The Chuhuertu granite is generally grayish white to pink in color, with medium-fine-grained hypidiomorphic granular structure and stripe structure, and massive texture (Figures 3A, B). This rock contains mainly alkali feldspar (55–60 vol.%), quartz (−20 vol.%), plagioclase (10–15 vol.%), biotite (−5 vol.%), and accessory magnetite. Mineral particle size is generally 0.2–5 mm. The alkaline feldspar is in the form of hypidiomorphic–xenomorphic plate-type, mainly composed of orthoclase, perthite, and microcline feldspar, with a particle size of 0.2–5 mm. Among them, perthite is primarily in the form of fine veins and branches arranged in the same direction (Figure 3D). The plagioclase is mainly andesine and oligoclase, in the form of euhedral-hypidiomorphic plate-type, with polysynthetic twin, occasionally with ring structure, and a particle size of 0.2–4 mm. Quartz is found as fill between plagioclase and alkali feldspar, with xenomorphic granular texture, a particle size of 0.2–5 mm, cracks, and wavy extinction. Biotite is distributed in a scale-blade shape with a diameter of 0.2–2 mm (Figures 3C, D).

In order to study the geochemical and chronological characteristics of the Chuhuertu granite, five geochemical samples were collected (Table 1), along with one (Table 2) zircon U-Pb sample, from different parts of the pluton on the basis of detailed field geological observations. The weathering degree of the samples was low, and free from oxidation, pollution, and obvious alteration. See Figure 2 for the sampling locations.

The whole-rock geochemical analyses were undertaken at the Central Laboratory of China Chemical Geology and Mine Bureau, Beijing, China. The zircon selection and target making were completed by the Beijing Kehui Testing Technology Co., Ltd., Beijing, China, and zircon U-Pb isotopic analyses were undertaken at the Beijing SHRIMP Center.

A Zsx Primus II wavelength dispersive X-ray fluorescence spectrometer (XRF) produced by RIGAKU, Japan, was used for the analysis of major elements in the whole rock; the X-ray tube was a 4.0 Kw end window Rh target. All major element analysis lines were kα, and the standard curve used the national standard material: the rock standard sample GBW07101-14. The relative standard deviation (RSD) was less than 2%.

Trace-element analyses of whole-rock were conducted on an Agilent 7700e ICP-MS. The detailed sample-digesting procedure was as follows: 1) Sample powder (200 mesh) was placed in an oven at 105°C for drying for 12 h; 2) an amount of 50 mg of sample powder was accurately weighed and placed in a Teflon bomb; 3) a measure of 1 mL of HNO₃ and 1 mL of HF were slowly added into the Teflon bomb; 4) the Teflon bomb was enclosed in a stainless steel pressure jacket and heated to 190°C in an oven for >24 h; 5) after cooling, the Teflon bomb was opened and placed on a hotplate at 140°C and evaporated to incipient dryness, and then, 1 mL HNO₃ was added and evaporated to dryness again; 6) a measure of 1 mL of HNO₃, 1 mL of MQ water, and 1 mL of internal standard solution at 1 ppm were added, and the Teflon bomb was resealed and placed in the oven at 190°C for >12 h; 7) the final solution was transferred to a polyethylene bottle and diluted to 100 g via the addition of 2% HNO₃.

The zircon U-Pb dating method used was LA-ICP-MS in-situ U-Pb isotope dating, and the test instrumentation included a NWR193 laser ablation system and Jena PQMS inductively coupled plasma mass spectrometry. The depth of laser denuded zircons was 20–40 μm. The diameter of the denudation beam spot was 35 μm. The international standard zircon 91,500 was used as the external zircon age standard for isotopic fractionation correction, and the element content was corrected with the standard sample NIST SRM610 as the external standard and ²⁹Si as the internal standard. The test results were processed and mapped using ICPMSDataCal software (Liu et al., 2008), and the ²⁰⁸Pb correction method was used to correct the common lead (Anderson, 2002).

The Chuhuertu granite has LOI content ranging from 0.55 to 0.87 wt.% (Table 1), consistent with petrographic evidence for partial alteration. In addition, the elements Li, Rb, Cs, K, U, and P may be readily mobilized in rocks during hydrothermal alteration (Hart et al., 1999; Bach et al., 2001). Consequently, in the following descriptions and discussion, the major element contents were normalized to 100 wt.% on an anhydrous basis, considering all Fe as FeO; we preferentially used elements that are relatively insensitive to alteration, such as SiO₂, Al₂O₃, TiO₂, MgO, Fe₂O₃, high-field-strength elements (HFSE), rare-earth elements (REE), Th, and Ba.

Chemical data for five whole-rock samples of the Chuhuertu granite are presented in Table 1. The samples showed high SiO₂ (73.04–76.01 wt.%) and K₂O (4.25–5.44 wt.%) content. At the same time, the samples had high Na₂O (3.42–4.69 wt.%) content, and the high total K₂O + Na₂O content (7.81–9.50 wt.%) suggested material rich in alkali. The SiO₂ versus Na₂O + K₂O – CaO diagram (Figure 4A) indicates samples with calcic-alkali to alkali features. The samples had low CaO (0.20–0.98 wt.%), MgO (0.10–0.22 wt.%), P₂O₅ (0.03–0.06 wt.%), and TiO₂ (0.11–0.19 wt.%) content. The Al₂O₃ content of samples was high, at 12.86–13.45 wt.%. In the A/CNK versus A/NK diagram (Figure 4B), samples fall into quasi-aluminous and peraluminous areas, with A/CNK values (molar Al₂O₃/(CaO + Na₂O + K₂O) = 0.96–1.11) and A/NK values (molar Al₂O₃/(Na₂O + K₂O) = 1.00–1.24) showing weakly peraluminous characteristics.

The total amount of rare earth elements (ΣREE) found in the samples was in the range of 109.35–164.73 ppm, with an average of 136.75 ppm (Table 1). The normalized REE distribution diagram of chondrites indicates “Seagull type” (Figure 5A), with obvious negative Eu anomaly, and δEu of 0.30–0.75, with an average value of 0.51. The REE curve for the samples has a generally gentle right dip, the light REE (LREE) appears slightly enriched, and the heavy REE (HREE) curve is relatively flat. The negative Eu of the samples is abnormally significant, suggesting that the pluton may be affected by the fractional crystallization of plagioclase or related to residual plagioclase in the source area (Eby, 1992). In terms of trace elements, the samples were enriched in Rb, Th, U, K, and Ga, and depleted in Ba, Sr, P, and Ti (Figure 5B), showing the unique component characteristics of A-type granite (Eby, 1992).

The representative zircon U-Pb data are listed in Table 2. Most zircon grains separated from the granite sample TW04 showed similar crystal shapes and subhedral to euhedral morphologies, with no evidence of resorption or inherited cores. The zircon grains were transparent, colorless, and showed oscillatory zoning, consistent with a magmatic origin (Figure 6A). All zircon grains had high Th/U ratios (0.41–0.60); Table 2), supporting a magmatic origin (Hoskin and Schaltegger, 2003). Ten analyses of zircons from sample TW04 yielded concordant zircon ²⁰⁶Pb/²³⁸U ages, ranging from 152 ± 2 to 158 ± 2 Ma. These data yielded a weighted-mean age of 155.6 ± 1.6 Ma (MSWD = 1.4) (Figures 6B, C), interpreted as the crystallization age of the Chuhuertu granite.

The Chuhuertu granite was characterized by high concentrations of SiO₂ and K₂O + Na₂O, but low CaO, MgO, P₂O₅, and TiO₂, with obvious negative Eu anomaly. The trace elements Rb, Th, K, Ta, and Hf were relatively enriched, while the elements Ba, Sr, P, and Ti were relatively depleted, and all had high (Na₂O + K₂O)/CaO (15.08–21.39) and TFeO/MgO (29.16–57.82) values. At the same time, the Chuhuertu granite had high Ga (16.70–19.86 ppm) values, and its 10000Ga/Al value was 2.52–2.90. Except for sample XT02-1, the samples had greater than the lower limit of 2.6 for A-type granite. In the four widely used 10000Ga/Al versus K₂O + Na₂O, (N₂O + ka₂O)/CaO, K₂O/MgO, and Zr granite discrimination diagrams (Figure 7), except for sample XT02-1 in Figure 7D, the samples fall into the A-type granite area. The aforementioned geochemical characteristics show that the Chuhuertu granite has the characteristics of A-type granite. However, the DI of the samples was found to be 92.05–96.36, and some samples plot in the fractional granite field (Figure 7D), which may be a result of crystallization fractionation. Crystallization fractionation could easily modify the geochemical composition of granite, and most of the highly fractionated granites have pseudo A-type features. Therefore, it remains unclear whether crystallization fractionation has taken place in Chuhuertu granite.

It is not a straightforward task to distinguish the highly fractionated from A-type granites, especially aluminous A-type granites. Whalen et al. (1987) found that A-type granites are low in Al but high in Ga and Zr concentrations, and then proposed that 10000Ga/Al = 2.6 and Zr = 250 ppm might indicate boundaries between A-type and other types of granites (A-type granite has 10000Ga/Al ratio > 2.6 and Zr > 250 ppm). However, quite a few highly fractionated granites also have a high 10000Ga/Al ratio similar to A-type granites (Perez-Soba and Villaseca, 2010; Breiter et al., 2013). Likewise, A-type granite, if intensively fractionated, is geochemically overlapping with highly fractionated granite (King et al., 2001). To address this problem, we adopted the classification method proposed by Wu et al. (2017). First, A-type granite is characterized by its high temperature of magma. However, the crystallization temperature of magma tends to decrease in the process of magmatic differentiation. Second, fractional crystallization will shift rock type from A-type to highly fractionated granite, as shown in Figure 7D. On the contrary, the fractionation of I- or S-type granites by crystal differentiation would increase their 10000Ga/Al ratios, and thus, another kind of evolutionary trend is observed (Figure 7D). Therefore, A-type granite can be effectively distinguished from highly fractionated I- or S-type granites, although the former occasionally shows an intensive magmatic differentiation. The data from samples from the Chuhuertu granite clearly showed an A-type granite differentiation trend (Figure 7D), indicating its A-type affinity. In addition, highly fractionated granitic magma is easily contaminated by country rocks due to its relatively prolonged crystallization time (Wu et al., 2017). During the early stage, the surrounding rocks are continuously assimilated by the magma, but the fractional crystallization leads to the surrounding rock being rarely observed in this granite. However, as the temperature of the magma gradually decreases with differentiation, the residual melt is not able to digest the surrounding rocks as assimilation proceeds. At the same time, the residual melt is low in capacity for movement, resulting in abundant xenocrysts of the surrounding rocks in the final solidified granite. Although the Chuhuertu granite has undergone crystallization differentiation, there are few xenocrysts in the pluton. Therefore, the Chuhuertu granite may have been in the early stage of contamination when it crystallized. At the same time, in the A/CNK versus A/NK diagram (Figure 4B), the samples fall into quasi-aluminous and peraluminous areas, with A/CNK values (0.96–1.11) and A/NK values (1.00–1.24), indicating that the A-type granite should be weakly peraluminous, which is consistent with the geochemical characteristics of A-type granite in the eastern segment of CAOB (Liu et al., 2005; Zhou et al., 2010; Zhang et al., 2012; Shi et al., 2014; Xue et al., 2015).

A-type granites can be formed in a variety of petrogenetic processes, which can be summarized into the following three types: mixing of crust-derived acidic magma and mantle-derived basic magma (Griffin et al., 2002; Barbarin, 2005); low degree partial melting or crystallization differentiation of mantle-derived basaltic magma (Turner et al., 1992; Lee and Bachmann, 2014); and partial melting of crust-derived materials (Chappell and White, 2001; Sisson et al., 2005). The Chuhuertu granite has high SiO₂ content and narrow range. There are no contemporaneous mafic rocks or mafic enclaves, and Chuhuertu granite does not have the characteristics of crystallization fractionation of mantle-derived alkaline basaltic magma and mixing of mantle-derived magma and crust-derived magma formed by deep melting. The lower REE content and right-dipping REE diagram (Figure 5A) of the Chuhuertu granite indicate that there is no residual garnet in the source area. The strong depletion of Sr and Ba indicates that the residual phase in the source area contains feldspar, and the strong depletion of Sr and Eu (Figure 5B) reveals that the plagioclase in the source area is the main residual phase, reflecting that the source area is a shallow low pressure area (<10 kbar) without garnet residue and enriched plagioclase (Rapp and Watson, 1995), which should be the petrogenesis of partial high-temperature melting of intermediate basic crust in neoaccretion under low pressure and subsequent crystallization differentiation. These characteristics of Chuhuertu granite reveal that the area is in an extensional environment, similar to the petrogenesis of the A-type granite in the eastern segment of CAOB in mid-eastern Inner Mongolia (Zhang et al., 2012).

As for the high temperature anomaly required by the low pressure and high temperature melting of the shallow intermediate basic crust in neoaccretion, more and more scholars believe that it is related to the post orogenic mantle-derived basaltic magma underplating caused by the upwelling of mantle materials in the asthenosphere induced by the previous subducted slab breaking-off and gravity collapse (Jahn et al., 2000; Bonin, 2007; Wu et al., 2007; Zhang et al., 2012; Xue et al., 2015). In mid-eastern Inner Mongolia, Mesozoic aluminous A-type granites generally have a positive ε_Nd(t) value, which suggests that the underplating of mantle-derived magma related to slab break-off in the post-orogenic stage plays an important role in the formation of Mesozoic aluminum A-type granitic magma in mid-eastern Inner Mongolia (Jahn et al., 2000; Chen et al., 2008; Zhou et al., 2010; Shi et al., 2014; Xue et al., 2015). In terms of petrography, perthite (Figures 3C, D) is found locally in the samples and inherited zircon is not found in S-type granite. This reveals that the melt was high-temperature and water poor, and that the early magmatic crystallization occurred under high-temperature conditions. These characteristics also reflect the petrogenesis of partial high-temperature melting of the intermediate basic crust in neoaccretion under low pressure and its subsequent differentiation.

It is generally believed that A-type granite is formed under high temperature and low pressure, which generally occurs in the shallow middle and upper crust (Clemens et al., 1986). Therefore, the calculation of temperature and pressure conditions for the formation of Chuhuertu pluton indicates A-type granite from the side. At the same time, the temperature and pressure conditions of magma formation can be used to speculate on the depth of its magmatic source, thus providing constraints for the origin and evolution mechanism of magma.

For a long time, the driving mechanism of the evolution of the Northeast Asian continent has been generally manifested in the subduction of the Paleo-Pacific plate and the closure of MOO (Meng et al., 2011; Xu et al., 2013; Wang et al., 2015; Wu et al., 2017).

Recently, Ji et al. (2019) reported a Late Jurassic oceanic crust-derived low-K adakitic lava in the Hailar Basin and linked it to the flat-slab subduction of the Paleo-Pacific plate. The low-K adakitic lava was considered to represent the final pulse of slab-melt-related volcanism, which also suggests that the Hailar Basin might be the farthest position ever reached by the subducting Paleo-Pacific plate (Ji et al., 2019). When the flat-slab of the Paleo-Pacific plate subducted under the Hailar–Tamtsag Basin and the Greater Xing’an Range region with thick lithosphere, the increased subduction depth of the slab led to the eclogitization of the originally light oceanic crust, which increased the density of the subducted slab and returned it to negative buoyancy (Antonijevic et al., 2015). It was suggested that the flat-slab of the subducted slab was also unstable and decoupled from the overlying mantle wedge, which led to the rollback accompanied by the gravitational subsidence of the subducted slab (Antonijevic et al., 2015). The appearance of Late Jurassic A-type granites in the Hailar–Tamtsag Basin and the Greater Xing’an Range (Yang et al., 2014; Jiang et al., 2021) also suggests that the upwelling of asthenosphere mantle material caused by the slab rollback of the Paleo-Pacific plate in the Late Jurassic-Early Cretaceous provided some high-temperature conditions required for the formation of A-type granite magmas in the region (Ji et al., 2019).

In addition, many scholars believe that the destruction of NCC is related to the subduction of the Paleo-Pacific plate (Wu et al., 2005; Zhang et al., 2014; Zhu and Xu, 2019; Gao et al., 2021), and its remote effects caused large-scale Early Cretaceous massive magmatism and craton destruction of NCC, resulting in large-scale extension of the middle and lower crust (Zhang et al., 2009b; Zhang et al., 2014; Zhu and Xu, 2019; Gao et al., 2021). The lithospheric reconstruction of NCC by the Middle-Late Jurassic Paleo-Pacific plate was mainly limited to the eastern part of North China and the eastern part of Western Liaoning. A study of the lithofacies paleogeography showed that the Jurassic strata in the eastern part of NCC are obviously lost, which indicates that the area was in a tectonic setting of uplift in the Jurassic (Zhu and Xu, 2019), which was manifested by crustal thickening caused by subduction and compression, and formed a NE trending low-temperature and water-rich granite belt between the Jurassic granite in the eastern part of North China and the Zhangguangcai Ridge in NE China, which is parallel to the Paleo-Pacific subduction belt at the edge of Eurasia (Wu et al., 2005). The Chuhuertu granite is located in the Mesozoic magmatic active belt of the Greater Xing’an Range in the southeastern CAOB, and its emplacement age is Late Jurassic (ca. 156 Ma). Through petrology, mineralogy, and geochemistry, it is classified as A-type granite (water poor and formed under low pressure and high temperature). It is obvious that the Chuhuertu granite is unlikely to be the product of the subduction of the Paleo-Pacific plate in terms of space, time, and rock type.

The study area is located in southeastern Inner Mongolia-North Hebei. There are many Late Jurassic–Early Cretaceous volcanic eruption basins in the region. The crater is obviously distributed in the NE. The volcanic eruption basins are mainly depression type, and a large number of collapse calderas are developed inside. The long axis directions of the basins and the calderas are basically the same, both of which are in the NE, indicating that they are mainly controlled by large extension faults in the NE (Meng et al., 2011; Xu et al., 2013; Jiang et al., 2021). The direction is consistent with that of MOSB. At the same time, the volcanic rocks in the Late Jurassic and Early Cretaceous have similar distribution ranges, which are only distributed west of the Songnen Basin-North Hebei line (Figure 1C) (Wang et al., 2006; Zhang et al., 2008; Meng et al., 2011; Xu et al., 2011; Xu et al., 2013), and show more extensive features exposed to the west. These characteristics indicate that the Late Jurassic–Early Cretaceous volcanic rocks in the study area are remotely affected by the Mongol–Okhotsk tectonic system on the northwest side.

Chuhuertu A-type syenogranite is located in southeastern Inner Mongolia-North Hebei, and its emplacement age is Late Jurassic (ca. 156 Ma). Combined with a series of Late Jurassic A-type granites (Chen et al., 2008; Xie et al., 2012; Xue et al., 2015; Zhu et al., 2020) determined successively in the southeastern CAOB and NCC in recent years, the middle and lower crust in this area have been shown to have generally been in an extensional flow state in the Late Jurassic. In addition, alkaline-subalkaline volcanic rocks and rhyolites were widely exposed in the Greater Xing’an Range and the North Hebei-Western Liaoning during the Late Jurassic and Early Cretaceous (Meng et al., 2011; Li, 2012; Peng et al., 2012; Wang et al., 2013). The alkaline volcanic rock associations of the two periods of volcanic rocks indicate the regional extensional environment, which correspond to the extensional environment after A and B phases of Yanshan movement, respectively, suggesting that the two periods of magmatic events are related to the evolution of the Mongol–Okhotsk tectonic system (Zheng et al., 2000). Therefore, based on spatial fit and temporal correlation, Late Jurassic magmatism in the southeastern CAOB was mainly controlled by the Mongol–Okhotsk tectonic regime. During the Early Jurassic, the southern subduction of the Mongol-Okhotsk oceanic plate resulted in the formation of calc-alkali and mafic-intermediate volcanic rocks in the Ergun region (Figure 10A) (Xu et al., 2013). During the Middle Jurassic, the closure of the western MOO was completed, and a continental collision resulted in crustal thickening. The partial melting of the thickened crust formed S-type granite on the Xing’an Massif (Figure 10B) (Li et al., 2018a). During the Late Jurassic, with the “scissors” closure of the MOO from west to east, the subduction zone retreated in the NE trending, resulting in the subsequent rollback of the subduction slab, causing the collapse and delamination of the thickened crust, triggering the asthenospheric mantle upwelling, leading to the extension of the continental lithosphere (Tang et al., 2015; Zhang et al., 2018). The Chuhuertu A-type syenogranite in the study area is the product of the orogenic extension setting after the closure of MOO, and its emplacement under a certain extension event (Figure 10C). In the early Early Cretaceous, the rapid closure of the eastern MOO provided high-speed subduction, pushing the slab forward and causing flat-slab subduction of the oceanic plate. As the subduction slab continued to advance southward and the mantle wedge retreated, a set of basic to intermediate-acidic arc volcanic rocks were formed by partial melting of the mantle wedge in the Greater Xing’an Range (Figure 10D) (Meng et al., 2011; Xu et al., 2013). In the late Early Cretaceous, after the ocean closed, the convergence power of the subduction slab disappeared, and the subduction oceanic crust gradually collapsed from south to north due to the influence of gravitational potential energy, resulting in crustal thinning, mantle upwelling, and partial melting of crustal materials. A-type granite and rhyolite were formed in the Bahrain Left Banner-Zarut Banner in the southern segment of the Greater Xing’an Range (Figure 10E) (Zhang et al., 2020).