Geochronological, Geochemical and Sr-Nd-Pb-Hf Isotopic Constraints on the Petrogenesis of Pyroxene Diorites in the Sanchahe Iron-Gold Deposit, Western Shandong

: The Sanchahe iron-gold deposit in western Shandong province is located in the southeast of the North China Craton. The skarn type of iron-gold deposit (without copper) is located in the contact belt between pyroxene diorites and Ordovician limestone xenoliths. The zircon U-Pb age of pyroxene diorites is 138.4 ± 1.2 Ma belonging to the Early Cretaceous. Pyroxene diorites (Rittmann index σ = 2.2~2.5) belong to the calc-alkaline series and the SiO 2 content is 55~59%. Furthermore, pyroxene diorites are characterized by high Mg and Na contents and high contents of Sc, Cr, Co, and Ni, indicating that it has a geochemical afﬁnity with mantle-derived magma. All of the samples are enriched in light rare earth elements (LREEs) and depleted in heavy rare earth elements (HREEs), with a weak Eu-negative anomaly. They are also enriched in Cs, Ba and Sr and depleted in high ﬁeld strength elements (HFSEs) (Nb, Ta, Zr), indicating the involvement of some crustal materials. The whole-rock ( 87 Sr/ 86 Sr) I values are 0.70824~0.70891, and the ε Nd ( t ) values are from − 8.1 to − 12.5. Two-stage model ages of T DM2 are 1591~1928 Ma. The ε Hf(t) values range from − 7.0 to − 17.4, and the two-stage model ages of T DM2 range from 1629 to 2297 Ma, indicating that the magma was derived from Mesoproterozoic to Neoarchean crusts. Combined with Pb isotopic characteristics, pyroxene diorites may be generated by the partial melting of the enriched lithospheric mantle, which had been metasomatized by the ancient lower crust materials of the North China Craton and formed in the strong extensional tectonic setting linked with lithospheric thinning, which may be caused by the rollback of the Paciﬁc plate after subduction.


Introduction
Many large gold deposits have been discovered in the Jiaodong area, which is the most important gold metallogeny area in China [1]. The Yishu fault is a boundary fault between the western Shandong province and the Jiaodong area. Considering the similar tectonic setting and contemporary magmatic events between the western Shandong province and the Jiaodong area, although the number and scale of gold deposits are far less than those of the gold deposits in the Jiaodong area, the western Shandong province has the potential for the exploration of gold deposits. After the discovery of the Guilaizhuang gold deposit The western Shandong province in the southeastern North China Craton is to the northwest of the Sulu ultrahigh-pressure (UHP) metamorphic belt and is located at the Lankao-Liaocheng fault to the west, the Qihe-Guangrao fault to the north, the Tanlu fault to the east, and the Fengpei fault to the south [5,10] (Figure 1). The study area in this paper is in the middle of western Shandong province and to the northwest of the Taiwu depression. The crystalline basement in the study area is Neoarchean granitoids of the Taishan groups, whereas the sedimentary cover includes Cambrian-Ordovician limestone, Mesozoic clastic rock, and Cenozoic sediments. The faults in the study area are mainly in the direction of NNW, NE, and EW. Furthermore, the NNW trending fault is the main fault in the region. As an important NNW fault, the Tongyedian-Sunzu fault is about 130km long and is the space and channel of the magma. It is characterized by the multi-stage movements of faults, so the magmatic activities including Neoarchean granitoids and Yanshanian granitic complexes have a close relationship with the Tongyedian-Sunzu fault. As part of Yanshannian magmatic activities, Tietonggou granitic complexes with 7.5 km 2 of exposed area are near the Tietonggou area in the southeast of Laiwu and are at the junction of the Tongyedian-Sunzu fault and Qingnigou fault ( Figure 2). According to the field geological investigation, Tietonggou granitic complexes are composed of monzogranites, quartz diorite porphyrite, hornblende porphyrite, pyroxene diorite, hornblende, and olivine norite gabbro.  [11], (b) modified after [12]).  [11], (b) modified after [12]).

Deposit Characteristics
The Sanchahe deposit is in the north of the Tietonggou granitic complexes, and the ore deposit is under the Quaternary (Figure 2). The pyroxene diorites have a close relationship with the mineralization, and there is skarn metasomatism between the pyroxene

Deposit Characteristics
The Sanchahe deposit is in the north of the Tietonggou granitic complexes, and the ore deposit is under the Quaternary (Figure 2). The pyroxene diorites have a close relationship with the mineralization, and there is skarn metasomatism between the pyroxene diorites and marble. The Fe-Au skarn deposit is composed of Au, Fe, Ag, and S. The orebody is located near the contact zone between magmatic rocks and stratum, whereas the mineralization are mainly distributed in the skarn. The shape of the orebody is stably layered or lenticular, while the local shape of the orebody is consistent with the edge shape of xenoliths. Furthermore, the gold orebody-wrapped iron orebody is thick in the middle and annihilated to both sides ( Figure 3). The orebody inclines to the east with an average dip angle of 33 • . The average thickness of the gold orebody with an average grade of 4.54 ppm is about 4.65 m, whereas the average thickness of the iron orebody is 2.81 m. The ore types are composed of magnetically mineralized skarn and pyritized magnetically mineralized skarn. The ore minerals include magnetite, pyrite, and gold, whereas the gangue minerals are mainly diopside, tremolite, serpentine, biotite, etc. (Figure 4).

Analytical Methods
The zircon U-Pb dating and Lu-Hf isotope samples were taken from the ZK0003 borehole, the whole-rock major and trace elements, and the Sr-Nd-Pb isotope for test samples were taken from the ZK803, ZK004, and ZK005 boreholes. The rock is grayblack with granular structure, plagioclase is gray-white, accounting for about 70%, and hornblende is gray-black, accounting for about 30%; (b) Plagioclase (Pl, 55-60%): Polysynthetic twin are developed, and the particle size is generally 0.6-1.0 mm. hornblende (Hbl, 30-35%): there are two groups of oblique cleavage, and the particle size is generally 0.4-1.2 mm. Augite (Aug, 0.5%), short diameter, and the particle size is generally 0.1-0.3 mm.

Analytical Methods
The zircon U-Pb dating and Lu-Hf isotope samples were taken from the ZK0003 borehole, the whole-rock major and trace elements, and the Sr-Nd-Pb isotope for test samples were taken from the ZK803, ZK004, and ZK005 boreholes.

Zircon U-Pb Dating and Hf Isotope
The fresh samples without xenoliths (and after removing the weathered surface) were selected for crushing. After the pyroxene diorite was crushed manually, the zircons were separated by gravity separation and electromagnetic separation, and the zircon particles with better crystal shape and transparency were selected under binoculars. Then, they were glued on the glass slide with double-sided tape and collected. The collected zircons were put on the target ring, and after the epoxy resin was injected, it was cooled and polished to expose the zircon surface for the acquisition of transmitted light, reflected light, and cathodoluminescence (CL) images. Zircon U-Pb dating was performed using a combination of NWR 193 nm ArF excimer laser ablation system (Elemental Scientific Lasers, Omaha, NE, USA) and ThermoiCap-RQ quadrupole ICP-MS (Thermo Fisher Scientific, Bremen, Germany). In the experiment, helium was used as the carrier gas for the ablation material, and the laser frequency was 8 Hz, the energy was 5 J/cm 2 , the beam spot diameter was 32 µm, and the laser ablation time was 40 s. The test results were processed offline by ICPMSDataCal software (version 10.9, China University of Geosciences, Wuhan, Hubei, China). International standard zircon 91500 was used as the external standard to correct instrumental mass discrimination and elemental fractionation during the ablation and zircon PL was treated as quality control for geochronology (337.13 ± 0.37 Ma). The detailed experimental analysis steps and data processing methods can be found in references [13,14]. Sample processing and analytical testing were performed at Guangzhou Tuoyan Testing Technology Co., Ltd. (Guangzhou, Guangdong, China).
The zircon in situ Lu-Hf isotope test location is the effective point of the previous U-Pb dating. The Neptune multi-receiver inductively coupled plasma mass spectrometer (Thermo Fisher Scientific, Bremen, Germany) equipped with a 193 nm laser was used for the determination (Elemental Scientific Lasers, Omaha, NE, USA). For the detailed test process and analysis steps, please refer to reference [15]. During the analysis, a laser with a frequency of 9 Hz and an energy of 3.5 J/cm 2 ablated the zircons. The ablation diameter was 50 µm, and the ablation time was 40 s. During the testing process, for every five samples of zircon, one standard zircon (including GJ-1, 91500, Plešovice, Mud Tank, Penglai) was tested in turn to check the data quality of the zircon-Hf isotope ratio. For the calculation formula and the Hf isotopic method, please refer to reference [16]. Sample processing and analytical testing were performed at Guangzhou Tuoyan Testing Technology Co., Ltd.

Geochemistry
Sample processing and the major and trace elements analysis were performed at Guangzhou Tuoyan Testing Technology Co., Ltd. Major elements were determined using X-ray fluorescence spectrometry (XRF, M4 Tornado plus, Bruker, Germany). Trace and rare earth elements were analyzed by an inductively coupled plasma mass spectrometer (ICP-MS). The analytical precision and accuracy of major elements were better than 5%, and the analytical precision and accuracy of trace and rare earth elements were 5% to 10%. The detailed analysis steps of the experiment can be found in reference [17].
The whole-rock Sr-Nd-Pb isotopic analysis was completed at Beijing Zirconium Linghang Technology Co., Ltd. (Beijing, China). The model of the Sr-Nd isotope testing instrument was the thermal ionization mass spectrometer (TIMS-ICPMS). First, we weighed 50-100 mg of rock powder crushed to 200 mesh and placed it in a sample dissolving bomb, and dissolved it with 1.5 mL HF + 1.5 mL HNO 3 at 195 • C for 24 h. Then, we evaporated it to dryness at 115 • C, and added 1 mL HCl. Finally, AG50X8 Rb and Sr were separated by cation exchange resin, and Sm and Nd were separated by HDEHP leaching resin. The measured data of the samples were normalized with 86 Sr/ 88 Sr = 0.1194 and 146 Nd/ 144 Nd = 0.7219. The detailed experimental procedures and analysis methods can be found in the literature [17]. For the Pb isotope analysis, we first accurately weighed an appropriate amount (about 100 mg) of the sample into a polytetrafluoroethylene sample dissolution tank, added 2.5 mL of HF, 0.5 mL of HNO 3 , and 0.15 mL of HClO 4 , and placed it on an electric hot plate at 150 • C for one week for dissolving the sample. After the sample was dissolved, we evaporated the sample solution to dryness, dissolved the sample with 2 mL 3 M HNO 3 , centrifuged it for resin separation, used Sr special resin (2 mL) to separate and purify Pb, and used a 3 M HNO 3 and 6 N HCl acid process to separate and purify the Pb sample. During whole process of the lead, the background was less than 200 pg. The Pb isotope ratio was determined by a Triton thermal ionization mass spectrometer (TIMS) (Thermo Fisher Scientific, Bremen, Germany), and the lead isotope was measured using a silica gel emitter and a rhenium metal band, and the measurement temperature was about 1300 • C. The lead's standard material NBS981 was used to monitor the instrument status, and the international standard rock sample BCR-2 (basalt) was used to monitor the separation process during the experiment. The detailed analysis steps of the experiment can be found in reference [18].

Zircon U-Pb Ages
The CL images of zircons from the pyroxene diorites in the Sanchahe area are shown in Figure 6. The sizes of zircon grains are from 70 µm to 270 µm, whereas the shapes of zircons are mainly long or short and columnar. At the same time, the zircons show typical magmatic zonal structures [19]. The contents of Th and U are from 188 ppm to 633 ppm, and from 176 ppm to 456 ppm, respectively. The ratios of Th/U are from 1.0 to 1.4, indicating the features of magmatic zircons [20]. Twenty-five zircons from sample ZK0003B4 were selected for dating, and the data are listed in Table 1. The concordance degree of seven grains was below 90%, and the other eighteen grains yielded an average age of 138.4 ± 1.2 Ma (MSWD = 1.1, n = 18) (Figure 7). This Cretaceous age could represent the emplacement age of the gabbro diorites in the Sanchahe area. tank, added 2.5 mL of HF, 0.5 mL of HNO3, and 0.15 mL of HClO4, and placed it on an electric hot plate at 150 °C for one week for dissolving the sample. After the sample was dissolved, we evaporated the sample solution to dryness, dissolved the sample with 2 mL 3 M HNO3, centrifuged it for resin separation, used Sr special resin (2 mL) to separate and purify Pb, and used a 3 M HNO3 and 6 N HCl acid process to separate and purify the Pb sample. During whole process of the lead, the background was less than 200 pg. The Pb isotope ratio was determined by a Triton thermal ionization mass spectrometer (TIMS) (Thermo Fisher Scientific, Bremen, Germany), and the lead isotope was measured using a silica gel emitter and a rhenium metal band, and the measurement temperature was about 1300 °C. The lead's standard material NBS981 was used to monitor the instrument status, and the international standard rock sample BCR-2 (basalt) was used to monitor the separation process during the experiment. The detailed analysis steps of the experiment can be found in reference [18].

Zircon U-Pb Ages
The CL images of zircons from the pyroxene diorites in the Sanchahe area are shown in Figure 6. The sizes of zircon grains are from 70 μm to 270 μm, whereas the shapes of zircons are mainly long or short and columnar. At the same time, the zircons show typical magmatic zonal structures [19]. The contents of Th and U are from 188 ppm to 633 ppm, and from 176 ppm to 456 ppm, respectively. The ratios of Th/U are from 1.0 to 1.4, indicating the features of magmatic zircons [20]. Twenty-five zircons from sample ZK0003B4 were selected for dating, and the data are listed in Table 1. The concordance degree of seven grains was below 90%, and the other eighteen grains yielded an average age of 138.4 ± 1.2 Ma (MSWD = 1.1, n = 18) (Figure 7). This Cretaceous age could represent the emplacement age of the gabbro diorites in the Sanchahe area.   Note: The bold is the data of zircon samples with low concordance.

Major and Trace Elements
The samples ZK803, ZK004, and ZK005 were selected for major and trace elements analysis in the study. The data are provided in Table 2 and plotted in Figures 8 and 9.

Major Elements
The geochemical data of pyroxene diorites in the Sanchahe area are listed in Table 2. Although some minerals in the samples were altered, the loss (0.9~1.2%, <5%) indicated that the degree of weathering alteration was relatively low and had little effect on the determination of major elements. The samples with SiO 2 contents of 55.3~58.7% had intermediate magmatic characteristics and the contents of Al 2 O 3 , Na 2 O, K 2 O and Fe 2 O 3 were 15%, from 3.4% to 3.5%, from 2.1% to 2.4%, and from 7.0% to 8.9%, respectively. The contents of MgO were from 5.9% to 6.8% and the values of Mg # were 60% to 63%, indicating the features of high Mg diorites. Meanwhile, the total alkali contents (K 2 O + Na 2 O) were 5.5% to 5.9%. The ratios of Na 2 O/K 2 O were from 1.5% to 1.7%, and the Rittman indexes were from 2.2% to 2.5%, indicating the evolution trend of sub-alkaline series magma. The values of A/CNK were 0.8. Thus, all of the samples were weak peraluminous. The samples are plotted in the field of the diorites in the diagram of total alkali silicon (TAS) (Figure 8a), whereas all of the samples are plotted in the high-K calc-alkaline series field in the SiO 2 -K 2 O diagram (Figure 8b). the samples hada weak negative anomaly of Eu. In the primitive mantle normalized spi-der diagram (Figure 9b), the samples were enriched in LILEs (Rb, Cs, Ba, Sr), and depleted in HFSEs (Nb, Ta, Zr). The samples had high contents of Sr and Yb: w(Sr) = 668~853 ppm, w(Yb) = 1.3~1.4 ppm. The samples also had high contents of Sc, Cr, Co, and Ni: w(Sc) = 19~24 ppm, w(Cr) = 289~334 ppm, w(Co) = 24~28 ppm, w(Ni) = 93~104 ppm. Furthermore, the values of Nb/Ta and Zr/Hf were from 15 to 17and from 38 to 87, respectively.

Trace Elements
The contents of REE from the samples were from 118 ppm to 128 ppm. Furthermore, the contents of LREEs were from 91 ppm to 102 ppm, and the contents of HREEs were from 26 ppm to 28

Sr-Nd-Pb Isotopic Features
The samples ZK002, ZK004 and ZK005 were selected for Sr-Nd-Pb isotope analysis in this study. The whole rock Sr-Nd-Pb isotopic compositions of those samples are listed in

Zircon Hf Isotopes
The sample ZK003 was selected for zircon Hf isotope analysis in the study. The locations of thirteen dating spots were the same or near the zircon U-Pb dating locations ( Figure 6). The values of ε Hf (t) were calculated based on the weighted average age of the samples. The data results have been listed in Table 4  6. Discussion

Geochronology
The Sanchahe Fe-Au deposit is a typical skarn deposit. Its mineralization is likely linked with the pyroxene diorites in the Sanchahe area. However, the chemical reaction strongly occurred between the pluton and surrounding rocks and promoted the exchange of Si, Al, Ca and Mg during the forming of skarn deposits. Thus, the forming age of the pluton could probably represent the forming age of the deposit. The hidden pluton of Sanchahe is located to the south of the Tietonggou complex pluton and is considered part of the Tietonggou complex pluton. However, many studies have been carried out on the geochronology of the Tietonggou complex. Considering that the Jurassic Santai group was intruded by the Tietonggou complex, the forming age of the Tietonggou complex pluton was defined as the Early Cretaceous. Xu et al. [28] reported the biotite 40 Ar/ 39 Ar age of 133 Ma from pyroxene diorites. This age is consistent with the biotite 40 Ar/ 39 Ar age from the diorites reported by Wang et al. [29]. In addition, Yang et al. [5] reported the Early Cretaceous weighted mean ages (~134 Ma) of zircons with the method of LA-ICP-MS. The LA-ICP-MS zircon U-Pb dating of Sanchahe pyroxene diorites in Laiwu has been carried out in this paper. Our analyzed zircon grains are characterized by long or short columnar shapes, high Th/U ratios (1.0 to 1.4), low LREE/HREE ratios (>3), and the weak negative anomaly of Eu, indicating a magmatic origin. The sample yields a weighted mean age of 138.4±1.2 Ma, indicating the emplacement age of Sanchahe pluton is the Early Cretaceous. Considering the close temporal and spatial distribution relationships between the pluton and the skarn deposits, the forming age of Fe-Au skarn deposits is probably a little later than the forming age of the emplacement age of Sanchahe pluton. However, the emplacement age in this paper is a little older but close to the previously reported ages and is consistent with the ages reported in Jinan, Laiwu, Jinling, Tiezhai, and Yiyuan [30][31][32][33][34]. Based on the above chronological data, it is deemed that the emplacement age of the pluton is consistent with the age of large-scale magmatic events under the lithosphere thinning tectonic setting of the North China Craton.

The Nature of Source and Petrogenesis
High MgO, Cr, Sc, Co, and Ni contents and high Mg # values of our analyzed samples collectively indicate the Sanchahe pyroxene diorites were primarily derived from mantle source [35]. This view is further supported by their high Nb/Ta ratios (15 to 17) and Zr/Hf ratios (38 to 87), which are strongly resembling those of mantle-derived melts (Nb/Ta: 18; Zr/Hf: 36), but distinctly higher than those of crustal-derived magmas (Nb/Ta: 11; Zr/Hf: 11) [36]. The f Lu/Hf values (−0.89~−0.97) of our analyzed samples are much lower than those of the ferromagnetic crust(−0.34) and aluminosilicate crust(−0.72) [26,27]. This feature implies the Sanchahe pyroxene diorite originated from mantle.
The relatively low εHf(t) values, LREE and LILE enrichment characteristics of the Sanchahe pluton are very similar to the crustal source material composition, indicating that the source material may be the crustal source or mantle source that underwent subduction and metasomatism enrichments. Experiments have shown that the products of the partial melting of the crust are generally (high-K) calc-alkaline series magma [37], which is consistent with the characteristics of high-K calc-alkaline series pyroxene diorites in the Sanchahe area. Therefore, the Sanchahe pyroxene diorites magma source may have the addition of crustal materials. In addition, low Sm/Nd ratios (0.2) and the geochemical features of pyroxene diorites further indicate the existence of crustal materials in the magma source or during the magmatic evolution process. Furthermore, the high initial 87 Sr/ 86 Sr values (0.70824~0.70891) and low ε Nd (t) values (from −8.1 to −12.5) indicate the existence of crustal material. The lithospheric mantle metasomatism by melting material of the subducted oceanic plate is excluded, since the latter has a relatively low initial 87 Sr/ 86 Sr value and a relatively high εNd(t) value [38]. The 176 Hf/ 177 Hf of zircons are from 0.28220 to 0.28250, whereas the values of ε Hf (t) are in the range of −7.0 and −17.4. All of the spots are plotted between the lower crust and chondrite evolution line in Figure 10a and are plotted between the crust and chondrite evolution line in Figure 10b. However, the ε Hf (t) values are not consistent with the ε Hf (t) values(ε Hf (t) > 0) [39] of the magma from the partial melting of the oceanic crust. At the same time, the Nd two-stage model ages are from 1591 to 1950 Ma, whereas the zircon Hf two-stage model ages are from 1639 Ma to 2297 Ma.
These models ages imply that the source or evolutionary process of primitive magma could involve ancient crust materials. 87 Sr/ 86 Sr value and a relatively high εNd(t) value [38]. The 176 Hf/ 177 Hf of zircons are from 0.28220 to 0.28250, whereas the values of εHf(t) are in the range of −7.0 and −17.4. All of the spots are plotted between the lower crust and chondrite evolution line in Figure 10a and are plotted between the crust and chondrite evolution line in Figure 10b. However, the εHf(t) values are not consistent with the εHf(t) values(εHf(t) > 0) [39] of the magma from the partial melting of the oceanic crust. At the same time, the Nd two-stage model ages are from 1591 to 1950 Ma, whereas the zircon Hf two-stage model ages are from 1639 Ma to 2297 Ma. These models ages imply that the source or evolutionary process of primitive magma could involve ancient crust materials.   (Figure 11). Furthermore, the Nd T DM1 ages are from 1576 Ma to 2108 Ma, similar to the ages of the lower crust in the North China Craton. Thus, we conclude that the crustal materials in the mantle were from the lower crust.  There are two main mechanisms for the transformation of the lithospheric mantle by lower crustal materials, including (1) the transformation by crustal materials occurring in the magma source area [38], and (2) the assimilation and contamination by continental crust materials during the process of magma ascending and emplacement [41,42]. Essen- Figure 11. Initial ( 87 Sr/ 86 Sr) i vs. ε Nd (t) diagram for the Sanchahe pyroxene diorite. (Tietonggou data according to reference [5]).
There are two main mechanisms for the transformation of the lithospheric mantle by lower crustal materials, including (1) the transformation by crustal materials occurring in the magma source area [38], and (2) the assimilation and contamination by continental crust materials during the process of magma ascending and emplacement [41,42]. Essentially, the former mechanism occurs in the magma source and finally forms a homogeneous chemical system. However, the other mechanism occurred at the depth of crust when mantle-derived magma rises and invaded the lower crust. As a result, the composition of magma was changed significantly [43]. Furthermore, the high 87 Sr/ 86 Sr values and low ε Nd (t) values indicate the continental crusts were mixed in the magma source. In addition, there are no obvious linear relationships between ε Nd (t) and SiO 2 and between ( 87 Sr/ 86 Sr) i and SiO 2 (Figure 12), indicating limited crustal contamination. Involvement of crustal materials in the formation of Sanchahe pyroxene diorites mainly occurred in its mantle source.  Research has shown that Rb, Sr, and Ba have great compatibility in amphibole, whereas Rb and K have greater compatibility in phlogopite [44]. Thus, the ratios of Ba/Rb and Rb/Sr could indicate whether the melt source contains amphibole (Ba/Rb > 20, Rb/Sr < 0.1) or phlogopite (Ba/Rb < 20, Rb/Sr > 0.1). The ratios of Rb/Sr and Ba/Rb are from 0.04 to 0.10, and from 12 to 22, respectively. These ratios indicate that amphiboles play an important role and phlogopite plays a secondary role during the formation process of magma melting ( Figure 14). Furthermore, the amphiboles and phlogopites provide the most incompatible elements for the refractory ancient lithospheric mantle. In a word, the pyroxene diorites are probably the result of partial melting of mantle peridotite with amphibole and phlogopite. Figure 13. Pb isotopic compositions of the Sanchahe pyroxene diorite (Tietonggou data according to [5]). NHRL, Northern Hemisphere Reference Line.
Research has shown that Rb, Sr, and Ba have great compatibility in amphibole, whereas Rb and K have greater compatibility in phlogopite [44]. Thus, the ratios of Ba/Rb and Rb/Sr could indicate whether the melt source contains amphibole (Ba/Rb > 20, Rb/Sr < 0.1) or phlogopite (Ba/Rb < 20, Rb/Sr > 0.1). The ratios of Rb/Sr and Ba/Rb are from 0.04 to 0.10, and from 12 to 22, respectively. These ratios indicate that amphiboles play an important role and phlogopite plays a secondary role during the formation process of magma melting ( Figure 14). Furthermore, the amphiboles and phlogopites provide the most incompatible elements for the refractory ancient lithospheric mantle. In a word, the pyroxene diorites are probably the result of partial melting of mantle peridotite with amphibole and phlogopite.

Tectonic Setting
The Early Cretaceous emplacement age (138.4 ± 1.2 Ma) of pyroxene diorites in the Sanchahe area is consistent with the ages of Early Cretaceous magmatic activities distributed widely in the North China Craton and has similar geochemistry and Sr-Nd-Pb-Hf isotopic characteristics, suggesting that the Early Cretaceous magmatic activities formed in the same tectonic setting. The Early Cretaceous igneous rocks are a set of bimodal rock assemblages in the eastern North China Craton [45]. The emplacement ages of bimodal rocks are contemporaneous with the peak time of lithospheric thinning in the eastern North China Craton and are also consistent with the forming ages of the metamorphic core complex, extensional sedimentary basins, A-type granites, and alkaline magmatic rocks [46][47][48], suggesting they formed in an extensional tectonic setting. The Sanchahe pyroxene diorites temporally coincident with widespread bimodal magmatism, rift basin development, and exhumation of metamorphic core complexes, combined marking the peak period of the late Mesozoic large-scale lithospheric thinning in the North China Craton [49,50]. Previous studies suggested that the lithospheric thinning of the North China Craton in the Late Mesozoic was caused by the rollback subduction of the Paleo-Pacific plate towards Eurasia [51,52].
We suggest that the Sanchahe pyroxene diorite was a product of lithospheric destruction of the NCC, which was ultimately triggered by westward subducting paleo-pacific plate. Delamination of lower crust enriched and hydrated the ancient sub-continental lithospheric mantle beneath the Laiwu area. Retreat of the subducting lithospheric plate was accompanied by up welling of hot asthenosphere, leading to extensive partial melting of the fertile mantle lithosphere and formation of voluminous hydrous basaltic magmas. Such basaltic magmas ascent with intense differentiation and fractional crystallization eventually yield intermediate melts from which the Sanchahe pyroxene diorite formed.

Conclusions
The emplacement age of pyroxene diorites in the Sanchahe area is 138.4 ± 1.2 Ma, belonging to the Early Cretaceous, and is consistent with the ages of diorite rocks in western Shandong province, whereas the forming age of the Fe-Au deposit was later than the emplacement age of the pluton.
The pyroxene diorites have high Mg # , and all of the samples are characterized by high LREEs/HREEs ratios with enrichment in LILEs but depletion in HFSEs. The Sr-Nd-Pb-Hf isotopic features indicate that the magma source of the pyroxene diorites could be partial melting of the enriched lithospheric mantle dominated by amphibole and reformed by the

Tectonic Setting
The Early Cretaceous emplacement age (138.4 ± 1.2 Ma) of pyroxene diorites in the Sanchahe area is consistent with the ages of Early Cretaceous magmatic activities distributed widely in the North China Craton and has similar geochemistry and Sr-Nd-Pb-Hf isotopic characteristics, suggesting that the Early Cretaceous magmatic activities formed in the same tectonic setting. The Early Cretaceous igneous rocks are a set of bimodal rock assemblages in the eastern North China Craton [45]. The emplacement ages of bimodal rocks are contemporaneous with the peak time of lithospheric thinning in the eastern North China Craton and are also consistent with the forming ages of the metamorphic core complex, extensional sedimentary basins, A-type granites, and alkaline magmatic rocks [46][47][48], suggesting they formed in an extensional tectonic setting. The Sanchahe pyroxene diorites temporally coincident with widespread bimodal magmatism, rift basin development, and exhumation of metamorphic core complexes, combined marking the peak period of the late Mesozoic large-scale lithospheric thinning in the North China Craton [49,50]. Previous studies suggested that the lithospheric thinning of the North China Craton in the Late Mesozoic was caused by the rollback subduction of the Paleo-Pacific plate towards Eurasia [51,52].
We suggest that the Sanchahe pyroxene diorite was a product of lithospheric destruction of the NCC, which was ultimately triggered by westward subducting paleo-pacific plate. Delamination of lower crust enriched and hydrated the ancient sub-continental lithospheric mantle beneath the Laiwu area. Retreat of the subducting lithospheric plate was accompanied by up welling of hot asthenosphere, leading to extensive partial melting of the fertile mantle lithosphere and formation of voluminous hydrous basaltic magmas. Such basaltic magmas ascent with intense differentiation and fractional crystallization eventually yield intermediate melts from which the Sanchahe pyroxene diorite formed.

Conclusions
The emplacement age of pyroxene diorites in the Sanchahe area is 138.4 ± 1.2 Ma, belonging to the Early Cretaceous, and is consistent with the ages of diorite rocks in western Shandong province, whereas the forming age of the Fe-Au deposit was later than the emplacement age of the pluton.
The pyroxene diorites have high Mg # , and all of the samples are characterized by high LREEs/HREEs ratios with enrichment in LILEs but depletion in HFSEs. The Sr-Nd-Pb-Hf isotopic features indicate that the magma source of the pyroxene diorites could be partial melting of the enriched lithospheric mantle dominated by amphibole and reformed by the delaminated crust of the North China Craton and the subduction plate of the Yangtze Craton.
The Early Cretaceous pyroxene diorites formed in an extensional tectonic setting of the eastern North China Craton. The rollback subduction of the Pacific plate is probably the genetic mechanism for the extensional tectonic setting. The Sanchahe pyroxene diorite was a product of lithospheric destruction of the NCC.