Geochemical constraints on the origin of Early Cretaceous alkaline intrusions and its tectonic implication, Sulu Orogenic Belt, Eastern North China Craton

Shen Liu · Caixia Feng · Yan Fan · Kairui Tai · Tianjing Gao ·Ian M. Coulson

Abstract Post-orogenic alkaline intrusions from the Sulu Orogenic Belt of eastern North China Craton consist of A-type granites.In this study,we report U-Pb zircon ages,geochemical data, Sr-Nd-Pb, and zircon Hf isotopic data for these rocks. The LA-ICP-MS U-Pb zircon analyses yield consistent ages ranging from 127.1±2.4 to 119.5±4.8 Ma for four samples. The alkaline rocks are characterized by high total alkalis (K2O + Na2O =8.32-10.11 wt%),light rare-earth element enrichment,and heavy rare-earth element depletion,with a wide range(La/Yb)N values(20-48),moderate negative Eu anomalies(Eu/Eu* = 0.50-0.74), enrichment in large-ion lithophile elements(LILEs,i.e.,Rb,Th,U and Pb),and depletion in Ba,Sr and high field strength elements (HFSEs, i.e., Nb, Ta,and Ti),high(87Sr/86Sr)i ranging from 0.708 to 0.7089,low εNd (t) values from -19.4 to -16.8, (206Pb/204Pb)i =16.751 - 16.935, (207Pb/204Pb)i = 15.381-15.535, (208-Pb/204Pb)i = 37.472-37.838, negative εHf (t) values between -21.3 and -25.7 for the magmatic zircons, and larger TDM2 model ages from 2.5 to 2.8 Ga. These results suggest that the rocks were derived from a common enriched lithospheric mantle source that was metasomatized by foundered lower crustal eclogitic materials before magma generation. Furthermore, the geochemical and isotopic feature implies that the primary magma of these rocks originated through partial melting of ancient lithospheric mantle that was variably hybridized by melts derived from lower crust eclogite. These rocks in this study may have been generated by subsequent fractionation of potassium feldspar, plagioclase, ilmenite, and/or rutile. However,negligible crustal contamination occurred during the diagenesis process.

Keywords Post-orogenic magmatism · Alkaline rocks ·Contamination · Sulu Orogenic Belt · North China Craton

1 Introduction

As a typical Archaean craton, the North China Craton(NCC) exhibits distinct characteristics from other cratons(Menzies et al. 2007; Griffin et al. 1998; Menzies and Xu 1998;Xu et al.1998a,b;Fan et al.2000;Zheng et al.2001,2006; Gao et al. 2002, 2004; Zhang et al. 2002, 2003; Wu et al. 1998, 2003, 2005a, b, 2006; Chen et al. 2004; Wilde et al. 2004; Zhai et al. 2007). For example, strong lithospheric destruction has occurred since the Mesozoic.Nevertheless, there are still many controversial issues concerning the destruction of the NCC, such as the destruction time, space, mechanism, and control factors(Wu et al. 2006; Yang et al. 2006; Menzies et al. 2007;Zhai et al.2007;Zhang et al.2007;Zheng et al.2007;Wu et al. 1998; Liu et al. 2008a, b, 2009; 2013c). The main reasons for the above disagreement are the lack of understanding of Mesozoic lithospheric mantle properties and deep processes beneath the NCC(Wu et al.1998).In recent years, it is generally accepted that the NCC was in an extensional tectonic background during Mesozoic (Wu et al.1998).Therefore, it is particularly important to study the chronology, elements, and isotope geochemistry of mantle-derived magmatic rocks produced in an extensional setting (e.g., mafic dykes, carbonates and alkaline rocks).Mesozoic mafic dykes and carbonates are widely spread in the NCC. More than 200 dykes has been found, and the mafic dykes are mostly distributed along the NE,NW,and EW directions, the length of them is between 10 and 35 km, and the width is more than 8.0 km (Cheng et al.1998; Zhang and Sun 2002; Liu et al. 2004a, b, 2005a, b,2006,2008a,b,2009,2010a,b,2012a,b,2013a,b,d,2014,2015,2016,2017a,b,2018;Yang et al.2004, 2012,2013a,b;Tang et al.2014;Guo et al.2016;Shao et al.2003,2005;Yan et al. 2000, 2007; Qiu 1993; Ying et al. 2004).

In contrast, as one special petrographic type, alkaline rocks belong to the alkaline-peralkaline magmatic rocks.They are characterized by unsaturated silicic acid, high alkali content,obvious feldspar,and alkaline dark minerals,and invisible quartz (Harker 1896; Daly 1914; Wright 1969; Irvine and Baragar 1971). Furthermore, alkaline rocks can be subdivided into ultrabasic rocks (e.g., Kimberlite, carbonite, aegirinite, and cabernet sauvignon),basic rocks (e.g., alkaline gabbro and basalt), medium rocks (e.g., alkaline normal-alkaline coarse-nepheline normal-sounding rock), acidic rocks (e.g., alkaline granite,alkaline rhyolite), and alkaline dykes. In general, alkaline rocks are often late products of the magmatic activity in the mantle source region. The study of alkaline rocks thus has important implications for the development, evolution and dynamic processes of orogenic belts(Wu 1966;Ding 1989;Li 1991;Eby 1992;Qiu 1993;Kogarko et al.1995;Li et al.1999;Zhou et al.1995,2009;Zheng and Bian 1996;Zhang and Xie 1997; Tan 1997; Xu et al. 1998a, b, 2017; Xie et al.1999,2006;Altherr et al.2000;Mingram et al.2000;Wang et al. 2000; Han 2000; Yan et al. 2000, 2002, 2007;Zhang 2001; Zhang et al. 2002, 2005, 2015; Wu et al.2002;Litvinovsky et al.2002;Chen et al.2003,2013;Ren et al.2004;Yang et al.2005a;Huang et al.2005, 2016;Ke et al. 2006; Liu et al. 2006, 2008a, 2013c; Wang et al.2009,2010a,b;Yu et al.2010;Chen and Jiang 2011;Kim et al. 2016). However, alkaline rock has not received due attention for a long time because of of its small distribution area (Yan et al. 2002).

Fig. 1 a Simplified tectonic map of the Sulu Orogenic Belt, eastern NCC (Guo et al. 2004). b The geologic map of study areas and the distributions of the alkaline intrusions

Fig. 2 Representative CL images of zircon grains and zircon LA-ICP-MS zircon U-Pb concordia diagrams for representative A-type granites(JN02, DD02, and LST01) and (LC01) from the Sulu Orogenic Belt, eastern NCC

In this study, we report new results from LA-ICP-MS zircon U-Pb, major and trace element geochemistry, Sr-Nd-Pb isotopic studies, and zircon Hf data for four representative felsic plutons from central Sulu Orogenic Belt. These allow us to: (1) document the reliable age and geochemical characteristics of these rocks; (2) investigate their magma source(s) and origin; and (3) define the tectonic implication during the Early Cretaceous in the study areas.

2 Geological setting and petrography

As the world’s largest high-pressure (HP)-ultra-high pressure (UHP) Orogenic rock area, the Sulu Orogenic Belt is located in the eastern part of the NCC. Currently, it has been accepted as the eastern part of Qinling-Dabie collisional Orogenic Belt between the north China and Yangtze Blocks in the Triassic (e.g., Yin and Ni 1993; Ye et al.1996a, b, 2000; Jahn et al. 1996; Zheng et al. 2002),generally,which can be divided into Orogenic parts,i.e.,an HP blueschist unit to the south and a UHP Orogenic granitic gneiss, granulite and subordinate eclogite, schist,amphibolite, marble, and quartzite association unit to the north(Zhai et al.2000;Chen et al.2003;Guo et al.2004).The northern belt also includes Mesozoic Laiyang Basin.In addition,the Sulu orogenic belt is the region with the postcollision magmatic activity during the ultrahigh-pressure and high-pressure rock (e.g., Chung et al. 2005; Dilek and Altunkaynak 2007), especially the Mesozoic magmatism(225-110 Ma; Zhao et al. 1997; Zhou and Lu 2000; Fan et al. 2001; Chen et al. 2003; Zhou et al. 2003; Guo et al.2004, 2005, 2006; Huang et al. 2005; Zhang et al. 2005;Meng et al. 2005; Yang et al. 2005a, b; Hou et al. 2007;Zhang and Zhang 2007; Liu et al. 2008a, b, 2009, 2013c,2014; Zhao and Zheng 2009; Zhang et al. 2010; Zhang 2010). Geological data indicate that Mesozoic alkaline complex and alkaline rocks are widespread in Sulu Orogenic Belt(Jiazishan,Jiaonan,Wulian,Juxian,Dadian,and Junan;Shandong Province Geology and Mineral Resources Bureau 1991;Han 2000).At present,investigation on some alkali rock mass (e.g., Jiaonan, Jiazishan, Juxian, Junan)has been carried out (215-115 Ma; Yang et al. 2005a;Zhang et al. 2005; Xie et al. 2006; Liu et al. 2006, 2008a;Wang et al. 2009; Chen and Jiang 2011). Nevertheless, a large number of controversies still exist for the alkaline rocks from the Sulu Orogenic Belt (e.g., the origin and evolution, the genetic mechanism, and tectonic implication).

1σ2222122222333222 b/23 8U206P 119118119119119122122122121121120122122 3222323223222322 123 1σ91197127991014221011 123123123123123123123123123123123123127127127127126127 11 b/23 5U207P 12 112 112 111 511 511 611 612 112 812 911 911 9 610951065119551569917227 12 3 15 81 181515211418182027 858451 120 83361717 356 12 412 012 312 412 013 212 114 314 214 5 17 3 Age (Ma)6P b 1σb/20207P 03 12871 0.0003 431 0.0003 129 0.0003 124 0.0002 352 0.0004 523 0.0003 564 0.000303 106 04 262 05 295 05 555 05 555 98 119 16 315 882 119 17 398 119 68 121 1615 3881 119 78 119 28 373 144 17122932 586882 143 04 351 1σ0.00 0.00 0.00 0.00 0.00 0.0003 455 04 545 03 267 03 392 04 351 03 455 04 352 04 544 03 266 03 391 03 385 05 275 03 446 04 553 03 428 05 376 03 565 03 426 0.00 CC 8U elt, eastern N b/23 20 6P .0186.0186.0187.0187.0186.0191.0191.0191.019.0189.0188.0191.0192 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00.0193 rogenic B 1σ0.0106 0 0.0118 0 0.0105 0 0.0081 0 0.0128 0 0.0075 0 0.0101 0 0.0098 0 0.0115 0 0.0158 0 0.0251 0 0.0128 0 0.0126 0.0193.0193.0192.0193.0193.0193.0192.0193.0192.0193.0193.0193.0201.0199.0199.0199.0198.0199 0.0115 0 5U ulu O 7P b/23 2628 265263186188212209248261241359346246 0.0066 0 0.0115 0 0.0105 0 0.0058 0 0.0115 0 0.0068 0 0.0056 0 0.0115 0 0.0105 0 0.0056 0 0.0058 0 0.0172 0 0.0068 0 0.0098 0 0.0105 0 0.0191 0 0.0245 0 0.0075 0 283 m the S 20 42 0.1 73 0.1 41 0.1 32 0.1 51 0.1 31 0.1 38 0.1 36 0.1 45 0.1 62 0.1 96 0.1 46 0.1 46 0.1246304255242283246268303256242243375508265501485525491 65 0.1 e rocks fro σ pic ratios 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.0028 0.1 45 0.1 43 0.1 25 0.1 65 0.1 26 0.1 25 0.1 45 0.1 42 0.1 25 0.1 24 0.1 65 0.1 28 0.1 36 0.1 41 0.1 72 0.1 08 0.1 28 0.1 0.00 6P b 1 alin b/20 85568561616161758215228686 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.01 0.00 35 e data for the alk h Isoto 20 7P 0.04 0.05 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.05 0.05 0.05 0.056283164453562358316454517586154428953 0.05 2T 8U/23 isotop 23 1.45 1.44 0.85 0.73 0.78 0.73 0.76 0.82 0.76 0.86 0.58 0.56 0.74 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.04 0.05 0.05 0.05 0.05 0.88 U-Pb Th/U 1.37 1.34 1.33 1.40 1.32 1.37 1.38 1.27 1.40 1.20 1.77 1.56 1.36 0.83 1.16 0.78 0.76 0.68 0.66 0.93 0.78 0.92 0.76 0.83 0.78 0.72 0.83 0.63 0.82 0.86 0.75 1.35 P-MS 15.4 14.8 11.28.26 6.88 5.15 5.24 6.93 5.41 5.09 3.25 6.35 15.8 1.34 1.33 1.32 1.32 1.41 1.56 1.40 1.52 1.36 1.38 1.34 1.30 1.40 1.27 1.63 1.24 1.23 1.38 16.5 A-IC Pb U 94 ble 1 Z ircon L hot T 586 428 636 478 488 368 382 272 295 223 235 171 231 168 286 226 242 173 208 174 166355 228 766 565 16.3 17.6 11.4 17.36.85 8.43 9.25 5.54 13.55.86 10.63.13 7.23 5.68 7.74 5.78 6.49 6.94 855 635 1 Ta Sp LS T0 01 12345678910111213LC 638 475 896 675 456 345 881 665 308 218 431 276 366 261 244 161 465 342 255 185 434 323 133 102 316 225 223 176 378 232 219 176 255 208 305 221 02 12345678910111213DD 123456

1σ22222222222323223222 b/23 8U206P 126124124123129129124124124124124124124123123123123123125125 1σ851366955556566656665 b/23 5U207P 12 811 8 312 74 135 23 8 122 11 875 138 16 6 124 78 118 78 118 75 118 78 120 98 120 65 118 85 118 75 123 85 118 75 118 86 118 98 120 85 118 78 120 Age (Ma)6P b 1σb/20207P 04 292 03 381954542 03 3858 03 33527 8530 5 29 0 04 0.00 0.00 0.00 0.00 0.000303 305 03 302 03030304 282 03 295 0403 290 03 306 03 298 03 277 0303 281 1σ0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 b/23 8U 94959595959593929292939696 20 6P .0197.0194.0194.0192.0203.0202 0.01 0.01 0.01 0.01.0194 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 1σ0.0092 0 0.0062 0 0.0142 0 0.0063 0 0.0065 0 0.0096 0 0.006 0.006 0.006 0.006 0.0072 0 0.006 0.006 0.006 0.006 0.006 0.006 0.007 0.006 0.006 b/23 5U 358422282222454302232225424232324232525 20 7P 38 0.1 25 0.1 55 0.1 25 0.1 25 0.1 35 0.1 28 0.1 26 0.10.123 0.126 0.129 0.123 28 0.1 0.0028 0.1 0.0024 0.1 0.00 0.0025 0.0025 0.1 0.0026 0.1 0.00 0.0025 0.1 2626 0.1 262525 0.1 σ pic ratios b 10.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 6Pb/20 21 h Isoto 42796132692423541862192225212523612119 20 7P 0.05 0.05 0.04 0.04 0.05 0.04 0.05 0.05 0.05 0.05 0.04 0.05 0.05 0.05 0.05 0.05 0.05 0.04 0.05 0.05 2T/23238U 0.69 0.82 0.83 0.96 0.86 0.94 4.16 2.15 1.56 1.65 1.62 1.63 1.62 1.61 1.58 1.63 2.13 2.12 2.12 2.12 Th/U 1.56 1.29 1.39 1.10 1.22 1.14 4.12 2.72 1.03 1.20 1.62 1.26 1.97 3.90 1.28 1.73 2.49 3.64 2.84 2.76 Pb 5.02 8.49 4.85 7.73 8.78 9.36 53.5 46.5 12.5 27.2 21.3 17.38.9 11.3 18.4 26.4 32.4 ed U 60 ble 1 con tinu h 228 146 356 275 216 155 233 212 336 275 337 295 14 62 355 643 236 458 445 412 73364 225 425 338 238 121 526 135 161 173 135 183 106 536 215 575 158 12 65 445 116 872 316 Ta Sp ot T 02 789101112JN 1234567891011121314

Table 2 Major element concentrations (wt %) for the alkaline rocks from the Sulu Orogenic Belt, eastern NCC; LOI = loss on ignition,Mg# = 100 Mg/(Mg + Fe) in atomic proportions, RV = recommended values, MV = measured values

The studied area is located in the central section of the Sulu Orogenic Belt from Jiaonan to Lashantou (Jiaonan,Liangcheng, Dadian, and Lanshantou; Fig. 1), and the alkaline rocks mainly include quartz-monzonites (JN1,JN2, JN3, JN4, JN5, JN6, JN7, JN8, DD1, DD2, DD3,DD4, DD5, DD6, DD7, DD8, DD9, LST1, LST4, LST5,LST6, LST7, LST8, LST9) from Jiaonan, Dadian, and Lanshantou, and the A-type granites (LC1, LC3, LC4,LC5, LC7, LC8) from Liangcheng.

Table 3 Trace element compositions (in ppm) of the alkaline rocks from from the Sulu Orogenic Belt, eastern NCC

Table 3 continued

Table 3 continued

2.1 Jiaonan quartz-monzonites

The Jiaonan quartz monzonitic intrusion (～390 km2)mainly intruded into Archean or lower Proterozoic gneisses and is the largest intrusion in the study area.It is associated with the Yanshanian granites(Fig. 1).These alkaline rocks are light grey and characteristically medium- to coarsegrained with granular and porphyritic textures. They are dominated by K-feldspar (40 %-43 %), quartz (11.0 %-14.0 %), andesine (30 %-33 %), minor amphibole, and biotite (2.0 %-4.0 %). Accessory minerals include apatite,zircon, magnetite, and titanite.

2.2 Dadian quartz-monzonites

Dadian A-type granites outcrop over ca. 180 km2; these alkaline rocks also intruded into the Archean or lower Proterozoic gneisses and were intruded by the Yanshanian granite (Fig. 1). The rocks are commonly light grey, medium- to coarse-grained with granular and porphyritic textures. In addition, the Dadian A-type granites consist predominantly of K-feldspar(40%-45 %),quartz(10.0%-14.0 %), andesine (32 %-34 %), diopside (8.0 %-9.0 %),subordinate (～ 2.0 %-3.0 %) amphibole and biotite, and accessory minerals including apatite,zircon,magnetite,and titanite.

2.3 Lanshantou quartz-monzonites

The Lanshantou A-type granitic intrusion (～160 km2)mainly intruded into Archean or lower Proterozoic gneisses. They are light grey and characteristically medium- to coarse-grained with granular and porphyritic textures.Furthermore, these alkaline rocks are dominated by K-feldspar (40 %-44 %), quartz (10.0 %-15.0 %), andesine (30 %-35 %), diopside (8.0 %-10.0 %), subordinate(～2.0 %) amphibole and biotite, and accessory minerals including apatite, zircon, magnetite, and titanite.

2.4 Liangcheng A-type granites

The Liangcheng A-type granites(～85 km2)also intruded into Archean or lower Proterozoic gneisses (Fig. 1).These alkaline rocks are commonly light grey to pink in color and composed of quartz (25 %-33 %), perthite (30 %-45 %),albite (An 0-5) (15.0 %-20 %), and minor muscovite.Accessory minerals include zircon, magnetite, and apatite.

3 Analytical procedures

Thirty samples from the alkaline rocks were collected for this study (Fig. 1). Zircon grains were separated from four samples (JN02, LC01, DD02, and LST01) using conventional heavy liquid and magnetic techniques at the Langfang Regional Geological Survey, Hebei Province, China.After separation and mounting, the morphology and internal structure of the zircon were imaged using transmitted and reflected light and by cathodoluminescence(CL)techniques at the State Key Laboratory of Continental Dynamics, Northwest University (Fig. 2). Prior to zircon U-Pb dating, grain mount surfaces were washed in dilute HNO3and pure alcohol to remove any potential lead contamination. Zircon U-Pb and207Pb/206Pb weighted average ages were determined by LA-ICP-MS (Table 1;Fig. 2) using an Agilent 7500a ICP-MS instrument equipped with a 193 nm excimer laser at the State Key Laboratory of Continental Dynamics,Northwest University.The zircon standard 91500 was used for quality control, and a NIST 610 standard was used for data optimization. A spot diameter of 24 μm was used during analysis,employing the methodologies described by Liu et al. (2010a, b, c).Common Pb correction was undertaken following the approach of Andersen (2002), and the resulting data were processed using GLITTER and ISOPLOT (Table 1;Fig. 2). Uncertainties on individual LA-ICP-MS analyses are quoted at the 95 % (1σ) confidence level.

Fig. 3 Classification of the quartz monzonites and A-type granites from the Sulu Orogenic Belt on the basis of a the TAS diagram.All the major element data have been recalculated to 100%on a LOI-free basis(Middlemost 1994;Le Maitre 2002);b Na2O versus K2O diagram,showing the alkaline rocks to be shoshonitic (Middlemost 1972); c Al2O3/(Na2O + K2O) molar versus Al2O3/(CaO + Na2O + K2O)molar plot. Most samples fall in the metaluminous field except some A-type granites straddle the metaluminous and peralkaline boundary; d (10000) * Ga/Al versus Zr plot, all the sample fall in the A-type field. Legends in other figures are the same as in this figure

Major oxides were analyzed with a PANalytical Axiosadvance X-ray fluorescence spectrometer (XRF) at the State Key Laboratory of Ore Deposit Geochemistry(LODG), Institute of Geochemistry, Chinese Academy of Sciences. Fused glass disks were used and the analytical precision as determined on the Chinese National standard GSR-3 was better than 5.0 % (Table 2). Loss on ignition was obtained using 1.0 g powder heated up to 1100 °C for 1 h. Trace elements were performed with an ELAN 6000 ICP-MS at the LODG, following procedures described by Qi et al. (2000). The discrepancy between triplicate analyses is less than 5.0 % for all elements. Analyses of international standards OU-6 and GBPG-1 are in good agreement with recommended values (Table 3).

For Rb-Sr and Sm-Nd isotope analyses, sample powders were spiked with mixed isotope tracers, dissolved in Teflon capsules with HF + HNO3acids, and separated by conventional cation-exchange technique (Zhang et al.2001). Isotopic measurements were performed using a Finnigan Triton Ti thermal ionization mass spectrometer at the LODG. Procedural blanks yielded concentrations of ＜200 pg for Sm and Nd and ＜500 pg for Rb and Sr, and mass fractionation corrections for Sr and Nd isotopic ratios were based on86Sr/88Sr = 0.1194 and146Nd/144Nd =0.7219,respectively.Analysis of the NBS987 and La Jolla standards yielded values of87Sr/86Sr = 0.710246 ± 16(2σ), and143Nd/144Nd = 0.511863 ± 8 (2σ), respectively.

Fig. 4 Chondrite-normalized REE and b Primitive mantle-normalized multi-element variation diagrams for the alkaline rocks from the Sulu Orogenic Belt, eastern NCC. Concentrations are normalized to chondrite composition of Sun and McDonough (1989)

In-situzircon Hf isotopic analyses were performed on a Nu Plasma HR MC-ICP-MS equipped with a GeoLas 2005 193 nm ArF-excimer laser-ablation system.Analyses were carried out using a spot size of 44 μm and He was also used as a carrier gas. The laser repetition rate is 10 Hz and the energy density applied is 15-20 J/cm-2. During the analysis, the176Hf/177Hf ratio of the standard zircon (91500)was 0.282295 ± 0.000027 (n = 14, 2σ), which is in good agreement with the recommended176Hf/177Hf ratio within 2σ (0.2823075 ± 58, 2σ; 0.282015 ± 0.000029, 2σ)(Griffin et al.2006;Wu et al.2006).All the above analysis was performed at the Key state Laboratory of Continental Dynamics, Northwest University, Xi’an, China.

4 Results

4.1 Zircon U-Pb ages

Euhedral zircons in samples JN02, LC01, DD02, and LST01 are clean and prismatic and show clear oscillatory magmatic zoning (Fig. 2). Fourteen zircon grains from sample JN02 yielded a weighted mean206Pb/238U age of 121.3 ± 5.7 Ma (1σ, 95 % confidence interval; Table 1;Fig. 2a).Thirteen zircon grains from sample LC01 yielded a weighted mean206Pb/238U age of 123.0 ± 1.2 Ma (1σ;95 %confidence interval;Table 1;Fig. 2b).Twelve zircon grains from sample DD02 yielded a weighted mean206Pb/238U age of 127.1 ± 2.4 Ma (1σ; 95 % confidence interval; Table 1; Fig. 2c). Thirteen zircon grains from sample LST01 yielded a weighted mean206Pb/238U age of 119.5 ± 4.8 Ma (1σ; 95 % confidence interval; Table 2;Fig. 2d). These new-age data provide the best estimates of the crystallization ages of alkaline rocks within the study area. No major zircon inheritance was observed in any of the samples.

4.2 Major and trace element geochemistry

Geochemical data for the A-type granites from Sulu Orogenic Belt are listed in Tables 2 and 3. The alkaline rocks have a wide range of chemical compositions, with SiO2-= 60.76-73.48 wt %, TiO2= 0.18-0.76 wt %, Al2O3-= 13.71-16.43 wt %, Fe2O3= 1.58-6.26 wt %,MnO = 0.06-0.15 wt %, MgO = 0.31-2.75 wt %, CaO =1.08-4.42 wt %, Na2O = 3.63-4.43 wt %, K2O= 4.68-5.71 wt %, and P2O5= 0.05-0.51 wt %. These rocks are relatively high in total alkalis,with K2O + Na2O range from 8.32 to 10.11 wt%.All alkaline rocks plot in the alkaline field on the total alkali-silicon (TAS) diagram(Fig. 3a).All samples straddle the shoshonitic series in the Na2O versus K2O (Fig. 3b). In a plot of molar ratios of Al2O3/(Na2O + K2O) and Al2O3/(CaO + Na2O + K2O),the alkaline rocks are all metaluminous (Fig. 3c). In addition,the 10,000 × Ga/Al ratios of the A-type granites range from 2.75 to 3.45. In the Ga/Al versus Zr discrimination diagram (Fig. 3d) of Whalen et al. (1987), the felsic rocks are all classified as A-type granite. The A-type granites display an inconspicuous correlation between MgO,Al2O3,Fe2O3,CaO,TiO2,P2O5,Zr,Sr,Ba,and SiO2Na2O,K2O,and Rb (not shown).

The A-type granites are all characterized by LREE enrichment and HREE depletion, with a wide range (La/Yb)Nvalues (20-48) and moderate negative Eu anomalies(Eu/Eu* = 0.50-0.74) (Fig. 4a). In the primitive mantlenormalized trace element diagrams, the A-type granites show enrichment in LILEs (i.e., Rb, Th, U, and Pb) and depletion in Ba, Sr and REEs (i.e., Nb, Ta, and Ti)(Fig. 4b).

4.3 Sr-Nd-Pb isotopes

d)i εNd(t)- 17.5- 17.6- 17.7- 17.6- 17.7- 17.6- 19.1- 19.3- 19.4- 19.3- 17.5- 17.7- 17.5- 17.6- 17.4- 16.8- 16.8- 16.8- 17.2 10-1 2 year-1 d/14 4N 787579760092847058 2σ (87Sr/86Sr)i (14 3N 0.511584 0.5115 0.5115 0.5115 0.511573 0.5115 0.5115 0.5114 0.5114 0.511488 0.511576 0.511569 0.511575 0.5115 0.511591 0.511623 0.511624 0.511624 0.5117= 6.54 ×λSm 0.70 8448 0.70 8497 0.70 8430 0.70 8446 0.70 8480 0.70 8469 0.70 8548 0.70 8549 0.70 8546 0.70 8544 0.70 8107 0.70 8226 0.70 8247 0.70 8177 0.70 8130 0.70 8867 0.70 8825 0.70 8821 0.70 8817 7) and d ±9101010812101010999810999710 4Nd/14 1653 1652 1648 1651 1649 1644 1572 1563 1557 1564 1648 1645 1648 1646 1662 1716 1719 1721 1758teiger and Jäger 197 14 3N 0.51 0.51 0.51 0.51 0.51 0.51 0.51 0.51 0.51 0.51 0.51 0.51 0.51 0.51 0.51 0.51 0.51 0.51 0.51 1 year-1 (S 4N dm/14 10-1 7S 642165980271281091404 140.08 0.09 0.09 0.09 0.09 0.08 0.08 0.08 0.09 0.09 0.08 0.09 0.08 0.09 0.09 0.11 0.12 0.12 0.13 1.42 ×CC d (ppm)72.5 68.3 73.8 76.2 66.4 83.5 40.5 23.6 25.2 36.4 78.8 75.3 74.5 74.2 83.6 20.4 21.3 21.6 23.7 rogenic B elt, eastern N m (ppm) N.4101011111011.6.3.5.5.75.98 3.43 3.75 5.51 1111101112.4.4.8.2.54.03 4.27 4.43 5.24 ulu O ± 2σ S 2343412323243242043 s of the alkaline rocks from the S 87Sr/86Sr 0.70 8955 1 0.70 9033 1 0.70 9063 1 0.70 9074 1 0.70 9005 1 0.70 8984 1 0.71 1508 1 0.71 2655 1 0.71 2148 1 0.71 2853 1 0.70 9305 1 0.70 9308 1 0.70 9293 1 0.70 9201 1 0.70 9244 1 0.70 9515 1 0.70 9465 1 0.70 9436 1 0.70 9478 1 Reservoir values and decay constants of λRb =/8 6Sr rm 87Rb 0.29 4 0.31 1 0.36 7 0.36 4 0.30 4 0.29 9 1.69 3 2.34 9 2.06 1 2.46 5 0.66 3 0.59 9 0.57 9 0.56 7 0.61 7 0.38 1 0.37 7 0.36 2 0.38 9nifo rite Uhond position r (ppm)1132 1165 1072 1065 1161 1153236143215148812846865858864 1025 1006 1023 1012 pic com ge (Ma) R b (ppm) S 115125136134122119138116153126186175173168184135131128136 ere calculated using C r-Nd isoto 1.31.3 1.31.31.31.33.03.03.03.07.17.17.17.17.19.59.59.59.5s warti 1978)Table 4 S le A 12121212121212121212121212121211111111 position Samp JNJNJNJNJNJNLC 1256781358 LC LC LC DD 1 DD 3 DD 4 DD 5 DD 9 LS T1 LS T5 LS T6 LS T8 Th e com(L ugmair and H

b)ib/204Pb)i (208P 37.545 37.631 37.727 37.628 37.630 37.631 37.472 37.472 37.474 37.473 37.836 37.838 37.835 37.832 37.837 37.835 37.832 37.834 37.836 b/20 4Pb)i (207P 15.383 15.382 15.424 15.381 15.419 15.421 15.475 15.474 15.476 15.476 15.444 15.446 15.445 15.446 15.444 15.433 15.433 15.435 15.535 b/20 4P 858486 b (20 6P 16.882 16.880 16.883 16.8 16.8 16.8 16.751 16.753 16.754 16.751 16.935 16.930 16.934 16.930 16.935 16.924 16.926 16.922 16.923 h/20 4P232T 11.0 1315.2.97.17.07.6 39373522.7.9.3.6 23211219131820.1.0.7.7.0.2.7 3421.0.6 4P b/20235U 0.0163 0.0323 0.0345 0.0145 0.0156 0.0182 0.0313 0.0315 0.0522 0.0216 0.0227 0.0222 0.0159 0.0383 0.0152 0.0229 0.0310 0.0473 0.0311 4P b 8U/20 C 232.24.54.82.02.22.54.34.37.23.03.13.12.25.32.13.24.36.54.3 elt, eastern NC h (ppm)8.26 4.65 5.35 4.27 4.16 4.25 12.2 12.5 10.4 16.2 16.9 15.39.41 6.35 9.26 13.3 12.4 12.45.61 rogenic B b (ppm) T 47.6 22.3 21.4 38.2 37.5 35.2 19.4 20.8 18.6 45.3 46.5 46.4 47.2 20.5 45.3 46.4 38.1 23.2 16.5 ulu O ) P positions of the alkaline rocks from the S (ppm 1.75 1.62 1.66 1.25 1.32 1.44 1.37 1.48 2.19 2.21 2.36 2.31 1.68 1.76 1.54 2.38 2.65 2.46 1.15 4P b Ub/20208P 37.611 37.711 37.823 37.671 37.672 37.678 37.715 37.705 37.691 37.616 37.982 37.971 37.915 37.957 37.914 37.943 37.955 38.036 37.836 4P bb/20207P 15.385 15.3862 15.428 15.383 15.421 15.423 15.479 15.478 15.483 15.479 15.447 15.449 15.447 15.451 15.446 15.436 15.437 15.441 15.535 4P bb/20 61 pic com 20 6P 16.924 16.965 16.974 16.923 16.925 16.934 16.834 16.837 16.893 16.811 16.997 16.991 16.978 17.036 16.974 16.983 17.006 17.045 16.923 b isoto ge (Ma)1.31.3 121212121212121212 1.3 1.31.31.33.03.03.03.07.17.17.17.17.19.59.59.59.5 Table 5 P le A 12121212121211111111 Samp 13459 JN 1 JNJNJNJNJNLC 256781358 LC LC LC DD DD DD DD DD LS T1 LS T5 LS T6 LS T8

Fig. 5 Variations in initial 87Sr/86Sr versus εNd (t) values for the alkaline rocks from the Sulu Orogenic Belt,eastern NCC.The studied rocks analyzed during this study plot within the enriched mantle source field

Fig. 6 208Pb/204Pb and 207Pb/204Pb versus 206Pb/204Pb diagrams for the alkaline rocks. Fields for I-MORB (Indian MORB) and P&NMORB(Pacific and North Atlantic MORB),OIB,NHRL and 4.55 Ga geochron are after Barry and Kent 1998; Zou et al. 2000), and Hart(1984), respectively

Sr, Nd, and Pb isotopes for the A-type granites from Sulu Orogenic Belt are listed in Tables 4 and 5. The studied alkaline rocks show very uniform(87Sr/86Sr) i ranging from 0.708 to 0.7089 and relatively small variation in initial εNd(t) values from - 19.4 to - 16.8, which suggests a common source region. In addition, the Sr-Nd isotopic compositions (Fig. 5) are comparable to those of the Mesozoic medium-acid rocks, granitoids, gabbros, lamprophyres, adakites, and alkaline rocks in Sulu-Dabie Orogenic Belt (Zhao et al. 1997; Zhou and Lu 2000; Fan et al. 2001; Zhang et al. 2005; Meng et al. 2005; Huang et al.2005;Yang et al.2005a;Liu 2004;Liu et al.2005a,b,2006, 2008a, 2009, 2011, 2012a, b, 2013a, b, c, d, 2014;Zhang et al.2010;Chen and Jiang 2011;Wang et al.2013).The Pb isotopic ratios in the alkaline rocks are characterized by206Pb/204Pb = 16.751-16.935,207Pb/204Pb =15.381-15.535,208Pb/204Pb = 37.472-37.838 (Table 5).They are significantly different from those of the Yangtze lithospheric mantle ((206Pb/204Pb = 17.649-18.603,207Pb/204Pb = 15.422-15.623,208Pb/204Pb = 37.674-38.521;Chen et al.2001;Yan et al.2003,2005;Yang et al.2004; Wang et al. 2005) and are identical to those of alkaline complex and mafic rocks from the central NCC and Sulu-Dabie Orogen (206Pb/204Pb ＜17.5; Yan et al.2003; Zhang et al. 2004; Xie et al. 2006), having a clear EM1 affinity (Fig. 6).

4.4 Zircon Hf isotopes

Zircon Hf isotope results are listed in Table 6. Nineteen spot analyses were obtained for sample JN02, yielding relatively uniform εHf(t) values of between - 21.8 and- 25.7, corresponding to TDM2model ages of between 2652 and 2792 Ma (Table 6; Figs. 7, 8, 9), and giving an average of εHf(t) = - 23.9 and TDM2= 2683 Ma. Twenty spot analyses were obtained for sample LC01;they show a narrow range of εHf(t) values - 23.4 and - 24.3, corresponding to TDM2model ages between 2632 and 2684 Ma(Table 6; Figs. 7, 8, 9), and yielded a mean εHf(t) = - 23.7 and TDM2= 2668 Ma.Nineteen spot analyses were obtained for sample DD02. The determined εHf(t)values vary between - 21.3 and - 22.4, corresponding to TDM2model ages in the range from 2525 Ma and 2595 Ma(Table 6; Figs. 7, 8, 9). These nineteen spots gave a mean εHf(t) = - 21.9 and TDM2= 2562 Ma. Nineteen spot analyses were obtained for sample LST01, giving εHf(t)values between- 21.1 and- 25.5,corresponding to TDM2model ages between 2504 and 2595 Ma (Table 6; Figs. 7,8, 9), thus yielding an average of εHf(t) = - 22.0 and TDM2= 565 Ma.

5 Discussion

5.1 Source, fractional crystallization, and crustal assimilation

The studied alkaline rocks are characterized by similar patterns of rare earth and trace elements (Table 3; Fig. 4a,b), comparable Sr-Nb-Pb isotopic composition (Tables 4,5;Figs. 5,6),large TDM2model ages(2.8-2.5 Ga;Table 6)and negative εNd(t) (from - 19.4 to - 16.8; Table 4) and εHf(t)(between- 25.7 and- 21.3;Table 5),implying the ancient enriched and mixed source(crustal and lithospheric mantle).This further supported by their relatively high Mg#values (27-49; Table 2; Rapp et al. 1999), Ce/Pb average ratios (10.4; Rudnick and Fountain 1995), Nb/Ta ratios(29-58;Table 3;Dostal and Chatterjee 2000),Zr/Hf ratios(32-40; Table 3; Green 1995), and lower Nb/U average ratios (3.53; Table 3; Rudnick and Fountain 1995). Moreover, in the major element versus MgO diagrams (not shown), the studied rocks plot along an extension of the homochronous mafic rocks from Sulu Orogenic Belt trend(Fan et al. 2001; Yang et al. 2005a; Liu 2004; Liu et al.2005a,b,2006,2008a,2009,2011,2012a,b,2013a,b,c,d,2014), indicating they were the result of crystal fractionation of a mafic magma similar to the parental magma of the mafic rocks.The studied alkaline rocks thus were products of a mafic magma derived from a mantle source, but with some crustal involvement.

Table 6 Hf isotopic compositions of representative alkaline samples in the the Sulu Orogenic Belt, eastern NCC

Table 6 continued

Fig. 7 Plots of zircon ages versus εHf(t)values(JN02,LC01,DD02,and LST01) for the alkaline rocks from Sulu Orogenic Belt, eastern NCC

As an important geological process, at present, fractional crystallization has received a certain degree of attention. The studied samples are characterized by higher Rb (125-212 ppm) and Th values (25.2-77.5 ppm;Table 3),207Pb/204Pb (15.381-15.535; Table 5), low Sr/Y ratios (10.3-21.6; Table 2), implying the fractional crystallization is relatively obvious, on the plots between Ba,Sr, and Eu anomalies (Fig. 10a, b). The parent magma of the studied granites occurred fractionation of potassium feldspar and plagioclase and the plagioclase fractionation is further supported by the Eu negative anomalies (Eu/Eu* = 0.50-0.74) (Fig. 4a). This is also evident in the Ba versus Rb/Sr diagrams (Fig. 11). In general, negative Ti anomalies in all felsic rocks (Fig. 4b) agree with the fractionation of Fe-Ti oxides such as rutile and ilmenite.In contrast,the LC1-LC8 sample experienced a higher degree of separation and crystallization in the genetic process(Fig. 3b, c; Wu et al. 2015).

Assimilation, crystallization fractionation (AFC), or magma mixing is usually postulated to explain the occurrence of felsic rocks (e.g., Depaolo 1981; Marsh 1989).The major and trace element characteristics indicate that crustal contamination is negligible. Generally, for felsic rocks,the AFC process and magma mixing would result in a possible linear correlation between MgO versus (87-Sr/86Sr)iand εNd(t) (Fig. 12), these correlations, however,are not observed in the studied alkaline rocks(not shown),indicating that magma evolution is not significantly affected by crustal contamination of magma mixing.

5.2 Genetic mechanism

As discussed above, the alkaline rocks in this study are derived from the partial melting of an ancient and enriched mantle source. A dynamic mechanism, however, is required to decipher the origin of the enriched mantle.Currently, at least three competing mechanisms can be envisaged (Yang et al. 2005a, b; Zhang et al. 2005; Liu et al. 2006, 2008a, 2013c): (1) partial melting of lowercrustal rocks under fluxing of volatile (e.g., Lubala et al.1994); (2) partial melting of the subducted lithospheric mantle of the Yangtze Craton that has experienced extensive fractionation and minor contamination by crustal material (Yang et al. 2005a); (3) partial melting of the metasomatized mantle,subsequent fractionation of mantlederived magma with or without crustal contamination(e.g.,Sutcliffe et al. 1990; Lynch et al. 1993; Liu et al. 2008a,2013c). The studied alkaline rocks are characterized by relatively higher εNd(t) values than lower-crustal values published for the NCC(Jahn et al.1999),ruling out the first genetic mechanism for these rocks. At present, it is generally accepted that there is no basic lower crust in eastern China(Gao et al.1998),however,the first genetic model so far cannot give a reasonable explanation. As discussed above, the studied alkaline rocks were derived from the partial melting of an ancient enriched and mixed source.The parental magma of them thus in this study were derived from the melting of the lithospheric mantle of the NCC, and some ancient continental crust materials were involved during magma ascent by crustal contamination or the source region due to metasomatism. In the Late Mesozoic (160-120 Ma), lithospheric mantle beneath the North China and Yangtze Craton shared similar trace element characteristics,but different Sr-Nd-Pb isotopic ratios(Xie et al.2006).As discussed above,the Pb isotopic ratios of the studied alkaline rocks from Sulu Orogenic Belt(Table 5) are significantly different from those of the Yangtze lithospheric mantle, and are similar to those of mafic rocks from the central NCC, implying that the studied rocks are not derived from the subducted Yangtze plate, but from the overlying NCC. Therefore, the second genetic model can be eliminated in this study.Owing to its higher density than that of lithospheric mantle peridotite by 0.2-0.4 g cm-3, foundering of lower crustal eclogites into underlying convecting mantle has been proposed to play a role in plume magmatism, crustal evolution and formation of chemical heterogeneities within the mantle (Arndt and Goldstein 1989;Kay and Kay 1991;Rudnick and Fountain 1995;Jull and Kelemen 2001;Escrig et al.2004;Gao et al.2004;Elkins-Tanton 2005;Lustrino 2005;Anderson 2006;Gao et al. 2008). This mechanism currently has been reasonably used for explaining the origin of some Mesozoic igneous rocks from the NCC (e.g., Liu et al. 2008a, b,2009,2010a,b,2012a,b,2013a,b,c,d,2014,2015,2017a,b, 2018). The Pb-isotopic differences observed in the studied rocks from Sulu Belt are not related to the subducted Yangtze lithosphere but result from the contribution made by the ascending asthenospheric mantle following Late Mesozoic lithospheric thinning (200-120 Ma; Yang et al.2004).This also can be used to interpret the origin of the Early Cretaceous mafic rocks from Dabie terrane(Huang et al. 2003). Consequently, we suggest that the Sr,Nd,Pb,and zircon Hf isotopic compositions of the alkaline rocks from central Sulu Orogenic Belt is not due to the involvement of subducted Yangtze lithosphere, but essentially caused by the upwelling of asthenospheric mantle following foundering of lower crustal materials during the Late Triassic (＜200 Ma) and Early Cretaceous (120 Ma;Guo et al.2004).Nevertheless,the genetic mechanism and process for the studied alkaline rocks must be given a reasonable explanation (Fig. 13).

Fig. 8 Histogram of zircon εHf (t) values for the alkaline rocks from the Sulu Orogenic Belt, eastern NCC

Nowadays, it is generally believed that between ～240 and 220 Ma (Zhang et al. 2005; Yang et al. 2005b; Liu et al. 2008a, b, c, 2009, 2013c), the continual collision between the NCC and Yangtze Craton induced a thickened crust(e.g.,Liu et al.2008a,b,2009,2013c),as well as the peak metamorphism and rapid exhumation of the HP-UHP Orogenic terrane (Guo et al. 2006); during ～ 225-205 Ma, the break-off of the subducted Yangtze plate occurred (Chen et al. 2003); between ～205 and 185 Ma, further thickened crust appeared due to intracontinental compression between the NCC and Yangtze Craton (e.g., Liu et al. 2008a, b, 2009, 2013c).Subsequently, at ～ 185-165 Ma, asthenospheric upwelling, uplift of Sulu Orogenic Belt, lithospheric collapse,extension and thinning, and emplacement of alkaline magmas occurred beneath the Sulu-Dabie Orogenic Belt and eastern NCC (150-110 Ma; e.g., Li et al. 2002; Liu et al. 2008a, b, 2009, 2013c). And then decompression melting of the thickened crust produced primary melts,which underwent fractionation(i.e.,potassium feldspar and plagioclase, rutile and ilmenite) without crustal contamination to produce the alkaline rocks intrusions in the study areas.

Fig. 9 Histogram of zircon TDM model ages for the alkaline rocks from the Sulu Orogenic Belt, eastern NCC

6 Tectonic implications

In general, A-type granites occur in the lithospheric extensional setting, and their research has been highly concerned(Liu et al.2008a;Jiang et al.2012;Zhang et al.2020).Previous studies have shown that(Jiang et al.2012;Zhang et al.2020),A-type granites can be divided into two types: non-orogenic and post-orogenic (A1 and A2). The A1-type granites were formed in the extension stage after the stability of the continental lithosphere, while the A2-type granites were a sign of the end of orogeny(Eby 1992).The study of A-type granite thus is of great significance in tectonic indication. Currently, it is generally accepted that there exist three stages of magmatism after the UHP metamorphic event (240-220 Ma; Yang et al. 2005b), i.e.,Late Triassic, Late Jurassic, and Early Cretaceous. Early Cretaceous magmatic activities (130-110 Ma) were widespread in Sulu Orogenic Belt, petrological, geochronological and geochemical data emphasize that the Early Cretaceous magmatism from the Dabie-Sulu Orogenic Belt has been proposed to result from partial melting of a metasomatized lithospheric mantle source post-collisional extension and thinning after the collision of the NCC and Yangtze Craton (Zhao et al. 1997; Ma et al. 1998; Jahn et al. 1999; Fan et al. 2001; Chen et al. 2001, 2004; Yan et al.2003,2005;Li et al.2004;Yang et al.2004,2005a,b;Wang et al. 2005; Huang et al. 2005; Zhang et al. 2005,2010; Xie et al. 2006; Liu et al. 2006, 2008a, b, 2009;Zhang 2010; Zhao and Zheng 2009; Wang et al. 2009,2010a,b;Chen and Jiang 2011),displaying features similar to other alkaline rocks referred to in the literature as post-collisional (Sylvester 1989) or post-orogenic/anorogenic syenites (Bonin 1990). The exact mechanism of lithospheric extension and thinning, however, remains controversial. Firstly, previous studies have shown that all Mesozoic magmatism in eastern China was the result of a back-arc extensional setting due to the subduction of Izanagi Plate beneath the East Asian continent (Chen et al.2004). Nevertheless, the Early Cretaceous was a period when the Izanagi Plate primarily moved toward the north or north-northeast, thence, providing little chance for inducing broad back-arc extension in the study areas(Maruyama and Send 1986; Kimura et al. 1990; Li et al.2004; Liu et al. 2008a). In addition, another model also is recommended (Houseman et al. 1981), i.e., lithospheric extension and thinning have been induced by convective instability of a thickened mantle boundary layer.However,this model has been excluded due to the absence of magmatism (185-165 Ma) in the Sulu Orogenic Belt (Zhao et al. 1997; Fan et al. 2001; Chen et al. 2003; Guo et al.2004,2005,2006;Huang et al.2005;Yang et al.2005a,b;Liu et al. 2008a, b, 2009). Currently, lithosphere removal in eastern China has been well recognized and verified(Menzies et al. 1993; Menzies and Xu 1998; Griffin et al.1998;Wu et al.2003;Gao et al.2002),as discussed above,lithospheric foundering can be explained as lithospheric removal, undoubtedly, this model (lithospheric removal)would result in lithospheric extension, thinning and coeval magmatism during Early Cretaceous (130-110 Ma)beneath the Sulu Orogenic Belt (Kay and Kay 1993; Liu et al. 2008a, b, 2009).

Fig. 10 Plots of a EuN/Eu* versus Sr, and b Ba for the A-type granites. Mineral fractionation vectors calculated using partition coefficients from Schnetzier and Philpotts(1970).Tick mark indicate percentage of mineral phase removed, in 10% intervals. Pl-plagioclase; Kf-potassium feldspar

Fig. 11 Plots of Ba versus Rb/Sr for the A-type granites

Fig. 12 Plots of MgO versus (87Sr/86Sr) i and εNd (t) for the A-type granites

Fig. 13 Plots of Zr versus SiO2 for the A-type garnites

7 Conclusions

Several conclusions can be drawn from the petrological and geochemical studies of the studied alkaline rocks:

1. LA-ICP-MS U-Pb zircon dating results indicate that the Jiaonan, Liangcheng, Dadian, and Lanshantou alkaline intruded at 121.3, 123.0, 127.1, and 119.5 Ma, respectively. They are all relatively high in total alkalis (K2O + Na2O = 8.32-10.11 wt %);enriched in LREE, LILE (Rb, Th, and U), and Pb and depleted in Nb, Ta, and Ti; and moderate negative Eu anomalies (Eu/Eu* = 0.50-0.74).

2. The alkaline rocks in this study are derived from the partial melting of an ancient and enriched mantle source. The parental magma originated by partial melting of hybridized mantle derived from foundered lower crustal eclogites. Subsequent fractionation of potassium feldspar, plagioclase, Fe-Ti oxides (i.e.,rutile, ilmenite). The zircon saturation temperatures(TZr) of the A-type granites are 830-907 °C, which approximately represents the crystallization temperature of the magma in this study.

3. It is proposed that the occurrence of lithospheric extension and thinning beneath the Sulu Orogenic Belt is related to lithospheric removal due to foundering(removal) of the lower crust.

AcknowledgementsThis research was supported by the National Natural Science Foundation of China(41373028 and 41573022).The authors thank Honglin Yuan for assistance during zircon Hf isotope analyses and Liang Qi for assistance during Sr-Nd-Pb isotope and trace element analyses, and the zircon U-Pb dating.