As already mentioned, the EXOB is located in a region with a complicated overprinting and interaction of the Palaeozoic Palaeo-Asian tectonic–metallogenic domain and the Mesozoic Western Pacific margin tectonic–metallogenic domain (Figure 1A). The tectonic evolution of the EXOB has been especially studied in many studies, which we summarize here. The EXOB has undergone two stages of tectonic evolution in different tectonic settings (Wang YB. et al., 2016). In the Paleozoic, tectonism and magmatism were controlled by the evolution of the Paleo-Asian Ocean between the Siberia and the North China cratons (Wang and Mo, 1995). The EXOB was developed in this stage, accompanied by the amalgamation of several microcontinental blocks in NE China (Zeng et al., 2012; Ouyang et al., 2013): the Erguna block in the northwest, the Xing’an and Songliao blocks in the centre, the Liaoyuan block in the southeast and the Jiamusi-Khanka block in the northeast. Final closure of the Paleo-Asian Ocean marked by suturing between the Songliao block and the Liaoyuan block is believed to have taken place along the Solonker-Xar Moron-Changchun suture zone at ca. 250 Ma (Xiao et al., 2003; Wu et al., 2011; Ouyang et al., 2013). However, other studies suggest that this event may have occurred at ca. 230 Ma (Zhou and Wilde, 2013). Since the Early Jurassic, the tectonic framework of northeastern China has been dominated by the subduction of the Paleo-Pacific oceanic plate in the east (Wu et al., 2011; Zeng et al., 2012; Xu et al., 2013a; Zeng et al., 2013), which induced extensive magmatic activities throughout the EXOB (Wu et al., 2011). The deposits in the study area mainly belong to the Yanshanian intermediate-acid magmatism Au, Ag, Cu, Pb, Zn (Mo) metallogenic series, which is controlled by structure, rock strain and subvolcanic porphyry. The complex geological evolution has resulted in intense magmatism as well as extensive precious and non-ferrous metal mineralization in this area, especially epithermal gold mineralization (Ge et al., 2007; Zeng et al., 2011; Zeng et al., 2012; Ouyang et al., 2013).

A lot of gold deposits (Figure 1B), including epithermal, metamorphic hydrothermal, magmatic hydrothermal, orogenic, porphyry and skarn gold deposits, have been discovered in recent years in Heilongjiang province. In the porphyry-epithermal metallogenic system, porphyry deposits are enriched in intrusive rocks, and epithermal deposits cover the top of intrusions, and the metallogenic age of the latter is generally 10∼20 Ma later than the former (Wang et al., 2012). The low-density gas fluid in the porphyry metallogenic domain can carry ore-forming metals to migrate to the epithermal metallogenic domain, and gradually transform into low-temperature and low-salinity liquid fluid through gas phase contraction (Ni et al., 2020). The porphyry - epithermal deposit in Xinchenggou area is mainly Jinchang copper - gold deposit. In Jinchang deposit, there is a transformation from veinlet-impregnated ore body to cryptoexplosive breccia ore body, that is the transformation from porphyry gold-copper deposit (No.18 copper-gold deposit) to epithermal high-sulfide deposit (No.1 breccia gold deposit) (Wu, 2018). Among these gold deposits, epithermal gold deposits (including Wulaga, Dong’an, Sandaowanzi and Sipingshan epithermal gold deposits; Wang YB. et al., 2016) account for more than 60% of present estimates of the gold resource in Heilongjiang province and these deposits are believed to be associated with Early Cretaceous volcanic and subvolcanic intrusions (Wang et al., 2016a; and references therein). In the past, there have been many studies on single deposit, but few studies on the distribution characteristics and metallogenic rules of the whole metallogenic zone.

The Xingchenggou area tectonically belongs to the Khanka block in the EXOB (Figure 1; Figure 2). It is adjacent to the Songnen-Zhangguangcai Range along the Jiayin-Mudanjiang Fault to the west, the Sikhote-Alin Orogenic Belt along the Central Sikhote-Alin Fault to the east, and the Jiamusi Terrain along the Dunhua-Mishan Fault to the north. The Khanka Terrain mainly comprises the Precambrian metamorphic complex, early Paleozoic -Cenozoic sedimentary cover, Mesozoic granite and Late-Middle Permian-Late Triassic igneous rock (Wang F. et al., 2016). The Mesozoic and Cenozoic strata in the study area belong to marginal-Pacific strata, including the Late Triassic Luoquanzhan Formation (T₃ l), Early Cretaceous Muling Formation (K₁ m), and Miocene-Pliocene Chuandishan Formation (Nc) (Figure 3). In terms of the rock assemblages, most of them are tuff, sandstone and basalt. Magmatic activities in the area mainly occurred during the Late Triassic-Early Jurassic. Petrographically, most of them are medium-coarse-grained syenogranite, and to a lesser extent diorite porphyry dykes. The major deformation structure are faults and there are no noticeable folds. The faults are dominated by the NE-extending, high-angle thrust fault structure which was formed before the Late Triassic. The emplacement of the Mesozoic magmatic rocks is controlled mainly by these faults. It also can be observed that the early Paleozoic strata and granitic plutons are locally cut by faults.

The relationship between the Khanka Terrain and the Jiamusi Terrain is still debated. Based on zircon dating and whole-rock geochemical results for Permian volcanic rocks at the eastern and southeast Jiamusi Terrain, Meng et al. (2008)proposed that the Paleo-Asian Ocean between Jiamusi Terrain and Khanka Terrain disappeared in 268 Ma, followed by collision between the two terrains. Zhou et al. (2012) studied the geochronology of the Hutou complex rocks and discovered that the early Paleozoic magma and metamorphic event recorded by Hutou complex rocks in Khanka Terrain were consistent with the age of the Mashan Group in the western Jiamusi Terrain. They thus argued that the Khanka Terrain and the adjacent Jiamusi Terrain had an obvious structural affinity and had the same geological evolution history (Yang et al., 2012). Many other studies have compared the geographic features of the Carboniferous, Permian and Triassic strata and fossils distributed in the Khanka Terrain and South China Plate (Kobayashi, 1997; Kobayashi 2003). It is suggested that the Khanka Terrain was an exotic terrane with a South China Plate affinity (Zhang, 1997; Zhang, 2004), and might have belonged to the northeast part of the South China Plate (Li et al., 2017a; Guo et al., 2017).

Many epithermal deposits (e.g., Jinchang, Naozhi and Ciweigou gold deposits) have been discovered in the region adjacent to the Xinchenggou area (Qi et al., 2005), which show similar geological features. For example, in the Jinchang gold deposit and Naozhi copper-gold deposit granitoids mainly consists of granite, granite porphyry, syenogranite and monzogranite, and the main altered minerals are pyrite, sericite, feldspar, chlorite, epidote and so on in these two deposits (Zhu et al., 2003; Li, 2021). Beresitization (pyrite - sericite - silicification) is closely related to epithermal hydrothermal mineralization, which belongs to direct prospecting indicator. Meanwhile, there are also good geophysical and geochemical anomaly characteristics and mineralogical alteration characteristics in the study area. Sericite, adularia, argillite, chlorite, pyrite and other alteration minerals are found in the fractures of the granite body in the study area (Figure 4). It shows that the study area has the potential to search for epithermal hydrothermal deposits.

Meng (2012) determined that the zircons U-Pb age of granodiorite-porphyry directly related to mineralization in Yanghuidongzi copper deposit in the northern part of the study area is 194.8 ± 1.9 Ma, which is close to the regional magmatic rock age and roughly indicates the metallogenic age of Yanghuidongzi copper deposit.

The main strata in the study area are the volcanic rocks of the Late Triassic Luoquanzhan Formation (T₃ l), whose formation ages are 206–213 Ma (Zhao et al., 2013; Wei, 2021). The petrogeochemical characteristics indicate that the volcanic rocks in the study area are partial melting products of crust. The volcanic rocks of the Luoquanzhan Formation mark the regional tectonic transformation from the end of Xing`anling-Mongolian orogeny to the initial of Pacific plate subduction to the Eurasia plate (Zhao et al., 2013). The preconcentration of gold element occurs in primary magma which formed Luoquanzhan Formation and provided part of ore material to Jinchang Cu-Au deposit (Zhao et al., 2013).

The main magma emplacement ages of granodiorite and monzonite, which are closely related to mineralization of Jinchang Cu-Au deposit, are 213 ± 1 Ma and 204. 8 ± 1.1 Ma, respectively. The magma evolution spans the Late Triassic and Early Jurassic (223–200 Ma), and has experienced multiple stages of rapid decompression and cooling, constant temperature and constant pressure crystallization (Han et al., 2015). The age of intrusive rocks in the Jinchang deposit is 163–203 Ma (Mu et al., 2000; Men, 2008), while the metallogenic age is 190–210 Ma (Men, 2008). Similar to these deposits, the granitoids in the Xinchenggou area are petrographically mainly composed of Mesozoic syenogranite and monzogranite. The formation ages of syenogranite and monzogranite are 222–190 Ma (the peak value is about 201 Ma) and 234–191 Ma (the peak value is about 205 Ma) (Han et al., 2010), indicating that syenogranite was emplaced slightly later than monzogranite (Zhang, 2016).

The ore-forming potential of the Mesozoic syenogranite and monzogranite in the Xinchenggou area is demonstrated by soil geochemical data (Hao and Wang 2017). The results of soil data in the Xinchenggou area show noticeable anomalies of Au, Ag, Cu, Mo, Bi, Sb and Sn, which are assemblages of medium-low temperature ore-forming elements common in epithermal deposits (Figure 5). The metal anomaly is mainly located in the contact zone of these Mesozoic syenogranite and the Late Triassic Luoquanzhan Formation tuff. The good ore-forming potential in the Xinchenggou area is further verified by the geophysical surveys (Figure 6), which have shown that there are good induced polarization (IP) anomalies and high magnetic anomalies around these plutons (Wang, 2018).

According to the major element analytical result, the SiO₂ content of granite in the study area is between 71.23% and 76.78%. Their aluminum is weakly supersaturated (Al₂O₃ = 11.98%–13.52%) MgO is 0.10%–0.90%, whereas the contents of TiO₂, Fe₂O₃ and P₂O₅ are low. In QAP diagram (Figure 8), the studied samples all plot into the syenogranite and monzogranite regions, which is consistent with the results based on petrography.

In the SiO₂-K₂O diagram (Figure 9A), all the samples are mainly located in the high-K calc-alkaline series region. K₂O+Na₂O ranges from 6.30% to 8.39% and the K₂O/Na₂O ratio is greater than 1.0. The aluminum saturation index (ACNK) of granites is 1.02∼1.20, with six of seven samples characterized by ACNK<1.1. In A/CNK-A/NK diagram, the point of monzogranite is located in the weak peraluminous area, and the syenogranite is mainly located in the peraluminous area (Figure 9B). The differentiation index of granites (DI) is 79.67–94.15, indicating that the fractional crystallization effect of granite magma is relatively strong in this area. The solidification index of granites (SI) is 0.98–8.17, reflecting that the magmatic differentiation degree is high.

Rare earth elements results (Figure 10A) show that the total REE contents are relatively low (∑REE = 72.35 × 10⁻⁶∼217.64 × 10⁻⁶ and ∑LREE/∑HREE is 3.35–9.71). LREE is comparatively enriched, whereas HREE is relatively depleted. La_N/Yb_N is 2.74–11.37 (the average value:6.27), indicating that light and heavy rare earth differentiation is strong. All intrusions are characterized by light rare earth element (LREE)-enriched patterns, as manifested in the chondrite-normalized rare earth element (REE) diagram. The Ce anomaly is not obvious, with δCe being 0.96–1.11. The Eu anomaly is noticeable with δEu characterized by 0.14–0.83.

In the primitive mantle-normalized trace element diagram (Figure 10B), high-field-strength elements like Nb, Ta, P and Ti are comparatively depleted, and the depletion of P and Ti indicates that apatite and ilmenite present obvious fractional crystallization in the magmatic evolution process.

The negative anomalies of Nb, Ta, P and Ti, and the negative anomalies of large-ion lithophile element Sr, indicate the attribute of island arc magma (Li et al., 2019; Zhong et al., 2021a; Zhong et al., 2021b). It is suggested that the magma source area was contaminated and metasomatized by crustal materials or subduction residual oceanic crust fluids (Fitton et al., 1991).

The depletion of Nb and Ta reflects that the magma originates from crust or suffers from strong contamination of crust materials (Zhu et al., 2022). Large-ion lithophile elements like Rb, Ba, K and Hf are relatively enriched. The comparative enrichment of the strongly incompatible element Rb indicates that strong differentiation might happen during ascending of magmas (He et al., 2014).

Nb/Ta ratios range from 6.18 to 26.33, with an average value of 15.18, which are lower than the average value of 16.2 in the upper crust of eastern China and 18 in the primitive mantle (Sun and McDonough, 1989). These data indicate that the granitic magma source in the study area may derive from the upper mantle or lower crust, and a small amount of crustal material was involved during the formation process (Zhao et al., 1997; Hao and Wang 2017).

During magmatic evolution, although both the abundance of Nb and Ta increased, Ta increased more quickly than Nb. Therefore, the Nb/Ta ratio gradually decreased from early to late magmatic evolution (Chen et al., 2021). The Nb/Ta value of syenogranite in the study area is from 6.18 to 8.25, while the Nb/Ta value of monzogranite is from 24.5 to 26.33 (Jing et al., 2015), which also shows that the evolution stage of syenogranite in this area is later than that of monzogranite.

Granitic rocks are commonly divided into I-, S- and A-types. A-type granites typically contain high-temperature anhydrous phases such as pyroxene and fayalite (e.g., King et al., 1997). The syenogranite and monozogranite that we studied contain minor biotite but lack pyroxene or fayalite. Furthermore, the syenogranite and monozogranite all belong to the high-K series and most have low Zr, Nb, and 10,000 × Ga/Al ratios, distinguishing them from A-type granites (Figure 11A; Whalen et al., 1987). Moreover, most of the samples we studied fall within the field of I- and S-type granites (Figure 11B). Chappell and White (1974) proposed that the boundary between S-type and I-type granites can be drawn at an A/CNK ratio of 1.1. Most of samples in this study are characterized by A/CNK ratio < 1.1, thus not consistent with that the studied samples are S-type rocks. Besides, S-type granites generally contain primary Al-rich minerals such as cordierite or muscovite, which were not observed from granites in the Xinchenggou area. Geochemically, S-type and I-type granites can also be distinguished by the relationship between SiO₂ and P₂O₅: P₂O₅ increases with SiO₂ for S-type granites whereas decreases for I-type granites. The observed negative relationship between SiO₂ and P₂O₅ thus indicates a petrogenesis of I-type granites (Figure 11C; Chappell and White, 1992; Chappell, 1999). Besides, syenogranite and monzogranite in the Xinchenggou area are characterized by relative enrichments of LREEs and depletion of HFSEs (e.g., Nb, Ta, and Ti), relatively flat HREE patterns, and negative Eu anomalies, consistent with the geochemical characteristics of subduction-related I-type magmas. In summary, these several lines of evidence suggest that the syenogranite and monzogranite in the Xinchenggou area are (slightly fractionated) I-type granites rather than A-type or S-type granites.

According to previous studies (Wang, 2018), the monzogranite in the epithermal Jinchang gold deposit near the study area has the geochemical characteristics of an active continental margin. In the tectonic discriminant diagram, these samples plot into volcanic arc granite region, which has the tectonic background related to ocean subduction and the characteristics of I-type granite. This is confirmed by the Nb-Y and Rb-(Yb + Ta) diagrams of Pearce et al. (1984) (Figure 12). In Figure 12, almost all syenogranite and monzogranite samples plot within volcanic arc granite region. In the Sr/Y-Y tectonic environment discrimination diagram, the intrusions in the study area are mainly located in the classic island arc granite region (Figure 13A). In the Na₂O/K₂O-SiO₂ diagram, two intrusions in the study area are mainly located in the active continental margin area (Figure 13B), indicating that two intrusions were formed in the island arc environment of the active continental margin.

Combined with previously published studies in the EXOB, we propose the following geodynamic models for the Xinchenggou granites and the related epithermal mineralization.

Many gold and copper (gold) deposits are discovered near the study area (Table 3; Figure 1), and the major deposit type is epithermal -porphyry type; the mineralogenetic epoch mainly centers on 120–210 Ma, mainly mid-late Yanshanian period. This period is the important Cu metallogenic epoch (110–200 Ma) in Northeast China region, including one of the four age intervals of Mesozoic epithermal Au mineralization in Eastern China (144–135 Ma) (Lu et al., 2016). The weighted average age of zircon in dioritic porphyry and granite porphyry closely related to mineralization in the Jinchang gold deposit near the study area is 203 Ma, and the metallogenic time is speculated to 190–210 Ma (Men, 2008), it is Late Triassic-Early Jurassic.

During the Mesozoic, the Pacific Plate subducted beneath the Eurasian continental plate, forming a large-scale continental margin plate tectonic-magmatic metallogenic belt. Mesozoic epithermal gold deposits near the study area, such as Tuanjiegou gold deposit, Jinchang copper and gold deposit, formed in this tectonic-magmatic metallogenic belt of this tectonic background (Wan, 2013; Wu, 2018).

Located in Jilin-Heilongjiang provinces metallogenic belt in the east of Central Asian Orogenic Belt, the study area is in the superimposition and transition region of the E-W-trending Paleozoic Paleo-Asian Ocean Metallogenic Domain and the Mesozoic-Cenozoic Coastal Pacific Metallogenic Domain of NNE-orientated structure (Liu et al., 2004). The geological evolution of this area is complex and the minerogenetic condition is superior. Magmatic activities in the area are frequent, mainly dominated by the magmatism of the Mesozoic. A relatively favorable geological environment is provided for the mineralization of endogenous metal. Volcanic edifice and fault structure are developed, and channels and space are provided for ore-forming fluid migration and ore storage.

Generally speaking, the granites in study area have the transitional characteristics of I- and S-type granites, and the formation might be related to the tectonic environment of volcanic arc and tectonic event of syn-collision (Zhang, 2012). The magma generation type is transitional crustal syntectic type, and the syntectic type magma often appears in the active continental margin zone (Xu et al., 1983). The high Sr/Y granitoids of 251–245 Ma were found in the Solonker-Xar Moron River area on the southern margin of the Central Asian Orogenic Belt (Li et al., 2017b), and the adakitic granites were found in the central part of Jilin Province (Wang et al., 2015b). The geochronology, petrology and geochemistry of granitic rocks in the eastern segment of the Central Asian Orogenic Belt indicate that the collision between the North China Craton and the Khanka-Jiamusi Massif occurred during the Late-Permian to Early-Triassic along with the closure of the Paleo-Asian Ocean (Yang et al., 2018). A scissor-like closure of the Paleo-Asian Ocean from west to east between the Songnen-Zhangguangcai Range block and North China Plate took place during the Middle-Permian to Middle-Triassic (Figures 14A, B) (Wang et al., 2015a; b). Continental collision happened between the Songnen-Zhangguangcai Range block and North China Craton (Qi et al., 2005), forming the syn-collisional S granite. The collision and extrusion led to strong stacking and thickening of continental crust and lithosphere in the northeast. After extrusion and proliferation, extension-thinning was started.

At present, the timing of the subduction of the Pacific plate beneath the Eurasian plate is widely debated. Most researchers believe that the subduction of the Pacific plate beneath the Eurasian plate began during or before the Late Triassic (Isozaki et al., 2010; Wu et al., 2011; Li et al., 2012; Wilde Simon, 2015; Han et al., 2020; Liang et al., 2021). However, a few scholars believe that the subduction began in the Early Jurassic (Zhou et al., 2006; Xu et al., 2013b; Wang et al., 2015). Through deep analysis, this paper suggests that the subduction time is Late Triassic (Figures 14C, D). Due to this subduction, the Mudanjiang Ocean between the Songnen-Zhangguangcai Range blocks and the Khanka-Jiamusi block split anew in the Middle-Late Triassic (Xu et al., 2012). The 217–201 Ma of A-type volcanic rocks are developed in Yanbian-Dongning area (Wang et al., 2015a), and bimodal association of volcanic rocks are present in the region. The opening of NEE-NE trending Mesozoic-Cenozoic basin group (Yang et al., 2015; Yang et al., 2018), and the emplacement time of the Changchun-Yanji suture zone is 240–220 Ma (Cao et al., 2020), indicating that the Yanbian-Dongning area in the Late-Triassic was a back-arc extensional tectonic environment after the closure of the Paleo-Asian Ocean (Wang et al., 2015a). Meanwhile, as affected by post-orogenic extension, a large number of magmatic-hydrothermal metallogenic systems were developed, causing the formation of epithermal deposits (Qi et al., 2005; Zhang et al., 2008), such as the Jinchang gold deposit (Figures 14C, D).

In the Early Jurassic, the onset of subduction of the Paleo-Pacific Plate beneath the Eurasia Plate (Jing et al., 2015) led to the closure of the Mudanjiang Ocean between the Songnen-Zhangguangcai Range block and the Khanka-Jiamusi block. Early Jurassic calc-alkaline volcanic rocks in the eastern margin of Jilin-Heilongjiang provinces (Xu et al., 2013a; Wang Z. H. et al., 2017), the 205–160 Ma of calc-alkaline intrusive rocks (Yang et al., 2018), the low temperature dynamo metamorphism (193 Ma of biotite Ar-Ar age) of Huangsong Group in the eastern margin of Songnen-Zhangguangcai Range (Yu et al., 2015), and the Heilongjiang Complex with a metamorphic age of 175–186 Ma were all the products of the subduction of the Paleo-Pacific plate under Eurasia (Zhou et al., 2009; Wang Z. W. et al., 2017; Yang et al., 2018). Due to multiple tectonic compression, syn-collision S-type granites and calc-alkaline I-type granites were formed (Figures 14E, F) (Yang et al., 2017; Yang et al., 2018).

The Rb-Sr age of biotite in quartz diorite of Taiping Mountain near the study area is 250 Ma (HBGRM, 1979) the K-Ar age of biotite in granitic pluton of Taiping Mountain is 221.5 Ma (HBGRM, 1986); the weighted average age of zircon in granite porphyry of Jinchang gold deposit is 203 Ma (Men, 2008). Moreover, the intrusive rock near the study area can be divided into three invasion stages including Late Variscan, Early Indosinian and Late Indosinian. Some rocks also have the characteristics of S-type granite (HBGRM, 1986), showing that the granite in this area is characterized by multiple phases, multiple stages and compound origin.