Geochronology and geochemistry of magmatic rocks in the Dongzi–Changhanboluo Pb–Zn ore district in Chifeng, Inner Mongolia, and their relationship with metallogenesis

Abstract

Bulk-rock elements, isotopes, and zircon U–Pb ages are reported for magmatic rocks in the Dongzi–Changhanboluo Pb–Zn ore district in Chifeng, Inner Mongolia, China. Zircon U–Pb dating identified four stages of magmatism: Late Silurian gabbroic diorite (~ 420 Ma), Middle Permian monzonite (~ 274 Ma), Late Jurassic quartz porphyry and ignimbrite, breccia tuff (153–158 Ma) and Early Cretaceous andesitic porphyrite (~ 127 Ma). Integrating field observations, geochronology, and element and isotope geochemistry indicated a complex petrogenetic history of the magmatic rocks. The gabbroic diorite may have been sourced from EM1-type mantle. The source of the monzonite may have been mantle metasomatized by melt from the subducting plate. The Jurassic volcaniclastic rocks formed in a medium-pressure, high-temperature environment, possibly in the background of crustal thickening in a syncollisional stage and an early postcollisional stage. During this process, shaly sedimentary rocks were brought into the deep crust and heated, followed by the rapid isostatic uplift of the crust, which caused partial melting of the sedimentary rocks. Quartz monzonite porphyry and quartz porphyry formed by partial melting of mantle metasomatized by subducted sediments, but the quartz porphyry experienced high-degree differentiation and evolution. The andesitic porphyrite has characteristics similar to those of Permian monzonite, indicating that its source area was also the zone of mantle metasomatized by subducted sediment. The late Silurian and Permian magmatic rocks in this area most likely formed against a continental arc background related to the subduction of the Paleo-Asian Ocean Plate beneath the North China Plate. The Late Jurassic magmatic rocks suggest that the northern margin of the North China Craton may have been in a post-collisional setting during the Late Jurassic, with no obvious crustal thinning. The Cretaceous andesitic porphyrite may have formed against the background of lithospheric extension and thinning. According to the comprehensive analysis of geological characteristics, diagenetic and metallogenic epochs, and Pb isotope data, the formation of ore bodies in the Dongzi–Changhanboluo ore district was closely related to the Jurassic quartz porphyry.

1 Introduction

The northern margin of the North China Craton (NCC) has undergone a complicated and long-term evolutionary process, forming a variety of different types of deposits (Rui et al. 1994; Chen et al. 2003; Zhang et al. 2009a; Liu et al. 2010). The Chifeng area is located in the eastern part of the northern margin of the NCC, which hosts many gold deposits (e.g., Honghuagou and Lianhuashan), Mo deposits (e.g., Giguanshan, Nianzigou, and Chehugou), and Pb–Zn deposits (e.g., Dongzi and Xiaoyingzi) (Fig. 1b) and has become one of the most important ore clusters in eastern China. Many studies have been carried out on the Mo and Au deposits to identify the geological characteristics, ore genesis, and corresponding geodynamic settings (Qin et al. 2009; Liu et al. 2010; Zhang et al. 2010b; Sun 2013; Wu et al. 2014; Shu et al. 2015; Wang et al. 2017). However, the Pb–Zn deposits have received little attention, and only a small amount of research has been carried out (Chen et al. 2003; Tang 2013; Xie et al. 2016a, b). The Pb–Zn deposits (e.g., Dongzi, Changhanboluo, Xishuiquan, Paoshouyingzi, Xiaoyingzi, and Yujiawopu) are distributed near the Shaolang River fault, which is a secondary fault of the Xilamulun fault, and form the Shaolanghe Pb–Zn–Ag metallogenic belt (Fig. 1b).

Fig. 1

Geological sketch map of the northern margin of the North China Craton, showing the location of major Mo and Pb–Zn deposits (after Liu et al. 2010). Serial numbers of Pb–Zn deposits: 1-Changhanboluo; 2-Dongzi; 3-Paoshouyingzi; 4-Shanzuizi; 5-Tianchungou; 6-Xishuiquan; 7-Dongshuiquan; 8-Xiaoyingzi; 9-Mujiadian; 10-Yujiawopu

The Dongzi and Changhanboluo deposits are located in the western part of the Shaolanghe Pb–Zn–Ag metallogenic belt and have been classified as hydrothermal vein-type deposits that formed at 153 Ma and 157.6 Ma, respectively (Chen et al. 2003; Xie et al. 2016a). Previous studies have mainly focused on the geological characteristics and ore-forming fluids, which suggest that the deposits are mostly controlled by fault structures and that the mixing of hot magmatic fluids and meteoric water resulted in the rapid precipitation of ore-forming materials (Xie et al. 2016a; Wang et al. 2017). However, note that extensive magmatic rocks are developed in the Dongzi–Changhanboluo Pb–Zn ore district. These magmatic rocks can provide important constraints on magmatic processes and tectonic settings. However, to date, the petrogenesis and tectonic settings of the magmatic rocks in this area remain poorly constrained. In addition, the period of magmatism related to mineralization is ambiguous. In this paper, we report new geochronological and geochemical data for the magmatic rocks of the Dongzi–Changhanboluo Pb–Zn ore district in Chifeng. These results can help us better understand the timing of magmatism and Pb–Zn mineralization in this area and elucidate the petrogenesis of these intrusions and their relationship with metallogenesis.

2 Regional geological setting

The Dongzi–Changhanboluo Pb–Zn ore district belongs to the Bainaimiao arc on the northern margin of the NCC (Fig. 1a), which is bounded by the Xilamulun fault to the north and the Chifeng-Bayan Obo fault to the south. Magmatism of the Hercynian, Indosinian, and Yanshanian periods developed in much of the region, forming a large number of Mo, Pb, and Zn deposits (Fig. 1b). Some thorough and detailed studies on the Mo deposits in this area concluded that the mineralization was closely related to the synchronous intermediate-acid magmatism. The diagenesis and mineralization are mainly concentrated in the three periods of 258–210, 185–150, and 140–110 Ma, which correspond to the three processes of post-collisional extension, tectonic stress transformation, and intracontinental extension (Zhang et al. 2010a).

The Pb–Zn deposits mainly occur in the Shaolanghe Pb–Zn–Ag metallogenic belt, where more than ten small and medium-sized Pb–Zn deposits have been formed (Fig. 1c). The Dongzi and Changhanboluo deposits are located in the western part of the Shaolanghe Pb–Zn–Ag metallogenic belt. The main strata exposed in the study area include Ordovician–Silurian amphibolite and gneiss, which are scattered in the Dongzi and Yujiawopu ore districts; Carboniferous sandstone, andesite, etc.; a Permian volcanic-sedimentary clastic rock series, members of which are widely distributed and act as important host rocks; a Jurassic volcanic-sedimentary rock series, such as tuff, breccia tuff, ignimbrite, and others; and Tertiary porous and dense massive basalt with mudstone (Fig. 1c).

Fault structures are the main structural type in this area, followed by fold structures. The main fold structure is the Dongzi–Xiaoyingzi E-W-trending compound anticline, which has undergone several stages of tectono-magmatic activity. The faults in the study area are well-developed, with E-W, N-W, and N-E directions, and form a lattice fault system. The E-W-trending Shaolanghe fault and its secondary faults on both sides controlled the locations of the Pb–Zn–Ag ore veins, while the NE-trending faults controlled the distribution of magmatic rocks.

Multiple magmatic events took place in this region during different episodes, leading to the formation of the Herwusu monzonite, K-feldspar granite and granite porphyry, quartz porphyry, diorite porphyry, and syenite porphyry occurring as batholith, stocks, and dikes. The Herwusu monzonite intruded into the Permian System, and a large number of Pb–Zn deposits or ore sites are distributed around the monzonite batholith (Fig. 1c). These Pb–Zn deposits formed in the Jurassic and corresponded to the structural system transition process in the northern margin of the NCC (Rui et al. 1994; Chen et al. 2003; Wang 2016; Xie et al. 2016a).

3 Deposit geology and petrographic characteristics

The Dongzi deposit is a representative deposit of the Shaolanghe Pb–Zn–Ag metallogenic belt and was classified as an epithermal deposit in previous studies (Xie et al. 2016b). The strata in the Dongzi Pb–Zn mining area are mainly Permian andesite. The ore bodies, which lie concealed between gabbroic diorite and andesite units, are obviously controlled by faults. The main magmatic rocks are the Herwusu monzonite and the Dongzi gabbroic diorite. Some enclosed ancient amphibolite strata can also be seen in the gabbroic diorite body (Fig. 2a). There are two main ore bodies in the mining area: 1 and 1-1. The No. 1 ore body is the main E-W-trending ore body and has good continuity along a length of approximately 1530 m. The No. 1-1 ore body is a vein parallel to the No. 1 orebody, which also occurs in a vein shape. The metallic minerals are pyrite, galena, and sphalerite, followed by chalcopyrite, arsenopyrite, and others. The wall rock alteration is dominated by medium- to low-temperature linear alteration, such as silicification, sericitization, carbonation, and chloritization, which has developed on both sides of the hydrothermal vein.

Fig. 2

Geological maps of Dongzi (a), the Changhanboluo ore district (b) and the 750 m ichnography of the Changhanboluo mining area (c)

The Changhanboluo deposit is located south of the Dongzi deposit (Fig. 1c). The strata that crop out in the mining area are mainly the Permian Yujiabeigou Formation andesite and andesitic basalt, which are exposed in the northeastern part of the mining area; the Jurassic Baiyingaolao Formation tuff and ignimbrite, which are widely exposed in the northwest and central parts of the mining area; and the Tertiary Hannuoba Formation basalt and mudstone, which are exposed in the middle part of the mining area (Fig. 2b). The intrusive rocks in the mining area include the Herwusu monzonite, which occurs as a batholith and overlaps the Permian andesite strata from east to west, a small number of quartz monzonitic porphyry veins filled along N-E or E-W faults, quartz porphyry intruded between the Herwusu monzonite and andesite and andesitic porphyrite veins that penetrate the quartz porphyry (Fig. 2b, c). The ore bodies lie concealed to the south of the mining area and mainly occur in quartz porphyry as veins. The main metallic minerals are galena, sphalerite, and pyrite, followed by chalcopyrite, arsenopyrite, aerosite, tetrahedrite, and others.

Extensive magmatic activity occurred in the Dongzi–Changhanboluo ore district and formed various rock types, which are mainly composed of the Dongzi gabbroic diorite, Herwusu monzonite, Changhanboluo quartz porphyry, and andesite porphyrite. In addition, there are some volcaniclastic rocks, such as ignimbrite and breccia tuff in the northwestern part of the Changhanboluo mining area, which are intruded by quartz monzonite porphyry veins.

The Dongzi gabbroic diorite and andesite are the main host rocks of the ore bodies, and we can see some hydrothermal veins intruded into the andesite (Fig. 3a). The Dongzi gabbroic diorite is a medium-grained rock with a hypidiomorphic granular texture (Fig. 3d). The rock is composed of plagioclase (~ 70 vol%), pyroxene (~ 15 vol%), and hornblende (~ 12 vol%), with minor amounts of quartz and apatite (~ 3 vol%). The plagioclase is plate-shaped, measures 0.2 × 1.0–0.6 × 2.0 mm, and forms subhedral laths with occasional albite and Carlsbad-albite-type combined twinning. The pyroxene and hornblende are idiomorphic–hypidiomorphic granular with granularities between 0.3 and 1.5 mm.

Fig. 3

Field photographs and photomicrographs of magmatic rocks in the Dongzi–Changhanboluo ore district. a Dongzi hydrothermal veins penetrated fractures in andesite; b Changhanboluo ore veins penetrated fractures in monzonite and caused chlorite alteration; c quartz porphyry with disseminated mineralization and hydrothermal vein penetration; d Dongzi gabbroic diorite; e Herwusu monzonite; f quartz porphyry; g rhyolite ignimbrite; h quartz monzonite porphyry; i andesitic porphyrite. Mineral abbreviations: Bi, biotite; Hb, hornblende; Kf, K-feldspar; Pl, plagioclase; Px, pyroxene; Py, pyrite; Q, quartz; Ser, sericite. a The Dongzi gabbroic diorite (DZ18-1); b the Herwusu monzonite (Zk003-104.5); c1, c2 the quartz porphyry (CHBL-1); d the ignimbrite (BD-10); e the quartz monzonite porphyry (BD-23); f, g the andesitic porphyrite (CH-CM-2)

The Herwusu monzonite are fresh and there are also hydrothermal veins penetrating into the monzonite (Fig. 3b). The monzonite displays a monzonitic texture composed of plagioclase (~ 40 vol%), K-feldspar (~ 40 vol%) and minor quartz (~ 5 vol%), amphibole (~ 8 vol%), and biotite (~ 6 vol%). The accessory minerals are apatite, titanite, zircon, etc. (Figure 3e). The K-feldspars have xenomorphic granular textures and are mostly altered to kaolinite. The plagioclases are euhedral and tabular, measuring 0.8 × 1.6–1.6 × 3.2 mm, with polysynthetic twins and zoned textures, and a portion are replaced by sericite. Anhedral plagioclase is embedded in the K-feldspar crystals.

The quartz porphyry invaded between the andesite and the Herwusu monzonite as stocks, and it often shows disseminated mineralization with hydrothermal vein penetration (Fig. 3c). The rock has a porphyritic texture, with quartz as the phenocrysts. The quartz phenocrysts (~ 30 vol%) occur with a xenomorphic granular texture, measure 0.4–2.4 mm in diameter and are mostly eroded, with embayed shapes. The matrix is composed of felsic minerals with spherulitic textures and is generally altered to sericite.

The rhyolite ignimbrite shows a welded tuff texture and pseudo flow structures and is composed of a large number of rock fragments (~ 60 vol%), vitroclastic fragments (~ 25 vol%), and a small amount of crystal fragments (~ 15 vol%). The rock fragments are banded and flame-like at both ends and show the same direction as that of the vitroclastic fragments, which look like fine stripes and earthworms. The crystal fragments are mainly composed of quartz, plagioclase, biotite, etc. (Fig. 3g).

The Changhanboluo quartz monzonite porphyry shows a porphyritic texture and contains ~ 60 vol% phenocrysts by volume (Fig. 3h). The phenocrysts consist of platy plagioclase (~ 35 vol%), K-feldspar (~ 20 vol%), and subordinate biotite (~ 5 vol%). The plagioclase phenocrysts generally form subhedral laths with occasional albite and Carlsbad-albite-type combined twinning. Some of these laths display distinct zoning. The K-feldspar is mostly granular, and some are metasomatized to chlorite in the middle. The biotite is reddish-brown, indicating high Ti content, and some are decomposed to form rutile.

The andesitic porphyrite invaded the quartz porphyry as veins, with obvious differences in occurrence. The rock is porphyritic with an almond structure. The phenocrysts and matrix are plagioclase and elongated plagioclase microlites, respectively (Fig. 3i).

4 Samples and analytical methods

4.1 Zircon U–Pb dating

Zircons for U–Pb dating were selected from the magmatic rocks of the Dongzi–Changhanboluo ore district (Dongzi diorite DZ18-1; Changhanboluo quartz monzonite Zk003-104.5; ignimbrite BD-10; quartz monzonite porphyry BD-23; andesite porphyry CH-CM-2 and quartz porphyry CHBL-1). These zircons were separated using conventional heavy liquid and magnetic techniques and then handpicked under a binocular microscope to select euhedral and translucent zircons for analysis. The zircons were mounted in epoxy and then polished and gold-coated for subsequent cathodoluminescence (CL) imaging and zircon U–Pb analyses. CL images were obtained at the Electron Microprobe Laboratory of the Institute of Geology, Chinese Academy of Geological Sciences (CAGS). Two methods, laser-ablation inductively coupled plasma mass spectrometry (LA-ICP-MS) and secondary-ion mass spectrometry (SIMS), were used for zircon U–Pb dating.

LA-ICP-MS analyses were performed on the exposed zircon surfaces using a 213 laser ablation system coupled with an Agilent 7500a quadrupole ICP-MS instrument at the Key Laboratory of Orogenic Belt and Crustal Evolution, Beijing University, China. The zircon standard PLE was used as a calibration standard, while the zircon standard QH was used as a supplementary calibration standard. The laser beam had a diameter of 32–60 μm, depending on the sample grain size, was operated at 5 Hz and 12 J/cm² and was limited to an analytical cycle of 0.2 s. Ages were calculated from the U and Th decay constants recommended by Steiger and Jäger (1977). All data were processed using the Glitter 4.4 software, and the reported ages were calculated using Isoplot 3.0.

SIMS zircon U–Pb analyses were conducted using the CAMECA IMS-1280 ion microprobe at the Institute of Geology and Geophysics, Chinese Academy of Sciences, Beijing, China. U–Th–Pb ratios and absolute abundances were determined relative to the zircon standard 91500 (Wiedenbeck et al. 1995), and the instrumental performance and detailed analytical procedures were similar to those described by Li et al. (2009). Isotopic compositions were corrected for common Pb using the measured ²⁰⁴Pb. Uncertainties concerning individual analyses in data tables are reported at the 1σ level and mean ages for pooled U/Pb analyses are quoted at the 95 % confidence interval. Data reduction was performed using Isoplot 3.0 (Ludwig 2003).

4.2 Major and trace element determinations

The major and trace element compositions were determined at the Analytical Laboratory of the Beijing Research Institute of Uranium Geology, China National Nuclear Corporation (CNNC). The major element compositions were analyzed using an automatic X-ray fluorescence (XRF) spectrometer (Philips PW2404). The contents of volatiles (e.g., CO₂ and H₂O) were determined by measuring the reduction in weight that resulted after heating the samples at 1050 °C for 30 min. The analytical precision for the major oxides SiO₂, Al₂O₃, MgO, CaO, and TFe₂O₃ was better than 0.1 wt%, and that for the major oxides MnO, Na₂O, K₂O, P₂O₅,and TiO₂ was better than 0.01 wt%. Trace elements, including rare earth elements (REEs), were determined using an ELEMENT ICP-MS. The measurement precision for the trace elements was generally better than 3 wt%.

4.3 Sr–Nd–Pb isotope analysis

The Sr–Nd–Pb isotope analyses were conducted using an ISOPROBE-T thermoelectric ionization mass spectrometer (TIMS) at the Analytical Laboratory of the Beijing Research Institute of Uranium Geology, CNNC. The Sr–Nd isotopic ratios were corrected for mass fractionation by normalization to ⁸⁶Sr/⁸⁸Sr = 0.1194 and ¹⁴⁶Nd/¹⁴⁴Nd = 0.7219. The international standard samples NBS-987 and JMC were employed to evaluate the instrument stability during the period of data collection. The measured values for the NBS-987 Sr standard and the JMC Nd standard were ⁸⁷Sr/⁸⁶Sr = 0.710250 ± 7 (2SD) and ¹⁴³Nd/¹⁴⁴Nd = 0.512109 ± 3 (2SD), respectively. The Pb isotopic analytical procedures were described by Liu et al. (2014). The analytical precision was 0.005% for ²⁰⁸Pb/²⁰⁶Pb.

5 Results

5.1 Zircon U–Pb dating

Zircon U–Pb dating of samples from the Dongzi deposit (sample DZ18-1) and Changhanboluo deposit (samples Zk003-104.5, BD-10, BD-23, and CH-CM-2) was undertaken using LA-ICP-MS, whereas zircons of sample CHBL-1 from the Changhanboluo deposit were analyzed using SIMS U–Pb dating because of the small size of the zircon particles. The results are given in Table 1 and shown in Fig. 4.

Table 1 LA–ICP–MS and SIMS zircon U–Pb data for magmatic rocks from the Dongzi–Changhanboluo ore district

Fig. 4

Zircon U–Pb concordia diagrams for magmas in the Dongzi–Changhanboluo ore district

5.1.1 The Dongzi gabbroic diorite

Sample DZ18-1 is a gabbroic diorite collected from the Dongzi deposit. Zircons from this sample are short columnar or long columnar, with particle sizes varying from 50 to 300 µm, and the ratio of the long to the short axis is mostly between 1:1 and 3:1. The zircons have oscillatory growth zoning, which is typical of magmatic zircons (Koschek 1993). The ²⁰⁶Pb/²³⁸U ages from 20 zircon analyses yield a weighted mean ²⁰⁶Pb/²³⁸U age of 419.6 ± 5.5 Ma [mean square weighted deviation (MSWD) = 1.16] (Fig. 4a), representing the emplacement age of the gabbroic diorite.

5.1.2 The Herwusu monzonite

Sample Zk003-104.5 is a monzonite collected from drill core Zk003 in the Changhanboluo deposit. This sample yields euhedral zircons with oscillatory growth zoning and Th/U ratios of 0.91–1.89, which are indicative of an igneous origin (Koschek 1993; Hoskin and Black 2000). The ²⁰⁶Pb/²³⁸U ages obtained from 25 analyses range between 270 and 276 Ma, yielding a weighted mean ²⁰⁶Pb/²³⁸U age of 274.0 ± 1.0 Ma (MSWD = 0.33) (Fig. 4b), which is interpreted to represent the formation time of the monzonite.

5.1.3 The quartz porphyry

Sample CHBL-1 is a quartz porphyry collected from the tunnel of the Changhanboluo deposit. Zircons from this sample are euhedral-subhedral with oscillatory growth zoning and have high Th/U ratios (0.13–1.05), suggesting a magmatic origin (Hoskin and Black 2000). The ²⁰⁶Pb/²³⁸U ages from 19 zircon analyses range from 149 to 240 Ma, yielding two concordant groups of ages at 229.8 ± 6.3 Ma (MSWD = 4.3, n = 11) and 158.1 ± 3.4 Ma (MSWD = 4.0, n = 8) (Fig. 4c1, c2). These data indicate that the quartz porphyry formed at 158.1 ± 3.4 Ma, with 229.8 ± 6.3 Ma representing the age of captured or inherited zircons entrained by the quartz porphyry.

5.1.4 The ignimbrite and quartz monzonite porphyry

Samples BD-10 and BD-23 are an ignimbrite and a quartz monzonite porphyry, respectively, collected from the western part of the Dongzi–Changhanboluo ore district and contain euhedral–subhedral zircons with oscillatory zoning that have Th/U ratios of 0.39–7.37, indicating a magmatic origin (Koschek 1993; Hoskin and Black 2000). The ²⁰⁶Pb/²³⁸U ages from the zircon analyses of ignimbrite and quartz monzonite porphyry yield weighted mean ²⁰⁶Pb/²³⁸U ages of 153.4 ± 0.7 Ma (MSWD = 1.2) and 153.2 ± 0.7 Ma (MSWD = 1.0) (Fig. 4d, e), respectively, which are interpreted as the crystallization ages of the ignimbrite and quartz monzonite porphyry.

5.1.5 The andesitic porphyrite

Sample CH-CM-2 collected from the tunnel of the Changhanboluo deposit is an andesitic porphyrite that cuts through the Changhanboluo Pb–Zn ore veins. Zircons from this sample are subhedral with oscillatory zoning and have high Th/U ratios (0.43–1.08), suggesting a magmatic origin (Hoskin and Black 2000). The ²⁰⁶Pb/²³⁸U ages from 10 analyses yield a weighted mean ²⁰⁶Pb/²³⁸U age of 127.4 ± 1.9 Ma (MSWD = 4.7) (Fig. 4f, g), which is interpreted as the crystallization age of the andesitic porphyrite. Most of the other analyses show the occurrence of Pb loss and may not be significant.

The above zircon U–Pb ages suggest that the Early Paleozoic–Mesozoic magmatism in the study area can be subdivided into four stages: Late Silurian (~ 420 Ma), Middle Permian (~ 274 Ma), Late Jurassic (153–158 Ma) and Early Cretaceous (~ 127 Ma).

5.2 Major and trace elements

The major and trace element geochemistry of the Early Paleozoic–Mesozoic magmas of the Dongzi–Changhanboluo ore district is given in Table 2.

Table 2 Whole-rock chemical analyses for the magmatic rocks in Dongzi–Changhanboluo ore district

5.2.1 Late Silurian magmas

The Late Silurian magmas in the study area are the Dongzi gabbroic diorites. They have SiO₂ = 50.77–52.88 wt%, TiO₂ = 0.73–0.85 wt%, Al₂O₃ = 19.72–20.14 wt%, Mg^# [Mg^# = 100 Mg/(Mg + TFe²⁺)] = 47–48, and Na₂O + K₂O = 4.61–4.93 wt% (Table 1), and they belong to the subalkaline (Fig. 5a) and calc-alkaline series (Fig. 5b).

Fig. 5

SiO₂ versus total alkali (Na₂O + K₂O) (a) and SiO₂ versus K₂O (b) diagrams for the Early Paleozoic–Mesozoic magmas in the study area. The boundary lines in the SiO₂ versus (Na₂O + K₂O) and SiO₂ versus K₂O diagrams are from Irvine and Baragar (1971) and Peccerillo and Taylor (1976), respectively. Abbreviations: CH, Changhanboluo mining area; DZ, Dongzi mining area; GD, gabbroic diorite; MD, monzodiorite; MZ, monzonite; QM, quartz monzonite

The gabbroic diorites have 131.34–135.52 ppm of total REEs and display coherent REE patterns characterized by relatively light rare earth element (LREE) enrichment and heavy rare earth element (HREE) depletion ((La/Yb)_N = 12.58–14.46), with weak positive Eu anomalies (δEu = 1.05–1.19) (Fig. 6a). They are enriched in large ion lithophile elements (LILEs), such as Rb, Ba, K, and LREEs, but depleted in high-field strength elements (HFSEs), such as Nb, Ta, Ti, and HREEs (Fig. 6b).

Fig. 6

Chondrite-normalized REE patterns (a, c, e, g) and primitive mantle-normalized trace element spidergrams (b, d, f, h) for the Early Paleozoic–Mesozoic magmas in the study area. Chondrite and primitive mantle normalization values are from Boynton (1984) and McDonough et al. (1992), respectively. Abbreviations: CH, Changhanboluo mining area; DZ, Dongzi mining area

5.2.2 Middle Permian magmas

The Middle Permian magmas are the Herwusu monzonites in the study area. The monzonites have compositional ranges of 57.68–59.87 wt% SiO₂ and 16.15–16.93 wt% Al₂O₃. They are enriched in Na₂O + K₂O (6.83–7.76 wt%), with Na₂O/K₂O ratios ranging from 1.01 to 1.19. All samples plot in the monzonite field on the total alkali-silica (TAS) diagram (Fig. 5a) and the high-K calc-alkaline to shoshonitic series area on the SiO₂ versus K₂O diagram (Fig. 5b).

The monzonites are enriched in LREEs but depleted in HREEs (Fig. 6a), with total REEs of 176.95–221.73 ppm and slightly negative Eu anomalies (δEu: 0.69–0.87). All samples are enriched in LILEs, such as Rb, Ba, Th, U, K, and LREEs, but have negative anomalies for HFSEs (e.g., Nb, Ta, P, Ti, and HREEs) compared to the primitive mantle (Fig. 6b). At the same time, the monzonites present high Sr contents (542–703 ppm, average 606 ppm) and Sr/Y values (18–29, average 22).

5.2.3 Late Jurassic magmas

The Late Jurassic magmas in the study area consist of a series of quartz porphyry, ignimbrite, breccia tuff, and quartz monzonite porphyry.

The quartz porphyries have high SiO₂ (75.76–78.76 wt%), K₂O (3.91–5.90 wt%) and MgO (11.89–13.26 wt%) contents and low CaO (0.16–0.91 wt%) and TiO₂ (0.051–0.057 wt%) contents. All the samples plot in the granite field on the TAS diagram (Fig. 5a) and the high-K calc-alkaline and shoshonite series on the SiO₂ versus K₂O diagram (Fig. 5B). These quartz porphyries have “Seagull”-type REE patterns, with very low ΣREE (78.29–102.51 ppm) and strongly negative Eu anomalies (δEu = 0.05–0.08) (Fig. 6e). Additionally, the rocks show high Rb (180–228 ppm), Ga (14.4–17.5 ppm), Zr (74.9–90.8 ppm), Nb (25.4–34.3 ppm), and Y (25.8–37.4 ppm) contents and low Sr (24.7–52.8 ppm), Ba (27.6–182 ppm), Cr (average 2.0 ppm), and Ni (average 1.5 ppm) contents and exhibit strongly negative Ba, Sr, P, and Ti anomalies on the spider diagram of trace elements (Fig. 6f).

The ignimbrite and breccia tuff in the western part of the Dongzi–Changhanboluo ore district have high SiO₂ (77.32 and 78.87 wt%) and low MgO (0.16 and 0.08 wt%) contents. Their A/CNK [molar Al₂O₃/(CaO + K₂O + Na₂O)] values are 1.11 and 1.12, respectively, which are indicative of peraluminous granites (Maniar and Piccoli 1989). All the samples plot in the granite field on the TAS diagram (Fig. 5a) and the high-K calc-alkaline series on the SiO₂ versus K₂O diagram (Fig. 5b). The ignimbrite and breccia tuff have lower REEs (119.02 and 124.41 ppm), Yb (0.98 and 2.19 ppm) and δEu (0.24, 0.12). In addition, the ignimbrite and breccia tuff show similar trace element partition models, except for strongly negative Ba, Sr, P, and Ti anomalies. Overall, the REE and trace element distribution patterns of ignimbrite and breccia tuff are similar to those of quartz porphyry (Fig. 6e, f).

The quartz monzonite porphyry has SiO₂ = 67.07–69.04 wt%, TiO₂ = 0.46–0.52 wt%, and Al₂O₃ = 16.09–16.35 wt% and belongs to the high-K calc-alkaline series (Fig. 5b). The samples have total REE contents (ΣREE) ranging from 98.50 to 141.48 ppm and show distinct REE enrichment [(La/Yb)_N = 15.63–20.28] and positive Eu anomalies (δEu = 1.34–1.41) on chondrite-normalized REE diagrams (Fig. 6e). On primitive mantle-normalized spidergrams, the samples exhibit similar trace element patterns, with depletions in Nb, Ta, P, and Ti and enrichments in Ba and Eu (Fig. 6f).

5.2.4 Early Cretaceous magmas

The Early Cretaceous magmas formed an andesitic porphyrite that cuts through the ore veins in the Changhanboluo deposit and underwent obvious carbonation alteration. The andesite porphyrite has SiO₂ = 56.25–58.96 wt% and Al₂O₃ = 16.46–17.53 wt%. The samples have lower ΣREE (150.68–161.19 ppm) and weak negative Eu anomalies (δEu = 0.85–0.89) on the chondrite-normalized REE diagram (Fig. 6g). On the primitive mantle-normalized spidergram, the samples exhibit enrichments in LILEs but depletions in HFSEs, with weak depletions in Nb, Ta, Sr, and Ti (Fig. 6h).

5.3 Sr–Nd isotopes

The whole-rock Sr–Nd isotope data are listed in Table 3 and plotted in Fig. 7. The initial ⁸⁷Sr/⁸⁶Sr ratios and the ε_Nd(t) values are calculated according to their zircon U–Pb ages.

Table 3 Sr–Nd isotope compositions for the magmatic rocks in Dongzi–Changhanboluo ore district

Fig. 7

(⁸⁷Sr/⁸⁶Sr)_i versus ε_Nd(t) diagram for magmas in the Dongzi–Changhanboluo ore district. The Sr–Nd isotope data of the lower and upper crust are derived from Jahn et al. (1999)

The Late Silurian Dongzi gabbroic diorites show homogeneous Sr–Nd isotope compositions characterized by (⁸⁷Sr/⁸⁶Sr)_i = 0.706338–0.706347 and ε_Nd(t) values from − 15.22 to − 14.78. The Middle Permian Herwusu monzonites have (⁸⁷Sr/⁸⁶Sr)_i ratios of 0.704974 to 0.705002 and ε_Nd(t) values of − 8.18 to − 7.47. The Late Jurassic magmas of quartz porphyry and quartz monzonite porphyry show similar Sr–Nd isotope data characterized by (⁸⁷Sr/⁸⁶Sr)_i = (0.705422–0.707590) and ε_Nd(t) values from − 8.36 to − 6.89. The Early Cretaceous andesitic porphyrites are similar to the Late Jurassic magmas, with (⁸⁷Sr/⁸⁶Sr)_i = 0.706133–0.706225 and ε_Nd(t) values from − 9.29 to − 8.89.

6 Discussion

6.1 Petrogenesis

6.1.1 Late Silurian magmas

The Late Silurian gabbroic diorites have low SiO₂ contents (50.77–52.88 wt%) and high Mg^# (47–48), CaO (7.27–8.10 wt%), and Na (Na₂O/K₂O = 3.28–4.82) contents, which indicate the partial melting of ultramafic rocks from the mantle. The mantle is usually divided into the asthenospheric mantle and lithospheric mantle. Different mantle sources show different geochemical properties. The ratios of LILEs to HFSEs can trace the sources of magmas. For example, La/Nb < 1.5 and La/Ta < 22 imply asthenospheric mantle, while La/Nb > 1.5 and La/Ta > 30 indicate lithospheric mantle (Fitton et al. 1988; Leat et al. 1988). The gabbroic diorite displays obviously higher La/Nb (5.21–5.91) and La/Ta (120.2–133.6) than the values of the asthenospheric mantle. Together with the enrichment in LILEs and depletion of HFSEs and HREEs, these data suggest that the source of the gabbroic diorites was not the asthenospheric mantle but the enriched continental lithospheric mantle (Song et al. 2015).

Compared with the major and trace element compositions of the intrusion, the radiogenic isotope ratios, which are less affected by magma evolution processes such as fractional crystallization, can be more useful for interpreting magma sources (e.g., Li et al. 2014, 2015). The gabbroic diorites have relatively homogeneous ε_Nd(t) values from − 15.22 to − 14.78 and (⁸⁷Sr/⁸⁶Sr)_i ratios between 0.706338 and 0.706347. In the (⁸⁷Sr/⁸⁶Sr)_i–ε_Nd(t) diagram, all the samples plot near the enriched mantle 1 (EM1) field. Therefore, we conclude that the gabbroic diorite may have been sourced from EM1-type mantle.

6.1.2 Middle Permian magmas

The Middle Permian magmatic rocks are the Herwusu monzonite batholiths in the study area. The rocks are characterized by high alkali contents (K₂O + Na₂O > 5 wt%) and K₂O/Na₂O values (K₂O/Na₂O > 0.5 and 1.0 when SiO₂ = 50 and 55 wt%, respectively) and low TiO₂ contents (mostly < 1.3 wt%). The contents of Al₂O₃ are higher and have a large range of variability (mostly between 14 and 19 wt%). The rocks show strong enrichment in LILEs (such as average values of Rb = 91.7, Ba = 962, and Sr = 606, in ppm) and LREEs, which are characteristics of the shoshonitic rock series (Morrison 1980). In Fig. 8, the samples are all located in the shoshonitic series area. The shoshonitic series was limited to volcanic rocks in the past (Morrison 1980). Current studies show that the potassium basalt series is not confined to volcanic rocks but also includes intrusive rocks (Fowler and Henney 1996; Eklund et al. 1998). Therefore, the monzonite in this area belongs to the shoshonitic series.

Fig. 8

(After Turner et al. 1996)

K₂O versus Na₂O diagram for the Middle Permian Herwusu monzonite.

The shoshonitic rocks mainly originated from the metasomatic mantle enriched in potassium and LILEs, which are associated with subduction (Foley and Peccerillo 1992). Such rocks are mostly formed in island arc or continental arc settings but can also occur in intraplate and post-collision settings (Muller and Groves 1995; Deng et al. 2004; Li et al. 2006; Xia et al. 2006). The shoshonitic rocks associated with island arcs are highly enriched in LREEs and LILEs and significantly depleted in Nb, Ta and Ti (Li et al. 2001). The shoshonitic rocks in an oceanic plate have trace element characteristics of oceanic island basalts (OIB) (Weaver 1991), while those in a continental plate have both island arc and OIB trace element characteristics (Chung et al. 2001; Zou et al. 2003). The monzonites have LREEs and LILEs enrichments and Nb, Ta, and Ti depletions. These characteristics are all typical island arc magmatic characteristics, and in the (La/Yb)_N–Yb_N and Sr/Y–Y diagrams (Fig. 9), the samples all plot in the classic island arc field.

Fig. 9

(After Defant and Drummond 1990 and Martin et al. 2005)

Sr/Y–Y and (La/Yb)_N–Yb_N diagrams for the Middle Permian Herwusu monzonite.

The monzonites also have homogeneous ε_Nd(t) values from − 8.18 to − 7.47 and (⁸⁷Sr/⁸⁶Sr)_i ratios from 0.704974 to 0.705002, which are higher ε_Nd(t) and lower (⁸⁷Sr/⁸⁶Sr)_i values than those of theLate Silurian gabbroic diorites. The Sr–Nd isotope compositions of the monzonites represent the slightly enriched mantle-derived material in this region. Because of the similar partition coefficients of Nb and U, the Nb/U ratio can reflect the characteristics of the source region (Chung et al. 2001). The Herwusu monzonites have Nb/U ratios (3.7–7.4, average 5.6) lower than those of mid-ocean ridge basalts (MORB) and OIB (average 50) and the silicate Earth (average 32) and slightly lower than that of the upper crust (average 9). In Fig. 10a, most of the samples plot in the arc volcanic rock field. The lower Nb/U ratios may indicate fluid metasomatism in the subduction zone. In addition, the Ce/Pb ratio of monzonite (3.86–5.15, average 4.57) is significantly lower than those of oceanic basalts (average 25) and the silicate Earth (average 11) and slightly higher than that of the upper crust (average 3.2). In Fig. 10b, the samples are located above the upper crust field, near the arc volcanic rock field. Therefore, the monzonite in this area is suggested to have originated mainly from the metasomatic mantle rich in potassium and LILEs related to subduction. Against the background of subduction, the source may have been the mantle zone of metasomatism, where fluid/melt was released by the subducting plate. The higher Al₂O₃ content (16.15–16.93 wt%), Th/Ce ratio (0.10–0.14) and Th/Nb ratio (0.72–1.14) and lower Pb/Nd ratio (0.36–0.44) and Ce/Pb ratio (3.86–5.15) all indicate the addition of subducted sediments in the mantle source area.

Fig. 10

(After Chung et al. 2001)

Nb/U-Nb (a) and Ce/Pb–Ce (b) diagrams for the Middle Permian Herwusu monzonit.

Al₂O₃ content is an important indicator of pressure. The pressure is less than 16 kbar when Al₂O₃ < 15 wt% and the residues are amphibole, plagioclase and orthopyroxene. The pressure is greater than 16 kbar when Al₂O₃ > 15 wt% and the residual phases are clinopyroxene, amphibole, plagioclase and garnet (Rapp et al. 1991). When the pressure is between 20 and 30 kbar, the melt equilibrates with garnet-bearing, plagioclase-free amphibolite-, granulite- or eclogite-facies rocks (Rapp et al. 1991; Rapp and Watson 1995). The content of Al₂O₃ in the monzonite (16.15–16.93 wt%) is greater than 15 wt%, which indicates that the formation pressure of the monzonite was > 16 kbar. The Sr/Y ratio is mainly controlled by the relative contents of garnet, amphibole and plagioclase in the residual phase. A high content of garnet and low contents of amphibole and plagioclase in the residual phase lead to a high Sr/Y ratio (Springer and Seck 1997). High Sr/Y ratios (average 22) in the monzonite may indicate the existence of garnet but not plagioclase in the residual phase, which is consistent with the weak Eu anomaly (Fig. 6c). Therefore, the monzonite may have been formed under moderate or high pressure, which is consistent with its high Sr and Yb characteristics (Zhang et al. 2006).

6.1.3 Late Jurassic magmas

The Late Jurassic magmatic rocks are widely distributed in this area, forming a large range of intermediate-acid volcaniclastic rocks and subvolcanic veins (including ignimbrite, breccia tuff, and quartz monzonite porphyry veins) in the western part of the study area and quartz porphyry in the Changhanboluo mining area.

Granitic rocks should originate mainly from the partial melting of crustal rocks, regardless of how strongly the geochemistry is imprinted by the mantle because the partial melting of mantle peridotite cannot directly produce granitic melts (Liu et al. 2002). The ignimbrite and breccia tuff have similar geochemical characteristics that may indicate the same magma source. They all have high SiO₂ contents, low CaO and MgO contents, and high aluminum saturation index (ASI) values (average 1.1), which mark peraluminous rocks. Corundum occurs as a CIPW normative mineral, and the content is more than 1%, which is consistent with the characteristics of typical S-type granite (ASI > 1.1 and the CIPW normative corundum content > 1%, Chappel and White 1974). The source area may have been argillaceous rock with low CaO/Na₂O ratios (average 0.08, < 0.3) (Sylvester 1998).

The Al₂O₃/TiO₂ ratio in peraluminous granite can be used as an indicator of the partial melting temperature of source rocks. Previous studies show that when the ratio of Al₂O₃/TiO₂ is less than 100, the partial melting temperature of the source region is greater than > 875 °C, while when the ratio of Al₂O₃/TiO₂ is greater than 100, the partial melting temperature of the source region is less than 875 °C; thus, the ratio of Al₂O₃/TiO₂ is inversely related to the formation temperature (Sylvester 1998). The ratios of Al₂O₃/TiO₂ of the volcanic clastic rocks in this area are between 66 and 77 (all less than 100), indicating that the partial melting temperature in the source may have been higher than 875 °C. By combining this information with the whole-rock saturation temperature of zirconium (840–879 °C), we can conclude that the volcaniclastic rocks are high-temperature peraluminous granites. The enrichments in LREEs and depletions in HREEs and Y(Yb) indicate the existence of garnet, while the strong depletion in Sr and Eu indicates the existence of plagioclase in the residual phase. Xiong et al. (2005) showed that when the melting pressure exceeds a certain range, the contents of some trace elements in the melt are changed, such as the pressure-sensitive elements Y and Yb (mainly related to garnet). Zhang et al. (2006) also noted that the formation pressure can be inferred from the composition of the residual phase and the contents of Sr and Yb in granite. The low Sr content (16.8–101, average 58.9 < 400, in ppm) and low Yb content (0.98–2.19, average 1.59 < 2, in ppm) of volcaniclastic rocks indicate that they were formed under moderate pressure (0.8–1.5 GPa), with garnet + plagioclase as the main residual phase, and the corresponding crustal thickness was approximately 40–50 km (Zhang et al. 2010c). Therefore, in this paper, we infer that the Jurassic volcaniclastic rocks are medium-pressure and high-temperature granites, which may have formed against the background of crustal thickening in a syncollisional stage or the early postcollisional stage. During this process, muddy sedimentary rocks were brought into the deep crust and heated (including radioactive decay and heat generation), followed by a rapid isostatic uplift of the crust, which caused the partial melting of the sedimentary rocks due to decompression.

In addition to volcaniclastic rocks, the Sr–Nd isotope compositions of the quartz monzonite porphyry and quartz porphyry are similar to those of Permian magmatic rocks [ε_Nd(t) = −8.36 to −6.65; (⁸⁷Sr/⁸⁶Sr)_i = 0.705422–0.707590], indicating source areas similar to those of the Late Permian monzonite. Quartz monzonite porphyry also has low Sr contents (116–246, average 209 < 400, in ppm) and low Yb contents (0.94–1.21, average 1.06 < 2, in ppm). The enrichment in LREEs and depletion in HREEs and Y(Yb) indicate the existence of garnet, while the depression between middle rare earth elements (MREEs) and HREEs in the REE distribution curve and Ho_N < Yb_N (MREE depletion) suggests that hornblende may have been an important residual phase. The positive Eu and Ba anomalies may indicate that there was no residual plagioclase in the source area. Therefore, the quartz monzonite porphyry may have been formed by partial melting of the mantle metasomatized by subducted sediment, and the residual phases in the source area were mainly garnet and hornblende but not plagioclase.

The quartz porphyry is characterized by high silicon and alkalis, low calcium, magnesium and iron, and significant depletion of Ba, Sr, P, Ti, and Eu, which may indicate a high degree of differentiation and evolution. However, due to the similar characteristics of highly evolved I- and S-type granites (SiO₂ > 74 wt%) and A-type granites, accurately distinguishing their types by chemical composition is particularly important. Quartz porphyry has a relatively low FeO*/MgO ratio (4.06–7.44), despite its high alkali content, which is different from that of A-type granite (FeO*/MgO > 10, Whalen et al. 1987). The Ga content of quartz porphyry is relatively low, with a 10,000*Ga/Al ratio of 2.17–2.49 (average 2.28), which is lower than that of A-type granite (Whalen et al. 1987). The contents of Zr, Nb, Ce, and Y are relatively low, with the content of Zr + Nb + Ce + Y from 166 to 186 ppm, which is significantly lower than the lower limit for A-type granite (350 ppm) recommended by Whalen et al. (1987). In the discriminant diagrams based on Ga/Al proposed by Whalen et al. (1987) (Fig. 11a, b), all the samples plot in non-A-type granite fields. In the discriminant diagrams for A-type and differentiated granites, the samples are mainly located in highly differentiated granite fields (Fig. 11c, d). Experiments by Clemens et al. (1986) and Watson and Harrison (1983) show that the distribution coefficient of zirconium D^Zr zircon/melt is a function of the principal component parameter M = (Na + K + 2Ca)/(Si*Al) and the melt temperature. Given the M value and Zr content of the whole rock, the saturation temperature of zircon (close to the liquidus temperature) can be calculated, indicating the minimum initial temperature of magma in the source area. The saturation temperature of quartz porphyry is 757–818 °C, with an average of 780 °C, which is lower than the average value of 839 °C (n = 55) for aluminous A-type granite in the Lachlan fold belt, Australia (King et al. 1997). Therefore, in this paper, we consider that the quartz porphyry is not an A-type granite but a highly differentiated granite. The significant depletions in Sr and Eu are due to the crystallization of plagioclase. The obvious Ba anomaly is more closely related to the crystallization of potassium feldspar (Han et al. 1997). The obvious P depletion may be caused by apatite separation and crystallization, while the Ti depletion may be caused by ilmenite, titanite and rutile separation, and crystallization, but rutile has a high distribution coefficient for Nb and Ta, and its separation and crystallization can also result in Nb and Ta depletion. However, the Nb and Ta depletions are not obvious on the trace element spidergrams, so the Ti depletion may have been caused by the separation and crystallization of ilmenite and rutile.

Fig. 11

Ce- and Zr-10,000*Ga/Al (a, b) and FeO*/MgO- and (K₂O + Na₂O)/CaO–Zr + Nb + Ce + Y (c, d) diagrams for the quartz porphyry of the Changhanboluo mining area (after Whalen et al. 1987). I S M, I-, S-, and M-type granites, respectively; FG, fractionated I-type granites; OGT, unfractionated I-, S- and M-type granites

Therefore, in this paper, we infer that the quartz monzonite porphyry and quartz porphyry were formed by partial melting of a mantle source area metasomatized by subducted sediments, but the quartz porphyry experienced a high degree of differentiation and evolution. The Jurassic volcaniclastic rocks were formed in a medium-pressure and high-temperature environment, possibly against the background of crustal thickening in a syncollisional stage and the early postcollisional stage. During this process, muddy sedimentary rocks were brought into the deep crust and heated, followed by the rapid isostatic uplift of the crust, which caused partial melting of the sedimentary rocks.

6.1.4 Early Cretaceous magmas

Cretaceous magmatism is mainly manifested by the latest andesitic porphyrite dike intruded in quartz porphyry. The andesitic porphyrite shows a serious carbonation alteration with a high loss on ignition (7.64–9.68 wt%). The major elements, such as K₂O and Na₂O, are greatly affected by the alteration; therefore, the REEs and trace elements are mainly used for analysis in this paper.

The andesitic porphyrite is enriched in LILEs (such as Rb, U, and K) and LREEs but depleted in HREEs and HFSEs (such as Nb, Ta, and Ti). The main formation mechanisms for these features include the partial melting of crustal materials, partial melting of mantle-derived materials accompanied by strong crustal contamination, and partial melting of enriched mantle formed by fluid/melt metasomatism in the subduction zone.

The low SiO₂ content (average 57.36 wt%) of andesitic porphyrite suggests that it cannot have been derived from the partial melting of crustal materials. Nb/U, Ta/U, and Ce/Pb can usually be used as indicators to judge crustal contamination. The Nb/U, Ta/U, and Ce/Pb ratios of the samples are 6.2, 0.3 and 4.6, respectively, which are lower than the crustal values (Nb/U = 12.1, Ta/U = 1.1, and Ce/Pb = 4.1) (Taylor and McLennan 1995) (except that the Ce/Pb values of andesitic porphyrite are higher than the Ce/Pb crustal values). Therefore, large-scale crustal contamination may not have occurred during the formation of the andesitic porphyrite. However, considering the ancient zircons found in the andesitic porphyrite, we consider that these zircons were captured during magma ascent. The formation of andesitic porphyrite involved contamination by crustal materials, but the degree of contamination was low. The andesitic porphyrite shows REE characteristics, trace element patterns and isotope geochemistry similar to those of the Permian monzonite, which indicates that the former may have also been formed by the partial melting of mantle metasomatized by subducted sediment.

6.2 Tectonic evolution

Tectonic evolution can be recorded by the compositions of associated magmatic rocks, so we can reconstruct the tectonic history of the study area and provide support for tectonic evolution in the northern margin of the NCC by the rock assemblages outlined above.

6.2.1 Late Silurian and Middle Permian subduction

As mentioned above, both Late Silurian gabbroic diorite and Middle Permian monzonite have geochemical characteristics of volcanic arc magmatic rocks, indicating a subduction background. The points are located in the volcanic arc field in the discrimination of trace element maps (Fig. 12a, b), and in the La/Nb-Ba/Nb diagram, the samples are also located in the arc volcanic field (Fig. 12c). The HFSE ratios with Yb normalization can reduce the effects of partial melting and fractional crystallization, so these ratios should closely reflect the source region characteristics. The Th/Yb-Nb/Yb diagram distinguishes continental arc magmas from oceanic arc magmas (Pearce and Peate 1995). In Fig. 12d, the gabbroic diorite and monzonite samples are all located in the continental arc area. Therefore, the Late Silurian and Permian magmatic rocks in this area are most likely to have formed against a continental arc background, which was related to the subduction of the Paleo-Asian Ocean Plate beneath the North China Plate.

Fig. 12

Discrimination diagrams of the tectonic setting for the gabbroic diorite and monzonite. a Diagram of Yb versus Ta (after Pearce et al. 1984); b diagram of Y + Nb versus Rb (after Pearce 1996); c diagram of Ba/Nb versus La/Nb (after Jahn et al. 1999; Fan et al. 2004); d diagram of Nb/Yb versus Th/Yb (after Pearce and Peate 1995). Abbreviations: syn-COLG, syncollisional granites; VAG, volcanic arc granites; ORG, oceanic ridge granites; WPG, within-plate granites; Dupal OIB, ocean island basalts with Dupal anomaly

The northern margin of the NCC underwent a long geological evolution process from the Paleozoic to Mesozoic (Meng 2003; Xiao et al. 2003; Liu et al. 2005). Late Paleozoic granitoids are also abundant along the northern margin of the NCC. They range from gabbro, diorite, and granodiorite to tonalite in composition and constitute a plutonic complex that was continuously emplaced from the Late Carboniferous to the Early Permian (324–274 Ma; Zhang et al. 2007a, 2009a, b, c). In the Early Paleozoic, the Paleo-Asian Ocean subducted beneath the North China Block and formed the Early Paleozoic arc accretion series (490–446 Ma) (Xiao et al. 2003; Windley et al. 2007). With the continued subduction of the Paleo-Asian Ocean, the Dongzi gabbroic diorite was formed in the subduction setting. During the Middle-Late Paleozoic, Andean-type continental arc magmatism (390–270 Ma; Zhang et al. 2007b, c, 2009a) developed on the northern margin of the NCC. The monzonite was formed against the Permian subduction background, and the source may have been a mantle zone metasomatized by melt released from the subducting plate. The Permian mineralization in the northern margin of the NCC formed deposits such as Haolibao, Chehugou, Hadamiao, and Bilihe. Among them, the Re–Os isochron age of molybdenite in the Haolibao deposit is 265 Ma (Zeng et al. 2013); the Rb–Sr isochron age of veinlet chalcopyrite in the Chehugou deposit is 256 Ma (Wan et al. 2009); and the Re–Os weighted mean age and isochron age of molybdenite are 258 and 250 Ma, respectively (Liu et al. 2010; Meng et al. 2013). The zircon LA-ICP-MS U–Pb age of the ore-forming porphyry in the Hadamiao gold deposit is 271 Ma (Lu et al. 2009); the Re–Os isochron age of molybdenite in the Bilihe gold deposit is 273 Ma (Qing et al. 2011), and the zircon U–Pb isotope age of the granodiorite porphyry-monzogranite porphyry is 284–280 Ma (Lu et al. 2012). These Late Paleozoic metallogenic events provide evidence for the Late Paleozoic convergent plate tectonic system in the northern margin of the NCC and establish a theoretical foundation for the exploration of Late Paleozoic deposits.

6.2.2 Late Jurassic collision

Late Jurassic magmatism mainly formed ignimbrites, breccia tuffs, quartz monzonite porphyry veins and quartz porphyry veins in this area. The rocks may have been formed against the background of crustal thickening in a syncollisional stage or an early postcollisional stage. Li et al. (2007) noted that the collision between the North China Plate and the Siberian Plate occurred in the Middle and Late Permian and probably lasted to the Middle Triassic, which may have been a relatively long geological process. The mingling of all kinds of ancient animals in the middle Permian marked the beginning of the collision. The appearance of molasse deposits in the Early-Middle Triassic and the occurrence of magmatism in the Middle Triassic indicated the occurrence and end of the collision, respectively. Zhang et al. (2009a) also noted that the final collision between the North China Plate and the Siberian Plate may have occurred between 270 and 250 Ma. Wan et al. (2009) claimed that the closure of the Paleo-Asian Ocean north of the Chifeng area may have occurred before 260 Ma, possibly during the Middle Permian. Therefore, we infer that the northern margin of the NCC may have been in a post-collisional setting during the Late Jurassic (153–158 Ma), with no obvious crustal thinning. The extension stage and compression-extension transition stage of post-collision are the peak periods of large-scale mineralization (Wang and Xu 2006). Since the Early and Middle Jurassic, the northern margin of the NCC has been in the transitional system between the Paleo-Asian Ocean tectonic domain and the Paleo-Pacific Ocean tectonic domain. Under the influence of multidirectional tectonic stresses, strong magmatic activity developed, accompanied by a large number of polymetallic deposits (Wu and Sun 1999; Mao et al. 2005; Zhang et al. 2010b), such as the Jiguanshan molybdenum deposit and Nianzigou molybdenum deposit. Among them, the Re–Os isochron age of the Jiguanshan porphyry molybdenum deposit is 155 Ma (Wu et al. 2011), while the Re–Os isochron age of the Nianzigou quartz vein molybdenum mineralized deposit is 154 Ma (Zhang et al. 2009d), which means that both formed in the postorogenic extensional tectonic setting (Chen et al. 2008). Most deposits in the Shaolanghe Pb–Zn–Ag metallogenic belt were formed in the Jurassic, and the formation of deposits may have been related to magmatism in this period.

6.2.3 Early Cretaceous extension

The regional tectonic system in the northern margin of the NCC has undergone an important transformation (Zhao et al. 2004) from an E-W-trending tectonic system in the Early Mesozoic to a NE- to NNE-trending system in the Late Mesozoic-Early Cenozoic. For example, Ren et al. (1997) found that the E-W-trending Middle-Late Jurassic Tiaoijishan Formation volcanic rocks and the Tuchengzi Formation red strata discovered in northern Hebei and southern Inner Mongolia are overlapped by the NE- to NNE-trending Early Cretaceous Zhangjiakou Formation and Yixian Formation volcanic rocks, which suggests that the Jurassic-Cretaceous was an important transitional period for the dynamic structure in the eastern part of the NCC. In addition, many NE- to NNE-trending faults have been formed. Li (2013) believed that the time of transition from an E-W to a NE-NNE tectonic system in eastern China was 143–120 Ma. The Cretaceous andesitic porphyrite may have been formed against the background of lithospheric extension and thinning. The upwelling of asthenospheric mantle and the increase in the geothermal gradient resulted in the partial melting of the low-temperature melting components (water and LILE-enriched metasomatic mantle) in the upper lithospheric mantle. During the Cretaceous Period, the northern margin of the NCC underwent intense mineralization and formed deposits such as Xiaodonggou and Hongshanzi. The Xiaodonggou deposit formed at 135 Ma, while the Hongshanzi deposit formed at 132 Ma (Nie et al. 2007). In the northern part of the Xilamulun fault, many Mo and Sn polymetallic deposits have also been found, such as Banlashan, Haolibao, Aolunhua, Huanggang, and Anle. Among these deposits, the age of metallogenic granodiorite in the Banlaishan deposit is 132–133 Ma (Zeng and Liu 2010; Zhang et al. 2010d), the molybdenite Re–Os isochron age of the Aolunhua deposit is 129–132 Ma (Shu et al. 2009; Ma et al. 2009), the molybdenite Re–Os age of the Huanggang deposit is 135 Ma (Zhou et al. 2010), and the age of the Anle deposit is 133 Ma (Wang et al. 2001).

6.3 The relationship with metallogenesis

The Dongzi and Changhanboluo Pb–Zn deposits are located south of the Shaolanghe fault. The ore bodies are obviously controlled by the faults and mainly occur in veins. The ore types are mainly massive, banded, and breccia. The metallic minerals mainly consist of pyrite, galena, and sphalerite, followed by chalcopyrite, freibergite, arsenopyrite, etc. Nonmetallic minerals mainly include quartz, calcite, muscovite, and feldspar. Wall rock alteration mainly developed on both sides of the ore veins and mostly consists of silicification, sericitization, carbonation, chloritization, and other medium- to low-temperature hydrothermal alterations. Previous studies on fluid inclusions of the two deposits show that the ore-forming fluids were characterized by a low temperature and low salinity (Xie et al. 2016a; Wang et al. 2017). The above geological characteristics show that the Dongzi and Changhanboluo deposits are hydrothermal vein deposits related to magmatism. According to the above studies, at least four stages of magmatism occurred in this area: Late Silurian (~ 420 Ma), Middle Permian (~ 274 Ma), Late Jurassic (153–158 Ma), and Early Cretaceous (~ 127 Ma). Which stage of magmatism was related to the formation of the deposits?

From the geological characteristics, the gabbroic diorite and monzonite in the Dongzi–Changhanboluo mining area are the main host rocks of the ore bodies, but their alteration and mineralization are weak, so they cannot possibly be the ore-forming geological bodies of the deposits. The Early Cretaceous andesitic porphyrite is clearly younger than the deposits and can be seen to cut through ore bodies; therefore, it is not related to mineralization. The Jurassic quartz porphyry is the main host rock in the Changhanboluo mining area, with sericitization and pyritization alteration, which may be closely related to the formation of the deposit. However, in the Dongzi deposit, the main host rock is diorite and andesite with weak alteration. We did not find quartz porphyry with noticeable alteration. Therefore, we speculated that the ore-forming geological body of the Dongzi deposit should be remote Jurassic magmatic rocks. Based on combining this information with the geochronology studies, the ore-forming ages of the Dongzi and Changhanboluo deposits are 153 Ma and 157 Ma (Chen et al. 2003; Xie et al. 2016a), respectively. The gabbroic diorite (420 Ma) and monzonite (274 Ma) were obviously earlier than the ore-forming age of the Dongzi and Changhanboluo deposits, while the Early Cretaceous andesite porphyrite (127 Ma) was obviously later than the ore bodies. Therefore, according to the diagenetic and ore-forming ages, the formation of the deposits was mainly related to the Jurassic magmatism (153–158 Ma). In addition, from the analysis of Pb isotopes (Table 4), the similar Pb isotope ratios of ores from the Dongzi and Changhanboluo deposits indicate that the two deposits have the same metallogenic magma source. The Pb isotope ratios show (²⁰⁶Pb/²⁰⁴Pb)_i of 17.181–17.374, (²⁰⁷Pb/²⁰⁴Pb)_i of 15.384–15.476 and (²⁰⁸Pb/²⁰⁴Pb)_i of 37.425–37.823, which are obviously higher than those of the Dongzi gabbroic diorite and the monzonite batholith [(²⁰⁶Pb/²⁰⁴Pb)_i of 16.128–16.920, (²⁰⁷Pb/²⁰⁴Pb)_i of 15.333–15.382, and (²⁰⁸Pb/²⁰⁴Pb)_i of 36.763–36.923]. The Pb isotope ratios of the ores are similar to those of the Jurassic magmas, such as quartz porphyry, quartz monzonite porphyry and ignimbrite. In Fig. 13, we can see that the Pb isotope ratios of the ores are plotted in the same range as that of quartz porphyry. Therefore, from the comprehensive analysis of geological characteristics, diagenetic and ore-forming ages and Pb isotopes, the formation of ore bodies in the Dongzi–Changhanboluo ore district was closely related to the Jurassic quartz porphyry.

Table 4 Lead isotopic analyses for the magmatic rocks and ores in Dongzi–Changhanboluo ore district

Fig. 13

The (²⁰⁶Pb/²⁰⁴Pb)_i versus (²⁰⁷Pb/²⁰⁴Pb)_i (a) and (²⁰⁸Pb/²⁰⁴Pb)_i (b) diagrams for the ores and magmas in the Dongzi–Changhanboluo ore district

7 Conclusions

Based on the zircon U–Pb ages and the element and isotope geochemical data presented above, we draw the following conclusions:

• 1. The zircon U–Pb dating identifies four stages of magmatism in the Dongzi–Changhanboluo ore district: Late Silurian gabbro diorite (~ 420 Ma), Middle Permian quartz monzonite (~ 274 Ma), Late Jurassic quartz porphyry, ignimbrite and breccia tuff (153–158 Ma), and Early Cretaceous andesitic porphyrite (~ 127 Ma).

• 2. The gabbroic diorite may have been sourced from EM1-type mantle. The monzonite source may have been a mantle zone of metasomatism from fluid/melt released by the subducting plate. The Jurassic volcaniclastic rocks were sourced from the partial melting of the sedimentary rocks. Quartz monzonite porphyry and quartz porphyry formed from the partial melting of a mantle source area metasomatized by subducted sediments, but the quartz porphyry experienced high-degree of differentiation and evolution. The andesite porphyrite has characteristics similar to those of the Permian monzonite sourced from a mantle zone metasomatized by subducted sediment.

• 3. The late Silurian and Permian magmatic rocks in this area most likely formed against a continental margin arc background, which was related to the subduction of the Paleo-Asian Ocean Plate beneath the North China Plate. The Late Jurassic magmatic rocks suggest that the northern margin of the NCC may have been in a post-collisional setting during the Late Jurassic, with no obvious crustal thinning. The Cretaceous andesite porphyrite may have been formed against the background of lithospheric extension and thinning.

• 4. According to the comprehensive analysis of geological characteristics, diagenetic and metallogenic epochs and Pb isotope data, the formation of ore bodies in the Dongzi–Changhanboluo ore district was closely related to the Jurassic quartz porphyry.