The anomalous lithium isotopic signature of Himalayan collisional zone carbonatites in western Sichuan, SW China: Enriched mantle source and petrogenesis

Available online at www.sciencedirect.com

ScienceDirect Geochimica et Cosmochimica Acta 159 (2015) 42–60 www.elsevier.com/locate/gca

The anomalous lithium isotopic signature of Himalayan collisional zone carbonatites in western Sichuan, SW China: Enriched mantle source and petrogenesis Shihong Tian a,b,⇑, Zengqian Hou c,⇑, Aina Su d, Lin Qiu e, Xuanxue Mo b, Kejun Hou a, Yue Zhao a, Wenjie Hu f, Zhusen Yang a a

MLR Key Laboratory of Metallogeny and Mineral Assessment, Institute of Mineral Resources, Chinese Academy of Geological Sciences, Beijing 100037, PR China b School of Earth Science and Mineral Resources, China University of Geosciences, Beijing 10083, PR China c Institute of Geology, Chinese Academy of Geological Sciences, Beijing 100037, PR China d Institute of Hydrogeology and Environmental geology, Chinese Academy of Geological Sciences, Shijiazhuang 050061, Hebei Province, PR China e Department of Geology, University of Maryland, College Park, MD 20742, USA f Jiangxi Provincial Institute of Geological Survey, Nanchang 330030, Jiangxi Province, PR China Received 28 June 2014; accepted in revised form 14 March 2015; Available online 23 March 2015

Abstract Lithium concentrations and isotopic compositions of 38 carbonatites and associated syenites from the Maoniuping, Lizhuang, and Dalucao in western Sichuan, along with previously published and new Pb–Sr–Nd–C–O isotope data and whole-rock analyses, are used to constrain their mantle source and genesis. Carbonatites and syenites are characterized by extremely varying Li concentrations (0.8–120 ppm) and highly variable Li isotopic compositions ( 4.5& to +10.8&). Among them, the majority of the carbonatites and syenites have d7Li values between +0.2& and +5.8&, which overlap with the reported values for MORB and OIB; 3 carbonatites have higher d7Li values between +8.7& and +10.8&; 5 carbonatites and 4 syenites have lighter d7Li values between 4.5& and 0.3&. These highly variable d7Li compositions could not have been produced by diffusive-driven isotopic fractionation of Li and thus may record the isotopic signature of the late Proterozoic subcontinental lithospheric mantle (SCLM). This paper demonstrates the existence of anomalous d7Li within the late Proterozoic subcontinental lithospheric mantle, suggesting that the ancient SCLM beneath western Sichuan was modified by interaction with fluids derived from the subducted oceanic crust and marine sediments. The modeling curves of fluids derived from a dehydrated slab (ratios: AOC80–SED20 to AOC40–SED60) with a representative mantle composition can account for the majority of lithium compositional variations. Some samples with unusual Pb– Sr–Nd–O isotopic compositions and highly variable d7Li compositions are affected by significant involvement of marine sediments in their source region, not contaminated by crustal materials. The carbonatites and syenites in western Sichuan were

⇑ Corresponding authors at: MLR Key Laboratory of Metallogeny and Mineral Assessment, Institute of Mineral Resources, Chinese Academy of Geological Sciences, Beijing 100037, PR China (S. Tian), Institute of Geology, Chinese Academy of Geological Sciences, Beijing 100037, PR China (Z. Hou). E-mail addresses: [email protected] (S. Tian), [email protected] (Z. Hou), [email protected] (A. Su), [email protected] (L. Qiu), [email protected] (X. Mo), [email protected] (K. Hou), [email protected] (Y. Zhao), [email protected] (W. Hu), [email protected] (Z. Yang).

http://dx.doi.org/10.1016/j.gca.2015.03.016 0016-7037/Ó 2015 Elsevier Ltd. All rights reserved.

S. Tian et al. / Geochimica et Cosmochimica Acta 159 (2015) 42–60

43

generated by the partial melting of subcontinental lithospheric mantle, which was metasomatized by the Li-rich fluids derived from the subducted oceanic crust and marine sediments. This melting was most likely triggered by a Cenozoic asthenospheric mantle diapir related to Indian-Asian continental collision and post- or late-collisional stress relaxation in the Oligocene. Ó 2015 Elsevier Ltd. All rights reserved.

1. INTRODUCTION Lithium stable isotopes are increasingly being used to study recycling processes in the mantle (e.g., Aulbach and Rudnick, 2009; Vlaste´lic et al., 2009; Tang et al., 2014), because of their moderate incompatibility during magmatic processes (Ryan and Langmuir, 1987), strong fluid mobility (You et al., 1996), and highly variable Li isotopic compositions (up to 80&) in terrestrial samples (Tomascak, 2004). In contrast to the relatively uniform lithium isotopic composition of the convecting mantle (+2& to +6&, as inferred from relatively pristine olivine and fresh MORB; Chan et al., 1992; Moriguti and Nakamura, 1998; Tomascak, 2004; Magna et al., 2006a; Tomascak et al., 2008), the lithospheric mantle has highly variable d7Li values (40&; Tomascak, 2004), primarily because subducted sediments, altered basaltic and ultramafic rocks on the ocean floor have high Li contents (up to 80 ppm) and diverse Li isotopic compositions (d7Li = 12& to +21&) (Chan et al., 1992; Bouman et al., 2004; Rudnick et al., 2004) and therefore fluid/melt derived from the subducting slabs can strongly affect Li isotopic compositions in the overlying mantle wedge (Magna et al., 2006b, 2008; Tomascak et al., 2000). Equilibrium fractionation of lithium isotopes occurs in igneous systems at the low temperatures associated with pegmatite formation (Teng et al., 2006a), but not significantly at the high temperatures and high degrees of partial melting associated with basalt generation and differentiation (>1050 °C; Tomascak et al., 1999). This indicates that Li isotopes undergo negligible fractionation during magmatic processes, suggesting that the Li isotopic compositions of mantle-derived magmas may directly record the compositions of their source materials (e.g., Zhang et al., 2010; Tang et al., 2012, 2014; Su et al., 2012). Several studies have suggested that the anomalous d7Li values in the mantle, including high d7Li values (up to 7.9&) of HIMU lavas (Chan et al., 2009; Vlaste´lic et al., 2009), and low d7Li values (lower than 27.0&) of mantle-derived minerals (Tang et al., 2012, 2014) and lavas (Agostini et al., 2008), are related to isotopic fractionation associated with fluid loss during the dehydration of subducted oceanic crust (e.g., Zack et al., 2003; Elliott et al., 2004; Wunder et al., 2007). However, recent analyses of peridotites and phenocrysts within lavas indicate that Li isotopes can be strongly fractionated during diffusive kinetic fractionation associated with high-temperature igneous processes (e.g., Lundstrom et al., 2005; Teng et al., 2006b; Marschall et al., 2007; Rudnick and Ionov, 2007). This indicates that it is important to determine whether the non-MORB-like d7Li values of the mantle-derived samples are the result of

kinetic effects or the recycling of crustal components (Aulbach and Rudnick, 2009). Carbonatites are mantle-derived, intraplate magmas that provide a means of documenting the isotopic secular evolution of Earth’s mantle (Halama et al., 2008, 2011), and are usually interpreted to reflect the recycling of oceanic lithosphere (Hou et al., 2006a). This indicates that carbonatites may also provide useful evidence of crustal recycling. In addition, carbonatites form from extremely low-temperature, low-density, and low-viscosity magmas compared to typical silicate melts (Genge et al., 1995), which means that these carbonatitic magmas rapidly rise to Earth’s surface and undergo only limited interaction with continental crustal material (Bell and Tilton, 2002). This implies that carbonatites can provide robust evidence for the composition of the lithospheric mantle. Moreover, carbonatite-related rare-earth-element (REE) deposits are the most significant source of the world’s REEs, accounting for 50% of all global REE reserves (Hou et al., 2015). A wide variety of carbonatite-associated REE deposits in China, including Bayan Obo (the world’s largest LREE– Fe–Nb deposit; Yang et al., 2009; Lai et al., 2012; Ling et al., 2013), Maoniuping (the giant LREE deposit; Hou et al., 2009) and numerous medium-large LREE deposits (Ying et al., 2004; Hou et al., 2009; Xu et al., 2010), account for approximately 65% of China’s REE reserves. Carbonatites in rift zones are generally thought to be derived from the melting of sublithospheric mantle triggered by mantle plume activity (Bell and Simonetti, 2010); however, the genesis of the carbonatite in a collisional zone remains unclear. Halama et al. (2007, 2008) presented the Li isotopic compositions of carbonatites and associated silicate rocks in rift zones, but to date no Li isotope data have been reported from carbonatites and associated silicate rocks in a collisional zone. Here, we present the first anomalous Li isotope compositions of carbonatites and spatially associated syenites from three districts in the eastern IndianAsian collisional zone (EIACZ), and use these data to provide new constraints not only on the nature of the source region and the petrogenesis of these carbonatitic magmas but also on the processes that occurred during crust–mantle interaction related to the Indian-Asian continental collision. In addition, lithium stable isotopes are an effective method to make a difference between the collisional zone carbonatites and rift zone carbonatites. 2. GEOLOGICAL BACKGROUND AND SAMPLES Cenozoic carbonatite–alkalic complexes within the EIACZ are located on the western margin of the Yangtze

44

S. Tian et al. / Geochimica et Cosmochimica Acta 159 (2015) 42–60

95°

100°

105° Tianshan Mts

35°

Asia

N South China

Yushu

Nangqian

Batang 30°

Fig.1

Fa ul t

Chengdu

t ul Fa

Gongjue

India Lo th ngm ru e st ns be ha lt n

ng jia Li gtan Ba

Xi an sh ui he

Jia li Fa ult

Ganluo Fault Lijiang

Xiaojiang Fault

Lanping

Fig.2

Indian Continent

South China

Dali

t

25°

Fa

ul

Re

g

d

Ri

ao l

ig

on

ve

r

G

Fa

ult

Ai

lao

sh

an

sh

ea

LanpingSimao fold belt

r

zo

ne

Thrust belt

Fold

Strike-slip fault

Tertiary basin

CretaceousJurassic basin

Potassic intrusion

Lamprophyre district

Carbonatite alkalic complex

Fig. 1. Cenozoic tectonic map of eastern Tibet (Wang et al., 2001), showing the location of the Himalayan potassic rock belt (Zhang and Xie, 1997; Chung et al., 1998), a shoshonitic lamprophyre district (Guo et al., 2005), and a belt of carbonatite–alkalic complexes (Yuan et al., 1995) that define a semi-continuous Cenozoic igneous province within the eastern Indian-Asian collisional zone.

S. Tian et al. / Geochimica et Cosmochimica Acta 159 (2015) 42–60

Craton in southern China (Fig. 1). The basement of the Yangtze craton consists of Archean high-grade metamorphic rocks and Proterozoic metasedimentary rocks, which are covered by Phanerozoic clastic and carbonate sequences (Luo et al., 1998). The Phanerozoic cover units in this area are dominated by Silurian–Triassic clastic-carbonate sequences and a mid-Permian flood basalt sequence that contains minor amounts of picrites and picritic basalts. The former covers the majority of the western margin of the Yangtze Craton, whereas the latter forms a large igneous province with an area of 500,000 km2 (Xu et al., 2001), which is most likely related to the Emeishan mantle plume (Chung and Jahn, 1995; Xu et al., 2001; Hou et al., 2006b). This mantle plume was associated with the formation of the Panxi paleo-rift zone (Zhang et al., 1988), which is bounded by the N–S-striking Yalongjiang and Ganluo faults (Fig. 2).

45

The 65–50 Ma Indian-Asian collision in southern Tibet (Yin and Harrison, 2000; Zhong et al., 2001; Mo et al., 2003) defines the western margin of the Yangtze Craton and forms a N–S-oriented collisional orogenic belt, i.e., the Jinpingshan Orogen, which lies along the eastern margin of the Tibetan plateau (Fig. 2; Luo et al., 1998). A series of Cenozoic strike-slip faults (Fig. 1) accommodated and adjusted the stress and strain within the EIACZ that were produced by this collision (Wang et al., 2001). Accompanying with the Indian-Asian collision, Himalayan magmatic activity within the EIACZ forms a semi-continuous potassic igneous province in southwestern China (Fig. 1; Guo et al., 2005), including a 1000-km-long potassic rock belt (Zhang and Xie, 1997; Chung et al., 1998), a shoshonitic (calc-alkaline) lamprophyre district (Guo et al., 2005), and a 270-km-long belt of carbonatite– alkalic complexes in western Sichuan (Yuan et al., 1995).

102 Mianning

N

Muluozhai

Anninghe Fault

Ganl

uo

Xichang

Paleozoic-Mesozoic cover strata

Fault

Taihe Fault

28

Lizhuang

Precambrian metamorphic basement

Yalongjiang

Jinpingshan orogenic belt

Maoniuping

Butuo

Dechang Dalucao

28

Puluo

Cida

Xi

Baima

27 ng

jia

ao

27

Fa t

ul

Yanbian

Carbonatitealkalic complex Syenite intrusion

Panzhihua

Granite pluton 0

15

Strike-slip fault

30 km

102 Fig. 2. Sketch tectonic map showing the distribution of Himalayan carbonatite–alkalic complexes; the intrusion of these complexes was controlled by reactivated faults within the western Sichuan area (modified from Yuan et al., 1995).

46

S. Tian et al. / Geochimica et Cosmochimica Acta 159 (2015) 42–60

All of this magmatism within the EIACZ occurred over a relative short period of time (40–24 Ma, peaking at 35 Ma; Hou et al., 2006a). This study focuses on three representative carbonatite– alkalic complexes in western Sichuan, which from north to south are the Maoniuping, Lizhuang, and Dalucao complexes (Fig. 2). Carbonatites in these areas are present as sills, dikes, stocks, and hypabyssal intrusions within syenitic intrusions. All of the samples analyzed during this study were fresh and showed no evidence of significant hydrothermal alteration. The Maoniuping complex intrudes a Mesozoic granitic pluton and a nearby rhyolitic succession, and is dominated by pink calcite carbonatite and white coarse-grained calcite carbonatite, with 90% modal calcite and subordinate biotite, aegirine, aegirine–augite, riebeckite, arfvedsonite, and microcline. The syenites associated with these carbonatites are gray–white and fine-grained, and contain K-feldspar (Or > 97%), albite (Ab > 98%), and quartz (11–15%), with minor aegirine, phlogopite, biotite, and calcite (Yuan et al., 1995). The Sm–Nd isochron age of calcite from the carbonatite within the Maoniuping complex is 29.9 ± 1.7 Ma (Hu et al., 2012) and two syenites have yielded SHRIMP U–Pb zircon ages of 22.81 ± 0.31 Ma and 21.30 ± 0.40 Ma, respectively (Liu et al., 2014). The Lizhuang complex intrudes a Mesozoic granite intrusion and an overlying 1000-m-thick sequence of Silurian–Triassic carbonate and sandy mudstones, and consists of a NNW-SSE-striking syenitic stock and a series of carbonatitic sills. The least-altered carbonatites within the complex are yellow–brown and fine-grained, and contain calcite (82%) and minor arfvedsonite, aegirine, aegirine–augite, and biotite. The associated syenitic stock is dominated by a syenite phase that contains 75% modal microcline, 12% aegirine–augite, 10% quartz, and subordinate albite, arfvedsonite, and hematite. The Re–Os modal age of molybdenite from the carbonatite within the Lizhuang complex is 28.5 Ma and one syenite has yielded SHRIMP U–Pb zircon age of 27.41 ± 0.35 Ma (Liu et al., 2014). The Dalucao complex consists of a series of syenitic stocks, LREE-mineral-bearing carbonatitic breccia-pipes and carbonatite dikes, all of which intrude a Proterozoic quartz diorite pluton and a Triassic–Jurassic sedimentary sequence. The carbonatite dikes are white and mediumgrained, and contain calcite and minor arfvedsonite, fluorite, quartz, and bastnaesite. The carbonatites in the breccia-pipes are yellow–brown and fine-grained, and contain a hydrothermal bastnaesite, fluorite, and mica assemblage that overprints the original igneous carbonatite assemblage. The carbonatite and syenite within the Dalucao complex have yielded SHRIMP U–Pb zircon ages of 12.99 ± 0.94 and 14.53 ± 0.31 Ma, respectively (Tian et al., 2008), and two syenites have also yielded SHRIMP U–Pb zircon ages of 12.13 ± 0.19 Ma and 11.32 ± 0.23 Ma, repectively (Liu et al., 2014). 3. ANALYTICAL METHODS After petrographic examinations in thin section, selected least-altered whole-rock samples were crushed and powdered in an agate mill prior to geochemical and

Sr–Nd–Pb–Li isotopic analyses. Whole-rock analyses of these fresh rock samples were undertaken at the National Research Center for Geoanalysis, Chinese Academy of Geological Sciences (CAGS), Beijing, China. Major element concentrations were determined by X-ray fluorescence (XRF) with analytical uncertainties of 1–5%, whereas trace element (including the REE) concentrations were determined by inductively coupled plasma-mass spectrometry (ICP-MS) with a precision generally better than 5%. The ICP-MS analysis undertaken during this study used about 50 mg of powdered sample that was dissolved in a highpressure Teflon bomb for 24 h using a HF+HNO3 mixture. An internal standard solution containing the single element Rh was used to monitor signal drift during counting and a set of USGS and Chinese national rock standards, including BCR-1, BHVO-1, AGV-2, GSR-1, and GSR-3, were used for analytical calibration, yielding an analytical precision that was typically 2–5%. The C–O isotopic compositions of the carbonatites analyzed during this study used an approach where calcite separates were reacted with phosphoric acid at 25 °C to release CO2, and detailed procedures are given in Hou et al. (2006a). The Sr, Nd, and Pb isotopic analyses undertaken during this study used 50 mg samples that were digested in a HNO3–HF solution at 120 °C on a hot plate for 1–2 days until the samples were completely dissolved. Initial eNd values and 87Sr/86Sr ratios were calculated according to the ages of the corresponding carbonatites and syenites. Detailed procedures for sample dissolution, column chemistry, and instrumental analyses are also presented in Hou et al. (2006a). Separation of Li and precise lithium isotope analysis were undertaken at the Geochemistry Laboratory, University of Maryland, College Park, USA and at the MLR (Ministry of Land and Resources) Key Laboratory of Metallogeny and Mineral Assessment, Institute of Mineral Resources. Detailed procedures for sample dissolution, column chemistry, and instrumental analyses are given in Rudnick et al. (2004), Teng et al. (2004, 2006a), Tian et al. (2012), and Zhao et al. (2015). In brief, samples were initially dissolved in a HF–HNO3 solution before Li was purified using a cation exchange resin in a HCl medium, followed by in a HCl–ethanol medium. Lithium concentrations and isotopic compositions were determined at the University of Maryland using a Nu Plasma MC-ICP-MS. Each unknown was bracketed both before and after by measurements of the L-SVEC standard and the d7Li value of the unknown was calculated relative to the average of these two bracketing L-SVEC analyses. IRMM-016 and the in-house UMD-1 (a purified Li solution from Alfa AesarÒ) standard solutions were routinely analyzed during the course of each analytical session, yielding results for both standards (IRMM-016: d7Li = +0.2& ± 0.4&, 2r, n = 6; UMD-1: d7Li = +54.9& ± 0.8&, 2r, n = 7) that agree with previously published data (e.g., Rudnick et al., 2004; Halama et al., 2007, 2008). Two international rock standards were also analyzed during this study to evaluate the accuracy of these measurements; the basaltic BHVO-1 standard yielded a d7Li value of +4.2 ± 0.8& (2r, n = 4) and the andesitic AGV-1 standard yielded a d7Li value of +4.9 ± 0.7& (2r, n = 4). The

S. Tian et al. / Geochimica et Cosmochimica Acta 159 (2015) 42–60

47

external precision of Li isotopic analyses was determined by 2r values of repeat runs of pure Li standard solutions and rock solutions, yielding a precision of 6±1.0& (Teng et al., 2004). Li concentrations were determined by comparing signal intensities of sample solution with a 50 ppb L-SVEC standard solution and adjusting for sample weight. The accuracy of this method was established by Teng et al. (2004) as being ±5% based on isotope dilution methods, and the precision is less than ±10% (Teng et al., 2006b). Lithium isotopic compositions were determined at the Institute of Mineral Resources using a standard bracketing method and a Thermo Finnigan Neptune MC-ICP-MS instrument. The d7Li value of each unknown was calculated relative to the average of two bracketing IRMM-016 runs. During the course of this study, two international rock standards were analyzed to evaluate the accuracy of the measurements, with the basaltic BHVO-2 standard yielding a d7Li value of +4.3 ± 0.8& (2r, n = 18; Tian et al., 2012) and the andesitic AGV-2 standard yielding a d7Li value of +6.1 ± 0.4& (2r, n = 18; Zhao et al., 2015). The external precision of Li isotopic analyses was based on 2r values of repeat runs of pure Li standard solutions and rock solutions over a three-year period and is less than ±1.0& (Tian et al., 2012; Zhao et al., 2015). Li concentrations were determined by ICP-MS at the National Research Center for Geoanalysis, CAGS. The results of analyses of the same samples yielded almost consistent results between the MLR Key Laboratory of Metallogeny and Mineral Assessment and the Geochemistry Laboratory, University of Maryland (Table 1).

Carbonatites in the study area have extremely varying Li concentrations (0.8–120 ppm) that are similar to the range of concentrations reported for carbonatites within rift zones (Halama et al., 2008), even considering the uneven distribution of Li-bearing phases in carbonatites and the high incompatibility of Li in calcite (Halama et al., 2008). The Li concentrations of arfvedsonites and biotites from the carbonatites are unusually high (359.9–1925.5 ppm) compared to their corresponding whole-rock compositions. In contrast, the majority of the syenites analyzed in this study have Li concentrations between 17 and 39 ppm, which are similar to mafic silicate rocks within rift zones (Halama et al., 2008) and reflect the moderate incompatibility of Li in silicate magmas. The majority of the carbonatite samples (including arfvedsonite and biotite separates) and spatially associated syenites analyzed during this study have d7Li values between +0.2& and +5.8&, which overlap with the reported values for MORB and OIB (+1& to +7&; Fig. 4; Chan et al., 1992; Moriguti and Nakamura, 1998; Tomascak, 2004; Magna et al., 2006a; Tomascak et al., 2008). In contrast, three carbonatites have higher d7Li values between +8.7& and +10.8&, and five carbonatites (including arfvedsonite separates) and four syenites have lighter d7Li values between 4.5& and 0.3& (Fig. 4c). The fact that the arfvedsonites and biotites and their hosting carbonatites have similar Li isotopic compositions (within ±1& uncertainty) means that hereinafter we consider only the whole-rock data for the carbonatites.

4. RESULTS

5. DISCUSSION

Lithium concentrations and isotopic compositions of the western Sichuan carbonatites and associated syenites are given in Table 1, along with previously published (Hou et al., 2006a, 2015) and new Pb–Sr–Nd–C–O isotope data and whole-rock analyses. 4.1. Carbon and oxygen isotopes Most of the western Sichuan carbonatites have d13CV-PDB and d18OV-SMOW values of 8.8& to 4.4& and of +7.4& to +9.6&, respectively, and are generally similar to the C–O isotopic compositions of primary, mantle-derived carbonatites (Fig. 3a; Taylor et al., 1967; Keller et al., 2006). However, one carbonatite (LZ-01) from the Lizhuang has the higher d18OV-SMOW value (+10.5&) and the heavier d13CV-PDB value ( 3.9&). One carbonatite (DLC049-3) from the Dalucao has the higher d18OV-SMOW value (+13.1&) and the lighter d13CV-PDB value ( 9.9&). Three carbonatites (DLC016-2, DLC030-3, and DLC030-5) from the Dalucao have the higher d18OV-SMOW values (+11.7& to +19.7&) whereas the primary, mantle-derived d13CV-PDB values ( 7.9& to 4.7&). The syenites in the study area have a relatively small range in d18OV-SMOW values (+6.3& to +8.5&), which are slightly lighter than the calcites from the associated carbonatites (Fig. 3b).

4.2. Lithium concentrations and isotopic compositions

5.1. Lithium isotope fractionation during alteration/ weathering and magmatic differentiation The fact that Li is fluid mobile and the lighter isotope Li is preferentially retained in solid phases whereas 7Li preferentially moves into solution during weathering (Kisaku¨rek et al., 2004) means that Li isotopes can be fractionated by alteration or weathering. However, the carbonatite samples analyzed during this study are fresh and free of clay minerals, and contain primary igneous minerals, meaning that these samples are free of any effects of alteration or weathering. In addition, the majority of samples analyzed during this study plot within the known range of compositions of primary magmatic carbonatites (Fig. 3a), and have the mantle-like Sr–Nd isotopic compositions (this figure not shown; see Fig. 6 in Hou et al., 2006a), confirming that alteration or weathering has had little effect on the Li isotopic compositions of these rocks (Halama et al., 2007, 2008). Nevertheless, as a precaution, we have removed samples LZ-01 and DLC049-3 from further discussion as both samples have d13C and d18O values that fall outside the range of primary igneous carbonatites (samples in parentheses in Fig. 3a), although this may erroneously lead to the elimination of purely mantle derived carbonatites. Three samples (DLC016-2, DLC030-3, and DLC030-5) have elevated d18OV-SMOW values (Fig. 3a), 6

48

Table 1 Li concentrations, Li–O–C–Pb–Sr–Nd isotopic compositions and selected geochemical parameters of carbonatites and associated syenites from western Sichuan. Sample No.

Rock type

Material WR WR Arfvedsonite WR WR WR Arfvedsonite WR Arfvedsonite Arfvedsonite WR WR WR WR

Lizhuang district LZ-01 Carbonatite LZ-03 Carbonatite LZ-09 Carbonatite LZ-17 Carbonatite LZ-127 Carbonatite LZ-127 Carbonatite LZ-106 Carbonatite LZ09-02-1 Syenite

WR WR WR WR WR Biotite Biotite WR

Dalucao district DLC001-3 DLC038-1 DLC040-1 DLC099-2 DLC100-1 DLC102-1 DLC016-2 DLC030-3 DLC030-5 DLC049-3 DLC002-1 DLC002-2 DLC003-1-1

WR WR WR WR WR WR WR WR WR WR WR WR WR

Carbonatite Carbonatite Carbonatite Carbonatite Carbonatite Carbonatite Carbonatite Carbonatite Carbonatite Carbonatite Syenite Syenite Syenite

d7Li (&)

1.1a 49.4a 459.7a 4.2a 46.7a 28.9a 359.9a 43.5c 372.4c 420.8c 4.1a 34.5a 29.8c 32.5c

3.9a 2.9a 3.1a 0.5a/ 0.9a 1.1a/ 0.4a/ 1.1b 1.5b 0.4b 2.8a 0.2a/ 2.0b 2.7b

10.9a 0.8a 42.4a 65.0c 86.5c 1925.5a 464.5a 28.5c

2.3a 8.7a 4.4a 1.8b 1.5b 3.7a 2.0a 2.1b

3.1a 16.5c 14.0a 4.2a 18.4a 1.1a 92.1a 43.4a 120.0a 86.7c 38.5c 22.6c 28.7c

0.2b 0.3b 3.0b

0.3b/ 0.8b

10.8a/10.2b/11.5b 1.3b 3.2a/ 4.8b 4.6a 4.4a/ 3.2b 1.4a 1.9a 1.6a 1.8a 9.7b 0.3b 2.1b 2.3b

SiO2 (%)

Fe2O3 (%)

Rb (ppm)

d18O (&)

1.86e 7.06e

<0.05e 0.7e

2.5e 4.01e

7.9e 7.6e

1.33e 2.74e 1.79e

0.16e 0.22e <0.05e

5.83e 4.22e 5.34e

16.3e

2.74e

70.63d 70.92f 66.54d 52.86d

d13C (&)

208

207

206

(87Sr/86Sr)i

6.7e 6.5e

38.4553e 38.4512e

15.5947e 15.5929e

18.2157e 18.2310e

0.706146e 0.706146e

7.4e 7.5e 7.5e

6.6e 6.5e 6.6e

38.4563e 38.4604e 38.4527e

15.5952e 15.5938e 15.5935e

18.2202e 18.2236e 18.2206e

0.706169e 0.706181e 0.706155e

20e

7.6d

6.5d

38.4569e

15.5940e

18.2243e

0.706126e

0.26d 0.93f 0.21d 0.34d

206d 163f 196d 58.1d

7.5f 7.9f 6.8d 7.1d

38.7825d 38.473f 38.4528d 38.4763d

15.6084d 15.587f 15.5930d 15.5964d

18.3256d 18.233f 18.2168d 18.2560d

0.706447d 0.707038f 0.706053d 0.706129d

2.61f <0.01f 0.98d 0.66d 1.35d

1.38f 0.32f 0.09d 0.32d 0.63d

12.4f 1.03f 20.1d 11.6d 7.76d

10.5f 9.6f 9.0f 8.7f 8.8e

38.4342f 38.4013f 38.4084f 38.4243f 38.3573e

15.6017f 15.6014f 15.6025f 15.6038f 15.5873e

18.2201f 18.1965f 18.2010f 18.2069f 18.1926e

0.706305f 0.706713f 0.706997f 0.706314f 0.706195e

75.83e

1.06e

114e

7.4d

38.7801d

15.6134d

18.7035d

0.706330d

17.1e 1.94e 6.87e 9.34d 22.8d 1.65d 3.56d 7.83d 2.8d 7.95d 66.4e 67.4e 68.1e

0.58e 0.16e 1.47e 0.80d 1.53d 0.24d 0.86d 0.86d 1.03d 0.08d 1.90e 1.41e 1.83e

1.92e 10.4e 3.42e 4.58d 10.00d 1.86d 16.7d 7.69d 9.42d 4.78d 187e 296e 139e

8.5e 8.2e 8.2e 8.3d 8.2d 8.4d 17.7d 19.7d 11.7d 13.1d 7.0d 7.8d 8.5d

39.066e 38.888e 38.919e 38.715d 38.812d 38.591d 39.124d 38.362d 39.200d 38.801d 38.640d 38.801d 38.657d

15.708e 15.701e 15.713e 15.652d 15.683d 15.631d 15.779d 15.701d 15.802d 15.678d 15.634d 15.676d 15.634d

18.270e 18.270e 18.270e 18.230d 18.250d 18.185d 18.316d 18.256d 18.333d 18.234d 18.209d 18.251d 18.215d

0.707863e 0.707962e 0.707790e 0.707876d 0.707872d 0.707848d 0.707804d 0.707791d 0.707780d 0.707818d 0.707779d 0.707726d 0.707986d

3.9f 4.6f 4.4f 4.7f 4.5e

8.2e 8.7e 8.8e 8.5d 8.5d 8.5d 5.8d 7.9d 4.7d 9.9d

Pb/204Pb

Pb/204Pb

Pb/204Pb

(continued on next page)

S. Tian et al. / Geochimica et Cosmochimica Acta 159 (2015) 42–60

Maoniuping district MNP-118 Carbonatite MNP-125 Carbonatite MNP-125 Carbonatite MNP-129 Carbonatite MNP-131 Carbonatite MNP-147 Carbonatite MNP-147 Carbonatite MNP09-04-10 Carbonatite MNP09-04-10 Carbonatite MNP09-04-3 Carbonatite MNP-04 Syenite MNP-24 Syenite MNP09-04-2 Syenite MNP09-04-11 Syenite

Li (ppm)

Table 1 (continued) Sample No. DLC003-1-2 DLC006-3 DLC103-1

Rock type Syenite Syenite Syenite

Material WR WR WR

Li (ppm)

d7Li (&)

c

17.9 29.2c 20.6c

SiO2 (%)

b

e

5.8 3.0b 4.5b

63.1 61.2e 61.1e

Rock type

Material

143

Nd/144Nd

Carbonatite Carbonatite Carbonatite Carbonatite Carbonatite Carbonatite Carbonatite Carbonatite Carbonatite Carbonatite Syenite Syenite Syenite Syenite

WR WR Arfvedsonite WR WR WR Arfvedsonite WR Arfvedsonite Arfvedsonite WR WR WR WR

0.512463e 0.512487e

Lizhuang district LZ-01 LZ-03 LZ-09 LZ-17 LZ-127 LZ-127 LZ-106 LZ09-02-1

Carbonatite Carbonatite Carbonatite Carbonatite Carbonatite Carbonatite Carbonatite Syenite

Dalucao district DLC001-3 DLC038-1 DLC040-1 DLC099-2 DLC100-1 DLC102-1 DLC016-2 DLC030-3 DLC030-5 DLC049-3 DLC002-1

Carbonatite Carbonatite Carbonatite Carbonatite Carbonatite Carbonatite Carbonatite Carbonatite Carbonatite Carbonatite Syenite

e

2.95 5.29e 3.92e

Rb (ppm) e

189 251e 164e

d18O (&) d

7.9 8.2d 6.3d

d13C (&)

208

Pb/204Pb d

38.741 38.662d 38.494d

207

Pb/204Pb d

206

Pb/204Pb d

15.656 15.637d 15.571d

18.240 18.219d 18.135d

(87Sr/86Sr)i 0.707953d 0.707971d 0.707979d

Ba (ppm)

Nb (ppm)

Y (ppm)

Th (ppm)

2.7e 2.4e

25900e 27870e

4.69e 88.0e

232e 159e

104e 61.2e

0.512449e 0.512445e 0.512417e

3.0e 3.2e 3.7e

20580e 1071e 7547e

2.94e 1.27e 0.77e

134e 124e 131e

6.67e 2.31e 1.96e

0.512518e

1.6e

22630e

63.0e

84.4e

61.2e

0.512360d 0.512399f 0.512561d 0.512437d

4.7d 4.0f 0.7d 3.2d

6000d 1384f 14780d 32810d

21.4d 14.8f 49.9d 96.4d

65.2d 36.7f 34.5d 63.4d

69.6d 60f 41.5d 50d

WR WR WR WR WR Biotite Biotite WR

0.512372f 0.512412f 0.512432f 0.512441f 0.512506e

4.9f 4.2f 3.6f 3.2f 1.8e

60800f 45300f 41300d 36100d 34680d

3.42f 0.88f 4.98d 3.66d 11.6d

128f 94.5f 151d 102d 126d

48.7f 46.1f 347d 341d 174d

0.512436d

3.2d

1436e

45.8e

6.81e

101e

WR WR WR WR WR WR WR WR WR WR WR

0.512316e 0.512325e 0.512341e

6.0e 5.9e 5.6e

0.512186d 0.512337d 0.512339d 0.512336d 0.512346d 0.512342d

8.6d 5.7d 5.6d 5.7d 5.5d 5.5d

8970e 65780e 2031e 1830d 5121d 59640d 31670d 31650d 33820d 23090d 7663e

3.12e 40.6e 2.26e 9.42d 19.4d 5.16d 1.44d 1.31d 1.99d 4.85d 26.4e

60.1e 86.8e 544e 75.7d 520d 36.2d 159d 122d 144d 109d 11e

5.17e 9.15e 139e 9.23d 350d 2.51d 46.6d 45.7d 99.8d 29d 25.1e

eNd(t)

S. Tian et al. / Geochimica et Cosmochimica Acta 159 (2015) 42–60

Sample No. Maoniuping district MNP-118 MNP-125 MNP-125 MNP-129 MNP-131 MNP-147 MNP-147 MNP09-04-10 MNP09-04-10 MNP09-04-3 MNP-04 MNP-24 MNP09-04-2 MNP09-04-11

Fe2O3 (%)

49

S. Tian et al. / Geochimica et Cosmochimica Acta 159 (2015) 42–60

Notes: Replicate analyses of samples involved sample dissolution, column chemistry, and MC-ICP-MS analysis. a Analyzed at the Geochemistry Laboratory, University of Maryland, College Park, USA. b Analyzed at the MLR Key Laboratory of Metallogeny and Mineral Assessment, Institute of Mineral Resources, Chinese Academy of Geological Sciences, Beijing, China. c Analyzed at the National Research Center for Geoanalysis, Chinese Academy of Geological Sciences, Beijing, China. d This paper. e Hou et al. (2015). f Hou et al. (2006a). WR = whole rock.

DLC002-2 DLC003-1-1 DLC003-1-2 DLC006-3 DLC103-1

Syenite Syenite Syenite Syenite Syenite

WR WR WR WR WR

0.512336d 0.512343d 0.512366d 0.512337d

5.6d 5.5d 5.0d 5.6d

1719e 4094e 4999e 5354e 3516e

25e 32.6e 33e 16.3e 43.5e

14e 14.3e 31.8e 35.7e 31.3e

118e 56.6e 61.1e 34.2e 76.7e

50

but have Li and C isotopic compositions that are indistinguishable from “fresh” carbonatites (Fig. 3c); this may be consistent with alteration modeling, which suggests that low-temperature alteration can affect O isotopic compositions more than it can affect Li isotopic compositions (Halama et al., 2008), and thus these samples have Li isotopic compositions that are considered to be pristine (Halama et al., 2007). For the syenites, the freshest available materials were chosen for analyses; these samples have no petrographic evidence of alteration or weathering and have the mantle-like Sr–Nd (this figure not shown; see Fig. 6 in Hou et al., 2006a) and oxygen isotopic compositions (7.5&; Blattner et al., 1989, 2002), suggesting that the Li isotopic compositions of these samples were not significantly affected by alteration or weathering, although some minor weathering cannot be fully excluded. Furthermore, there exist almost positive d7Li–d18O (except samples DLC016-2, DLC030-3, and DLC030-5 in Fig. 3b) and d7Li–d13C (Fig. 3c) correlations for the carbonatites and associated syenites, both of which are contrary to the Halama et al.’s (2008) alteration model curves, which define negative correlations of d7Li with both d13C and d18O, indicating that the Li isotopic compositions of these samples were not related to low-temperature alteration. Lithium isotope fractionation during magmatic differentiation appears to be insignificant at the high temperatures (Tomascak et al., 1999). In addition, the Li isotopic compositions of carbonatites and spatially associated syenites from the western Sichuan area do not significantly correlate with the degree of magmatic differentiation, as inferred from various compositional parameters (e.g., Li, Rb, SiO2, and Fe2O3 concentrations; Fig. 6), indicating that their d7Li compositions are unlikely to have been affected by magmatic differentiation, as has been suggested by Teng et al. (2009). 5.2. Li isotope composition of the mantle source for the western Sichuan carbonatites The highly variable Li concentrations (0.8–120 ppm; Table 1) of carbonatites in the western Sichuan area make it unlikely to shift their d7Li by bulk crustal assimilation, since typical crustal rocks and mantle samples have low Li concentrations (averages of 35 ± 11 ppm, 13 ppm and 1–2 ppm for upper-crustal, lower-crustal rocks and mantle rocks, respectively; Jagoutz et al., 1979; Rudnick and Gao, 2003; Ottolini et al., 2004; Teng et al., 2004) and their concentration differences are not large, meaning that this shift might be very limited and it is impossible that the Li concentrations of carbonatites are more than 35 ppm, which is not the case for the samples from the study (Table 1). The continental crust appears to be isotopically light compared to the mantle (d7Li of 1.7&; Teng et al., 2009), suggesting that mantle-derived magmas (d7Li of 4&; Halama et al., 2008) that assimilate continental crust should shift toward lighter d7Li values. However, the lithium isotopic differences between crust and mantle are very small, yielding rather limited shift (Halama et al., 2008), which is contrary to the d7Li values (d7Li of 4.5 to +10.8&; Fig. 4) of the carbonatites within the western

S. Tian et al. / Geochimica et Cosmochimica Acta 159 (2015) 42–60

51

a

b

c

Fig. 3. Diagrams showing variations in (a) carbon vs. oxygen, (b) lithium vs. oxygen, and (c) lithium vs. carbon isotopic compositions of carbonatites and associated syenites within the western Sichuan area. The rectangular field encompasses the composition of “primary” igneous carbonatites, using published d18O, d13C values (Taylor et al., 1967; Keller et al., 2006) and d7Li values (Chan et al., 1992; Moriguti and Nakamura, 1998; Tomascak, 2004; Magna et al., 2006a; Tomascak et al., 2008). Symbols in parentheses denote samples that show evidence by having C and O isotopic compositions outside the field of typical mantle values. Note, however, that mantle-derived carbonatite magmas may show significant variations in O isotopic compositions and plot outside this field due to low-temperature alteration, but Li and C isotopic compositions are indistinguishable from “fresh” carbonatites (see text for discussion).

52

S. Tian et al. / Geochimica et Cosmochimica Acta 159 (2015) 42–60 60

OIB

a

MORB

50

Frequency

40 30 20 10 0 -4

-3 -2

-1

0

1

2

3

5

4

6

7

9

8

10 11 12

Average mantle value 14 Mafic silicate rocks

b

Carbonatites

12

Frequency

10 8 6 4 2 0

-4

-3 -2

-1

0

1

2

3 4 5 δ 7 Li (‰)

6

7

8

9

10 11 12

Average mantle value 11 10

Maoniuping Carbonatites

c

9

Syenites Lizhuang

Frequency

8

Carbonatites

7

Syenites

6 5

Dalucao Carbonatites

4

Syenites

3 2 1 0 -5 -4

-3 -2

-1

0

1

2

3

4

5

6

7

8

9

10 11 12

δ 7 Li (‰)

Fig. 4. Frequency distribution diagrams of d7Li values for (a) MORB (Elliott et al., 2006; Nishio et al., 2007; Tomascak et al., 2008) and OIB (Tomascak et al., 1999; Nishio et al., 2005), (b) carbonatites and mafic silicate rocks in rift zones (Halama et al., 2008), and (c) carbonatites and syenites from the study area (two samples with petrographic and C–O isotopic evidence for near-surface alteration are excluded, but four arfvedsonite and two biotite samples are included in this diagram; see text for discussion). The dashed line indicates the average mantle value of +4&.

Sichuan area. Halama et al. (2008) used modeled mixing between carbonatite magma and continental crustal material to suggest that variations of 2& in the d7Li composition of natrocarbonatites and calciocarbonatites would require 80% and 10% of assimilation of crustal material, respectively, both of which contrast sharply with the major

element compositions of these carbonatites. In addition, the low magmatic temperature of carbonatite magmas significantly inhibits the assimilation of crustal material (Halama et al., 2007, 2008), and the physical properties of carbonatite magmas, including their low density, low viscosity and consequent rapid ascent to Earth’s surface

S. Tian et al. / Geochimica et Cosmochimica Acta 159 (2015) 42–60

53

radiogenic isotopic compositions (Fig. 5) or chemical parameters (Fig. 6). This strongly suggests that the Li isotopic heterogeneity most likely reflects heterogeneity in the source rocks for these carbonatitic magmas, as is consistent with previous research into A-type granites (Teng et al., 2009). 5.3. Origin of the anomalous d7Li composition of mantlederived carbonatites and syenites in the western Sichuan area The young TDM age (0.91 Ga) of rocks in the study area coincides with the onset of initial subduction of the Proterozoic Pro-Tethyan oceanic plate, as is further evidenced by the Proterozoic Kangding arc granitoid batholiths (853–700 Ma) and contemporaneous arc volcanic rocks in the Yangtze Craton (Hou et al., 2006a), suggesting that the study area underwent an ancient oceanic slab subduction event (Xu et al., 1995; Luo et al., 1998). The subducted oceanic crustal material, associated pelagic/ terrigenous sediments, and fluids derived from the subducted slab could have significantly influenced the isotopic compositions of the subcontinental lithospheric mantle (Tatsumi et al., 1986). Collision-related carbonatites and spatially associated syenites in the western Sichuan area have d7Li values that range from significantly heavier to significantly lighter than the presumed values for the upper mantle, and have a larger range in Li isotopic compositions than has been previously observed for any carbonatites and associated silicate rocks on Earth. These Li isotopic data are consistent with observations from Cenozoic subduction-related calc-alkaline and ultrapotassic rocks in western Anatolia (about 15& units; Agostini et al., 2008). Previous research has suggested that anomalous d7Li values in the mantle relate to isotopic fractionation caused by fluid loss during the dehydration of subducted oceanic crust (e.g., Zack et al., 2003; Elliott et al., 2004; Wunder et al., 2007) or to diffusive kinetic fractionation during high-temperature igneous processes (e.g., Lundstrom et al., 2005; Teng et al., 2006b; Marschall et al., 2007; Rudnick and Ionov, 2007). The following sections address these two topics in detail.

Fig. 5. Diagrams showing variations in d7Li values compared to (a) Pb, (b) Sr, and (c) eNd(t) values for carbonatites and associated syenites within the western Sichuan area. Symbols are as in Fig. 3.

(Genge et al., 1995), are such that the time during which carbonatitic melts are able to react with crustal material is reduced (Halama et al., 2008). The Li isotopic compositions of the samples analyzed during this study vary by 15& and do not correlate with

5.3.1. Diffusion-driven isotopic fractionation The exceptionally high diffusivity of Li means that the Li isotope composition of mantle material can be affected by diffusive processes in the presence of a concentration or thermal gradient (Coogan et al., 2005). In addition, a mass difference of 16.7% leads to about 3% faster diffusion of 6 Li than 7Li (Richter et al., 2003). This has led to high temperature diffusion-driven fractionation of Li isotopes being used to explain the significant d7Li variation in pegmatitic country rocks (Teng et al., 2006b), within individual phenocrysts (Parkinson et al., 2007), and at intra–granular, inter-granular, and inter-sample scales in mantle-derived xenoliths (e.g., Lundstrom et al., 2005; Rudnick and Ionov, 2007). Halama et al. (2008) suggested two scenarios for a batch of ascending carbonatite magma, where either initially Lirich carbonatite magmas preferentially lose 6Li from the melt conduit into the country rock before being erupted

54

S. Tian et al. / Geochimica et Cosmochimica Acta 159 (2015) 42–60

with a slightly heavier isotopic composition than that of the initial melt, or Li-poor carbonatite magmas selectively gain 6 Li by diffusion from the wall rocks and hence acquire a lighter d7Li composition. The fact that the d7Li variations in the carbonatites within the study area do not correlate with variations in radiogenic isotope compositions (Fig. 5) suggests that these variations may be related to diffusion-driven fractionation (Halama et al., 2008). Nevertheless, the diffusional processes described above do not seem to adequately explain the extreme Li isotopic variations observed within magmatic complexes in the western Sichuan area. Richter et al. (2003) suggested that the rapid diffusion of Li could result in fast isotopic homogenization, especially when fluid phases or melts are involved (Lundstrom et al., 2005), indicating that kinetic Li isotopic fractionation is ephemeral (Agostini et al., 2008). In addition, the fact that the upper mantle contains low Li (1.5– 2 ppm; Jagoutz et al., 1979; Ottolini et al., 2004) and has

1000

a

a uniform overall isotopic signature (i.e., both MORBs and OIBs generally have d7Li between +3 to +5&; Elliott et al., 2006; Nishio et al., 2007; Tomascak et al., 2008) means that partial melting of the upper mantle, irrespective of process, should generate melts with d7Li compositions similar to that of other mantle-derived melts (Agostini et al., 2008). The fact that the majority of collision-related carbonatites and spatially associated syenites have d7Li compositions similar to the compositions of MORBs or OIBs (Fig. 4) is consistent with the fact that these carbonatites and associated syenites are, ultimately, mantlederived melts, and are indicative of a slab-derived flux that is either inadequately enriched in Li, or inadequately fractionated in terms of d7Li (or both), to significantly modify the Li isotopic signatures of these magmas, as has been suggested by Agostini et al. (2008). This indicates that the diffusive kinetic fractionation of Li isotopes has not significantly affected the d7Li compositions of the

1000

c

100

SiO2 (wt. %)

Li (ppm)

100

10

10

1

1 0.1

0.1 -6

-4

-2

0

2

4

6

8

10

1000

0.01 -6

12

b

-4

-2

0

2

4

6

8

10

1000

12

d

100

Fe2O3 (wt. %)

Rb (ppm)

100

10

10

1

1

0.1

0.1 -6

-4

-2

0

2

4 δ 7 Li (‰)

6

8

10

12

0.01 -6

-4

-2

0

2

4

6

8

10

12

δ 7 Li (‰)

Fig. 6. Diagrams showing variations in d7Li values compared to (a) Li, (b) Rb, (c) SiO2, and (d) Fe2O3 concentrations within carbonatites and associated syenites in the western Sichuan area. Symbols are as in Fig. 3.

S. Tian et al. / Geochimica et Cosmochimica Acta 159 (2015) 42–60

carbonatites and syenites in the study area. In addition, Elliott et al. (2006) concluded that any variations in d7Li compositions associated with diffusive kinetic fractionation should correlate with Li abundances. In contrast, no correlation is seen for the samples from the study area (Fig. 6a), which can therefore not be related to diffusive kinetic fractionation.

However, Marschall et al. (2007) suggested that dehydration can produce a d7Li decrease of only up to 3% within subducting oceanic crustal material and, as such, a significant amount of Li (about 45%) most likely remains in the slab to be subducted deeper into the mantle, indicating that the heavy Li isotopic signature of dehydrated recycled oceanic crustal material is partially preserved during subduction. This is also supported by the high d7Li (up to 7.9&) HIMU lavas (Chan et al., 2009; Vlaste´lic et al., 2009). Recycled oceanic crustal material has been thought to be responsible for the heavy d18O values associated with enriched MORB (Eiler et al., 2000), which is consistent with the generally elevated d18O of carbonatites and associated syenites in the western Sichuan area (Fig 3b). In addition, the carbonatites and associated syenites in the study area are characterized by negative eNd(t) ( 8.6 to 1.6), coupled with high (87Sr/86Sr)i (0.706053–0.707979), as well as a wide range of 207Pb/204Pb (15.571–15.802) and 208Pb/204Pb (38.3573–39.200) (Table 1; Fig. 5), suggesting the isotopic exchange of slab-derived fluids with the overlying mantle and the addition of a sedimentary component (Elliott et al., 2006), and heterogeneity in the source rocks for these magmas (Othman et al., 1989; Hou et al., 2006a; Yang et al., 2011). The majority of carbonatites and associated syenites in the western Sichuan area have consistent Nb/Y ratios (0.5–1.5) but highly variable Ba concentrations (Fig. 7a). Although processes such as assimilation-fractional crystallization (AFC), hydrothermanl alteration and/or small degrees of melting can change the concentrations of Sr, Ba, and high-field strength elements (HFSE: Nb, Ta, Ti, P, Zr, Hf), the presence of unusually high Sr, Ba, and REE contents in slightly-altered or fresh samples suggests that these secondary mobilities of elements are not significant and thus the enriched nature of carbonatites and associated syenites relates to fluid interaction. This metasomatism of the mantle involved fluids was mostly

5.3.2. Recycling of subducted material As mentioned above, we have excluded the possibility that the d7Li compositions of the carbonatites and associated syenites in the western Sichuan area are the result of the diffusive fractionation of Li isotopes during high-temperature igneous processes. This strongly suggests that the d7Li compositions of these carbonatites and associated syenites reflect the composition of the lithospheric mantle beneath this area, which in turn has been affected by the incorporation of recycled material. The d7Li compositions of carbonatites and associated syenites in the western Sichuan area are highly variable ( 4.5& to +10.8&; Fig. 4), especially when compared to the d7Li range of fresh MORB and OIB, and the published d7Li values for carbonatites and associated silicate rocks in rift zones (Fig. 4). This strongly indicates that the old lithospheric mantle beneath the study area had an anomalous d7Li composition; in addition, Lundstrom et al. (2005) and Tang et al. (2012, 2014) suggested that anomalous d7Li values could be retained in the mantle over relatively long timescales. Tang et al. (2014) suggested that anomalous d7Li values may be related to a subducted slab, as the fluids released by the dehydration of subducted slab material are rich in Li and 7Li, whereas residual slab material tends to be 7Li depleted (Zack et al., 2003; Wunder et al., 2006, 2007). This means that the low-d7Li component is subducted into the deeper mantle and forms a low-d7Li reservoir that could be the potential source of low-d7Li magmas (Zack et al., 2003; Wunder et al., 2006, 2007; Agostini et al., 2008).

1000000

a

Fluid-related enrichment

100000

b Fluid from altered oceanic crust

10000

100000

Fluid from subducted sediments

1000

10000 Melt-related enrichment Ba /Th

Ba (ppm)

55

1000

Marine sediments

OIBs 100

10

MORBs

MORBs

10

1 0

OIBs

100

1

2

4

6

8

10 Nb /Y

12

14

16

18 20

0.1 0.700

0.702

0.704

0.706

0.708

0.710

0.712

(87Sr / 86Sr)i

Fig. 7. Diagrams showing variations in (a) Nb/Y vs. Ba, and (b) 87Sr/86Sr vs. Ba/Th for carbonatites and associated syenites within the western Sichuan area; fields are from Hou et al. (2015) and symbols are as in Fig. 3.

56

S. Tian et al. / Geochimica et Cosmochimica Acta 159 (2015) 42–60

derived from subducted altered oceanic crust (Fig. 7b; e.g., Maoniuping and Lizhuang) or mostly from subducted marine sediments (Fig. 7b; e.g., Dalucao). The unusually high Ba/Th ratios and elevated 87Sr/86Sr values of some of the carbonatites and syenites from Dalucao also require significant involvement of marine sediments in the source region for the magmas that formed these units (Fig. 7b). Because of the complexity and range of subduction processes, it is difficult to model in detail the composition of subduction-modified mantle. Here, we provide an example using previously determined compositions of oceanic sediment (SED), altered oceanic crust (AOC), and mantle wedge (M) material (Table 2; Moriguti and Nakamura, 1998; Leeman et al., 2004; Moriguti et al., 2004). The modeling curves of fluids derived from a dehydrated slab (F1: AOC80–SED20 or F2: AOC40–SED60, respectively) with a representative mantle composition (M; Fig. 8a–c) do not perfectly mimic the compositions of the carbonatites and associated syenites in the western Sichuan area, but can account for the majority of their compositional variations, especially for the samples from Maoniuping and Lizhuang. However, some of the carbonatites and syenites from Dalucao with significant involvement of marine sediments in their source region (Fig. 7b), are characterized by

unusual Pb–Sr–Nd isotopic compositions (Fig. 5), and highly variable d7Li compositions ( 4.5& to +10.8&; Fig. 4), both of which lead to these samples far away from the modeling curves (Fig. 8a–c). This implies that the anomalous d7Li compositions of carbonatites and associated syenites in the western Sichuan area could not have been produced by diffusivedriven isotopic fractionation of Li, and are indicative of derivation from a region of the subcontinental lithospheric mantle that was modified by interaction at various mass ratios with fluids derived from subducted oceanic slab material associated with recycling of the oceanic crust and of marine sediments. 5.4. Geodynamic implications The carbonatites and associated syenites in the western Sichuan area have highly variable Li isotopic compositions (range of 15&) that are similar to the compositions of Cenozoic subduction-related calc-alkaline and ultrapotassic rocks in western Anatolia (Agostini et al., 2008). The d7Li compositions of the samples analyzed during this study do not correlate with variations in Pb–Sr–Nd isotopic compositions (Fig. 5) or in other chemical parameters

Table 2 The end-member compositions used in modeling. Reservoir component

Oceanic sediment (SED)

Altered oceanic crust (AOC)

Mantle wedge (M)

Li/ppm d7Li/& Pb/ppm 207 Pb/204Pb 208 Pb/204Pb Sr/ppm 87 Sr/86Sr

49a 3.0b 15b 15.7b 38.89c 330b 0.7090b

12b 12.0c 1b 15.49b 37.68c 130b 0.7040b

2b 2.0b 0.18b 15.53b 37.75c 18.2b 0.7028b

a b c

Moriguti and Nakamura (1998). Leeman et al. (2004). Moriguti et al. (2004).

15.84

0.710

39.5

a

15.80

b

c

SED

15.76 39.0 SED 40

40 80

15.64

90

15.60 15.56

F2

15.52

F1 M

70 80

38.5 90

38.0

F2

80 90

0.704

F2

F1

M

AOC

0

F1

70

0.706

AOC

15.48 15.44 -5

60 60

(86Sr/ 87Sr)i

Pb/ 204Pb

60 70

15.68

40

0.708

SED

208

207

Pb/ 204Pb

15.72

5

δ 7 Li (‰)

10

15

37.5 -5

M

AOC

0

5

δ 7 Li (‰)

10

15

0.702 -5

0

5

10

15

δ 7 Li (‰)

Fig. 8. Diagrams showing variations in d7Li compared to (a) 207Pb/204Pb, (b) 208Pb/204Pb, and (c) 87Sr/86Sr values for carbonatites and associated syenites within the western Sichuan area. Assumed compositions of the oceanic sediment (SED), altered oceanic crust (AOC), and mantle wedge (M) end-member components are given in Table 2. The solid curved line with tick marks represents bulk modeling between oceanic sediments (SED) and altered oceanic crust (AOC), whereas dashed curves represent the modeling of fluids derived from dehydrated slab material (F1: AOC80–SED20 or F2: AOC40–SED60, respectively) with a representative mantle composition (M). Symbols are as in Fig. 3.

S. Tian et al. / Geochimica et Cosmochimica Acta 159 (2015) 42–60

(Fig. 6), indicating that the Li isotopic heterogeneity likely reflects heterogeneity in the source for these magmas. The heterogeneous Pb–Sr–Nd–O isotopic compositions also support this viewpoint (Figs. 3b and 5). The data presented here indicate that the ancient subcontinental lithospheric mantle within the western Sichuan area was modified by interaction with fluids derived from a subducted oceanic slab and that the anomalous d7Li systematics of the western Sichuan area are a consequence of the unique geodynamic setting of this region. The presence of the Proterozoic Kangding granitoid batholiths and contemporaneous arc volcanic rocks associated with carbonatite-alkalic complexes in the study suggests the late Proterozoic subduction of an ancient oceanic plate beneath the Yangtze Craton (Xu et al., 1995; Luo et al., 1998). The Li-rich fluids from the subducted and recycled oceanic crust and marine sediments would have metasomatized the overlying SCLM, producing an enriched region of the mantle. The broadly coeval ages of the carbonatite–syenite complexes and associated potassic rocks and shoshonitic lamprophyres in the study area (Fig. 1) suggest that all of these units were related to the Cenozoic upwelling of an asthenospheric mantle diapir (Zhong et al., 2001). The formation of this diapir was probably triggered by Paleocene and later convergence and collision between the Indian continent and the Yangtze continental block (Zhong et al., 2001). This diapir is likely to have had an upright “tile sheet” shape and would have extended along the entirety of the western margin of the Yangtze Craton (Zhong et al., 2001). The upwelling of a Cenozoic asthenospheric mantle diapir and post- or late-collisional stress relaxation in the Oligocene caused partial melting of the overlying metasomatized and enriched SCLM underneath the EIACZ, generating the magmas that formed the Himalayan carbonatite–alkalic complex in this area (Hou et al., 2006a, 2009). 5.5. Mantle source differences between the collisional and rift zone carbonatites The Pb–Sr–Nd and C–O isotopic compositions of carbonatites in rift zones are indicative of derivation from mixed HIMU and EMI sources related to the subduction and recycling of oceanic lithospheric material (Bell and Tilton, 2001). Nevertheless, the Li isotopic compositions of Archean to Recent carbonatites (d7Li = +4.1& ± 1.3&; n = 23) overlap the range of compositions of modern mantle–derived rocks (MORB and OIB) and remain constant through time (Halama et al., 2008), suggesting that the bulk composition of subducted material does not deviate greatly from the average mantle value (d7Li = +4& ± 2&) or that crustal Li is effectively homogenized upon subduction into the mantle (Halama et al., 2008). This indicates that carbonatites derived from the sublithospheric mantle during mantle plume activity may not have been influenced by subduction processes and crustal recycling (Halama et al., 2011). However, the Pb–Sr–Nd and C–O isotopic compositions of carbonatites in collisional zones have shown that the least-contaminated carbonatites were derived from a

57

transitional source between EMI and EMII components, which formed as a result of the recycling of oceanic crust and a few percent (5–6%) of mixtures of pelagic and terrigenous sediments (Hou et al., 2006a). Nevertheless, the highly variable d7Li compositions of carbonatites and associated syenites indicate the existence of anomalous d7Li within the ancient subcontinental lithospheric mantle in the study area. The carbonatites and syenites in western Sichuan originated from subcontinental lithospheric mantle, which was metasomatized by the Li-rich fluids derived from the subducted oceanic crust and marine sediments (ratios: AOC80–SED20 to AOC40–SED60). Some of the carbonatites and syenites from Dalucao with unusual Pb–Sr–Nd–O isotopic compositions and highly variable d7Li compositions are affected by significant involvement of marine sediments in their source region, not strongly contaminated by crustal materials (Hou et al., 2006a). Therefore, lithium stable isotopes provide an effective method to identify the collisional and rift zone carbonatites and to study recycling processes in the mantle. 6. CONCLUSIONS

(1) The Li isotopic compositions of carbonatites and associated syenites in western Sichuan are highly variable ( 4.5& to 10.8&), which differ significantly from the range of d7Li values for fresh MORB and OIB and for carbonatites and associated silicate rocks within rift zones. (2) These highly variable d7Li compositions could not have been produced by diffusive-driven isotopic fractionation of Li and indicate the existence of ancient subcontinental lithospheric mantle with anomalous d7Li values, meaning that the ancient SCLM beneath western Sichuan was modified by interaction with fluids derived from the subducted oceanic crust and marine sediments. (3) The carbonatites and syenites in western Sichuan were generated by the partial melting of subcontinental lithospheric mantle, which was metasomatized by the Li-rich fluids originated from the subducted oceanic crust and marine sediments (ratios: AOC80– SED20 to AOC40–SED60). This melting was most likely triggered by a Cenozoic asthenospheric mantle diapir related to Indian-Asian continental collision at 65–50 Ma and post- or late-collisional stress relaxation in the Oligocene.

ACKNOWLEDGMENTS This research was financially supported by grants from the Ministry of Science and Technology of China (973 Project2011CB403106, 2011CB403104), the Natural Science Foundation of China (40973013, 41173003, 41373014, 41320104004), IGCP/SIDA–600, the Ministry of Land and Resources (201011027, 201011011), and the China Geological Survey (12120113016200). We would like to thank Prof. R. Rudnick, Prof. W. McDonough, Associate Prof. F. Z. Teng, and other staff of the Plasma Laboratory at the University of

58

S. Tian et al. / Geochimica et Cosmochimica Acta 159 (2015) 42–60

Maryland for support and analytical assistance during this study. Thanks to C. Q. Liu, Z. Q. Zhao, and Q. L. Wang of the Institute of Geochemistry, Chinese Academy of Sciences, China, for providing the L-SVEC standard material and advice on Li purification procedures. Y. H. Li, Z. Z. Li, D. Yang, X. K. Zhu, and S. H. Tang are also appreciated for help and advice on the Li purification procedures used during this study. We also thank the members of staff of the Maoniuping, Dalucao, and Lizhuang REE ore districts for assistance during fieldwork. D. C. Zhu and Z. D. Zhao are thanked for helpful discussions. We deeply acknowledge constructive comments and valuable suggestions by P. Tomascak and two anonymous referees on the earlier draft of this manuscript. This paper also benefited from general comments and helpful suggestions by Dr. D. Norman, Executive Editor of GCA and W. D. Sun, Associate Editor of GCA.

REFERENCES Agostini S., Ryan J. G., Tonarini S. and Innocenti F. (2008) Drying and dying of a subducted slab: coupled Li and B isotope variations in Western Anatolia Cenozoic Volcanism. Earth Planet. Sci. Lett. 272, 139–147. Aulbach S. and Rudnick R. L. (2009) Origins of non-equilibrium lithium isotopic fractionation in xenolithic peridotite minerals: examples from Tanzania. Chem. Geol. 258, 17–27. Bell K. and Tilton G. R. (2001) Nd, Pb and Sr isotopic compositions of East African carbonatites: evidence for mantle mixing and plume inhomogeneity. J. Petrol. 42, 1927–1945. Bell K. and Tilton G. R. (2002) Probing the mantle: the story from carbonatites. EOS Trans. AGU 273, 276–277. Bell K. and Simonetti A. (2010) Source of parental melts to carbonatites–critical isotopic constraints. Mineral. Petrol. 98, 77–89. Blattner P., Jin C. W. and Xu Y. (1989) Oxygen isotopes in mantle related and geothermally altered magmatites of the Transhimalayan (Gangdese) ranges. Contrib. Mineral. Petrol. 101, 438–446. Blattner P., Abart R., Adams C. J., Faure K. and Hui L. (2002) Oxygen isotope trends and anomalies in granitoids of the Tibetan plateau. J. Asian Earth Sci. 21, 241–250. Bouman C., Elliott T. and Vroon P. Z. (2004) Lithium inputs to subduction zones. Chem. Geol. 212, 59–79. Chan L. H., Edmond J. M., Thompson G. and Gillis K. (1992) Lithium isotopic composition of submarine basalts: implications for the lithium cycle in the oceans. Earth Planet. Sci. Lett. 108, 151–160. Chan H. L., Lassiter J. C., Hauri E. H., Hart S. R. and Blusztajn J. (2009) Lithium isotope systematics of lavas from the CookAustral Islands: constraints on the origin of HIMU mantle. Earth Planet. Sci. Lett. 277, 433–442. Chung S. L. and Jahn B. M. (1995) Plume-lithosphere interaction in generation of the Emeishan flood basalts at the PermianTriassic boundary. Geology 23, 889–892. Chung S. L., Lo C. H., Lee T. Y., Zhang Y. Q., Xie Y. W., Li X. H., Wang K. L. and Wang P. L. (1998) Dischronous uplift of the Tibetan plateau starting from 40 My ago. Nature 349, 769– 773. Coogan L. A., Kasemann S. A. and Chakraborty S. (2005) Rates of hydrothermal cooling of new oceanic upper crust derived from lithium-geospeedometry. Earth Planet. Sci. Lett. 240, 415– 424. Eiler J. M., Schiano P., Kitchen N. and Stolper E. M. (2000) Oxygen-isotope evidence for recycled crust in the sources of mid-ocean-ridge basalts. Nature 403, 530–534.

Elliott T., Jeffcoate A. and Bouman C. (2004) The terrestrial Li isotope cycle: light-weight constraints on mantle convection. Earth Planet. Sci. Lett. 220, 231–245. Elliott T., Thomas A., Jeffcoate A. and Niu Y. (2006) Lithium isotope evidence for subduction-enriched mantle in the source of mid-ocean-ridge basalts. Nature 443, 565–568. Genge M. J., Price G. D. and Jones A. P. (1995) Molecular dynamics simulation of CaCO3 melts to mantle pressures and temperatures: implications for carbonatite magmas. Earth Planet. Sci. Lett. 131, 225–238. Guo Z. F., Hertogen J., Liu J. Q., Pasteels P., Boven A., Punzalan L., He H. Y., Luo X. J. and Zhang W. H. (2005) Potassic magmatism in western Sichuan and Yunnan provinces, SE Tibet, China: petrological and geochemical constraints on petrogenesis. J. Petrol. 46, 33–78. Halama R., McDonough W. F., Rudnick R. L. and Bell K. (2008) Tracking the lithium isotopic evolution of the mantle using carbonatites. Earth Planet. Sci. Lett. 265, 726–742. Halama R., McDonough W. F., Rudnick R. L., Ionov D. A., Keller J. and Klaudius J. (2007) The Li isotopic composition of Oldoinyo Lengai: nature of the mantle sources and lack of isotopic fractionation during carbonatite petrogenesis. Earth Planet. Sci. Lett. 254, 77–89. Halama, R., McDonough, W. F., Rudnick, R. L. and Bell, K. (2011). The lithium isotopic signature of carbonatites. Goldschmidt conference abstracts, 965 Hou Z. Q., Tian S. H., Yuan Z. X., Xie Y. L., Yin S. P., Yi L. S., Fei H. C. and Yang Z. M. (2006a) The Himalayan collision zone carbonatites in western Sichuan, SW China: petrogenesis, mantle source and tectonic implication. Earth Planet. Sci. Lett. 244, 234–250. Hou Z. Q., Lu J. R. and Lin S. Z. (2006b) Heterogeneity of a Plume Axis: bulk–rock geochemical evidence from picrites and basalts in the Emei large igneous province, SW China. Int. Geol. Rev. 48, 1087–1112. Hou Z. Q., Tian S. H., Xie Y. L., Yang Z. S., Yuan Z. X., Yin S. P., Yi L. S., Fei H. C., Zou T. R., Bai G. and Li X. Y. (2009) The Himalayan Mianning–Dechang REE belt associated with carbonatite–alkalic complex in the eastern Indian-Asian collision zone, SW China. Ore Geol. Rev. 36, 65–89. Hou Z. Q., Liu Y., Tian S. H., Yang Z. M. and Xie Y. L. (2015) Formation of carbonatite-related giant REE deposits by the recycling of marine sediments. Sci. Rep.. Accepted. Hu W. J., Tian S. H., Wang S. P., Chen L. J., Su A. N., Yue L. and Wang Y. X. (2012) Sm–Nd isochron age of carbonatite from the Maoniuping REE deposit, Sichuan Province and their geological implications. Mineral. Resour. Geol. 26, 237–241 (in Chinese, with English abstract). Jagoutz E., Palme H., Baddenhausen H., Blum K., Cendales M., Dreibus G., Spettel B., Lorenz V. and Wa¨nke H. (1979) The abundances of major, minor and trace elements in the earth’s mantle as derived from primitive ultramafic nodules. Proc. Lunar Planet. Sci. Conf. 10th, 2031–2050. Keller J., Zaitsev A. N. and Wiedenmann D. (2006) Primary magmas at Oldoinyo Lengai: the role of olivine melilitites. Lithos 91, 150–172. Kisaku¨rek B., Widdowson M. and James R. H. (2004) Behavior of Li isotopes during continental weathering: the Bidar laterite profile, India. Chem. Geol. 212, 27–44. Lai X. D., Yang X. Y. and Sun W. D. (2012) Geochemical constraints on genesis of dolomite marble in the Bayan Obo REE–Nb–Fe deposit, Inner Mongolia: implications for REE mineralization. J. Asian Earth Sci. 57, 90–102. Leeman W. P., Tonarini S., Chan L. H. and Borg L. E. (2004) Boron and lithium isotopic variations in a hot subduction

S. Tian et al. / Geochimica et Cosmochimica Acta 159 (2015) 42–60 zone—the southern Washington Cascades. Chem. Geol. 212, 101–124. Ling M. X., Liu Y. L., Williams I. S., Teng F. Z., Yang X. Y., Ding X., Wei G. J., Xie L. H., Deng W. F. and Sun W. D. (2013) Formation of the world’s largest REE deposit through protracted fluxing of carbonatite by subduction–derived fluids. Sci. Rep. 3, 1–8. Liu Y., Hou Z. Q., Tian S. H., Zhang Q. C., Zhu Z. M. and Liu J. H. (2014) Zircon U–Pb ages of the Mianning–Dechang syenites, Sichuan Province, southwestern China: constraints on the giant REE mineralization belt and its regional geological setting. Ore Geol. Rev.. http://dx.doi.org/10.1016/j.oregeorev.2014.03.017. Lundstrom C. C., Chaussidon M., Hsui A. T., Kelemen P. and Zimmerman M. (2005) Observations of Li isotopic variations in the Trinity Ophiolite: evidence for isotopic fractionation by diffusion during mantle melting. Geochim. Cosmochim. Acta 69, 735–751. Luo Y. N., Yu R. L., Hou L. W., Lai S. M., Fu D. M., Chen M. X., Fu X. F., Rao R. B. and Zhou S. S. (1998) Longmenshan– Jinpingshan Intracontinental Orogenic Belt. Sichuan Science and Technology Publishing House, Chengdu, pp 1–171 (in Chinese, with English abstract). Magna T., Ionov D. A., Oberli F. and Wiechert U. (2008) Links between mantle metasomatism and lithium isotopes: evidence from glass–bearing and cryptically metasomatized xenoliths from Mongolia. Earth Planet. Sci. Lett. 276, 214–222. Magna T., Wiechert U. and Halliday A. N. (2006a) New constraints on the lithium isotope compositions of the Moon and terrestrial planets. Earth Planet. Sci. Lett. 243, 336–353. Magna T., Wiechert U., Grove T. L. and Halliday A. N. (2006b) Lithium isotope fractionation in the Southern Cascadia subduction zone. Earth Planet. Sci. Lett. 250, 428–443. Marschall H. R., Pogge von Strandmann P. A. E., Seitz H. M., Elliott T. and Niu Y. (2007) The lithium isotopic composition of orogenic eclogites and deep subducted slabs. Earth Planet. Sci. Lett. 262, 563–580. Mo X. X., Zhao Z. D., Deng J. F., Dong G. C. and Zhou S. (2003) Response of volcanism to the India-Asia collision. Earth Sci. Front. 10, 135–148 (in Chinese, with English abstract). Moriguti T. and Nakamura E. (1998) Across-arc variation of Li isotopes in lavas and implications for crust/mantle recycling at subduction zones. Earth Planet. Sci. Lett. 163, 167–174. Moriguti T., Shibata T. and Nakamura E. (2004) Lithium, boron and lead isotope and trace element systematics of quaternary basaltic volcanic rocks in northeastern Japan: mineralogical controls on slab–derived fluid composition. Chem. Geol. 212, 81–100. Nishio Y., Nakai S. I., Kogiso T. and Barsczus H. G. (2005) Lithium, strontium, and neodymium isotopic compositions of oceanic island basalts in the Polynesian region: constraints on a Polynesian HIMU origin. Geochem. J. 39, 91–103. Nishio Y., Nakai S. I., Ishii T. and Sano Y. (2007) Isotope systematics of Li, Sr, Nd, and volatiles in Indian Ocean MORBs of the Rodrigues Triple Junction: constraints on the origin of the DUPAL anomaly. Geochim. Cosmochim. Acta 71, 745–759. Othman D. B., White W. M. and Patchett J. (1989) The geochemistry of marine sediments, island arc magma genesis, and crust-mantle recycling. Earth Planet. Sci. Lett. 94, 1–2. Ottolini L., Le Fe`vre B. and Vannucci R. (2004) Direct assessment of mantle boron and lithium contents and distribution by SIMS analyses of peridotite minerals. Earth Planet. Sci. Lett. 228, 19–36. Parkinson I. J., Hammond S. J., James R. H. and Rogers N. W. (2007) High-temperature lithium isotope fractionation: insights

59

from lithium isotope diffusion in magmatic systems. Earth Planet. Sci. Lett. 257, 609–621. Richter F. M., Davis A. M., Depaolo D. J. and Watson E. B. (2003) Isotope fractionation by chemical diffusion between molten basalts and rhyolite. Geochim. Cosmochim. Acta 67, 3905–392. Rudnick R. L. and Gao S. (2003) Composition of the continental crust. In The Crust (eds. R. L. Rudnick, H. D. Holland and K. K. Turekian). Treatise on Geochemistry vol. 3, Elsevier, Oxford, pp. 1–64 Rudnick R. L. and Ionov D. A. (2007) Lithium elemental and isotopic disequilibrium in minerals from peridotite xenoliths from far-east Russia: product of recent melt/fluid-rock reaction. Earth Planet. Sci. Lett. 256, 278–293. Rudnick R. L., Tomasack P. B., Njo H. B. and Gardner L. R. (2004) Extreme lithium isotopic fractionation during continental weathering revealed in saprolites from South Carolina. Chem. Geol. 212, 45–57. Ryan J. G. and Langmuir C. H. (1987) The systematics of lithium abundances in young volcanic rocks. Geochim. Cosmochim. Acta 51, 1727–1741. Su B. X., Zhang H. F., Deloule E., Sakyi P. A., Xiao Y., Tang Y. J., Hu Y., Ying J. F. and Liu P. P. (2012) Extremely high Li and low d7Li signatures in the lithospheric mantle. Chem. Geol. 292– 293, 149–157. Tang Y. J., Zhang H. F., Deloule E., Su B. X., Ying J. F., Xiao Y. and Hu Y. (2012) Slab-derived lithium isotopic signatures in mantle xenoliths from northeastern North China Craton. Lithos 149, 79–90. Tang Y. J., Zhang H. F., Deloule E., Su B. X., Ying J. F., Santosh M. and Xiao Y. (2014) Abnormal lithium isotope composition from the ancient lithospheric mantle beneath the North China Craton. Sci. Rep. 4. http://dx.doi.org/10.1038/srep04274. Tatsumi Y., Hamilton D. L. and Nesbitt R. W. (1986) Chemical characteristics of fluid phase released from a subducted lithosphere and origin of arc magmas: evidence from highpressure experiments and natural rocks. J. Volcanol. Geotherm. Res. 29, 293–309. Taylor H. P., Frechen J. and Degens E. T. (1967) Oxygen and carbon isotope studies of carbonatites from the Laacher See district, West Germany and the Alno¨ district, Sweden. Geochim. Cosmochim. Acta 31, 407–430. Teng F. Z., McDonough W. F., Rudnick R. L., Dalpe´ C., Tomascak P. B., Chappell B. W. and Gao S. (2004) Lithium isotopic composition and concentration of the upper continental crust. Geochim. Cosmochim. Acta 68, 4167–4178. Teng F. Z., McDonough W. F., Rudnick R. L., Walker R. and Sirbescu M. L. C. (2006a) Lithium isotopic systematics of granites and pegmatites from the Black Hills, South Dakota. Am. Miner. 91, 1488–1498. Teng F. Z., McDonough W. F., Rudnick R. L. and Walker R. J. (2006b) Diffusion-driven lithium isotopic fractionation in country rocks of the Tin Mountain pegmatite. Earth Planet. Sci. Lett. 243, 701–710. Teng F. Z., Rudnick R. L., McDonough W. F. and Wu F. Y. (2009) Lithium isotopic systematics of A-type granites and their mafic enclaves: further constraints on the Li isotopic composition of the continental crust. Chem. Geol. 262, 370–379. Tian S. H., Hou Z. Q., Yang Z. S., Yang Z. M., Yuan Z. X., Wang Y. B., Xie Y. L., Liu Y. C. and Li Z. (2008) Zircon U-Pb ages, Hf isotopic compositions and geological significance. a case study of carbonatite and nordmarkite from the Dalucao REE deposit, Sichuan Province. Acta Petrologica Sinica 24, 544–554 (in Chinese, with English abstract). Tian S. H., Hou Z. Q., Su A. N., Hou K. J., Hu W. J., Li Z. Z., Zhao Y., Gao Y. G., Li Y. H., Yang D. and Yang Z. S. (2012)

60

S. Tian et al. / Geochimica et Cosmochimica Acta 159 (2015) 42–60

Separation and precise measurement of lithium isotopes in three reference materials using multi collector-inductively coupled plasma mass spectrometry. Acta Geologica Sinica (English edition) 86, 1297–1305. Tomascak P. B. (2004) Developments in the understanding and application of lithium isotopes in the earth and planetary sciences. In Geochemistry of Non-Traditional Isotope Systems (eds. C. Johnson, B. Beard and F. Albarede). Mineralogical Society of America, pp. 153–195. Tomascak P. B., Tera F., Helz R. T. and Walker R. J. (1999) The absence of lithium isotope fractionation during basalt differentiation: new measurements by multicollector sector ICP–MS. Geochim. Cosmochim. Acta 63, 907–910. Tomascak P. B., Ryan C. G. and Defant M. J. (2000) Lithium isotope evidence for light element decoupling in the Panama subarc mantle. Geology 28, 507–510. Tomascak P. B., Langmuir C. H., le Roux P. J. and Shirey S. B. (2008) Lithium isotopes in global mid–ocean ridge basalts. Geochim. Cosmochim. Acta 72, 1626–1637. Vlaste´lic I., Koga K., Chauvel C., Jacques G. and Te´louk P. (2009) Survival of lithium isotopic heterogeneities in the mantle supported by HIMU-lavas from Rurutu Island, Austral Chain. Earth Planet. Sci. Lett. 286, 456–466. Wang J. H., Yin A., Harrison T. M., Grove M., Zhang Y. Q. and Xie G. H. (2001) A tectonic model for Cenozoic igneous activities in the eastern Indian-Asian collision zone. Earth Planet. Sci. Lett. 88, 123–133. Wunder B., Meixner A., Romer R. L. and Heinrich W. (2006) Temperature-dependent isotopic fractionation of lithium between clinopyroxene and high-pressure hydrous fluids. Contrib. Mineral. Petrol. 151, 112–120. Wunder B., Meixner A., Romer R. L., Feenstra A., Schettler G. and Heinrich W. (2007) Lithium isotope fractionation between Li-bearing staurolite, Li-mica and aqueous fluids: an experimental study. Chem. Geol. 238, 277–290. Xu D., Chen Y., Zhang Y., Li W., Zhao B. and Luo T. (1995) The study of the age and origin of the Kangding granitic complex, western Yangtze craton. China. Geol. Rev. 41, 101–111 (in Chinese, with English abstract). Xu Y. G., Chung S. L., Jahn B. M. and Wu G. Y. (2001) Petrologic and geochemical constraints on the petrogenesis of PermianTriassic Emeishan flood basalts in southwestern China. Lithos 58, 145–168. Xu C., Kynicky J., Chakhmouradian A. R., Campbell I. H. and Allen C. M. (2010) Trace-element modeling of the magmatic evolution of rare-earth-rich carbonatite from the Miaoya deposit, Central China. Lithos 118, 145–155. Yang X. Y., Sun W. D., Zhang Y. X. and Zheng Y. F. (2009) Geochemical constraints on the genesis of the Bayan Obo

Fe–Nb–REE deposit in Inner Mongolia, China. Geochim. Cosmochim. Acta 73, 1417–1435. Yang K. F., Fan H. R., Santosh M., Hu F. F. and Wang K. Y. (2011) Mesoproterozoic carbonatitic magmatism in the Bayan Obo deposit, Inner Mongolia, North China: constraints for the mechanism of super accumulation of rare earth elements. Ore Geol. Rev. 40, 122–131. Yin A. and Harrison T. M. (2000) Geologic evolution of the Himalayan-Tibetan orogeny. Ann. Rev. Earth Planet. Sci. 28, 211–280. Ying J. F., Zhou X. H. and Zhang H. F. (2004) Geochemical and isotopic investigation of the Laiwu-Zibo carbonatites from western Shandong Province, China, and implications for their petrogenesis and enriched mantle source. Lithos 75, 413–426. You C. F., Castillo P. R., Gieskes J. M., Chan L. H. and Spivac A. J. (1996) Trace element behavior in hydrothermal experiments: implications for fluid processes at shallow depths in subduction zones. Earth Planet. Sci. Lett. 140, 41–52. Yuan Z. X., Shi Z. M., Bai G., Wu C. Y., Chi R. A. and Li X. Y. (1995) The Maoniuping rare earth ore deposit, Mianning County. Sichuan Province. Seismological Press, Beijing, pp 1–159 (in Chinese). Zack T., Tomascak P. B., Rudnick R. L., Dalpe´ C. and McDonough W. F. (2003) Extremely light Li in orogenic eclogites: the role of isotope fractionation during dehydration in subducted oceanic crust. Earth Planet. Sci. Lett. 208, 279– 290. Zhang Y. Q. and Xie Y. W. (1997) Geochronology of alkali–rich intrusions and Nd, Sr isotopic characteristics in AilaoshanJinsha River. Sci. China 27, 289–293 (in Chinese). Zhang Y. X., Luo Y. N. and Yang C. X. (1988) The panxi rift. Geological Publishing House, Beijing, pp. 224–270 (in Chinese). Zhang H. F., Deloule E., Tang Y. J. and Ying J. F. (2010) Melt/ rock interaction in remains of refertilized Archean lithospheric mantle in Jiaodong Peninsula, North China Craton: Li isotopic evidence. Contrib. Mineral. Petrol. 160, 261–277. Zhao Y., Hou K. J., Tian S. H., Yang D. and Su A. N. (2015) Study on measurements of lithium isotopic compositions for common standard reference materials using MC–ICP–MS. Rock Miner. Analy. 34, 28–39 (in Chinese, with English abstract). Zhong D. L., Ding L., Liu F. T., Liu J. H., Zhang J. J., Ji J. Q. and Chen H. (2001) The multi–direction layer and frame structure of rock layers in orogenic belt and its control on Cenozoic magmatic activities: an example from Sanjiang and its neighboring area. Sci. China 30, 1–8 (in Chinese). Associate editor: Weidong Sun