This article appeared in a journal published by Elsevier. The attachedcopy is furnished to the author for internal non-commercial researchand education use, including for instruction at the authors institution

and sharing with colleagues.

Other uses, including reproduction and distribution, or selling orlicensing copies, or posting to personal, institutional or third party

websites are prohibited.

In most cases authors are permitted to post their version of thearticle (e.g. in Word or Tex form) to their personal website orinstitutional repository. Authors requiring further information

regarding Elsevier’s archiving and manuscript policies areencouraged to visit:

http://www.elsevier.com/authorsrights

Author's personal copy

Geochronology and geochemistry of leucogranites from the southeastmargin of the North China Block: Origin and migration

Shuguang Li a,b,⁎, Shui-Jiong Wang a, Sushu Guo b, Yilin Xiao b, Yican Liu b, Sheng-Ao Liu a,Yongsheng He a, Junlai Liu a

a State Key Laboratory of Geological Processes and Mineral Resources, China University of Geosciences, 100083 Beijing, Chinab CAS Key Laboratory of Crust–Mantle Materials and Environments, School of Earth and Space Sciences, University of Science and Technology of China, Hefei 230026, China

a b s t r a c ta r t i c l e i n f o

Article history:Received 6 June 2013Received in revised form 5 August 2013Accepted 6 August 2013Available online 14 September 2013

Handling Editor: T. Kusky

Keywords:LeucograniteChannel flowContinental collisionDabie–Sulu orogenNorth China Block

Systematic geochronological, geochemical and geological data for the Late Jurassic granitic intrusions from theBengbu area in the southeastern margin of North China Block (NCB) are presented in this paper. The very lowFeO + MgO + TiO2 contents and ferromagnesian mineral abundance as well as low Al-saturation indexes(ASI = 0.92–1.09) of the rocks document that these granitic intrusions are metaluminous leucogranites. TheBengbu leucogranites contain abundant both inherited Neoproterozoic igneous and Triassic metamorphic zir-cons, showing a high agreement in ageswith the ultrahigh-pressure (UHP)metamorphic rocks from the adjacentDabie–Sulu orogen, but significant different from the country rocks. The magmatic overgrown zircon rims givethe ages of 167–148 Ma, consistent with the time of Jurassicmigmatization in the Sulu orogenic belt. Mineral in-clusions such as quartz, feldspar, apatite, titanite, biotite,muscovite and phengite (Si = 3.58)without coesite oc-curring within the Triassic metamorphic zircons suggest that the source rocks of the leucogranites might haveexperienced only high-pressure (HP) rather than UHP metamorphism. Furthermore, steep HREE patterns ofthe inherited metamorphic zircons, plus very low bulk REE contents and Sr–Nd–Pb isotopic characteristics, sug-gest that the leucogranites were most likely derived from partial melting of subducted felsic gneiss rather thanbasaltic eclogite in Dabie–Sulu orogen, with residual allanite in the source. Their low Rb, high Sr contents andlow Rb/Sr ratios suggest that the metaluminous leucogranites are derived from mica-poor orthogneiss. Meltingtemperature for the leucogranites is about 700–710 °C as estimated using combined Ti-in-zircon and zirconiumsaturation thermometries. Inherited metamorphic zircons yielded metamorphic Ti-in-zircon temperature of681 ± 60 °C, representing that for protolith-forming, which shows no significant difference from the meltingtemperature. H2O-present melting is thus required to generate the low-temperature melt. Sedimentary faciesanalysis of the Jurassic strata in the Hefei basin suggests that the Sulu orogen had been northward moved tothe east of the Bengbu uplift by the Tan–Lu Fault in the Late Jurassic, resulting into a contrasting topography inthis area and lateral pressure gradient in the local middle-lower crust. This lateral pressure gradient may drivethe partially molten crust (migmatite) in Sulu orogen westward flow into the middle-lower crust of the Bengbuuplift in NCB. Gravity then drove melt to ascent into the middle-upper crust of the Bengbu uplift to form theleucogranitic intrusions. The NWW–SEE distribution of the later Jurassic leucogranites in Bengbu uplift may indi-cate the direction of ancient channel flow.

© 2013 International Association for Gondwana Research. Published by Elsevier B.V. All rights reserved.

1. Introduction

Leucogranites as products of crustal melting are well developed incollisional orogen, which reveal when and where the crust has meltedsince continent–continent collision began. In the past few decades,channel flow of a weak crustal layer has been proposed to explain thethickening and outward growth of the Tibetan plateau aswell as surfacedeformation in other orogenic belts (e.g., Nelson et al., 1996; Royden

et al., 1997; Clark and Royden, 2000; Beaumont et al., 2001; Grujic,2006; Harris, 2007; Schulmann et al., 2008; Raimondo et al., 2009;Jamieson et al., 2011). Leucogranite has been considered as petrologicalevidence for this crustal flow model. For example, the High Himalayametasedimenatry rocks have been considered to be southward flow ofmid-crustal rocks beneath southern Tibet and its petrological evidencesmainly come from the High Himalaya leucogranites (e.g., Inger andHarris, 1993; Harris andMassey, 1994; Jamieson et al., 2011). All report-ed leucogranites, so far, are developed within the orogen belt and mostof them were derived from melting of their host rocks (e.g., Clarkeet al., 1993; Inger and Harris, 1993; Harris and Massey, 1994;Pressley and Brown, 1999; Nabelek and Bartlett, 1998; Van deFlierdt et al., 2003). Therefore, it is interesting to find out whether

Gondwana Research 26 (2014) 1111–1128

⁎ Corresponding author at: State Key Laboratory of Geological Processes and MineralResources, China University of Geosciences, 100083 Beijing, China.

E-mail address: lsg@ustc.edu.cn (S. Li).

1342-937X/$ – see front matter © 2013 International Association for Gondwana Research. Published by Elsevier B.V. All rights reserved.http://dx.doi.org/10.1016/j.gr.2013.08.019

Contents lists available at ScienceDirect

Gondwana Research

j ourna l homepage: www.e lsev ie r .com/ locate /gr

Author's personal copy

a melt-weakened crust or leucogranites as their melting productscan flow from one continent block to another during continentalcollision.

The Jingshan granitic intrusion in the Bengbu uplift is located inthe southeast margin of the North China Block (NCB), nearby theDabie–Sulu UHPM belt. It has been reported that the Jingshan intru-sion crystallized in the Late Jurassic and contains abundant inheritedzircons with Neoproterozoic magmatic and Triassic metamorphicages, which are comparable to those of ultrahigh pressure (UHP) meta-morphic rocks from the Dabie–Sulu orogen (W.L. Xu et al., 2005). In ad-dition, the inherited metamorphic zircons from the Jingshan intrusionhave extremely low and highly heterogeneous δ18O values rangingfrom −9.4‰ to 8.6‰, consistent with the values and variationsexhibited by metamorphic zircons from Dabie–Sulu orogen (Wanget al., 2013b). These observations aroused great interests of the geolog-ical society (W.L. Xu et al., 2005; Guo and Li, 2009a;Wang et al., 2009; Liet al., 2010;Wang et al., 2010; Yang et al., 2010;Wang et al., 2013b), be-cause it may contain important information on the crustal material mi-gration from the Dabie–Sulu orogen to the southeastmargin of theNCB.However, the origin of the Jingshan intrusion and their potential impli-cations to the evolution of Dabie–Sulu orogen are in debate (W.L. Xuet al., 2005; Guo and Li, 2009a; Wang et al., 2009, 2010; Yang et al.,2010). In addition, the protolith of the Jingshan intrusion and its meta-morphic andmelting conditions are unknown, despite their importanceto constrain the tectonic process related to the origin of the Jingshan in-trusion. Though all published data indicate the migration of subductedSCB crust to the Bengbu area, the mechanism of crustal migration andthe origin of the Jingshan intrusion need further studies.

In order to provide constraints on these controversial issues and tounderstand better the protolith,melting condition andmigrationmech-anism of leucogranites in continental collision zone, we have performeddetailed studies of zircon U–Pb dating, mineral inclusions and trace ele-ments in zircons, bulkmajor and trace elements, Sr–Nd–Pb isotope geo-chemistry, and fabric of the gneissosity of the leucogranites from theBengbu uplift. Based on the present data and regional geology analysis,we conclude that the Sulu orogen was transected and moved to thenorth by Tan–Lu Fault during the Jurassic and a lateral melt-weakenedcrust flow from the Sulu orogen has been westward injected into themiddle crust of theNCB in the Late Jurassic. The Bengbu leucogranitic in-trusions are formed by melt accumulation at a shallow level derivedfrom the crustal flow.

2. Geological background and samples

The Qinling–Dabie–Sulu orogen as a continental collision zone be-tween the South China Block (SCB) and the NCB (Fig. 1a) experienceda long period of convergence of the blocks from the Early Triassic tothe Late Jurassic (e.g., Lin et al., 1985; Li et al., 1993). During the colli-sion, the continental crust of the SCBwas subducted northward beneaththe NCB in the Triassic (e.g., Li et al., 1993). The protoliths of the UHPMrocks in the Dabie–Sulu orogen are mostly Neoproterozoic igneous bi-modal rocks with ages of ~750 Ma (Zheng et al., 2003), which experi-enced UHP metamorphism in the Triassic (242–225 Ma) (Li et al.,1989, 1992, 1993, 1994, 2000; Ames et al., 1993; Chavagnac and Jahn,1996; Hacker et al., 1998; Ayers et al., 2002; Li et al., 2004; F.L. Liuet al., 2005; Y.C. Liu et al., 2005; Liu et al., 2006; Liu et al., 2007, 2011;Liu et al., 2013). Studies on the cooling history of UHPM rocks fromthe Dabie–Sulu orogen suggest that the UHPM rock slices were ex-humed to the middle or lower crust level during the period of 226 ±3 Ma to 199 ± 3 Ma (Li et al., 2000; Li et al., 2003; F.L. Liu et al., 2008;Wang et al., 2012).

The Dabie–Sulu orogen was transected by the Tan–Lu sinistral strikeslip fault during the period from the Middle Triassic to the Early Creta-ceous with northwardmovement of the Sulu UHPmetamorphic terrane(Zhu et al., 2009; Fig. 1a). The magmatisms are different in the Dabieorogen and Sulu orogen, despite the consistent HP/UHP metamorphic

events in both terranes. In Dabie orogen, the Cretaceous magmatism in-cluding granitoids andmafic intrusions aswell asmigmatite are very de-veloped, but no magmatic activity has been observed until the EarlyCretaceous (e.g., Jahn et al., 1999; Li et al., 1999; Zhao et al., 2007a,b;Huang et al., 2008; He et al., 2011; Wang et al., 2013a; Fig. 1a). By con-trast, the Middle Triassic (225–205 Ma) syn-orogenic granitoids are de-veloped in eastern Sulu belt (Chen et al., 2003), beside the EarlyCretaceous magmatism. Recently, the Middle Triassic (219–215 Ma)and Late Jurassic (167–145 Ma) migmatites have been also recognizedin the whole Sulu belt suggesting two episodes of partial meting eventsin the Sulu orogenic crust (F.L. Liu et al., 2012; Fig. 1a).

The Bengbu area, situated ~150 km north to the Dabie orogen, is anuplifted region, in the southeasternmargin of the NCB (Fig. 1a), borderedby the Tan–Lu Fault to the east and Hefei basin to the south (Fig. 1b). Theexposedmetamorphic basement in Bengbu uplift is the Archean complexcalled as the Wuhe Group (Fig. 1b), which is mainly composed ofsupracrustal rocks, such as marble, meta-sediments and amphibolitelayers. Paleozoic sedimentary rocks are developed in its adjacent areasto the north and south (Fig. 1a). Its eastern and southern boundaries aretwo important Mesozoic tectonic zones in East China. The Tan–Lu Faulton the east is a continental scale fault in easternChina. Although the originof the Tan–Lu Fault remains controversial, it has been recently document-ed to be a syn-collisional transfer fault, which is initiated in the Middle–Late Triassic and experienced a large scale sinistral strike slip in the Juras-sic (Zhu et al., 2001, 2009). Since the Early Cretaceous, the Tan–Lu Faultwas transformed into a normal fault (Zhu et al., 2009) and became agraben-type basin covered by Cenozoic sediments (Fig. 1a). The Hefeibasin, a compressional foreland basin resulting from the convergence ofthe NCB and SCB in the Jurassic (Xue et al., 1999), is adjacent to theBengbu uplift to the south and the Tan–Lu Fault to the west (Fig. 1a).There are no Triassic deposits, but Jurassic foreland deposits occurred inthe Hefei basin (Liu et al., 2006; Zhu et al., 2010). However, the Hefeibasin was transformed into an extensional basin since Early Cretaceous,because the region was involved in younger extension during the periodof Cretaceous to Paleogene (Liu et al., 2006; Zhu et al., 2010). The Archeanmetamorphosed basement rocks (the Wuhe complex) exposed in theBengbu uplift experienced granulite-facies metamorphism in the EarlyProterozoic (1.8–1.9 Ga) accompanied by synchronous magmatism,which is comparable with other Early Proterozoic complex belts in theNCB (Guo and Li, 2009b; Liu et al., 2009).

A few episodic Mesozoic magmatisms in the Bengbu uplift havebeen identified. The Jurassic Jingshan granitic intrusion with weakgneissosity (Fig. 2) occurs in the west part of the Bengbu uplift. It in-truded into the Wuhe complex at ~160 Ma as suggested by zircon U–Pb dating data (W.L. Xu et al., 2005; Yang et al., 2010). Guo and Li(2009a) considered the Jingshan intrusion as leucogranite based on apetrochemical study, whereas others named it as either migmatiticgranite (W.L. Xu et al., 2005), or biotite–syenogranite (Yang et al.,2010), or monzogranite (Wang et al., 2010). The nomination ofleucogranite is mainly based on the very low Mg–Fe contents anddark-colored mineral abundance (see details in below) in Jingshan in-trusion. The intrusion is composed of gneissic leucogranite with a fewbiotite–garnet-rich restite and leucoaplite veins (Guo and Li, 2009a).Similar leucogranites have been identified elsewhere in the Tushan,Mayishan (Guo and Li, 2009a), and Laoshan (Wang et al., 2009) intru-sions, which together with the Jingshan intrusion comprise an NWW–

SEE trending leucogranitic belt in the Bengbu uplift and we call themas Bengbu leucogranites (Fig. 1b). Previous studies only reported zirconU–Pb ages of the Jingshan intrusion (W.L. Xu et al., 2005; Li et al., 2010;Wang et al., 2010; Yang et al., 2010; Wang et al., 2013a, 2013b) andLaoshan intrusion (Wang et al., 2009, 2013b). Whether or not theleucogranites in Tushan and Mayishan have the similar Jurassic intru-sive ages keeps unknown.

TheCretaceous granitic intrusions, including granodiorites and gran-ites, are distributed on the two sides of the Jurassic leucogranitic belt inthe Bengbu uplift (Fig. 1b; Yang et al., 2010; S.A. Liu et al., 2012).

1112 S. Li et al. / Gondwana Research 26 (2014) 1111–1128

Author's personal copy

Diabase, granite porphyry and diorite NNE-trending dykes cuttingacross the Jingshan leucogranite and some granodiorites are well de-veloped (Fig. 3; Yang et al., 2010; S.A. Liu et al., 2012). In contrast tothe Late Jurassic leucogranites, geochronological and geochemical

investigations of these Cretaceous granitic and mafic intrusionsshow that their inherited zircons and geochemical features are sim-ilar to those of the Archean–Paleoproterozoic basement rocks andmafic intrusions developed in the NCB but different from the UHPM

Fig. 1. a, Structural map showing the space relationship between the Bengbu Uplift, the Dabie–Sulu orogen and Hefei basin aswell as the Tan–Lu Fault (modified after Zhu et al., 2009; F.L.Liu et al., 2012), with the inset showing location of this area in east China. C–C′ shows location of the transection in Fig. 13. b, Geological map of the Bengbu Uplift showing the Jingshan,Tushan and Mayishan leucogranites intruded into the Archean Wuhe complex, the basement rocks of the North China Block (NCB).

1113S. Li et al. / Gondwana Research 26 (2014) 1111–1128

Author's personal copy

rocks in the Dabie–Sulu orogen (Yang et al., 2010; S.A. Liu et al.,2012).

Here, thirty-eight samples were collected from the Jingshan, Tushanand Mayishan intrusions in Bengbu uplift. The leucogranite fromJingshan are composed of quartz (~30 vol.%), orthoclase (25–32 vol.%),plagioclase (35–40 vol.%), and minor biotite (1–3 vol.%) and garnet(b1 vol.%). The garnets in Jingshan leucogranites are subdivided intothree types based on their textural features and chemical compositions(Xu et al., 2012). The Grt I is of peritectic garnet, whereas Grt II and IIIare probably the results of magmatic dissolution–precipitation process-es. The very low abundance of dark-colored minerals suggests that theJingshan intrusion is leucogranite (Nabelek et al., 1992; Clarke et al.,1993; Inger and Harris, 1993; Nabelek and Bartlett, 1998; van deFlierdt et al., 2003; Guo and Li, 2009a). The Tushan and the Mayishanleucogranites have a similar mineral assemblage with the Jingshanleucogranite but contain even few dark-colored minerals.

3. Analytical methods

Samples were carefully prepared to avoid contamination each otheror from tools and environment (See detail in Supplement 1 for samplepreparing method).

3.1. Bulk-rock major and trace elements

Major elementswere analyzed by the X-ray fluorescence (XRF) spec-trometer at the Institute of Geophysical and Geochemical ProspectingTechniques (IGGPT). FeO content was determined by wet chemistryand LOI (Loss on Ignition) was measured by gravimetric methods atthe IGGPT, respectively. Trace element concentrations of bulk sampleswere analyzed by using ICP-MS at the Key Laboratory of Continental Dy-namics of the Northwest University in Xi'an, China. For trace elementalanalysis, about 20–50 mg bulk-rock powder was dissolved in a Teflon

cap-screw bomb using super-pure HNO3 and HF at a temperature of190 °C for 48 h. The analyses were monitored by international standardreferences AGV-2, BCR-2 and BHVO-2, which shows that the analyticalerrors are less than 10%. The detailed procedure of analyses has been de-scribed inGao et al. (1999). Bulkmajor and trace elements data are listedin Supplementary Table S1.

3.2. Zircon U–Pb dating

Both optical photomicrographs and cathodoluminescience (CL) im-ages were taken for the selection of U–Pb dating spots. Internal zoningpatterns were recognized by CL images (Fig. 7), which were obtainedat the Institute of Mineral Resources, Chinese Academy of GeologicalSciences (CAGS).

Zircon U–Pb dating was performed by using the SHRIMP II at theBeijing SHRIMP Center, CAGS (Table 1) and Cameca IMS-1280 SIMS atthe Institute of Geology and Geophysics, Chinese Academy of Sciencesin Beijing (Table 2), respectively. Analytical procedures have been de-scribed in detail in Compston et al. (1992) and Williams (1998) forSHRIMP dating and Li et al. (2009) for SIMS dating. Data processingwas carried out using the Squid and Isoplot programs (Ludwig, 2001).Measured 204Pb was applied for the common lead correction. TheSHRIMP data and SIMS data are listed in Supplementary Tables S2 andS3, respectively. The errors given in Supplementary Tables S2 and S3for individual analyses are quoted at the 1σ level, whereas the errorsfor weighted mean ages are quoted at 2σ (95% confidence level).

3.3. Mineral inclusions in metamorphic zircon

Mineral inclusions were identified using the Raman spectroscopy atthe Analytical Centre, ChinaUniversity of Geosciences (Wuhan) and theContinental Dynamics Laboratory, CAGS, and/or substantiated using theelectron microprobe analyzer (EMPA) at the Analytical Centre, China

Fig. 2. Representative field photos of Jingshan pluton. (a), magmatic flow structure defined by the schlieren layering composed of restitic biotite and garnet; (b), local magmatic flow di-rection defined by the aligned biotite schlieren; (c) two stages of deformations recorded by the S-C fabric, showing that the D1 is typically magmatic and cut by the local shear plane de-fined by D2; (d), two stages of deformation recorded by the folds in Jingshan pluton. The leucogranite have foliation (D1) with strike of NW–SE, while the schlieren layer was folded bylater deformation (D2) with its axle strike NE.

1114 S. Li et al. / Gondwana Research 26 (2014) 1111–1128

Author's personal copy

University of Geosciences (Wuhan). The analytical conditions forRaman and EMPAwere the same as reported by S. Xu et al. (2005). Rep-resentative Raman spectra and electron microprobe analysis of mineralinclusions in zircon from Jingshan leucogranite are shown in Fig. 9b andTable 1.

3.4. Zircon trace element analyses

Zircon trace element analyses were performed by the LA–ICP–MS atthe State Key Laboratory of Continental Dynamics at the Northwest

University in Xi'an and the State Key Laboratory of Geological Processesand Mineral Resources, China University of Geosciences in Wuhan, re-spectively. Analytical spot sizes were 30 μm in diameter. Oxide produc-tion ratewas tuned to b0.5% ThO. TheNIST610 reference glasswas usedfor calibration, with working values recommended by Pearce et al.(1996). The average analytical error ranges from ca. ±10% for lightrare earth elements (LREE) to ca. ±5% for the other REE. Detail operat-ing conditions for the laser ablation system and the ICP-MS instrumentand data reduction follows Yuan et al. (2004) for analyses at theNorthwest University and Y.S. Liu et al. (2008) for those at the ChinaUniversity of Geosciences in Wuhan, respectively. Trace element dataof zircon are listed in Table 2 and Supplementary Table S4, and theirREE patterns are shown in Fig. 9c.

3.5. Sr–Nd–Pb isotope analysis

Sr–Nd–Pb isotopic ratios were measured on a Finnigan MAT-262thermal ionization mass spectrometer (TIMS) at the Laboratory for Ra-diogenic Isotope Geochemistry, Institute of Geology and Geophysics ofCAS, Beijing. Chemical procedures for Sr, Nd and Pb isotope analyseshave been described in Supplement 2. Procedural blanks were b200 pgfor Sr and≤50 pg for Pb and Nd. For themeasurements of isotopic com-positions, 143Nd/144Nd ratios were normalized to 146Nd/144Nd = 0.7219and 87Sr/86Sr ratios to 86Sr/88Sr = 0.1194. Pb standardNBS 981was usedto determine thermal fractionation. Measured isotopic ratios of sampleswere corrected with a value of 0.1% per atomic mass unit. Analyses of

Fig. 3.Geological mapwith fabric patterns in of Jingshan pluton. The solid line in the stereographic projection represents the D1 foliation, whereas the dotted line represents the D2 foliation.

Table 1Chemical compositions of biotite andphengicite as inclusions in the inheritedmetamorphiczircons from the Bengbu leucogranite.

Oxides (wt.%) js-1-1 js-1-2 js-1-3 Cations js-1-1 js-1-2 js-1-3

SiO2 36.76 37.37 54.63 Si 2.62 2.71 3.58TiO2 3.16 2.87 0.10 AlIV 1.26 1.21 0.42Al2O3 14.90 14.23 29.33 AlVI 1.85∑FeO 23.87 21.69 0.80 Ti 0.17 0.16 0.01MnO 0.67 0.85 0.11 Fe3+ 0.04MgO 6.54 7.14 – Fe2+ 1.43 1.31Na2O 0.07 0.34 0.18 Mn 0.04 0.01K2O 9.58 9.19 7.33 Mg 0.70 0.77 –

H2O⁎ 3.85 3.81 Na 0.01 0.05 0.02Total 99.40 97.49 92.48 K 0.87 0.85 0.61

Calculatedbased on O = 11

OH 2.00 2.00 2.00

⁎ Calculated H2O.

1115S. Li et al. / Gondwana Research 26 (2014) 1111–1128

Author's personal copy

standards during the analytical period are as follows: NBS987 gave87Sr/86Sr = 0.710254 ± 12 (n = 27, 2σ); JMC and AMES gave143Nd/144Nd = 0.511987 ± 7 (n = 8, 2σ) and 0.512145 ± 12(n = 15, 2σ), respectively. Details of chemical separation and mea-surement are described in Chen et al. (2000). Sr and Nd isotope analyt-ical results are listed in Table 3. Pb isotopic compositions of the samplesare listed in Table 4.

4. Results

4.1. Field geology of leucogranitic intrusions

The Jingshan leucogranite is characterized by weak gneissosity andthus is called as “migmatitic granite” (W.L. Xu et al., 2005; Wang et al.,2010; Yang et al., 2010). Wang et al. (2010) suggested that the “weakgneissosity” of Jingshan leucogranite could be a result of deformationfrom the movement of the Tan–Lu Fault. Nevertheless, no detailedworks so far on fabric of the gneissosity in Jingshan leucogranite havebeen carried out. In order to answer the question that whether theweak gneissosity indicate a subsolidus deformation of the Jingshan

leucogranite after its emplacement or alternatively, a flow foliationresulted from magmatic flow during its emplacement, a field foliationmapping on the Jingshan leucogranite has been conducted.

Our recent field mapping revealed that most outcrops of the plutonshow a moderate to strong foliation with weakly developed lineation.Based on field observation, two stages of deformation (D1 and D2)have been recognized (Fig. 2). Foliation of the first deformation (D1) iswell defined by the planar disposition of biotite, schlierens of restitic bi-otite and garnet (Fig. 2a, b), and somewhere the elongated recrystallizedquartz assemblages. Away from the wall rock (the Wuhe amphibolite),D1 strikes broadly NW–SE with a steep dip (Fig. 3) and lacks significantcrystalloblastic deformation (e.g., plagioclase preserve the magmaticzoning), indicating that the foliationwas acquired by pervasive deforma-tion of the magma before completely crystallized. This interpretation isfurther supported by the zircon U–Pb results got from the gneissicleucogranite and aplite veins (see later discussion). However, whereclose to the wall, D1 shows some solid-state deformation (e.g., recrystal-lization of quartz and plagioclase), and the strike is nearly parallel to thewall (Fig. 3). This fabric pattern is interpreted as that, with the ongoingascent and expansion of the magma flow, the foliation at the margin of

Table 2Trace element compositions (ppm) of the inherited metamorphic zircons from the Bengbu leucogranite.

Sample JS-1-8 JS-1-16 JS-1-17 JS-1-26 JS-1-19 JS-1-24 JS-1-28 JS-1-31

La b0.01 0.01 0.01 0.01 0.02 b0.01 b0.01 b0.01Ce 9.65 7.68 39.33 2.06 7.56 3.41 5.86 5.62Pr 0.02 0.00 0.03 0.00 0.02 0.01 0.00 0.00Nd 0.25 0.06 0.89 0.10 0.08 0.05 0.10 0.11Sm 0.39 0.26 2.39 0.13 0.47 0.14 0.89 0.29Eu 0.18 0.07 1.24 0.09 0.12 0.13 0.23 0.11Gd 2.73 2.38 15.68 1.22 2.25 0.88 6.34 2.70Tb 1.27 1.29 6.67 0.72 1.17 0.27 3.42 1.35Dy 19.6 21.3 92.4 14.4 19.1 2.55 54.9 23.1Ho 9.64 10.3 41.5 7.83 9.37 0.70 26.2 11.5Er 59.6 62.8 220 50.7 56.3 2.35 148 68.8Tm 19.1 19.3 60.3 16.8 17.3 0.55 42.7 22.1Yb 267 256 732 233 223 4.91 553 291Lu 55.8 48.4 138 43.3 43.2 0.73 98.9 55.6∑REE 446 430 1351 371 380 16.7 941 482Nb 25.4 24.5 31.4 5.29 29.5 0.83 79.3 22.4Ta 10.8 9.98 7.68 2.36 13.7 0.36 50.5 8.94Y 360 368 1326 299 334 23.3 861 404Hf 12,725 15,243 9719 12,148 12,026 12,016 12,038 14,612Sc 506 461 597 445 513 545 409 481Rb 0.22 0.26 0.57 0.25 0.15 0.04 0.44 0.22Sr 0.33 0.35 0.67 0.31 0.43 0.17 0.62 0.41Th 64.6 22.8 53.6 9.55 54.4 18.5 31.8 16.7U 826 406 619 598 715 247 755 348

Table 3Sr and Nd isotopic compositions of the Bengbu leucogranite.

Rock type Rb (ppm) Sr (ppm) 87Rb/86Sr 87Sr/86Sr 2 se 87Sr/86Sr(t) Sm (ppm) Nd (ppm) 147Sm/144Nd 143Nd/144Nd 2 se 143Nd/144Nd(t) εNd(t)

06JS-4 Leucogranites 115.2 413.8 0.8056 0.710467 15 0.708579 0.99 4.41 0.1354 0.511861 10 0.511713 −13.906JS-5 Leucogranites 111.9 382.5 0.8467 0.710487 12 0.708502 0.84 3.79 0.1342 0.511891 13 0.511744 −13.306JS-6 Leucogranites 115.5 486.6 0.6868 0.710311 14 0.708701 0.65 2.42 0.1611 0.511931 30 0.511755 −13.006JS-8 Leucogranites 118.6 424.3 0.8087 0.710385 13 0.708489 1.03 4.22 0.1472 0.511898 12 0.511737 −13.406JS-9 Leucogranites 100.3 386.3 0.7512 0.710360 15 0.708599 0.65 2.7 0.1450 0.511916 12 0.511757 −13.006JS-10 Leucogranites 119.4 410.2 0.8421 0.710461 11 0.708487 0.85 3.61 0.1431 0.511883 13 0.511726 −13.606JS-11 Leucogranites 125.8 244.8 1.4875 0.712421 14 0.708934 0.54 1.92 0.1715 0.511945 15 0.511758 −13.006JS-13 Leucogranites 104.8 413.8 0.7328 0.710267 12 0.708549 0.86 3.63 0.1424 0.511892 12 0.511736 −13.406JS-14 Leucogranites 110.8 154.9 2.0703 0.713364 14 0.708511 1.29 5.64 0.1380 0.511878 14 0.511727 −13.606TS-2 Leucogranites 132.7 324.0 1.1853 0.712328 13 0.709549 0.72 2.30 0.1891 0.511865 108 0.511653 −14.906TS-3 Leucogranites 114.7 149.3 2.2242 0.714425 15 0.709211 1.26 3.01 0.2534 0.511979 13 0.511653 −14.106TS-4 Leucogranites 131.3 315.6 1.2044 0.712106 12 0.709283 1.05 4.08 0.1553 0.511909 14 0.511653 −13.306TS-5 Leucogranites 241.8 375.5 1.1636 0.711737 13 0.709009 0.64 2.69 0.1433 0.511915 29 0.511653 −13.006MYS-1 Leucogranites 77.3 128.1 1.7466 0.714524 10 0.710430 0.94 2.71 0.2100 0.511981 12 0.511751 −13.106MYS-2 Leucogranites 72.0 172.5 1.2094 0.714525 12 0.711690 0.73 1.98 0.2219 0.511988 11 0.511745 −13.206MYS-3 Leucogranites 151.3 228.6 1.9162 0.713451 11 0.708959 0.46 1.75 0.1592 0.511926 15 0.511752 −13.106MYS-4 Leucogranites 110.5 327.1 0.9777 0.711155 15 0.708864 1.09 3.63 0.1815 0.511941 13 0.511743 −13.3

Notes: Parameters for calculation of εNd(t) and initial isotopes: (147Sm/144Nd)CHUR = 0.1967, (143Nd/144Nd)CHUR = 0.512638; λRb = 1.42 × 10−11 year−1, λSm = 6.54 × 10−12 year−1;t = 167 Ma.

1116 S. Li et al. / Gondwana Research 26 (2014) 1111–1128

Author's personal copy

magma around the wall will be controlled by the interface betweenthem. D2 is locally developed as small shear zones across the plutonand cut earlier D1 (Fig. 2c, d). The foliation of D2 strikes broadly NE–SW with the steep dips (Fig. 3). The wall rocks also record D2 on thefield. The orientation of D2 is parallel to the NE strike of the Tan–LuFault. Therefore, this observation proved an additional evidence forthat the sinistral strike slip Tan–Lu Fault was still active in the Late Juras-sic after the emplacement of the Jingshan intrusion.

The south boundaries of the Jinshan and Tushan intrusions are ex-posed showing magma intrusive relationships between the intrusionsand their wall rocks (amphibolite and gneiss of the Wuhe complex;Fig. 4). In contact zone, the branches of leucogranite occur as distinctdykes and veins emplaced subparallel (Fig. 4a) and perpendicular(Fig. 4b) to the foliation of the amphibolite inWuhe complex. This intru-sive relationship suggests that the Bengbu leucogranites are intrusionsnot migmatite. The amphibolites-facies metamorphism of the wallrocks suggests that the Bengbu leucogranite intrusions were emplacedinto the middle crust level in Late Jurassic.

4.2. Major and trace element compositions of whole rocks

Analytical results of major and trace elements of the Bengbuleucogranites are listed in Supplementary Table S1. The total 38 samplesfrom the Jingshan, Tushan and Mayishan intrusions show relativelyhigh SiO2 (71.40–75.44 wt.%), Al2O3 (13.41–15.33 wt.%) and Na2O +K2O (7.38–9.64 wt.%) contents, regular or slightly high CaO (0.55–2.28 wt.%) contents, but low Fe2O3 + FeO (≦1.20 wt.%), MgO(≦0.27 wt.%) and TiO2 (≦0.11 wt.%) concentrations (SupplementaryTable S1), consistent with the low ferromagnesian mineral abundancein the rocks. All samples plotted in granite field in the Na2O + K2O–SiO2 diagram with relative high Na2O + K2O content (Fig. 5a). Theyshow lowMg# (12–40) similar to those of High Himalaya leucogranites(HHL), Kerala Khondalite leucogranites in Southern India and a part of

South Mountain Batholith (SMB) leucogranites in Canada (Fig. 5b). Nocorrelation between MgO + FeOT and SiO2 contents has been observed(Fig. 5c), suggesting that their lowMgO + FeOT contents are not a resultof fractional crystallization of ferromagnesian minerals. Their averageAl2O3 contents are lower than those of the peraluminous High Himalayaleucogranites (Fig. 5d) and their Al-saturation indexes [ASI = mol.Al2O3/(CaO + K2O + Na2O)] range from 0.92 to 1.09 (SupplementaryTable S1), together suggesting that they are metaluminous granites.Their higher and variable Sr (124–704 ppm) and much lower Rb (54–154 ppm) contents and Rb/Sr ratio are different from those of theperaluminous High Himalaya leucogranites (HHL) and the South Moun-tain Batholith leucogranites (SMB) but are similar to those of the KeralaKhondalite leucogranites in Southern India (Fig. 5e, f).

The most notable feature of whole rock trace elements is the lowLREE, Th and Ti contents of the Bengbu leucogranites, resulting into spe-cial negative Th, La–Ce, Nd and Ti anomalies in the spider diagram(Fig. 6b). Their REE patterns are basically flat with concave and haveno significant Eu anomalies (Fig. 6a). In addition, their whole rock Zr–Hf contents are low although the Bengbu leucogranites contain abun-dant inherited zircons.

4.3. Zircon U–Pb chronology

Zircon CL images reveal that most zircons of the leucogranite sam-ples have core–mantle–rim structures (Fig. 7). The inherited coresshow clear oscillatory zoning, suggesting their magmatic origin(Rubatto and Gebauer, 2000). The mantle zones show homogeneousCL intensity, which are typical features for metamorphic zircons (Liuet al., 2006). Overgrowth rims show again clear oscillatory zoning, indi-cating a late stage magmatic crystallization.

The SHRIMP II zircon U–Pb dating results obtained in the early time(Supplementary Table S2) show that inherited zircon cores from theJingshan, Mayishan and Tushan intrusions give 206Pb/238U ages rangingfrom287 to 750 Ma, and define the discordant lineswith upper interceptages of 833 ± 98 Ma and 759 ± 31 Ma, respectively, for the Tushan andMayishan intrusions (Fig. 7), in agreement with the protolith age ofUHPM rocks from the Dabie–Sulu orogen. The discordant line definedby a few inherited cores of zircons from the Jingshan intrusion gives apoor-constrained but comparable upper intercept age of 785 ±260 Ma (Fig. 7).

Themantle zones of most inherited zircons from the Jingshan andTushan intrusions show ages ranging from 243.7 ± 6.6 Ma to202.9 ± 6.4 Ma, which define weighted mean 206Pb/238U ages of224.9 ± 6.2 Ma (MSWD = 4.1, n = 17) and 222.1 ± 7.1 Ma(MSWD = 5.1, n = 13) (Fig. 7), respectively. Only three individualages ranging from 211.4 ± 3.8 Ma to 195.2 ± 4 Ma for inherited man-tle zircons from Mayishan were obtained (Supplementary Table S2),those overlap the ages of the Jingshan and Tushan intrusions.

Table 4Pb isotopic compositions of the Bengbu leucogranite.

Sample Rock type Mineral 206Pb/204Pbi 207Pb/204Pbi 208Pb/204Pbi

06JS-4 Leucogranites K-feldspar 17.036 15.448 37.53706JS-5 Leucogranites K-feldspar 17.015 15.422 37.45106JS-6 Leucogranites K-feldspar 17.018 15.417 37.43506JS-10 Leucogranites K-feldspar 17.021 15.435 37.49506JS-11 Leucogranites K-feldspar 16.999 15.384 37.33606TS-2 Leucogranites K-feldspar 17.086 15.396 37.39606TS-3 Leucogranites K-feldspar 17.083 15.427 37.49506TS-4 Leucogranites K-feldspar 17.043 15.429 37.48206TS-5 Leucogranites K-feldspar 17.048 15.417 37.44706MYS-1 Leucogranites K-feldspar 17.122 15.440 37.54706MYS-2 Leucogranites K-feldspar 17.119 15.416 37.47106MYS-3 Leucogranites K-feldspar 17.129 15.428 37.51006MYS-4 Leucogranites K-feldspar 17.108 15.440 37.550

Fig. 4. Representative photos showing intrusive relationship between the Tushan leucogranite and wall rocks (the Wuhe complex). a, Leucogranite veins penetrate the amphibolite inWuhe complex subparallel to its foliations. b, A branch dyke of leucogranite intrude into the Wuhe complex perpendicular to the foliation of the wall rocks.

1117S. Li et al. / Gondwana Research 26 (2014) 1111–1128

Author's personal copy

The magmatic overgrowth rims of zircons from the Jingshanintrusion define a weighted mean 206Pb/238U age of 167.3 ± 5.8 Ma(MSWD = 2.9, n = 12; Fig. 7), representing the crystallization time ofthe leucogranite. Zircon rims from the Tushan and Mayishan intrusionsalso indicate late Jurassic intrusion time from 167 to148 Ma.

The Th–U diagram in Fig. 7 shows that most Triassic inherited zir-cons from the Bengbu leucogranites, including those of this study andthose published in literatures (W.L. Xu et al., 2005; Wang et al., 2010;Yang et al., 2010), have Th/U ratios lower than 0.1, comparable withthe metamorphic zircons from UHPM rocks of the Dabie–Sulu orogen.

Fig. 5. a, Total alkali versus SiO2 diagram showing that the Jingshan, Tushan andMayishan intrusions are granites; b, Mg# versus SiO2 diagram showing that the Mg# values of the Bengbuleucogranites are lower than those of SouthMountain Batholith leucogranites (SMB) and similar to those of HighHimalaya leucogranites (HHL) andKeralaKhondalite leucogranite, aswellas experimental melts of natural hydrous basalts at 1–4 GPa (Rapp et al., 1999), but much lower than those of high-magnesian melt derived from subducted slab (HMAs = High mag-nesium andesite from Setouchi, Japan; HMAa = High magnesium andesite fromW. Aleutians; SACA = S. Andes, Cook Isle, adakite); c and d, MgO + FeOT versus SiO2 and Al2O3 versusSiO2 diagrams (modified after Guo and Li, 2009a) document that the major element compositions of the Bengbu leucogranites are comparable with those of the HHL, SMB and KeralaKhondalite leucogranites; e and f, Rb/Sr versus Sr and Sr versus Rb diagrams showing higher Sr, lower Rb contents and lower Rb/Sr ratios of the Bengbu leucogranites than the HHLand SMB. See Supplementary Table S1 for data sources. The data of the HHL, SMB and Kerala Khondalite leucogranites are from Guo and Li (2009a) and references thereafter, Clarkeet al. (1993) and Braun et al. (1996). The Jingshan leucogranite labeled by the black circles are from present study. Other leucogranites in Bengbu uplift are from Guo and Li, 2009a.

1118 S. Li et al. / Gondwana Research 26 (2014) 1111–1128

Author's personal copy

In contrast, most magmatic overgrowth zircon rims from the Bengbuleucogranites have higher Th/U ratios (N0.1) comparable to those ofpost-collisional granites from the Dabie–Sulu orogen (Fig. 7). Fromthis viewpoint, we suggest that the Triassic inherited zircons in theBengbu leucogranites are of metamorphic origin related to the Triassiccontinental subduction of SCB.

Aiming to identify whether foliation D1 resulted from the defor-mation after Jingshan leucogranite emplacement or alternatively, bymagmatic flow during its emplacement, we did compare SIMS zir-con U–Pb dating of the foliated leucogranite (09JS-4) and the un-deformed leucoaplite vein (09JS-3) cutting the foliation of theleucogranite (Supplementary Table S3, Fig. 8). Two magmatic crys-tallization ages of 165.9 ±1.4 Ma and 162.9 ± 5.6 Ma for the foliat-ed leucogranite and the un-deformed leucoaplite were obtained,respectively (Fig. 8), which overlap within error limits. In addition,their inherited zircons also yield similar lower intercept ages of216 ± 15 Ma and 214 ± 16 Ma and upper intercept ages of825 ± 38 Ma and 770 ± 37 Ma (Fig. 8). The age consistency sug-gests that the foliated leucogranite and the un-deformed leucoapliteare products of the samemagma event, and that the latter could be amagma segregate vein formed in a shortly later stage. Therefore, D1should be a magmatic foliation formed in an early magma stage inwhich the Jingshan leucogranite emplaced into the present positioninstead of a subsolidus deformation after its emplacement.

4.4. Mineral inclusions and trace elements in the metamorphic zircons

Felsic mineral inclusions similar to those observed from Dabie–Sulugneisses, such as quartz, feldspar, apatite, titanite (Fig. 9a, b), garnet, bi-otite, muscovite and phengite (Si = 3.58; Table 1) have been identifiedin the inherited metamorphic zircons by Raman spectroscopy and elec-tron microprobe analyses. However, no UHP metamorphic or eclogite-facies mineral inclusions, such as coesite, omphacite and rutile, havebeen observed.

Trace elements of metamorphic zircons have been analyzed. Nine often analyses show steep heavy rare earth element (HREE) patternswithclear negative Eu anomalies (Fig. 9c), and high Nb (5.29–79.31 μg/g)and Ta (2.36–50.52 μg/g) contents (Table 3). On the other hand, oneanalysis shows a flat HREE pattern without Eu anomaly (Fig. 9c) andlow Nb and Ta contents (0.83 and 0.36 μg/g, respectively; Table 2).

4.5. Bulk-rock Sr–Nd–Pb isotopic compositions

The Bengbu leucogranites have homogeneous initial εNd(t) valueranging from−13.0 to−14.9 and large spread of initial 87Sr/86Sr valuesranging from 0.708487 to 0.711690 (Table 3). Except one sample withthe highest 87Sr/86Sr value plotted in the gneiss field, most samplesoverlap the fields of both gneiss and eclogite from the Dabie orogen(Fig. 10a). However, the common Pb data (Table 4) show that, for agiven 206Pb/204Pb ratio, most of K-feldspar samples from the Bengbuleucogranites have relatively high 207Pb/204Pb ratios that overlap thefields of gneisses but are distinct from those of eclogites from theDabie–Sulu orogen (Fig. 10b). These isotopic features suggest that theprotolith of the Bengbu leucogranites is subducted felsic gneisses ratherthan eclogite, consistentwith the conclusion drawn frommineral inclu-sions and trace element features of the inherited metamorphic zircons.

5. Discussion

5.1. Origin of the Bengbu leucogranites

This section will focus on the following issues: (1) what is the li-thology of their protolith, (2) where is the source rock derived from,(3) whether the source rock experienced UHP metamorphism or not,and (4) partial melting condition.

5.1.1. Protolith lithologyProtolith of leucogranites could be meta-sediments or meta-igneous

rocks, which may influence the major and trace element compositionsof leucogranites and melting condition. It is generally believed that thewidespread peraluminous leucogranites in the Himalayan orogenic beltare derived from the anatexis of metasediments of the subductedIndian slab (the High Himalayan Crystalline Series; e.g., Inger andHarris, 1993; Visona and Lombardo, 2002). While some other leuco-granites could be the products of partial melting of felsic orthogneiss(e.g., the Kerala Khondalite leucogranites in Southern India; Braun et al.,1996), or extensive fractional crystallization of granitic melt or late mag-matic and post-magmatic hydrothermal alteration (e.g., leucogranitesfrom the SouthMountain Batholith; Clarke et al., 1993). Therefore, a com-parison of major and trace elemental characteristics of the Bengbuleucogranites with those of the High Himalaya leucogranites (HHL) andSouth Mountain Batholith leucogranites (SMB) as well as the KeralaKhondalite leucogranites should be helpful for understanding theprotolith lithology of the Bengbu leucogranites.

Fig. 5c and d shows that MgO + FeOTotal, Al2O3 and SiO2 data of theBengbu leucogranites are mainly plotted in the HHL and SMBleucogranite fields and comparable with the Kerala Khondalite leuco-granites, though the HHL and SMB have somewhat higher Al2O3 andMgO + FeOTotal contents. The difference of Al2O3 contents betweenthe peraluminous HHL and SMB leucogranites and the metaluminousBengbu leucogranites (Fig. 5d) could be related to their different

Fig. 6. Chondrite-normalized REE patterns (a) and primitivemantle-normalized trace ele-ment spider diagrams (b) of the Bengbu leucogranites show that the leucogranites havemuch lower REE, Th and Ti contents than the Dabie–Sulu gneisses. Data for the Dabie–Sulu gneisses are from Li et al. (2000), Bryant et al. (2004), Zhao et al. (2007a, 2007b),Tang et al. (2008) and Xia et al. (2010).

1119S. Li et al. / Gondwana Research 26 (2014) 1111–1128

Author's personal copy

protoliths. Petrological studies indicate that the Rb-rich HHL leuco-granites was produced by muscovite dehydration partial melting ofmica-rich metapelites at relatively low temperatures (750–770 °C atP = 6–8 kbar; Patino Douce and Harris, 1998), while the Rb-rich SMBleucogranites were produced by interaction between a fluid and itshosting leuco-monzogranite or a highly fractionated melts that under-went late-magmatic to post-magmatic fluid alteration. In contrast, themetaluminous Bengbu leucogranites could be derived from the partialmelting of mica-poor orthogneiss, which is supported by their high Srand low Rb features. Fig. 5e and f shows that the metaluminous Bengbuleucogranites are characterized by higher Sr, lower Rb contents andlower Rb/Sr ratio compared to the HHL and SMB leucogranites (Fig. 5e,f). Furthermore, the absence of correlation between Sr and Rb contentsof the Bengbu leucogranites (Fig. 5f) suggests that their Rb content is in-dependent of plagioclase abundance. It is also interesting to note that theKerala Khondalite leucogranites in South India derived frompartialmelt-ing of orthogneiss also have features of higher Sr, lower Rb content andlower Rb/Sr ratio (Braun et al., 1996) (Fig. 5e, f). Therefore, unlike the

high Rb and low Sr peraluminous leucogranite represented by HighHimalaya leucogranites (HHL), the Bengbu low Rb and high Srmetaluminous leucogranites are derived from mica-poor orthogneiss.

5.1.2. Where were the source rocks derived from?Zircon U–Pb dating results clearly show that zircons from the Bengbu

leucogranites are totally different from those of their wall rocks (theWuhe complex). The latter gives much older metamorphic ages of1800 ± 15 Ma to 1918 ± 56 Ma and magmatic crystallization ages are2054 ± 22 Ma (Guo and Li, 2009b; Liu et al., 2009), which are similarto the metamorphic basement rocks in the NCB (Zhao et al., 2001).Therefore, the zircons with inherited ~750 Ma igneous cores and~225 Ma metamorphic mantles cannot be resulted from assimilation ofthe wall rocks but should be derived from the source rock of the Bengbuleucogranites. The consistency between the ages of inherited cores/man-tles of zircons from the Bengbu leucogranites and those from the UHPMrocks in the Dabie–Sulu orogen, suggests that the source rock of theBengbu leucogranites could be pieces of subducted continental crust of

Fig. 7. The zircon CL images show core (a)—mantle (b)— rim (c) inner-structure with spot locations and agesmarked. The zircon U–Pb age diagrams show zircon U–Pb ages for Jingshan(04JS-1), Tushan (04TS-2), andMayishan (04MYS-13) leucogranite intrusions, respectively. The zircon U–Pb ages were determined by SHRIMP II (Table 1) (Data are listed in Supplemen-tary Table S2). The zircon Th versus U diagram shows comparison between zircons from Bengbu leucogranites and Dabie–Sulu UHPM rocks as well as Cretaceous granitic intrusions (Seetext for detail explanation). Data sources: Th–U contents for Jingshan zircon domains are from this study (Table 1),W.L. Xu et al. (2005) and Li et al. (2010). Th–U contents for Dabie granitezircon are from Bryant et al (2004), Hacker et al. (1998), Huang et al. (2008), Wang et al. (2007), Xu et al. (2007), Zhao et al. (2007a, 2007b); Th–U contents for UHP zircon in Dabie–Suluare from Liu et al. (2006), Liu et al. (2007, 2011), Xia et al. (2010), Chen et al. (2010), Wang et al. (2012).

1120 S. Li et al. / Gondwana Research 26 (2014) 1111–1128

Author's personal copy

the SCB. The recent report of extremely low and highly heteroge-neous δ18O values for the inherited Triassic metamorphic zirconsfrom the Bengbu leucogranites, consistent with the values and vari-ations exhibited by metamorphic zircons from Dabie–Sulu orogen(Wang et al., 2013b), also supports this conclusion.

Moreover, the consistency between themagma crystallization age of167–148 Ma for the Bengbu leucogranite and migmatization time of167–145 Ma in Sulu orogen (F.L. Liu et al., 2012), and the absence of Ju-rassic magmatism in Dabie orogen suggest that the source rocks of theBengbu leucogranite is most likely to be derived from Sulu orogen in-stead of Dabie orogen.

5.1.3. Did the source rocks of the Bengbu leucogranites experience UHPmetamorphism?

This is another issue on the origin of the Bengbu leucogranites, be-cause the previous model for the origin of the Jingshan leucograniteproposed byYang et al. (2010) supposes a UHPM source rocks butwith-out evidence. Mineral inclusion assemblage inmetamorphic zircons is agood indicator for revealing zircon formation environments (e.g., F.L.Liu et al., 2005; Y.C. Liu et al., 2005; Liu et al., 2006; F.L. Liu et al.,2008; Wang et al., 2012). Studies on the Dabie–Sulu UHPM rocks sug-gested that coesite- or diamond-bearing UHP minerals, like coesite ordiamond + garnet + omphacite + rutile + phengite, are present asinclusions in UHP metamorphic zircons with ages of 238–225 Ma,whereas only amphibolite-facies mineral assemblage, such asamphibole + plagioclase + apatite, were identified as inclusionsin retrograde metamorphic zircons with ages of 219–199 Ma (Liu

et al., 2006, 2007; F.L. Liu et al., 2008; Wang et al., 2012). Our studyindicates that, only quartz, feldspar, apatite, titanite, biotite, musco-vite and phengite (Si = 3.58; Table 3) are present as mineral inclu-sions in the inherited metamorphic zircons from the Bengbuleucogranites. Among them, phengite is a HP metamorphic mineral,indicating that the source rocks might have experienced HPmetamorphism.

Trace elements of metamorphic zircons have been broadly used totrace coexisting minerals and thus metamorphic conditions (Bingenet al., 2004; Harley et al., 2007; Rubatto, 2002; Rubatto et al., 2009;Wang et al., 2012). Nine of ten analyses show steep heavy rare earth el-ement (HREE) patterns, suggesting that most of the metamorphic zir-cons crystallized in a garnet-poor environment (Rubatto, 2002). Inaddition, most of the steep HREE patterns show clear negative Eu anom-alies (Fig. 9c), suggesting that most of the metamorphic zircons crystal-lized in a plagioclase-coexisting environment. On the other hand, oneanalysis shows a flat HREE pattern without Eu anomaly (Fig. 9c) andlow Nb and Ta contents, suggesting a garnet- and rutile-rich butplagioclase-poor environment during its formation (Rubatto, 2002),i.e., probably from an eclogite. Overall, the mineral inclusions and traceelement features of the inheritedmetamorphic zircons from the Bengbuleucogranites suggest that the source of the rocks is mainly felsicgneisses with rare eclogites.

Comparing the mineral inclusion assemblage and REE patterns inmetamorphic zircons from the Bengbu leucogranites to those from theDabie–Sulu UHPM rocks, we suggest that the inherited metamorphiczircons from the former are neither UHP nor retrograde metamorphic

Fig. 8. The Concordia diagrams of zircon U–Pb dating by Cameca-1280 technique for the aplite (JS-3) and gneissic leucogranite (JS-4), with the inset of a geological sketch to describe thefield relationship between aplite sample and gneissic leucogranite sample (data are listed in Supplementary Table S3).

1121S. Li et al. / Gondwana Research 26 (2014) 1111–1128

Author's personal copy

zircons, but formed in the Triassic HP metamorphism during continen-tal collision. Therefore, the source rocks of the Bengbu leucogranitescould be a piece of shallowly underthrust continental crust or thickenedlower crust in mountain-root, and should not be deeply subducted con-tinental crust or exhumed UHP rocks as Yang et al. (2010) suggested.

The HP metamorphism in the Dabie–Sulu orogen occurred eitherbefore or after the peak UHP event, corresponding to different tectonicevolution stages. The ages of inherited metamorphic zircons from theBengbu leucogranites may provide constraints on the time frame ofthis issue. We have obtained two weighted mean 206Pb/238U ages of224.9 ± 6.2 Ma and 222.1 ± 7.1 Ma from the zircons, similar to re-ported age data of 217.1 ± 7.6 Ma and 221.5 ± 5.6 Ma (S. Xu et al.,2005; W.L. Xu et al., 2005; Wang et al., 2010; Yang et al., 2010). Allthese ages are younger than the peak metamorphic time (231–225 Ma) of the Dabie–Sulu UHPM rocks (e.g., Li et al., 1994, 2000;Ayers et al., 2002; F.L. Liu et al., 2008; Y.S. Liu et al., 2008), but consistentwith the initial cooling ages (219–214 Ma) of the Dabie–Sulu UHPMrocks (Li et al., 2000; Li et al., 2003; F.L. Liu et al., 2008). Therefore, thecontinental crust as the source rock of the Bengbu leucogranites shouldhave experienced HP metamorphism simultaneously with the initialexhumation of the UHPM slices in the Dabie–Sulu orogen. It might un-derthrust below the exhumedUHPM slices and be shortened and thick-ened in the mountain root. By analogy to Himalayan orogen, the

thickened lower crust could be up to 70 km in depth, thus it only expe-rienced HP but not any UHP metamorphism. Thickened lower crust(depth N50 km) of Dabie–Sulu orogen was existed till the Early Creta-ceouswhich is supported by the Cretaceous lowMg# adakitic granitoids(He et al., 2011). Therefore, we conclude that the protolith of theBengbu leucogranite may be derived from the thickened lower crustin Sulu orogen.

5.1.4. Partial melting conditionSince the gneisses from the Dabie–Sulu orogen are exhumed UHPM

rocks and a potential source for the Bengbu leucogranites, the muchlower LREE and Th contents of the Bengbu leucogranites relative togneisses from the Dabie–Sulu orogen (Fig. 6b) suggest that some LREEand Th enriched minerals could be residues during partial melting ofthe gneisses (Fig. 6b). It is recognized that allanite as LREE and Thenriched mineral is a common mineral phase in the Dabie gneisses(Carswell et al., 2000). The very low LREE and Th contents of the Bengbuleucogranites could be interpreted by allanite as residue in equilibriumwith the melt in the source, which suggests that the melting tempera-ture of the Bengbu leucogranites should be b800 °C, because allanitecrystallization in silica magma has a critical saturation temperature ofabout 800 °C (Chesner and Ettlinger, 1989).

In addition, although the Bengbu leucogranites contain abundantinherited zircons, their whole rock Zr–Hf contents are much lowerthan those of gneisses from the Dabie orogen (Fig. 6b), indicating a rel-atively lowmagma temperature (Miller et al., 2003). Using themodel of

Fig. 9. Representative zircon CL images (a) document quartz and titanite inclusions ininherited metamorphic zircons from Jingshan leucogranite with age and spot locationmarked. Diagram (b) shows representative Raman spectra of a titanite inclusion in meta-morphic zircons from Jingshan leucogranite. This spectrum also contains host zircon peaksat 438, 975 cm−1. Tit = titanite; Zr = zircon. Diagram (c) shows chondrite-normalizedREE patterns of ten metamorphic zircon domains for Jingshan intrusion.

Fig. 10. Sr–Nd–Pb isotopic diagrams of the Bengbu leucogranites (modified after Huanget al., 2008) showing that the Bengbu leucogranites has no difference from the Dabiegneisses, but are distinct from the Dabie eclogite.

1122 S. Li et al. / Gondwana Research 26 (2014) 1111–1128

Author's personal copy

zircon solubility in silicamelt given byWatson andHarrison (1983), zir-conium saturation temperatures (TZr) calculated from bulk rock Zr con-tents were obtained for the Late-Jurassic Bengbu leucogranites, theCretaceous granites in the Bengbu uplift and the Dabie post-collisiongranites (Fig. 11a). Since all these rocks contain inherited zircons, theTZr values provide the upper limit for the magma temperatures (Milleret al., 2003). Fig. 11a shows that the average TZr of the Bengbuleucogranites is 710 ± 18 °C, which is much lower than the averageTZr values of 759 ± 27 °C of the Bengbu Cretaceous granites and783 ± 37 °C of the Dabie post-collisional granites. Therefore, unlikethe Bengbu Cretaceous granites and the Dabie post-collisional granites,which are hot granites requiring heat input frommantle upwelling (Liuet al., 2010; He et al., 2011; S.A. Liu et al., 2012), the Bengbuleucogranites were formed by lower temperature melting (b710 °C),which requires fluid influx or dehydration (Miller et al., 2003).

While the TZr values provide the upper limit for the magma tempera-tures, temperature got from the Ti-in-zircon thermometrymay give a lowlimit for themagma temperature. 34 Late Jurassicmagmatic domains and38 Triassic metamorphic domains in zircons from the Jingshan leuco-granites were analyzed to determine their Ti contents, which showno difference between the zircons from gneissic leucogranite,leucogranitic veins and biotite-rich restite, respectively (see Supple-mentary Table S4). The Ti-in-zircon temperatures of magma andprotolith of the Jingshan leucogranites were calculated using thesedata. As there are no rutiles in the Jurassic magmatic zircons, we as-sume that the activity of Ti is 0.8 (Ferry and Watson, 2007). Fig. 11b

shows that the Late Jurassic magmatic zircons from the Jingshanleucogranite yield a Ti-in-zircon temperatures of 697 ± 62 °C, whichare also significantly lower than the Ti-in-zircon temperatures of784 ± 56 °C for the Bengbu Cretaceous granites (Fig. 11b). Althoughthis temperature estimation could be underestimated, by combinationwith the Zr saturated temperature (710 ± 20 °C) mentioned above,the magma/melting temperature for the Jingshan leucogranite is con-strained strictly at ca. 700–710 °C.

Fig. 11b also shows that 38 Triassic inheritedmetamorphic zircon do-mains from the Jingshan leucogranite give Ti-in-zircon temperatures of681 ± 60 °C, representing the metamorphic temperature of protolithof the Jingshan leucogranite at 220 Ma. This metamorphic temperatureis not significantly different from or even slightly lower compared tothe melting temperature (700–710 °C) at 160 Ma. To elevate theprotolith temperature about 30 °C during the time period of 60 Ma toachieve melting conditions is not difficult by radioactive decay of U, Thand K in the thickened middle-lower continental crust (Clark et al.,2011). It also means that no heat inputs are required during the meltingprocess of the Jingshan leucogranite.

Experimental studies show that dehydration-melting of meta-sedimental rocks to produce peraluminous leucogranite melt needsstarting materials having much higher muscovite + biotite contents(24–42%) and relatively higher temperature (T = 750–800 °C), whileadding H2O could lower the solidus, and thus melts can be producedfrom these starting materials at T ≤ 750 °C by H2O-fluxing (Vielzuefand Montel, 1994; Patino Douce and Harris, 1998; Sawyer et al., 2011).As discussed above, the protolith of the Bengbu leucogranites is mica-poor orthogneiss and their melting temperature is 700–710 °C signifi-cantly lower than 750 °C. In addition, their high Sr contents and lowRb/Sr ratios are not due to plagioclase accumulation because of the lackof positive Eu anomalies in their REE patterns (Fig. 6a). Therefore, thehigh Sr and low Rb/Sr features of the Bengbu leucogranites (Fig. 5e, f)suggest that their magma may not be produced by dehydration-melting but more likely to be formed by water-present melting ofmica-poor orthogneiss, because residue in this melting source containsmuscovite without plagioclase (Patino Douce and Harris, 1998).

5.2. Rheology of the source rocks

In order to understand the crustal migration from the Sulu orogen tothe NCB, it is necessary to discuss the rheology of protolith of the Beng-bu leucogranites. Viscosity of middle-lower crust is a key parameter forinjection of the Sulu crust into NCB crust, in the form of either a solid‘piston’ (hard injection; Zhao and Morgan, 1985) or a viscous crustalflow (soft injection; Westaway, 1995).

Trace element characteristics of the inherited metamorphic zirconsand bulk-rock Sr–Nd–Pb isotopic compositions of the rocks suggestthat protolith of the Bengbu leucogranites is mainly felsic gneisses rath-er than eclogite (see above), which is important for understanding therheology of protolith of the leucogranites. Asmentioned above, a crustalpartial melting event in the Late Jurassic was occurred in Sulu orogenbut absent in Dabie orogen (e.g., He et al., 2011; F.L. Liu et al., 2012). Ob-viously, given the presence of melt in felsic gneiss in the Sulu orogen,the viscosity of felsic thickened middle-lower crust in the Sulu orogenshould be low (b1019 Pa s) in Late Jurassic (Gerya et al., 2001). Thus,melt-weakened crust in Sulu orogen could flow outward in responseto the pressure gradients associated with mountain building (we willdiscuss it below),

5.3. Topography of the Bengbu uplift and its adjacent area in Jurassic

If the viscosity of felsic thickened middle-lower crust is lower than1019 Pa s in the Sulu orogen, it could be migrated as crustal flow drivenby lateral pressure gradient in the thickened middle-lower crustal layer(e.g., Bird, 1991). This lateral pressure gradient is caused by variations incrustal thickness above the low-viscosity layer (e.g., Bird, 1991). Thus, to

Fig. 11. a, Comparison of zircon saturation temperatures (TZr) between the Bengbu late Ju-rassic leucogranite and Cretaceous granites aswell as Dabie granites. The zircon saturationtemperatures (TZr) are calculated from bulk rock Zr contents of these granitic rocks. Datasources for the Bengbu late Jurassic leucogranite and Cretaceous granites are from thisstudy, Yang et al. (2010), and S.A. Liu et al. (2012). Data sources for the Dabie post-collisional granites are from those compiled by He et al (2011). b, Comparisons of Ti-in-zircon temperatures between magmatic zircons from the Jingshan late Jurassicleucogranite and Huanglishan Cretaceous granite in Bengbu uplift, respectively, and be-tween Triassic metamorphic zircons from the Jingshan leucogranite and Dabie–SuluUHPM rocks, respectively. Data sources for the Jingshan leucogranite are from this study(See Supplementary Table S4). Data sources for the Huanglishan granites are from S.A.Liu et al. (2012). Date sources for Dabie–Sulu UHPM rocks are from Chen et al. (2010),Chen et al. (2011) and Gao et al. (2011).

1123S. Li et al. / Gondwana Research 26 (2014) 1111–1128

Author's personal copy

recover topography of the Bengbu uplift and its adjacent area in the Ju-rassic is important for revealing if there was a lateral pressure gradienton the low-viscosity layer in the region. Because the Tan–Lu Fault is ini-tiated in theMiddle–Late Triassic and experienced a large-scale sinistralstrike slip in the Jurassic (Xu et al., 1987;Wang, 2006; Zhu et al., 2009),

the Sulu orogenic belt was translated to the north by the sinistral Tan–Lu Fault in Jurassic. In the Jurassic, the Hefei Basin was bounded by theDabie orogen on the south, the Sulu orogen on the east and the Bengbuuplift on the north (Fig. 1a). Thus, the Jurassic strata in the Hefei Basinmay provide information about topography of this region because theJurassic sediments in the Hefei Basin were derived from the surround-ing uplift regions.

On the basis of seismic interpretation and drill-hole data (Liu et al.,2006), the lower to upper Jurassic strata showsedimentary facies evolu-tion (Fig. 12). Fig. 12 shows that the Jurassic alluvial fan facies depositsonly developed in the south and east margins of the Hefei Basin,suggesting rugged and high topography to the south and east of theHefei Basin. The Jurassic molasse-type sediments in south margin ofthe basin contain eclogitic gravels, HP minerals (e.g., garnet andphengite) and Triassic zircons, which have been considered to be de-rived from the Dabie orogen (Li et al., 2005). It is reasonable to inferthat the Sulu beltwas located to the east of theHefei Basin in the Jurassicand provided themolasse-type sediments to theHefei Basin. Fig. 12 alsoshows a northward extension of the alluvial fan facies deposits in eastmargin of the Hefei Basin from the Early Jurassic to Late Jurassic,suggesting a northward movement of the Sulu belt in the Jurassic. TheSulu orogenic belt could have been moved to the east of the Bengbuarea (thatmay not uplifted yet) in the Late Jurassic (Fig. 12). In contrast,only anastomose river facies or fluvial facies deposits were developed inthe north andwest margins of the Hefei Basin (Fig. 12), suggesting a flatand non-high topography to the north and west of the Basin. If theabove recovered topography in the area surrounding the Hefei Basinin the Late Jurassic is correct, the lateral pressure gradient in a low-viscosity crustal layer from the Sulu orogen in the east to the Bengbuarea in the west must exist.

In addition, the northwardmovement of the Sulu belt in the Jurassicindeed supports that the Tan–Lu Fault is a syn-collisional transformfault (Zhang et al., 1984; Hsu et al., 1987; Zhu et al., 2009) but unlikelyto be a clockwise rotated suture line (Zhang, 1997; Gilder et al., 1999;Yang et al., 2010). Actually, the biggest problem for taking the Tan–LuFault as the suture line (Yang et al., 2010) is that there are no exhumedUHPM rocks along the middle section of the Tan–Lu Fault between theSulu and Dabie UHPM belts (Fig. 1a).

5.4. A crustal flow model for the origin of the Bengbu leucogranite

Before introducing our crustalflowmodel, we need to reevaluate theprevious model for the origin of the Jingshan leucogranite proposed byYang et al. (2010). Yang et al. (2010) suggested that NW subduction ofthe SCB beneath theNCB along the Tan–Lu Fault zone in the Triassic andthe upwelling of the asthenosphere after the slab break-off and delam-ination of the thickened NCC lithosphere caused partial melting of theexhumated SCB slab to form the Late Jurassic intrusion such as theJingshan leucogranite. In their cartoon, however, they show the partialmelting of SCB lower crust and the magma intruded into the NCBcrust to form the Late Jurassic Jingshan intrusion (Fig. 9d in Yang et al.,2010). Let alone the controversy about the rotated suture line modelfor the Tan–Lu Fault (Zhu et al., 2009 and references therein), Yanget al.'s model encounters the following difficulties.

(1) The model cannot explain why the Jurassic Bengbu leucogranitehave significant lowermelting temperatures than the Cretaceousgranites in the Dabie orogen and the Bengbu uplift, if partialmelting of SCB lower crust in the Late Jurassic was caused by up-welling of the asthenosphere.

(2) It is difficult to interpret why there is nomelting of lower crust ofthe NCB below the Bengbu area when the mantle upwelling oc-curred under the suture zone in the Late Jurassic (see Fig. 9d inYang et al., 2010).

(3) The delamination of the thickened lithosphere on the southeastmargin of the NCB occurred in the middle Cretaceous (~115 Ma),

Fig. 12. Sedimentary facies of the Jurassic strata in the Hefei Basin (modified after Liu et al.,2006) showing northward movement of the Sulu orogen belt during the Jurassic and to-pography of the area surrounding the Hefei basin.

1124 S. Li et al. / Gondwana Research 26 (2014) 1111–1128

Author's personal copy

which has been documented by the Cretaceous adakitic–basaltic–granitic magma sequence in the Bengbu area (S.A. Liu et al., 2012),rather than in the Late Jurassic as proposed by Yang et al. (2010).

(4) According to the model (Fig. 9e in Yang et al., 2010), the Late Ju-rassic leucogranite resulted frompartialmelting of the SCB lowercrust should be located in the east whereas the Cretaceous gran-ite derived from the NCB lower crust should be located in thewest, separated from each other. However, this is not the caseas shown in Fig. 1b, in which the Cretaceous plutons derivedfrom the NCB are distributed on both sides of the Jurassicleucogranite belt. In addition, if the continental crust of the SCBwas westward subducted along the Tan–Lu Fault as suggestedby Yang et al. (2010), the Late Jurassic granitic intrusions shouldbe developed parallel to the Tan–Lu Fault. However, the Late Ju-rassic leucogranitic intrusions are arranged vertical to the Tan–Lu Fault and form a NWW–SEE array (Fig. 1b) suggesting a chan-nel in which their source rocks derived from the SCB passedthough.

(5) Finally, the model assumed that the slab break-off in the Dabie–Sulu orogen occurred in the time period of 185 to 220 Ma (seeFig. 9c in Yang et al., 2010), which is contradictory to the slabbreak-off time of around 225 Ma supported by geochronologicalstudies on the syn-collisinal granite (Davies and Blanckenburg,1995; Sun et al., 2002; Chen et al., 2003), as well as the coolinghistory of the UHPM rocks in the Dabie–Sulu belt (Li et al.,2000; Li et al., 2003; F.L. Liu et al., 2008; Wang et al., 2012).

In contrast to the model of Yang et al. (2010), the present studyshows that all aspects related to the origin of the Bengbu leucogranites,including the felsic protolith and the SCB affinitywithout UHPmetamor-phism, low melting temperature and lowMg# melt, and the NWW–SEEdistribution of the Late Jurassic leucogranites in Bengbu uplift, suggestthat the source rocks of the Bengbu leucogranites could be injected

into the NCB middle-lower crust from the Sulu orogen instead ofexhumated SCB UHPM slab or subducted SCB lower crust beneath theNCB. In view of the low viscosity of the melt-weakened crust of Suluorogen in the Late Jurassic, as well as high topography to the east ofthe Bengbu area resulting into an east–west lateral pressure gradient,the crustal injection could be in the form of crustal flow. This interconti-nental crust flow model is described as follows.

The northward continental subduction of the SCB occurred in theEarly Triassic (245–226 Ma; Fig. 13a), followed by slab break-off andexhumation of UHPM slices in the Middle Triassic (225–213 Ma;Fig. 13b), resulted in a thickened crust in the Dabie–Sulu area. In the pe-riod ofMiddle Triassic to Jurassic, theDabie–Sulu orogenwas transectedby the Tan–Lu Fault with the Sulu terrane moved to the north (Fig. 13c;Zhu et al., 2009). Based on sedimentary facies analysis of the Jurassicstrata in the Hefei Basin, the Sulu terrane could be located to the eastof the Bengbu area as discussed above (Fig. 12). In the meanwhile,two partially crust melting events have occurred inmoving Sulu orogen(F.L. Liu et al., 2012). Lithospheric texture study on the Tan–Lu Faultzone suggests that the crust in the Sulu belt to the east is significantlythicker and deeper respectively than that of the NCB to the west (Zhuet al., 2002). Based on this study, the topography in the Sulu orogen ismuch higher than that in Bengbu area. Consequently, a cartoon transac-tion from west to east across the Bengbu area and Sulu belt to show achannel flow model in the Late Jurassic has been constructed(Fig. 13d). The crustal thickness of the Bengbu area in the Late Jurassiccould be up to ~50 km due to collision between the NCB and the SCBbut less than that of the Sulu orogen, which may be up to 70–80 kmsimilar to that of the Himalaya orogen. Because of the much higher to-pography in the Sulu orogen relative to the Bengbu area, a lateral pres-sure gradient in middle-lower crust from the Sulu orogen in the east tothe Bengbu area in the west must have present in Jurassic. Because thethickenedmiddle-lower crust in the Sulu belt had been partially meltedin Late Jurassic (F.L. Liu et al., 2012; S.A. Liu et al., 2012) and thusmust be

Fig. 13. (a) and (b) Cartoons showing continental subduction and UHPM rocks exhumation processes during collision between the SCB and NCB in the Early and Middle Triassic. (c) Acartoon showing the Su–Lu orogen was transected andmoved to the north by the Tan–Lu syn-collisional transform fault in the Late Triassic and Jurassic (after Zhu et al., 2009). (d) A car-toon transaction c′–c crossing the Bengbu area and Sulu belt from west to east (see Fig. 1a for the location) to show a crust flowmodel in this area in the Late Jurassic (see text for detailexplanation).

1125S. Li et al. / Gondwana Research 26 (2014) 1111–1128

Author's personal copy

a low-viscosity layer, this lateral pressure gradient might have driventhe melt-weakened felsic crust in the Sulu orogen to inject into themiddle-lower crust of the Bengbu area by the mechanism of channelflow (Bird, 1991; Fig. 14d). Meanwhile, the partial melt from the chan-nel flow ascended and intruded into the upper-middle crust as theleucogranite intrusions in the Bengbu area. Jamieson et al. (2011) indi-cated that this crustal flow may be transported hundreds of kilometerslaterally as mid-crustal channels. The NWW–SEE distribution (~70 kmin length) of the later Jurassic leucogranites in Bengbu uplift (Fig. 1b)may indicate the flow direction.

Although, we cannot prove the existence of the hidden crustal flowforwhich no surface representatives exist, we state only that the crustalflow or soft injection is the most likely model that can best explain thedata presented in this study. In order to verify this model more workstrying to find lower crustal xenolith in Mesozoic igneous intrusions inthe Bengbu area are needed.

6. Conclusions

Systematic geochronological, geochemical and geological data relat-ed to origin of the Bengbu leucogranites and their geological settings arereported in this paper. Based on these data, the following conclusionsare obtained.

(1) The Late Jurassic Jinshan, Tushan, Mayishan and Laoshan intru-sions NWW–SEE distributed in the Bengbu area of the NCB aremetaluminous leucogranites derived from thickened middle-lower crust of the Sulu orogenic belt. The source rock of the Beng-bu leucogranites is mainly felsic gneiss that has experienced HPmetamorphism in the Triassic at temperature of about 680 °Cbut not experienced UHPmetamorphism. The Late Jurassic crys-tallization ages of the Bengbu leucogranites are identical to themigmatite ages in the Sulu orogen, suggesting that their sourcerock was derived from the partially melted crust of the Suluorogen in Late Jurassic.

(2) The Bengbu leucogranites characterized by low Rb and high Srcontents as well as low REE contents without significant Euanomalies are derived frommica-poor orthogneiss and producedby water-present partial melting at low temperature (700–710 °C). Water-influx in the middle-lower crust but no heat-input is required for the melting.

(3) The sedimentary facies analyses of the Jurassic strata in the Hefeibasin suggests that the sinistral strike–slipmovement of the Tan–Lu Faultmoved the Sulu belt to the east of the Bengbu uplift in theLate Jurassic, which results in the contrast topography in theBengbu area and the neighboring Sulu orogen. Therefore, a lateralpressure gradient in middle-lower crust must exist from the Suluorogen in the east to the Bengbu area in the west in Late Jurassic.

(4) All aspects related to the origin of the Bengbu leucogranite requirea hidden partially melted source rocks from the Sulu orogen,which has been injected into the NCB crust. The crustal injectioncould be in the form of crustal flow because of its low viscosity inmelt-weakened crust and a lateral pressure gradient from theSulu orogen to the Bengbu area in the late Jurassic. Thus, the Juras-sic Bengbu leucogranite and themigmatite in Sulu orogen could bepotentially an example of ancient intercontinental channel flow.

Acknowledgments

We sincerely thank Xiaoming Liu for helps during major and traceelement analyses, Dunyi Liu, Hua Tao, Yanbin Wang, Yuruo Shi andYusheng Wan during zircon SHRIMP U–Pb analyses, Zhenyu Chen forelectron microprobe analyses, and the all staff member in the Sr–Nd–Pb isotope laboratory of Institute of Geology and Geophysics, ChineseAcademy of Sciences. We also thank Wenrong Cao and Shuangli Tangfor field assistance. The project was financially supported by grants

from the National Scientific Foundation of China (Nos. 40634023,90814008, 91014007, and 40921002). Discussions with Profs. YouxueZhang, Peizhen Zhang,Weidong Sun and FangHuanghelped to improvean early version of the manuscript.

Appendix A. Supplementary data

Supplementary data to this article can be found online at http://dx.doi.org/10.1016/j.gr.2013.08.019.

References

Ames, L., Tilton, G.R., Zhou, G., 1993. Timing of collision of the Sino–Korean and Yangtzecratons: U–Pb zircon dating of coesite bearing eclogites. Geology 21, 339–342.

Ayers, J.C., Dunkle, S., Gao, S., Miller, C.F., 2002. Constraints on timing of peak and retro-grade metamorphism in the Dabie Shan ultrahigh-pressure metamorphic belt, east-central China, using U–Th–Pb dating of zircon and monazite. Chemical Geology186, 315–331.

Beaumont, C., Jamieson, R.A., Nguyen, M.H., Lee, B., 2001. Himalayan tectonics explainedby extrusion of a low-viscosity crustal channel coupled to focused surface denuda-tion. Nature 414 (13), 738–742.

Bingen, B., Austrheim, H.K., Whitehouse, M.J., Davis, W.J., 2004. Trace element signatureand U–Pb geochronology of eclogite-facies zircon, Bergen Arcs, Caledonides of WNorway. Contribution to Mineralogy and Petrology 147, 671–683.

Bird, P., 1991. Lateral extrusion of lower crust from under high topography, in the isostaticlimit. Journal of Geophysical Research— Solid Earth and Planets 96 (B6), 10275–10286.

Braun, I., Raith, M., Kumar, G.R.R., 1996. Dehydration-melting phenomena in leptyniticgneisses and the generation of leucogranites: a case study from the Kerala KhondaliteBelt, Southern India. Journal of Petrology 37, 1285–1305.

Bryant, D.L., Ayers, J.C., Gao, S., Miller, C.F., Zhang, H.F., 2004. Geochemical, age, and isoto-pic constraints on the location of the Sino–Korean/Yangtze Suture and evolution ofthe Northern Dabie Complex, east central China. Geological Society of America Bulle-tin 116 (5–6), 698–717.

Carswell, D.A., Wilson, R.N., Zhai, M., 2000. Metamorphic evolution, mineral chemistryand thermobarometry of schists and orthogneisses hosting ultra-high pressureeclogites in the Dabieshan of central China. Lithos 52 (1–4), 121–155.

Chavagnac, V., Jahn, B.M., 1996. Coesite-bearing eclogites from the Bixiling complex,DabieMountains, China: Sm–Nd ages, geochemical characteristics and tectonic impli-cations. Chemical Geology 133, 29–51.

Chen, F.K., Hegner, E., Todt, W., 2000. Zircon ages, Nd isotopic and chemical compositionsof orthogneisses from the Black Forest, Germany— evidence for a Cambrian magmat-ic arc. International Journal of Earth Sciences 88, 791–802.

Chen, J.F., Xie, Z., Li, H.M., Zhang, H.D., Zhou, T.X., Park, Y.S., Ahn, K.S., Chen, D.G., Zhang, X.,2003. U–Pb zircon ages for a collision-related K-rich complex at Shidao in the Suluultrahigh pressure terrane, China. Geochemical Journal 37, 35–46.

Chen, R.X., Zheng, Y.F., Xie, L.W., 2010. Metamorphic growth and recrystallization of zircon:distinction by simultaneous in-situ analyses of trace elements, U–Th–Pb and Lu–Hf iso-topes in zircons fromeclogite-facies rocks in the Sulu orogen. Lithos 114 (1–2), 132–154.

Chen, Y.X., Zheng, Y.F., Chen, R.X., Zhang, S.B., Li, Q., Dai, M., Chen, L., 2011. Metamorphicgrowth and recrystallization of zircons in extremely 18O-depleted rocks duringeclogite-facies metamorphism: evidence from U–Pb ages, trace elements, and O–Hfisotopes. Geochimica et Cosmochimica Acta 75 (17), 4877–4898.

Chesner, C.A., Ettlinger, A.D., 1989. Composition of volcanic allanite from the Toba Tuff,Sumatra, Indonesia. American Mineralogist 74, 750–758.

Clark, M.K., Royden, L.H., 2000. Topographic ooze: building the eastern margin of Tibet bylower crustal flow. Geology 28 (8), 703–706.

Clark, C., Fitzsimons Ian, C.W., Healy, D., Harley, S.L., 2011. How does the continental crustget really hot? Elements 7, 235–240.

Clarke, D.B., Macdonald, M.A., Reynolds, P.H., Longstaffe, F.J., 1993. Leucogranites from theeastern part of the South Mountain Batholith, Nova Scotia. Journal of Petrology 34(4), 653–679.

Compston,W.,Williams, I.S., Kirschvink, J.L., Zhang, Z.,Ma, G., 1992. ZirconU–Pb ages for theearly Cambrian time-scale. Journal of the Geological Society of London 149, 171–184.

Davies, J.H., Blanckenburg, F.V., 1995. Slab breakoff: a model of lithosphere detachmentand its test in the magmatism and deformation of collisional orogens. Earth and Plan-etary Science Letters 129, 85–102.

Ferry, J.M., Watson, E.B., 2007. New thermodynamic models and revised calibrations forthe Ti-in-zircon and Zr-in-rutile thermometers. Contribution to Mineralogy and Pe-trology 154, 429–437.

Gao, S., Ling, W.L., Qiu, Y.M., Lan, Z., Hartmann, G., Simon, K., 1999. Contrasting geochem-ical and Sm–Nd isotopic compositions for Archean metasediments from the Konglinghigh-grade terrain of the Yangzte craton: evidence for cratonic evolution and redistri-bution of REE during crustal anatexis. Geochimica et Cosmochimica Acta 63,2071–2088.

Gao, X.Y., Zheng, Y.F., Chen, Y.X., 2011. U–Pb ages and trace elements in metamorphic zir-con and titanite from UHP eclogite in the Dabie orogen: constraints on P–T–t path.Journal of Metamorphic Geology 29 (7), 721–740.

Gerya, T.V., Maresch, W.V., Willner, A.P., Van Reenen, D.D., Smit, C.A., 2001. Inherent grav-itational instability of thickened continental crust with regionally developed low- tomedium-pressure granulite facies metamorphism. Earth and Planetary Science Let-ters 190 (3–4), 221–235.

Gilder, S.A., Leloup, H.L., Gourtillot, V., Chen, Y., Coe, R.S., Zhao, X., Xiao, W., Halim, N., Cogne,J.P., Zhu, R., 1999. Tectonic evolution of the Tancheng–Lujiang (Tan–Lu) Fault via middle

1126 S. Li et al. / Gondwana Research 26 (2014) 1111–1128

Author's personal copy

Triassic to Early Cenozoic paleomagnetic data. Journal of Geophysical Research 104,15365–15390.

Grujic, D., 2006. Channel flow and continental collision tectonics: an overview. From In:Law, R.D., Searle, M.P., Godin, L. (Eds.), Channel Flow, Ductile Extrusion and Exhuma-tion in Continental Collision Zones. Geological Society, London, Special Publication,268, pp. 25–37.

Guo, S.S., Li, S.G., 2009a. Petrochemical characteristics of leucogranite and a case study ofBengbu leucogranites. Chinese Science Bulletin 19, 23–30.

Guo, S.S., Li, S.G., 2009b. Zircon SHRIMP U–Pb ages constrained the early Proterozoicevents of metamorphism and magmatism on the south-eastern margin of theNorth China craton. Science in China Series D — Earth Sciences 39 (5), 1039–1045.

Hacker, B.R., Ratschbacher, L., Webb, L.E., Ireland, T., Walker, D., Dong, S.W., 1998. U/Pbzircon ages constrain the architecture of the ultrahigh-pressure Qinling–DabieOrogen, China. Earth and Planetary Science Letters 161 (1–4), 215–230.

Harley, S.L., Kelly, N.M., Möller, A., 2007. Zircon behaviour and the thermal histories ofmountain chains. Elements 3 (1), 25–30.

Harris, N., 2007. Channel flow and the Himalayan–Tibetan orogen: a critical review. Jour-nal of the Geological Society 164, 511–523.

Harris, N., Massey, J., 1994. Decompression and anatexis of Himalayan metapelites. Tec-tonics 13 (6), 1537–1546.

He, Y.S., Li, S.G., Hoefs, J., Huang, F., Liu, S.A., Hou, Z.H., 2011. Post-collisional granitoidsfrom the Dabie orogen: new evidence for partial melting of a thickened continentalcrust. Geochimica et Cosmochimica Acta 75, 3815–3838.

Hsu, K.J., Li, J., Chen, I., Wang, Q., Sun, S., 1987. A. M. C segor tectonic evolution of QinlingMountains, China. Eclogae Geol Helve 80, 735–752.

Huang, F., Li, S.G., Dong, F., He, Y.S., Chen, F.K., 2008. High-Mg adakitic rocks in the Dabieorogen, central China: implications for foundering mechanism of lower continentalcrust. Chemical Geology 255 (1–2), 1–13.

Inger, S., Harris, N., 1993. Geochemical constraints on leucogranite magmatism in theLangtang Valley Nepal Himalaya. Journal of Petrology 34 (2), 345–368.

Jahn, B.M.,Wu, F., Lo, C.H., Tsai, C.H., 1999. Crust–mantle interaction induced by deep sub-duction of the continental crust: geochemical and Sr–Nd isotopic evidence from post-collisional mafic–ultramafic intrusions of the northern Dabie complex. Chemical Ge-ology 157, 119–146.

Jamieson, R.A., Unsworth, M.J., Harris, N.B.W., Rosenberg, C.L., Schulmann, K., 2011. Crust-al melting and the flow of mountains. Elements 7, 253–260.

Li, S.G., Hart, S.R., Zheng, S., Liou, D., Zhang, G., Guo, A., 1989. Timing of collision betweenthe North and South China Blocks — Sm–Nd isotopic age evidence. Science in China.Series B 32, 1391–1400.

Li, S.G., Liu, D.L., Chen, Y.Z., Ge, N.J., 1992. The Sm–Nd isotopic age of coesite-bearingeclogite from the Southern Dabie Mountains. Chinese Science Bulletin 37, 1638–1641.

Li, S.G., Xiao, Y., Liu, D., Chen, Y., Ge, N., Zhang, Z., Sun, S.S., Cong, B., Zhang, R., Hart, S.R.,Wang, S., 1993. Collision of the North China and Yangtze Blocks and formation ofcoesite-bearing eclogites— timingandprocesses. ChemicalGeology 109 (1–4), 89–111.

Li, S.G., Wang, S.S., Chen, Y.Z., Liu, D.L., Qiu, J., Zhou, H.X., Zhang, Z.M., 1994. Excess argonin phengite from eclogite: evidence from dating of eclogite minerals by Sm–Nd, Rb–Sr and 40Ar/39Ar methods. Chemical Geology 112, 343–350.

Li, S.G., Hong, J.A., Li, H.M., 1999. U–Pb zircon ages of the pyroxenite–gabbro intrusions inthe Dabie Mountains and their geological implications. Geological Journal of ChinaUniversities 5 (3), 351–355 (in Chinese with English abstract).

Li, S.G., Jagoutz, E., Chen, Y.Z., Li, Q.L., 2000. Sm–Nd and Rb–Sr isotopic chronology andcooling history of ultrahigh pressure metamorphic rocks and their country rocks atShuanghe in the Dabie Mountains, Central China. Geochimica et CosmochimicaActa 64 (6), 1077–1093.

Li, Q., Li, S., Zheng, Y.F., Li, H., Massonne, H.J., Wang, Q., 2003. A high precision U-Pb age ofmetamorphic rutile in coesite-bearing eclogite from the Dabie Mountains in centralChina: a new constraint on the cooling history. Chemical Geology 200 (3), 255–265.

Li, X.-P., Zheng, Y., Wu, Y., Chen, F., Gong, B., Li, Y., 2004. Low-T eclogite in the Dabie ter-rane of China: petrological and isotopic constrains on fluid activity and radiometricdating. Contributions to Mineralogy and Petrology 148, 443–470.

Li, R.W., Wan, Y., Cheng, Z., Zhou, J., Li, S., Jin, F., Meng, Q.R., Li, Z., Jiang, M., 2005. Prove-nance of Jurassic sediments in the Hefei Basin, east-central China and the contribu-tion of high-pressure and ultrahigh-pressure metamorphic rocks from the DabieShan. Earth and Planetary Science Letters 231 (3–4), 279–294.

Li, X.H., Liu, Y., Li, Q.L., Guo, C.H., Chamberlain, K.R., 2009. Precise determination of Phan-erozoic zircon Pb/Pb age by multicollector SIMS without external standardization.Geochemistry, Geophysics, Geosystems 10, Q04010. http://dx.doi.org/10.1029/2009GC002400.

Li, Y., Han, F., Ling, M.X., Liu, J., Li, X.H., Li, Q.L., Sun, W.D., 2010. Geochronology, geochem-ical characteristics and geodynamic significance of the Jingshan and Tushan granitesin the Bengbu uplift. Geotectonica et Metallogenia 34 (1), 114–124 (In Chinese withEnglish abstract).

Lin, J.L., Fuller, M., Zhang, W.Y., 1985. Preliminary Phanerozoic polar wander paths for theNorth and South China blocks. Nature 313, 444–449.

Liu, F.L., Liou, J.G., Xu, Z.Q., 2005. U–Pb SHRIMP ages recorded in the coesite-bearing zircondomains of paragneisses in the southwestern Sulu terrane, eastern China: new inter-pretation. American Mineralogist 90, 790–800.

Liu, Y.C., Li, S.G., Xu, S.T., Jahn, B.M., Zheng, Y.F., Zhang, Z., Jiang, L.L., Chen, G., Wu, W.,2005. Geochemistry and geochronology of eclogites from the northern Dabie Moun-tains, central China. Journal of Asian Earth Sciences 25, 431–443.

Liu, G.S., Zhu, G., Niu, M.L., Song, C.Z., Wang, D.X., 2006. Meso-Cenozoic evolution of theHefei Basin (eastern part) and its response to activities of the Tan–Lu Fault zone. Chi-nese Journal of Geology 41 (2), 256–269 (in Chinese with English abstract).

Liu, Y.C., Li, S.G., Xu, S.T., 2007. Zircon SHRIMP U–Pb dating for gneisses in northern Dabiehigh TIP metamorphic zone, central China: implications for decoupling withinsubducted continental crust. Lithos 96 (1–2), 170–185.

Liu, F.L., Gerdes, A., Zeng, L.G., Xue, H.M., 2008. SHRIMP U–Pb dating, trace elements andthe Lu–Hf isotope system of coesite-bearing zircon from amphibolite in the SW SuluUHP terrane, eastern China. Geochimica et Cosmochimica Acta 72, 2973–3000.

Liu, Y.S., Hu, Z.C., Gao, S., Günther, D., Xu, J., Gao, C.G., Chen, H.H., 2008. In situ analysis ofmajor and trace elements of anhydrous minerals by LA–ICP–MSwithout applying aninternal standard. Chemical Geology 257, 34–43.

Liu, Y.C.,Wang, A.D., Rolfo, F., Groppo, C., Gu, X.F., Song, B., 2009. Geochronological and petro-logical constraints on Palaeoproterozoic granulite facies metamorphism in southeasternmargin of the North China Craton. Journal of Metamorphic Geology 27, 125–138.

Liu, S.A., Li, S.G., He, Y.S., Huang, F., 2010. Geochemical contrasts between early Cretaceousore-bearing and ore-barren high-Mg adakites in central-eastern China: implicationsfor petrogenesis and Cu–Au mineralization. Geochimica et Cosmochimica Acta 74,7160–7178.

Liu, Y.C., Gu, X.F., Li, S.G., Hou, Z.H., Song, B., 2011. Multistage metamorphic events ingranulitized eclogites from the North Dabie complex zone, central China: evidencefrom zircon U–Pb age, trace element and mineral inclusion. Lithos 122, 107–121.

Liu, F.L., Robinson, P.T., Liu, P.H., 2012. Multiple partial melting events in the Sulu UHP ter-rane: zircon U–Pb dating of granitic leucosomes within amphibolite and gneiss. Jour-nal of Metamorphic Geology 30, 887–906.

Liu, S.A., Li, S.G., Guo, S.S., Hou, Z.H., He, Y.S., 2012. The Cretaceous adakitic–basaltic–gra-nitic magma sequence on south-eastern margin of the North China Craton: implica-tions for lithospheric thinning mechanism. Lithos 134–135, 163–178.

Liu, Y.C., Wang, A.D., Li, S.G., Rolfo, F., Li, Y., Groppo, C., Gu, X.F., Hou, Z.H., 2013. Composi-tion and geochronology of the deep-seated xenoliths from the southeastern marginof the North China Craton. Gondwana Research 23, 1021–1039.

Ludwig, K.R., 2001. Users Manual for Isoplot/Ex (rev. 2.49): A Geochronological Toolkit forMicrosoft Excel. Special Publication, vol. 1a. Berkeley Geochronology Center (55 pp.).

Miller, C.F., McDowell, S.M., Mapes, R.W., 2003. Hot and cold granites? Implications of zir-con saturation temperatures and preservation of inheritance. Geology 31, 529–553.

Nabelek, P.I., Bartlett, C.D., 1998. Petrologic and geochemical links between the post-collisional Proterozoic Harney Peak leucogranite, South Dakota, USA, and its sourcerocks. Lithos 45, 71–85.

Nabelek, P.I., Russ-Nabelek, C., Denison, J.R., 1992. The generation and crystallization con-ditions of the Proterozoic Harney Peak leucogranite, Black Hills, South Dakota, USA:petrologic and geochemical constraints. Contrib Mineral Petrol Contribution to Min-eralogy and Petrology 110, 173–191.

Nelson, K.D., Zhao,W.J., Brown, L.D., 1996. Partially moltenmiddle crust beneath southernTibet: synthesis of project INDEPTH results. Science 274, 1684–1687.

Patino Douce, A.E., Harris, N., 1998. Experimental constraints on Himalayan anatexis.Journal of Petrology 39 (4), 689–710.

Pearce, N.J.G., Perkins, W.T., Westgate, J.A., Gordon, M.P., Jackson, S.E., Neal, C.R., Chenery,S.P., 1996. A compilation of new and publishedmajor and trace element data for NISTSRM 610 and NIST SRM 612 glass reference materials. Geostan Newslett 21, 115–144.

Pressley, R.A., Brown, M., 1999. The Phillips pluton, Maine, USA: evidence of heteroge-neous crustal sources and implications for granite ascent and emplacement mecha-nisms in convergent orogens. Lithos 46 (3), 335–366.

Raimondo, T., Collins, A.S., Hand, M., Walker-Hallam, A., Smithies, R.H., Evins, P.M.,Howard, H.M., 2009. Ediacaran intracontinental channel flow. Geology 37, 291–294.

Rapp, R.P., Shimizu, N., Norman, M.D., Applegate, G.S., 1999. Reaction between slab-derived melts and peridotite in the mantle wedge: experimental constraints at3.8 GPa. Chemical Geology 160 (4), 335–356.

Royden, L.H., Burchfiel, B.C., King, R.W., Erchie, W., Zhiliang, C., Feng, S., Yuping, L., 1997.Surface deformation and lower crustal flow in Eastern Tibet. Science 276, 788–790.

Rubatto, D., 2002. Zircon trace element geochemistry: partitioning with garnet and thelink between U–Pb ages and metamorphism. Chemical Geology 184, 123–138.

Rubatto, D., Gebauer, D., 2000. Use of cathodoluminescence for U–Pb zircon dating by ionmi-croprobe; some examples from the Western Alps. In: Pagel, M., Barbin, V., Blanc, P.,Ohnenstetter, D. (Eds.), Cathodoluminescence in Geosciences. Springer, Berlin,pp. 373–400.

Rubatto, D., Hermann, J., Berger, A., Engi, M., 2009. Protracted fluid-induced melting dur-ing Barrovian metamorphism in the Central Alps. Contribution to Mineralogy and Pe-trology 158, 703–722.

Sawyer, E.W., Cesare, B., Brown, M., 2011. When the continental crust melts. Elements 7,229–234.

Schulmann, K., Lexa, O., Štípská, P., Racek, M., Tajcˇmanová, L., Konopásek, J., Edel, J.-B.,Peschler, A., Lehmann, J., 2008. Vertical extrusion and horizontal channel flow of oro-genic lower crust: key exhumation mechanisms in large hot orogens? Journal ofMetamorphic Geology 26, 273–297.

Sun,W.D., Li, S.G., Chen, Y.D., 2002. Timing of synorogenic granitoids in the South Qinling,central China: constraints on the evolution of the Qinling–Dabie orogenic belt. Jour-nal of Geology 110, 457–468.

Tang, J., Zheng, Y.F., Wu, Y.B., Gong, B., Zha, X., Liu, X., 2008. Zircon U–Pb age and geo-chemical constraints on the tectonic affinity of the Jiaodong terrane in the Suluorogen, China. Precambrian Research 161, 389–418.

Van de Flierdt, T., Jung, S., Masberg, P., Hoffer, E., Schaltegger, U., Freidrichsen, H., 2003. Lowercrustal melting and the role of open-system processes in the genesis of syn-orogenicquartz diorite–granite–leucogranite associations: constraints from Sr–Nd–O isotopesfrom the Bandombaai complex, Namibia. Lithos 67, 205–226.

Vielzuef, D., Montel, J.M., 1994. Partial melting of metagreywackes. Part I. Fluid-absent ex-periments and phase relationships. Contribution to Mineralogy and Petrology 117,375–393.

Visona, D., Lombardo, B., 2002. Two-mica and tourmaline leucogranites from the Everest–Makalu region (Nepal–Tibet). Himalayan leucogranite genesis by isobaric heating?Lithos 62 (3–4), 125–150.

Wang, Y., 2006. The onset of the Tan–Lu Fault movement in eastern China: constraintsfrom zircon (SHRIMP) and 40Ar/39Ar dating. Terra Nova 18, 423–431.

1127S. Li et al. / Gondwana Research 26 (2014) 1111–1128

Author's personal copy

Wang, Q., Wyman, D.A., Xu, J.F., Jian, P., Zhao, Z.H., Li, C.F., Xue, W., Ma, J.L., He, B., 2007.Early Cretaceous adakitic granites in the Northern Dabie Complex, central China: im-plications for partial melting and delamination of thickened lower crust. Geochimicaet Cosmochimica Acta 71 (10), 2609–2636.

Wang, A.D., Liu, Y.C., Gu, X.F., Li, S.G., Xie, H.Q., 2009. Zircon SHRIMP U–Pb dating forgarnet-bearing gneissic granite at Laoshan, Bengbu: implications for recycling ofthe subducted continental crust of the South China Block. Journal of Mineralogyand Petrology 29 (2), 38–43 (in Chinese with English abstract).

Wang, X., Chen, J., Griffin, W.L., O'Reilly, S.Y., Huang, P.Y., Li, X., 2010. Two stages of zirconcrystallisation in the Jingshan monzogranite, Bengbu Uplift: implications for the syn-collisional granites of the Dabie–Sulu UHP orogenic belt and the climax of movementon the Tan–Lu Fault. Lithos 122, 201–213.

Wang, S.J., Li, S.G., An, S.C., Hou, Z.H., 2012. A granulite record of multistage metamor-phism and REE behavior in the Dabie orogen: constraints from zircon and rock-forming minerals. Lithos 136–139, 109–125.

Wang, S.J., Li, S.G., Chen, L.J., He, Y.S., An, S.C., Shen, J., 2013a. Geochronology and geo-chemistry of leucosomes in the North Dabie Terrane, East China: implication forpost-UHPM crustal melting during exhumation. Contribution to Mineralogy and Pe-trology. 165, 1009–1029.

Wang, S.J., Li, S.G., Liu, S.A., 2013b. The origin and evolution of low-δ18O magma recordedbymulti-growth zircons in granite. Earth and Planetary Science Letters 373, 233–241.

Watson, E., Harrison, T., 1983. Zircon saturation revistited: temperature and compositioneffects in a variety of crustal magma types. Earth and Planetary Science Letters 64,295–304.

Westaway, R., 1995. Crustal volume balance during the India–Eurasia collision and alti-tude of the Tibetan plateau: a working hypothesis. Journal of Geophysical Research100 (B8), 15173–15192.

Williams, I.S., 1998. U–Th–Pb geochronology by ionmicroprobe. Reviews in Economic Ge-ology 7, 1–35.

Xia, Q.X., Zheng, Y.F., Hu, Z., 2010. Trace elements in zircon and coexisting minerals fromlow-T/UHP metagranite in the Dabie orogen: implications for action of supercriticalfluid during continental subduction-zone metamorphism. Lithos 114 (3), 385–412.

Xu, J.W., Zhu, G., Tong, W.X., Cui, K.R., Liu, Q., 1987. Formation and evolution of theTancheng–Lujiang wrench fault system: a major shear system to the northern ofthe Pacific Ocean. Tectonophysics 134, 273–310.

Xu, S., Liu, Y.C., Chen, G., Ji, S., Ni, P., Xiao, W., 2005. Microdiamonds, their classificationand tectonic implications for the host eclogites from the Dabie and Su–Lu regionsin central eastern China. Mineralogical Magazine 69, 509–520.

Xu, W.L., Wang, Q.H., Yang, D.B., Liu, X.C., Guo, J.H., 2005. SHRIMP zircon U–Pb dating inJingshan “migmatitic granite”, Bengbu and its geological significance. Science inChina Series D — Earth Sciences 48 (2), 185–191.

Xu, H.J., Ma, C.Q., Ye, K., 2007. Early cretaceous granitoids and their implications for thecollapse of the Dabie orogen, eastern China: SHRIMP zircon U–Pb dating and geo-chemistry. Chemical Geology 240 (3–4), 238–259.

Xu, L.J., Xiao, Y.L., Wu, F., Li, S.G., Simon, K., Worner, G., 2012. Anatomy of garnets in a Ju-rassic granite from the south-eastern margin of the North China Craton: magmasources and tectonic implications. Journal of Asian Earth Sciences. http://dx.doi.org/10.1016/j.jseaes.2012.11.026.

Xue, A.M., Jin, W.J., Yuan, X.C., 1999. Tectonic evolution of Hefei basin, northern DabieMountain in the Mesozoic and the Cenozoic. Geological Journal of China University5, 157–163 (in Chinese with English abstract).

Yang, D.B., Xu, W.L., Wang, Q.H., Pei, F.P., 2010. Chronology and geochemistry of Mesozoicgranitoids in the Bengbu area, central China: constraints on the tectonic evolution ofthe eastern North China Craton. Lithos 114, 200–221.

Yuan, H.L., Gao, S., Liu, X.M., Li, H.M., Gunther, D., Wu, F.Y., 2004. Accurate U–Pb age andtrace element determinations of zircon by laser ablation–inductively coupled plasmamass spectrometry. Geostan Newslett 28, 353–370.

Zhang, K.J., 1997. North and south China collision along the eastern and southern NorthChina margins. Tectonophysics 270, 145–156.

Zhang, Z.M., Liou, J.G., Coleman, R.G., 1984. An outline of the plate tectonics of China. Geo-logical Society of America Bulletin 95, 295–312.

Zhao, W.L., Morgan, W.J., 1985. Uplift of Tibetan Plateau. Tectonics 4, 359–369.Zhao, G.C., Wilde, S.A., Cawood, P.A., Sun, M., 2001. Archean blocks and their bound-

aries in the North China Craton: lithological, geochemical, structural and P–Tpath constraints and tectonic evolution. Precambrian Research 107 (1–2),45–73.

Zhao, Z.F., Zheng, Y.F., Chen, R.X., Xia, Q.X., Wu, Y.B., 2007a. Element mobility inmafic andfelsic ultrahigh-pressuremetamorphic rocks during continental collision. Geochimicaet Cosmochimica Acta 71, 5244–5266.

Zhao, Z.F., Zheng, Y.F.,Wei, C.S., Wu, Y.B., 2007b. Post-collisional granitoids from the Dabieorogen in China: zircon U–Pb age, element and O isotope evidence for recycling ofsubducted continental crust. Lithos 93 (3–4), 248–272.

Zheng, Y.F., Fu, B., Gong, B., Li, L., 2003. Stable isotope geochemistry of ultrahigh pressuremetamorphic rocks from the Dabie–Sulu orogen in China: implications forgeodynamics and fluid regime. Earth-Science Reviews 62 (1–2), 105–161.

Zhu, G., Song, C.Z., Wang, D.X., Liu, G.S., Xu, J.W., 2001. Studies on Ar-40/Ar-39thermochronology of strike–slip time of the Tan–Lu Fault zone and their tec-tonic implications. Science in China Series D — Earth Sciences 44 (11),1002–1009.

Zhu, G., Song, S.-Z., Niu, M.-L., Liu, G.-S., Wang, Y.-S., 2002. Lithospheric textures of theTan–Lu Fault zone and its genetic analysis. Geological Journal of China Universities8 (3), 248–256 (in Chinese with English abstract).

Zhu, G., Liu, G.S., Niu, M.L., Xie, C.L., Wang, Y.S., Xiang, B.W., 2009. Syn-collisional trans-form faulting of the Tan–Lu Fault zone, East China. International Journal of Earth Sci-ences (Geol Rundsch) 98, 135–155.

Zhu, G., Niu, M., Xie, C., Wang, Y., 2010. Sinistral to Normal Faulting along the Tan‐Lu FaultZone: evidence for Geodynamic Switching of the East China Continental Margin. TheJournal of Geology 118 (3), 277–293.

1128 S. Li et al. / Gondwana Research 26 (2014) 1111–1128