diorite in the Nuri deposit.

No.	Pb (ppm)	Th (ppm)	U (ppm)	Th/U	207Pb/206Pb	1s%	207Pb/235U	1s%	206Pb/238U	1s%	207Pb/235U	2s (abs)	206Pb/238U	2s (abs)
1	2.1	339.9	310.8	1.1	0.0495	4.89	0.0977	4.65	0.0143	1.04	94.2	8.4	91.6	1.9
2	3.5	766.0	511.1	1.5	0.0469	4.22	0.0962	4.19	0.0148	0.86	92.9	7.4	94.6	1.6
3	2.9	590.4	418.4	1.4	0.0503	4.13	0.1023	4.19	0.0147	1.08	98.6	7.9	93.8	2.0
4	1.0	151.9	156.0	1.0	0.0435	7.52	0.0867	7.24	0.0144	1.77	86.4	12.8	92.4	3.3
5	1.6	249.4	237.8	1.0	0.0507	6.95	0.1050	6.54	0.0150	1.37	100.5	12.7	96.0	2.6
6	2.9	612.8	416.5	1.5	0.0508	4.46	0.1031	4.26	0.0147	1.11	99.3	8.1	94.3	2.1
7	3.5	1058.6	523.4	2.0	0.0440	4.79	0.0892	4.88	0.0145	1.08	86.4	8.2	93.0	2.0
8	0.9	168.4	140.3	1.2	0.0534	7.71	0.1038	6.87	0.0142	1.92	99.3	13.0	91.1	3.5
9	3.1	446.6	456.4	1.0	0.0457	3.99	0.0918	3.97	0.0145	1.13	88.9	6.8	92.8	2.1
10	3.4	683.0	503.0	1.4	0.0453	5.63	0.0904	5.54	0.0144	1.10	87.5	9.3	92.0	2.0
11	2.3	218.4	340.6	0.6	0.0454	5.61	0.0919	5.52	0.0146	1.10	88.8	9.4	93.6	2.0
12	1.1	123.5	162.7	0.8	0.0441	8.07	0.0870	7.87	0.0144	1.82	83.8	12.8	92.2	3.3
13	0.8	109.0	115.3	0.9	0.0427	9.00	0.0856	8.67	0.0141	1.78	82.3	13.7	90.2	3.2
14	1.5	168.3	214.9	0.8	0.0482	6.12	0.1000	5.93	0.0151	1.32	96.1	10.9	96.5	2.5
15	1.4	165.4	215.1	0.8	0.0467	8.19	0.0897	7.54	0.0145	1.42	86.3	12.4	92.7	2.6
16	4.4	885.5	653.5	1.4	0.0507	3.71	0.1033	3.62	0.0147	0.84	99.56	6.9	94.1	1.6
17	2.0	213.1	301.2	0.7	0.0463	5.56	0.0927	5.62	0.0145	1.08	89.5	9.6	92.7	2.0
18	2.0	233.3	285.4	0.8	0.0467	5.63	0.0969	5.41	0.0151	1.12	93.4	9.7	96.6	2.2
19	1.4	209.2	209.1	1.0	0.0484	6.97	0.0984	6.62	0.0148	1.34	94.5	12.0	94.8	2.5
20	3.0	346.5	458.3	0.8	0.0458	4.99	0.0911	4.82	0.0145	1.19	88.15	8.2	92.6	2.2
21	2.6	326.5	373.3	0.9	0.0448	4.41	0.0934	4.47	0.0151	1.24	90.3	7.7	96.4	2.4
22	3.1	286.9	454.6	0.6	0.0490	3.80	0.1001	3.89	0.0148	1.14	96.6	7.2	94.4	2.1
23	4.5	958.2	676.2	1.4	0.0474	3.93	0.0952	3.90	0.0145	0.72	92.1	6.7	92.7	1.3

).

A total of 23 zircons were dated using the U-Pb method in this study (Table 1), with U contents of 115.3 ppm–676.2 ppm, Th contents of 109 ppm–1058.6 ppm, and Th/U ratios of 0.6–2.0 (>0.1), indicating a magmatic origin of the zircons [69]. On the concordia diagram (Figure 6), the data points are relatively concentrated, yielding a weighted average ²⁰⁶Pb/²³⁸U age of 93.6 ± 0.4 Ma (n = 23 and MSWD = 2.1), indicating that the quartz diorite in the southern part of the Nuri ore district formed in the Late Cretaceous.

4.2. Whole-Rock Geochemistry

The major, trace, and rare earth element analysis results of the quartz diorite samples from the Nuri deposit are listed in

Table 2. Whole-rock major element (wt.%) and trace element (ppm) results of quartz diorite in the Nuri deposit.

Sample	4501–450.3	4501–451.7	4501–452.4	4501–453.6	4501–454.8
major element (wt.%)
SiO2	64.70	63.94	64.07	65.91	64.88
TiO2	0.72	0.65	0.75	0.55	0.61
Al2O3	15.54	15.33	14.14	14.71	15.03
TFe2O3	3.86	5.17	5.91	3.94	5.33
MnO	0.07	0.07	0.09	0.06	0.05
MgO	3.33	2.96	3.55	2.35	2.29
CaO	5.39	5.45	5.35	5.22	5.49
Na2O	3.05	3.20	3.73	3.97	3.36
K2O	1.60	0.95	1.07	1.10	1.17
P2O5	0.32	0.28	0.21	0.22	0.17
LOI	1.41	1.86	1.10	1.93	1.52
Total	99.99	99.86	99.97	99.98	99.91
Mg#	63.34	53.40	54.57	54.45	46.22
trace element (ppm)
V	122.00	121.55	103.53	93.46	86.80
Cr	116.00	69.50	118.31	66.30	72.90
Co	4.10	14.56	20.12	10.07	40.04
Ni	66.30	45.81	70.11	34.23	46.02
Ga	17.42	13.86	19.93	15.90	18.50
Rb	90.50	81.40	79.21	90.16	96.25
Sr	780.00	730.34	674.56	686.42	601.88
Y	10.26	11.50	10.70	13.50	12.50
Zr	40.30	59.50	68.41	61.90	70.20
Nb	4.52	5.95	6.98	5.90	10.20
Ba	318.00	151.33	203.19	166.54	182.67
La	23.99	27.89	30.71	26.39	27.03
Ce	47.29	51.82	55.33	50.51	46.19
Pr	5.62	5.97	6.11	5.38	5.53
Nd	23.49	22.53	24.12	19.96	20.43
Sm	3.78	4.45	4.7	3.84	3.82
Eu	1.08	1.03	1.25	0.92	1.11
Gd	2.92	3.49	2.99	2.92	2.97
Tb	0.40	0.49	0.50	0.43	0.47
Dy	1.88	2.72	2.69	2.35	2.69
Ho	0.35	0.51	0.51	0.45	0.52
Er	0.98	1.43	1.51	1.31	1.54
Tm	0.16	0.20	0.19	0.19	0.22
Yb	0.99	1.35	1.49	1.28	1.56
Lu	0.16	0.19	0.20	0.19	0.22
Hf	1.31	1.92	2.09	1.98	2.20
Ta	0.39	0.68	0.69	0.60	0.87
Pb	9.37	14.70	15.30	11.40	17.10
Th	8.66	8.54	9.66	9.41	7.82
U	2.22	2.26	3.07	2.67	2.42

Note: LOI = H₂O + CO₂; Mg# = Mg²⁺/(Mg²⁺ + Fe²⁺) × 100.

. The samples have an alumina saturation index (A/CNK) of 0.83~0.94 (averaging 0.89) and A/NK values of 1.90~2.43 (averaging 2.16), belonging to metaluminous granite (Figure 7a). On the SiO₂-K₂O diagram, the samples align within the low-K to medium-K calc-alkaline series (Figure 7b).

The chondrite-normalized REE patterns of the quartz diorite display an overall right-dipping trend (Figure 8a), characterized by a relative enrichment in LREE and a depletion in HREE, with LREE/HREE = 10.20~13.41, and (La/Yb)_N = 11.77~16.40. The samples exhibit a weak negative Eu anomaly (δEu = 0.77~0.97). Illustrated on the primitive mantle-normalized spider diagram (Figure 8b), the samples show a relative enrichment in large ion lithophile elements (LILEs) and a depletion in high field strength elements (HFSEs).

The quartz diorite in the Nuri ore district is distinguished by high SiO₂ (63.94 wt.%~65.91 wt.%), high Al₂O₃ (14.14 wt.%~15.54 wt.%), high Sr (601 ppm~780 ppm), high MgO (2.29 wt.%~3.55 wt.%), low Y (10.26 ppm~13.5 ppm), and low Yb (0.99 ppm~1.56 ppm) contents, indicating typical adakitic geochemical characteristics [1]. Moreover, on the (La/Yb)_N-Yb_N and Sr/Y-Y diagrams, the samples reside within the adakite field (Figure 9a,b), similar to the adakitic rocks in this ore district reported by previous studies.

4.3. Zircon Lu-Hf Isotopes

Zircon is a prevalent accessory mineral in igneous rocks and typically exhibits high Hf content but extremely low Lu content, resulting in very low ¹⁷⁶Lu/¹⁷⁷Hf ratios. All zircons analyzed in the quartz diorite in this study exhibit ¹⁷⁶Lu/¹⁷ ratios < 0.02 (Figure 10,

Table 3. Zircon Hf isotopic composition data for quartz diorite in the Nuri deposit.

No.	Age (Ma)	176Yb/177Hf	2σ	176Lu/177Hf	2σ	176Hf/177Hf	2σ	IHf	εHf (0)	εHf (t)	TDM (Ma)	TDMC (Ma)	f Lu/Hf
1	91.6	0.026918	0.000114	0.000594	0.000018	0.283104	0.000032	0.283103	11.7	13.7	206	276	−0.98
2	94.0	0.031483	0.000191	0.000527	0.000035	0.283092	0.000011	0.283091	11.3	13.4	223	302	−0.98
3	93.8	0.020268	0.000182	0.000664	0.000013	0.282993	0.000006	0.282992	7.8	9.8	364	530	−0.98
4	92.4	0.053948	0.000206	0.001756	0.000033	0.283018	0.000021	0.283015	8.7	10.6	338	477	−0.95
5	96.0	0.028073	0.000179	0.000615	0.000026	0.283124	0.000016	0.283123	12.5	14.5	178	228	−0.98
6	94.3	0.042756	0.000232	0.001161	0.000021	0.283107	0.000012	0.283105	11.9	13.9	205	270	−0.97
7	93.0	0.023031	0.000148	0.000762	0.000036	0.283034	0.000022	0.283033	9.3	11.3	307	437	−0.98
8	91.1	0.051896	0.000376	0.001868	0.000023	0.283014	0.000060	0.283011	8.6	10.5	345	487	−0.94
9	92.8	0.029022	0.000131	0.000644	0.000042	0.283096	0.000014	0.283095	11.5	13.5	218	295	−0.98
10	92.0	0.036752	0.000096	0.000701	0.000021	0.283101	0.000013	0.283100	11.6	13.6	211	284	−0.98
11	93.6	0.041246	0.000081	0.000834	0.000008	0.282901	0.000009	0.282900	4.6	6.6	495	738	−0.97
12	92.2	0.046546	0.000211	0.000986	0.000017	0.283019	0.000038	0.283017	8.7	10.7	330	472	−0.97

), indicating minimal accumulation of radiogenic Hf post-zircon formation. Consequently, these ratios effectively reflect the Hf isotopic composition characteristics of the magma during zircon formation [76]. Twelve zircon spots yielded initial ¹⁷⁶Hf/¹⁷⁷Hf ratios of 0.28290~0.28312 and εHf(t) values of +6.58~+14.52 (averaging 11.94). The corresponding depleted mantle model ages (T_DM) are 178 Ma~495 Ma, while the crustal model ages (T_DM^C) span from 228 Ma to 738 Ma.

5. Discussion

5.1. Magma Source

The Yarlung Zangbo oceanic plate began subducting northward beneath the Lhasa block in the Early Jurassic, and at about 65 Ma, the Neo-Tethys Ocean basin closed, and the Indian continent began colliding with the southern margin of the Asian continent [15,16]. The quartz diorite in the southern part of the Nuri ore district in this study formed in the Late Cretaceous (93.6 ± 0.4 Ma), temporally during the northward subduction stage of the Neo-Tethys oceanic crust.

The Late Cretaceous quartz diorites from the Nuri mining area exhibit high MgO contents (2.29–3.55 wt.%, averaging 2.90 wt.%) and elevated Mg# values (46.22–63.34, averaging 54.40), as well as compatible element concentrations of Cr (66.30–116.00 ppm, averaging 88.60 ppm) and Ni (34.23–70.11 ppm, averaging 55.49 ppm). These geochemical characteristics rule out the possibility of their origin from the partial melting of thickened mafic lower crust. High MgO contents and elevated concentrations of Cr and Ni are typically attributed to the interaction between ascending melts and the overlying mantle wedge during the subduction or delamination of the lower crust [5], often occurring in areas of subduction or lower crustal delamination [3,4,5,6]. While the lower crustal delamination model requires the presence of thickened continental crust [3,4], Zhu et al. [24] pointed out that the crustal thickening in the Gangdese belt began at 70~60 Ma based on whole-rock (La/Yb)_N ratios, significantly later than the emplacement time of the Late Cretaceous quartz diorite in the southern part of the Nuri ore district. Therefore, the lower crustal delamination model cannot be used to explain the magmatic origin of the quartz diorites in the Nuri mining area. Additionally, no contemporaneous ultramafic rocks or basalts associated with Late Cretaceous adakites were found in the study area. Previous studies have indicated that low-pressure plagioclase, amphibole, and fractional crystallization can result in a downward-concave pattern of rare earth element distribution between MREE and HREE, along with distinct negative Eu anomalies, while high-pressure amphibole and garnet fractional crystallization can lead to an increase in Y/Yb ratios in residual melts [41]. According to Figure 8a, the Late Cretaceous quartz diorites from Nuri do not exhibit the aforementioned rare earth element distribution pattern characteristics, lack significant negative Eu anomalies, and show little variation in Y/Yb ratios (7.18–10.55), and thus cannot be explained by mantle-derived basaltic magma crystallization and differentiation patterns. Therefore, the possible causative explanation is the partial melting of the subducted Neo-Tethyan oceanic slab.

Due to the very high closure temperature of the Lu-Hf isotope system in zircons, the Hf isotope ratios of zircons will not change with later partial melting or fractional crystallization, so the εHf(t) values of zircons represent the compositional characteristics of the magma source region [58]. The εHf(t) values of the zircons in the quartz diorite from the Nuri ore district are +6.58~+14.52, all positive values. On the εHf(t)—-t diagram, they all fall between the chondritic and depleted mantle evolution lines (Figure 10b); meanwhile, the two-stage model ages (T_DM^C) are 228 Ma–738 Ma. This also indicates that the quartz diorite in the Nuri ore district originated from the depleted mantle or juvenile crust newly accreted from the depleted mantle [60,76].

Partial melting experiments on granite indicate that the potassium content of the initial material has a significant influence on the potassium content of the derived melt [77,78]. The whole-rock geochemical analysis results show that the quartz diorite in the Nuri ore district possesses typical adakitic geochemical characteristics, with relatively low K₂O contents (0.95%~1.60%, averaging 1.18%), which is close to the average K₂O content (1.72 wt.%) of adakites formed by slab melting, and much lower than the K₂O contents (2.9 wt.%~4.1 wt.%) of adakites formed by the partial melting of the lower crust [3,79].

As shown in Figure 11a,b, the quartz diorite samples in the Nuri ore district mostly fall in the field of adakites formed by the partial melting of subducted oceanic crust, very similar to the distribution characteristics of adakites reported by many scholars [24,25,72] that were formed by the partial melting of subducted oceanic crust, and obviously different from the adakites reported by [19,23] that were formed by the partial melting of thickened lower crust. Experimental petrology shows that slab melts will be rapidly consumed in the overlying mantle wedge and have difficulty rising to the shallow crust [80]. Only when the slab melt/mantle wedge ratio is very high can the slab melt successfully pass through the overlying mantle wedge [5]. High-SiO₂ adakites (HSA) and low-SiO₂ adakites (LSA) are respectively used to represent slab melts that have undergone metasomatism with the overlying mantle wedge and adakitic melts formed by the partial melting of the mantle wedge that has been metasomatized by slab melts [75]. The samples in this study mainly fall into the region of HSA in the discrimination diagrams (Figure 11c,d), combined with the results of Hf isotopic analysis, fully indicating that they originated from the partial melting of the subducted slab and underwent interaction reactions with the mantle wedge during ascent.

5.2. Geodynamic Setting

During the Late Cretaceous (109~76 Ma), there was a magmatic flare-up event in the Gangdese belt, with the widespread development of adakites as well as mantle-derived normal arc magmas [26,27,30]. Large-scale magmatism usually corresponds to specific tectono-magmatic events. However, the triggering mechanism of the Late Cretaceous magmatic flare-up remains uncertain, with three subduction-related geodynamic models proposed, including (1) the mid-ocean ridge subduction model [25,28,33]; (2) the low-angle or flat subduction model [29,30]; (3) and the slab rollback model [27,30,31,32,35].

The low-angle or flat subduction of oceanic plates squeeze out the mantle wedge above the subduction zone or leave only a small amount of mantle wedge material, which contradicts the widespread distribution of mantle-derived arc magmatic rocks from Xigaze to Linzhi [27,30,31]. Therefore, the low-angle or flat subduction model is not a reasonable explanation for the Late Cretaceous magmatic flare-up in the Gangdese belt. Considering that the Early Cretaceous magmatic rocks are mainly distributed in the northern and central Lhasa terranes, while the southern Lhasa terrane mainly develops Late Cretaceous magmatic rocks, Ma et al. [30,32] proposed that the subducted Neo-Tethys oceanic slab underwent slab rollback in the Cretaceous. However, based on the latest research on the spatiotemporal distribution, lithological assemblage, and geochemical characteristics of the Mesozoic magmatism in the Lhasa terrane, Wu et al. [73] found that there are two opposite subduction zones beneath the Lhasa terrane (i.e., the northward subduction of the Neo-Tethys Ocean and the southward subduction of the Bangong Lake–Nujiang Tethys Ocean), and the Early Cretaceous magmatism in the northern and central parts of the Lhasa terrane may be controlled by the slab rollback and breakoff processes of the southward subducting Bangong Lake–Nujiang Tethys oceanic plate [81]. Chen et al. [34], in their study of Late Cretaceous adakites in the northern Nuri area, attributed the Late Cretaceous adakites on the southern margin of the Gangdese belt to slab rollback. If the dynamic mechanism of slab rollback is valid, then prior to the Late Cretaceous, oceanic crust was subducted at low angles or normal angles, followed by the slab reversing its subduction direction. This process would initiate magmatism in the northern part of the Lhasa terrane, gradually progressing southward with rocks of increasingly younger ages. However, according to the formation ages and spatial locations of the Late Cretaceous magmatic rocks in the Gangdese metallogenic belt, spatially from north to south, the ages of the magmatic rocks of this period do not show a clear continuous change trend (Figure 12). Therefore, mid-ocean ridge subduction may be the most reasonable genetic model.

Under the context of oceanic ridge subduction, high heat flow through a slab window can lead to the partial melting of the oceanic crust (and a small amount of subducted sediment) at the margins of the slab window [28], forming adakitic magmas under amphibolite–eclogite facies conditions. Zhang et al. [28] proposed that between Milin and Lilong in the southern Gangdese belt, charnockites with adakitic characteristics, dated as 90–86 Ma, are products of the subduction of the Neo-Tethyan mid-ocean ridge. The formation age of the quartz diorites from the southern segment of the Nuri mining area obtained in this study is 93.6 ± 0.4 Ma, which is very close to the subduction time (105 ± 10 Ma) [25,28]. Additionally, numerous studies have confirmed large-scale magmatism in the Gangdese metallogenic belt during 109–76 Ma [6,28,31,34,39,41,58,61,73]. Therefore, this study suggests that Late Cretaceous adakites in the Gangdese belt are likely products of oceanic ridge subduction.

5.3. Metallogenesis

Pan et al. [54] proposed that the tectonic evolution of the Neo-Tethys in the Qinghai–Tibet Plateau is mainly manifested as the opening, closure, and subduction and consumption of the oceanic lithosphere of two Neo-Tethys oceans, as well as the Mesozoic intracontinental tectono-magmatic activities superimposed on the Paleo-Tethys tectonic belt, based on the research results of basic sciences such as geochronology, paleontology, geochemistry, and geophysics. The two Tethys oceans developed on the northern and southern margins of the Lhasa block, respectively, with the former represented by the Bangong Lake–Nujiang ocean basin, whose opening and extension led to the separation of the Lhasa block from the Qiangtang block by the sea, and the latter represented by the Yarlung Zangbo ocean basin, whose northward subduction and consumption formed the Gangdese volcanic magmatic arc similar to the Andean-type volcanic magmatic arc [47,48,51]. A large number of studies have confirmed that the Yarlung Zangbo ocean basin developed on the passive continental margin of the northern margin of the Tethys Himalayas and the Gondwana continent, roughly opened in the Jurassic, subducted and consumed in the Cretaceous, and roughly completed the closure process at the end of the Late Cretaceous [80]. During the demise of the Neo-Tethys, the partial melting of the oceanic crust and mantle wedge beneath the continental plate provided a geological environment for the formation of Cu-Au polymetallic deposits related to oceanic crust subduction and intermediate-acidic magmatic activities [15,16,57]. The Nuri ore district is located along the southern margin of the Gangdese belt and the Yarlung Zangbo suture zone. In the Late Cretaceous, the Bangong Lake–Nujiang ocean basin had already closed [18]. Therefore, both the tectonic location and magmatic activity timing of the Nuri ore district indicate a close relationship with the northward subduction of the Neo-Tethys oceanic crust. The northward subduction and consumption of the Yarlung Zangbo back-arc basin in the Cretaceous and the dynamic process of arc–arc collision had important influences and constraints on the tectono-magmatic-metallogenic activities of the Kelu–Chongmuda Cu-W-Mo metallogenic sub-belt on the southern margin of the Gangdese belt [15,16,34,61,73].

Numerous studies have shown that adakites are closely related to mineralization, especially adakites related to plate subduction [51,52,53]. In the Kelu–Chongmuda Cu-W-Mo metallogenic sub-belt on the southern margin of the Gangdese belt, Liang et al. [81] reported that the ore-forming porphyry of the Sangbujiala skarn Cu-Au deposit has adakitic geochemical characteristics and formed at 92.1 ± 0.6 Ma; Jiang et al. [81] studied the Kelu Cu-Au deposit and found that the 90 Ma adakitic intrusions are closely related to Cu-Au mineralization; and Zhang [60] revealed that the skarn Cu-Fe ore segment in the northern part of the Nuri ore district is closely related to the adakitic tonalite formed at 93.0 ± 1.4 Ma. In addition, many researchers have also reported that a large number of Late Cretaceous adakitic intrusions and volcanic rocks developed in the Nuri ore district [25,34,61,73]. It is worth noting that in some drill holes in the Nuri ore district, independent Cu orebodies in the Late Cretaceous quartz diorite can reach 10.69 m, with an average Cu grade of 0.37 wt.% (according to the internal company report). The above research results indicate that during the Late Cretaceous, the southern margin of the Gangdese belt likely developed a strong tectono-magmatic-metallogenic event related to the subduction of the Neo-Tethys Ocean.

Numerous low-temperature geochronological studies on the Gangdese batholith in the southern Lhasa terrane have found that the eastern Gangdese metallogenic belt experienced significant uplift-erosion during the Oligocene and Miocene (29 Ma to 9 Ma) or even earlier [82]. Currently, pre-collisional deposits are rarely found in the eastern Gangdese metallogenic belt, most likely because most of the deposits have been eroded away, rather than the absence of strong mineralization during this period. Therefore, geological prospectors should pay close attention to the mineralization during this period.

6. Conclusions

(1) The quartz diorite in the Nuri Cu-W-Mo deposit along the southern margin of the eastern Gangdese belt formed in the Late Cretaceous (93.6 ± 0.4 Ma) and exhibits typical adakitic geochemical characteristics.

(2) The Late Cretaceous adakites in the Nuri ore district were derived from the partial melting of the subducted Neo-Tethys oceanic slab, with subsequent interaction with the mantle wedge, as evidenced by their low K₂O, high Mg, Cr, and Ni contents, and positive zircon εHf(t) values (+6.58 to +14.52).

(3) The Late Cretaceous magmatic flare-up and associated adakitic magmatism in the Gangdese belt were likely triggered by the subduction of the Neo-Tethys mid-ocean ridge, which induced the partial melting of the subducted oceanic slab and overlying mantle wedge.

(4) The close temporal and spatial relationship between the Late Cretaceous adakitic intrusions and Cu mineralization in the Nuri ore district probably suggest a strong tectono-magmatic-metallogenic event related to the Neo-Tethys subduction during this period.