Chemical Geology 344 (2013) 1–11

Contents lists available at SciVerse ScienceDirect

Chemical Geology

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

U–Pb isotope and trace element analysis of columbite-(Mn) and zirconby laser ablation ICP–MS: Implications for geochronology of pegmatiteand associated ore deposits

Xiao-Dong Deng a,b, Jian-Wei Li a,b,⁎, Xin-Fu Zhao b, Zhao-Chu Hu a, Hao Hu b,David Selby c, Zorano S. de Souza d

a State Key Laboratory of Geological Processes and Mineral Resources, China University of Geosciences, Wuhan 430074, Chinab Faculty of Earth Resources, China University of Geosciences, Wuhan 430074, Chinac Department of Earth Sciences, University of Durham, Durham DH1 3LE, United Kingdomd Department of Geology, Universidade Federal do Rio, Grande do Norte, Natal 1524, Brazil

⁎ Corresponding author at: Faculty of Earth Resources, CWuhan 430074, China. Tel.: +86 1387111 2076; fax: +8

E-mail address: jwli@cug.edu.cn (J.-W. Li).

0009-2541/$ – see front matter © 2013 Elsevier B.V. Allhttp://dx.doi.org/10.1016/j.chemgeo.2013.02.014

a b s t r a c t

a r t i c l e i n f o

Article history:Received 5 December 2012Received in revised form 17 February 2013Accepted 18 February 2013Available online 26 February 2013

Editor: K. Mezger

Keywords:Columbite-(Mn)ZirconPegmatiteU–Pb datingLA-ICP-MS

U–Pb isotopes and trace elements of columbite-(Mn) and zircon from an early Cretaceous pegmatite dike inthe Xiaoqinling district, North China Craton, were analyzed using laser ablation inductively coupled plasmamass spectrometry (LA-ICP-MS) to better understand the suitability of these minerals for U–Pb geochronol-ogy when they contain unusually high U and Th. Columbite-(Mn) grains have high W, Ti, U, Th, and REE con-tents and yield concordant U–Pb age of 143 ± 1 Ma (2σ, n = 10) that is interpreted as the emplacement ageof the pegmatite dike. In contrast, zircon grains from the same dike show three distinct U–Pb age populations.Nine out of seventeen zircon grains analyzed have textural features typical of magmatic zircons and yield aweighted mean 206Pb/238U age of 143 ± 1 Ma (2σ, n = 9), identical to that of columbite-(Mn) and thus con-strain the timing of pegmatitic magmatism. The second population of zircon is characterized by corroded andzoned textures with the geochemical affinities of magmatic zircon. These zircon grains have a weightedmean207Pb/206Pb age of 1879 ± 19 Ma (2σ, n = 5) and are considered to be inherited components derived fromPaleoproterozoic basement rocks that are widely distributed in the Xiaoqinling district. A third zircon popu-lation is characterized by high porosity and abundant Th–U-rich mineral inclusions (e.g. thorite, uraniumoxides), and yield a younger U–Pb age of 127 ± 3 Ma (2σ, n = 3). These younger zircon grains have elevatedHf, Ca, P, Nb, Ta, and Ti contents and much higher Th/U, LREE/MREE, and LREE/HREE ratios compared to the143 Ma zircon. The textural and geochemical data for the 127 Ma zircon grains indicate that they wereformed by the hydrothermal alteration of precursor zircon crystallized from the pegmatitic magmas, presum-ably associated with pervasive hydrothermal flow that led to the formation of numerous early Cretaceousgold deposits in the Xiaoqinling district. Results from this study demonstrate that columbite-(Mn) fromthe pegmatite dike in the Xiaoqinling district has been resistant to post-magmatic hydrothermal alterationthat, however, have disturbed the U–Pb isotopes in some zircon grains from the same dike. Consequently,we propose that columbite-(Mn) is an ideal target for U–Pb dating of pegmatite and associated ore deposits.

© 2013 Elsevier B.V. All rights reserved.

1. Introduction

Pegmatites commonly form in thewaning stage of magma evolutionby fractional crystallization of volatile-richmagmas (London, 2005), andserve as important sources for strategicmetals (e.g. Li, Be, Cs, Ta, Nb, andrare earth elements (REEs); Linnen et al., 2012), as well as high-qualitygem minerals (e.g. beryl, tourmaline, topaz, spodumene, and spessar-tine; Simmons et al., 2012). Ages of pegmatite have traditionally beendetermined by U–Pb dating of zircon formed during the crystallization

hina University of Geosciences,6 27 6788 5096.

rights reserved.

of pegmatitic magmas (e.g. Tapponnier et al., 1990; Schärer et al.,1994). Zircons from pegmatite may have extremely high U and Thcontents that potentially cause radioactive damage to the mineralstructure. In this case, later hydrothermal activities may lead to incon-gruent dissolution and re-crystallization of zircon and consequentlythe disturbance of the U–Pb isotope systematics in the mineral(Geisler et al., 2002, 2007; Kusiak et al., 2009). This would hinder theability to precisely date the formation age of pegmatites and associatedore deposits by U–Pb zircon geochronology.

Columbite-groupminerals are another common accessory phases inpegmatites (Černý and Ercit, 1989), and widely occur in granite-relatedhydrothermal Sn, W, and REE deposits as well (Lerouge et al., 2007;Beurlen et al., 2008). They are also present in alkaline and carbonatitic

2 X.-D. Deng et al. / Chemical Geology 344 (2013) 1–11

intrusions (Möller, 1989). Columbite-groupminerals usually have rela-tively high U, but low common Pb contents, and therefore can be anideal target for U–Pb dating of pegmatites and related mineral deposits(Romer and Wright, 1992; Romer and Smeds, 1994; Romer andLehmann, 1995; Romer and Smeds, 1996, 1997). Smith et al. (2004)and Melcher et al. (2008) successfully dated, for the first time,columbite-group minerals from pegmatites in the Superior Province ofCanada and Africa (Ghana, Rwanda, Congo, and Namibia), using theLA-MC-ICP-MS and LA-ICP-MS methods, respectively. In this paper,we present a comparative geochemical and U–Pb geochronologicalstudy of columbite-(Mn) and zircon from a pegmatite dike in theXiaoqinling district, North China Craton (NCC). Our results reveal thatsome zircon grains from this pegmatite are inherited components andothers have been intensively affected by post-pegmatite hydrothermalalteration, yielding ages that are significantly younger than the trueemplacement age of the dike. In contrast, coexisting columbite-(Mn)grains survived the hydrothermal alteration and provide reliablechronological constraints for the pegmatitic magmatism. The presentstudy therefore highlights the utilization of columbite-group mineralsas an alternative U–Pb geochronometer to precisely date the formationof pegmatites and associated ore deposits, notably in case the coexistingzircons are subjected to radioactive damage due to high U and Thcontents.

2. Geological setting

The Xiaoqinling district is located along the southern margin ofthe NCC and is bounded by the Taiyao Fault to the north and theXiaohe Fault to the south (Fig. 1). The district is dominated lithologi-cally by amphibolite facies metamorphic rocks of the late Archean toearly Paleoproterozoic Taihua Group consisting mainly of amphibo-lite, gneiss, and migmatite with zircon U–Pb ages of 2.6 to 2.3 Ga (Liet al., 2007; Xu et al., 2009). A number of plutonic intrusions, rangingin composition from monzogranite through biotite granite to pegma-titic granite and pegmatites, were emplaced into the Taihua Group

Fig. 1. Simplified geological map showing the distribution of the G1 and G2 pegmatite dikChina Craton (modified from Li et al., 1996).

(Fig. 1). The biotite granite, which is distributed along the XiaoheFault, has zircon U–Pb ages of ~1.7 Ga (Nie et al., 2001), whereasthe samples of pegmatitic granite have zircon U–Pb ages of ~2.0 Ga(Hu and Lin, 1989). The monzogranites consist of, from west to east,the Huashan, Wenyu, and Niangniangshan plutons (Fig. 1), whichwere emplaced at ~146 Ma, 141–138 Ma, and 143–135 Ma, respec-tively, as constrained by in situ zircon U–Pb analysis (Mao et al.,2010; Li et al., 2012a). Widespread diabase dikes in the distinctformed in two episodes, at ~1.9 to 1.8 Ga and 140–125 Ma (Li et al.,2012a). The Xiaoqinling district contains numerous gold deposits,mostly occurring in the flanks of three NWW-oriented folds (Fig. 1).Comprehensive geochronological studies (molybdenite Re–Os andmica 40Ar/39Ar) indicate that gold mineralization occurred in therange of 154 to 118 Ma (Li et al., 2012a, 2012b).

Numerous pegmatite dikes intrude the Taihua Group and the latePaleoproterozoic pegmatitic granite (Fig. 1). These pegmatites can beclassified into two main groups, hereafter termed as G1 and G2 dikes(Fig. 1). The G1 dikes generally strike 275–300° and typically containK-feldspar, quartz, biotite, and tourmaline. These pegmatites arethought to be related to regional migmatization and magmatism dur-ing the 2.0–1.8 Ga interval (Li et al., 2007). The G2 pegmatite dikesmostly strike northeast and surround the Wenyu pluton (Fig. 1).Individual G2 dikes occur as relatively flat bodies that are 1–3 mwide and 5–100 m long, and frequently crosscut Paleoproterozoicdiabase dikes and pegmatitic granites (Fig. 1). The G2 pegmatitesconsist mainly of albite and quartz, with minor amounts of muscovite,biotite, and garnet. Age constraints on the G2 pegmatites are lacking.

3. Sample description

Samples used for U–Pb geochronology were collected from a G2pegmatite emplaced in the southern margin of the Wenyu pluton(Fig. 1). The samples consist of albite (50 vol.%), quartz (43 vol.%),muscovite (4 vol.%), biotite (2 vol.%), and garnet (1 vol.%) (Fig. 2).Quartz ranges in size from 0.3–2 cm, whereas albite crystals are

es around the Wenyu monzogranite pluton in the Xiaoqinling district, southern North

Fig. 2. Photograph showing mineralogical and textural features of a G2 pegmatiteused for this study. Abbreviations: Qz: quartz, Ab: albite, Ms: muscovite, Bt: biotite,Grt: garnet.

3X.-D. Deng et al. / Chemical Geology 344 (2013) 1–11

commonly 1–5 cm across. The dike shows weak zonation: the centralpart of the dike hosts abundant large crystals (up to 5–6 m), which,however, are few in the marginal zone. Garnets are characteristicallyfine-grained, generally b0.2 cm in diameter (Fig. 2). Hydrothermalsericite is locally present along the cleavages of or micro-fracturesin albite. Columbite-(Mn), zircon, monazite, and thorite are commonaccessory minerals (Fig. 3).

The geochemical composition of the pegmatite samples is summa-rized in the Supplementary data (Tables S.1 and S.2). The rocks areslightly peraluminous with aluminum-saturation index [Al / (2(Ca−1.67P) + Na + K)] of 0.97–1.21. They are enriched in incompatibleelements including Rb (689–1179 ppm), Th (52.3–85.5 ppm), U (20.2–39.5 ppm), Nb (328–530 ppm), Ta (9.8–17.1 ppm), Zr (142–185 ppm),Hf (12.9–17.4 ppm), but have very low MgO (0.03–0.06 wt.%), Sr(2.35–7.30 ppm) and Ba (5.63–18.2 ppm) (Table S.2). The K/Rb valuesare relatively low (35–40), indicating that the pegmatitewas crystallizedfrom a highly fractionated magma. In addition, the rocks contain 99.8 to173.9 ppm REEs, with significant negative Eu anomalies (Eu* = 0.013–0.03; Tables S.1 and S.2). The geochemical signatures suggest that theG2 pegmatite can be classified as NYF pegmatite (cf. Černý and Ercit,2005).

4. Analytical methods

Thin sections of the samples were first investigated by transmitted-light microscopy to determine the mineralogy, textural relationships,and extent of post-emplacement hydrothermal alteration. Zircon andcolumbite-(Mn) grains were separated using conventional heavy-liquid and magnetic methods, and then handpicked under a binocularmicroscope. These grains were mounted in epoxy and polished to ex-pose their interiors. The polished grains were examined using opticalmicroscopy and scanning electron microscopy (SEM) equipped withenergy dispersive spectrometry (EDS). Back-scattered electron (BSE)and cathodoluminescence (CL) images were used to characterize themorphology and internal structure of zircon and columbite-(Mn),using a FEI Quanta200 environmental SEM and a MonoCL detector ona JXA-8100 electron microscope hosted in the State Key Laboratory ofGeological Processes and Mineral Resources (GPMR), China Universityof Geosciences, Wuhan.

Major element analyses of columbite-(Mn) were carried out on aJAX 8230 Superprobe at the Center for Material Research and Analysis,Wuhan University of Technology. Operating conditions included anacceleration voltage of 20 kV and a beam current of 30 nA. The electronbeam was 5 μm in diameter. The counting time was 30 s on-peak and

10 s for off-peak background measurements. The following standardswere used: Ca5(PO4)3F (Ca), (Mn,Ca)SiO3 (Mn), Fe2O3 (Fe), Nb (Nb),Ta (Ta), TiO2 (Ti), U (U), Sc (Sc), SnO2 (Sn), W (W), ZrO2 (Zr), andY5PO14 (Y).

U–Pb isotopes and trace elements of zircon and columbite-(Mn)were simultaneously analyzed at GMPR using an Agilent 7500a ICP–MS apparatus coupled with a GeoLas 2005 laser-ablation systemwith a DUV 193 nm ArF-excimer laser (MicroLas, Germany). Detailedanalytical procedures and data reduction are available in Liu et al.(2008, 2010a) and are briefly summarized here. A spot size of32 μm was used for all analyses. Argon was used as the make-upgas and mixed with the carrier gas (helium) via a T-connector beforeentering the ICP. Nitrogen was added into the central gas flow(Ar + He) of the Ar plasma to decrease the detection limit and im-prove precision, which increases the sensitivity for most elementsby a factor 2 to 3 (Hu et al., 2008). Each analysis incorporated a back-ground acquisition of 20–30 s (gas blank) followed by 50 s data ac-quisition. Zircon 91,500 was used as a calibration standard for massdiscrimination and U–Pb isotope fractionation. Time-dependent driftof U–Th–Pb isotopic ratios were corrected using a linear interpolation(with time) for every five analyses according to the variations of91,500 (Liu et al., 2010b). Preferred U–Th–Pb isotopic ratios usedfor 91,500 are fromWiedenbeck et al. (1995). The precision and accu-racy of U–Pb dating with this technique have been evaluated by com-parison with the TIMS data of zircon standard GJ-1 (Jackson et al.,2004). In this study, the minor, non-radiogenic isotope 204Pb was an-alyzed as a monitor of common lead, and the signals of the radiogenicisotopes 206Pb, 207Pb, and 208Pb were corrected in proportion to theirrelative abundances in common lead (Stacey and Kramers, 1975). Theisotope 202Hg was measured simultaneously and used to correct forthe 204Hg isobaric interference on 204Pb. This approach has beenshown to be effective in correcting minor common Pb (Storey et al.,2006; Li et al., 2010). Trace elements were calibrated againstmultiple-reference standards (NIST SRM610 and BCR-2G) combinedwith internal standardization (Liu et al., 2010b). Off-line selectionand integration of background and analyzed signals, and time-driftcorrection and quantitative calibration for trace element analysesand U–Pb dating were performed by ICPMSDataCal (Liu et al., 2008,2010a). Uncertainties of preferred values for the external standard91,500 were propagated into the ultimate results of the samples.Concordia diagrams and weighted mean calculations were madeusing Isoplot/Ex_ver3 (Ludwig, 2003).

5. Results

5.1. Geochemistry of columbite-(Mn)

Columbite-(Mn) occurs as euhedral grains enclosed in albite, quartz,or garnet (Fig. 3A–C). Individual grains are 20–100 μm wide and 50–400 μm long, without obvious zoning and lacking mineral inclusions(Fig. 3D). Twelve electron microprobe analyses on 5 grains revealrelatively homogeneous compositions, with 10.27–12.67 wt.% MnO(excepting one analysis that has 17.17 wt.% MnO), 2.87–9.87 wt.%FeO, 68.13–73.96 wt.% Nb2O5, and 1.30–7.56 wt.% Ta2O5 (Table 1). Inaddition, the mineral contains a significant amount of TiO2 (1.12–4.76 wt.%), WO3 (up to 4.26 wt.%), UO2 (0.05–1.35 wt.%), and Y2O3

(0.35–1.26 wt.%). Other minor elements include SnO2 (b0.17 wt.%),ZrO2 (b0.79 wt.%), and Sc2O3 (b0.39 wt.%). In the columbite–tantalitequadrilateral diagram, the analyses plot in the columbite-(Mn) field(Fig. 4A). The compositions show a slight deviation from the idealtrend defined by the substitution: (Fe, Mn)2+ + 2(Nb, Ta)5+ = 3(Ti,Sn, U, Sc, Zr)4+ (Fig. 4B; Černý et al., 1985), likely due to the presenceof significant amounts of other elements (e.g., Th, Y, and REEs) inthe mineral. Titanium correlates positively with U, Sc, Zr, and Sn(Fig. 5A–D), demonstrating that these cations behave like Ti duringthe crystallization of columbite-(Mn). The tetravalent cations combined

Fig. 3. BSE images showing the compositional and textural features of columbite-(Mn) from the G2 pegmatite. (A–C) columbite-(Mn) included in quartz, albite, and garnet. Otheraccessory minerals in equilibration with garnet include zircon, monazite, and thorite. (D) Compositionally and texturally homogeneous columbite-(Mn). Abbreviations: Ab: albite,Bt: biotite, Col: columbite-(Mn), Grt: garnet, Ms: muscovite, Moz: monazite, Qz: quartz, Zr: zircon.

Table 1Electron microprobe analyses (wt.%) of columbite-(Mn) from the G2 pegmatite.

Sample 1#-1 1#-2 2#-1 2#-2 3#-1 3#-2 4#-1 4#-2 4#-3 5#-1 5#-2 5#-3

CaO 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00FeO 2.87 7.50 7.90 8.76 9.70 9.47 8.18 8.23 8.30 8.58 9.66 9.87MnO 17.17 12.50 12.67 11.95 10.27 10.38 12.41 12.34 12.12 11.60 10.50 10.54TiO2 1.12 1.58 1.56 1.50 4.50 4.76 2.48 2.59 2.70 4.43 4.17 4.24Nb2O5 69.25 68.13 72.80 72.98 69.92 70.24 73.85 73.96 73.00 72.07 71.16 71.24Ta2O5 6.81 7.56 3.66 3.33 1.57 1.38 1.63 1.60 1.60 1.35 1.32 1.30SnO2 0.01 0.04 0.02 0.00 0.02 0.14 0.06 0.04 0.01 0.17 0.15 0.09WO3 4.02 4.26 1.81 1.67 0.56 0.53 0.76 0.65 0.68 0.53 0.00 0.38UO2 0.05 0.11 0.14 0.09 0.94 1.35 0.37 0.40 0.51 0.82 0.71 0.70ZrO2 0.00 0.00 0.11 0.00 0.79 0.75 0.33 0.24 0.45 0.70 0.70 0.72Y2O3 0.35 0.63 1.09 1.15 1.06 0.75 1.26 1.17 1.08 1.12 0.66 0.68Sc2O3 0.07 0.10 0.03 0.00 0.37 0.34 0.09 0.09 0.10 0.34 0.39 0.39Total 101.70 102.42 101.76 101.43 99.69 100.09 101.41 101.31 100.56 101.70 99.43 100.15Ca 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000Fe 0.137 0.357 0.370 0.412 0.456 0.444 0.380 0.383 0.389 0.395 0.454 0.461Mn 0.831 0.603 0.602 0.569 0.490 0.493 0.585 0.581 0.575 0.541 0.500 0.499Ti 0.048 0.068 0.066 0.063 0.190 0.201 0.104 0.108 0.114 0.183 0.176 0.178Nb 1.788 1.753 1.847 1.855 1.779 1.779 1.857 1.859 1.849 1.794 1.808 1.799Ta 0.106 0.117 0.056 0.051 0.024 0.021 0.025 0.024 0.024 0.020 0.020 0.020Sn 0.000 0.001 0.000 0.000 0.000 0.003 0.001 0.001 0.000 0.004 0.003 0.002W 0.059 0.063 0.026 0.024 0.008 0.008 0.011 0.009 0.010 0.008 0.000 0.006U 0.001 0.001 0.002 0.001 0.012 0.017 0.005 0.005 0.006 0.010 0.009 0.009Zr 0.000 0.000 0.003 0.000 0.022 0.021 0.009 0.007 0.012 0.019 0.019 0.020Y 0.011 0.019 0.032 0.034 0.032 0.022 0.037 0.035 0.032 0.033 0.020 0.020Sc 0.004 0.005 0.001 0.000 0.018 0.016 0.004 0.004 0.005 0.016 0.019 0.019Total 2.984 2.987 3.006 3.010 3.031 3.024 3.017 3.016 3.018 3.023 3.030 3.032Mn/(Mn + Fe) 0.859 0.628 0.619 0.580 0.518 0.526 0.606 0.603 0.597 0.578 0.524 0.519Ta / (Ta + Nb) 0.056 0.063 0.029 0.027 0.013 0.012 0.013 0.013 0.013 0.011 0.011 0.011

4 X.-D. Deng et al. / Chemical Geology 344 (2013) 1–11

5X.-D. Deng et al. / Chemical Geology 344 (2013) 1–11

(Ti + Sn + U + Zr + Sc) decrease with an increasing Ta / (Ta + Nb)(Fig. 5E), indicating that these cations are preferentially partitioned intoearlier-formed minerals such as zircon and Ta-poor columbite-(Mn)(Černý et al., 1986; Ercit, 1994; Linnen and Keppler, 1997). Columbite-(Mn) also contains variable WO3 (up to 4.26 wt.%) that correlates posi-tively with Ta (Fig. 5F), suggesting thatW behaves similarly to Ta duringthe crystallization of the mineral.

Trace element concentrations of the columbite-(Mn) determinedby LA-ICP-MS are presented in Table 2. The contents of U rangefrom 1713 to 19,310 ppm (Table 4), consistent with the results ofelectron microprobe analysis (Table 1). Thorium concentrations arebetween 36.7 and 1190 ppm, and correlate positively with U(Fig. 5G). The Th vs U correlation suggests that F, rather than Cland/or CO2, was the dominant complexing agent both for Th and U(Keppler and Wyllie, 1990). This in turn indicates that columbite-(Mn) likely formed from F- and Li-dominated fluids as shown in theTa / (Ta + Nb) vs Mn / (Mn + Fe) diagram (Fig. 4A). In addition,the columbite-(Mn) is relatively enriched in MREE and HREE, withLREE/MREE and LREE/HREE ratios of 0.3–0.5 and 0.7–1.1, respectively.The total REE contents correlate positively with U (Fig. 5H), indicatingthat REEs may enter the columbite-(Mn) structure by an euxenite-typesubstitution mechanism (A2+ + B5+ → REE3+ + Ti4+; Graupner etal., 2010). Chondrite-normalized REE patterns (Fig. 5I) are character-ized by significant enrichment of MREE, prominent negative Eu anom-alies (Eu* = 0.002–0.008), and weak positive Ce anomalies (Ce*=0.92–4.07).

5.2. Geochemistry of zircon

Based on their morphology and textures, zircon grains from thepegmatite can be classified into three types. Type 1 consists ofeuhedral to subhedral grains that are 100–200 μm long with length/width ratios of 1:1–2:1. They are semitransparent, pale to gray, ho-mogenous (Fig. 6A), and characterized by dark cores and irregularmagmatic overgrowth zones in CL images (Fig. 6B). These zircon crys-tals have low Hf (1.1–1.8 wt.%), P (144–860 ppm), Ca (b80.5 ppm),Ti (1.1–3.3 ppm), Nb (1.2–25.4 ppm) and Ta (1.2–9.8 ppm) withNb/Ta ratios of 1.0–2.6 (Table 3; Fig. 7A–C). Thorium and U concen-trations are relatively low, with Th/U ratios of 0.2–0.6 (Fig. 7D). Inaddition, they have relative low LREE/HREE, LREE/MREE, and MREE/HREE ratios ranging from 0.01 to 0.04, 0.05 to 0.19, and 0.20 to 0.28,respectively (Fig. 7E and F). Chondrite-normalized REE patternsdisplay obvious negative Eu anomalies and variable Ce anomalies(Fig. 7G). In the Ce* vs (Sm/La)N and (Sm/La)N vs La diagrams, type

Fig. 4. (A) Ta / (Nb + Ta) vs Mn / (Mn + Fe) diagram for columbite-group minerals (modiratios) ternary plots of columbite-(Mn) from the pegmatite dike. The line denotes the ideal trend(Černý et al., 1985).

1 zircon grains fall in or close to the magmatic zircon field as definedby Hoskin (2005) (Fig. 7H and I).

Type 2 zircon consists of euhedral to subhedral crystals that are200–400 μm long with the aspect ratio of 2:1 to 4:1. Most grains arebrown to dark, homogenous (Fig. 6C), and black in CL images(Fig. 6D). These grains have abnormally high Hf (4.1–20.8 wt.%) andrelatively low P (284–1260 ppm), Ca (15.5–703 ppm), Ti (1.9–9.2 ppm), Nb (32.0–202 ppm) and Ta (28.4–137 ppm), with Nb/Taratios of 0.6–1.8 (Table 3; Fig. 7A–C). Thorium and U concentrationsare extremely variable, ranging from 261 to 15,800 ppm and 3270to 71,180 ppm, respectively (Tables 3 and 4), corresponding to lowTh/U ratios at 0.1–0.2 (Fig. 7D). Type 2 zircon grains have relativelylow LREE/HREE (0.04–0.19) and LREE/MREE (0.03–0.13) ratios, buthigh MREE/HREE ratios (0.35–2.68) (Fig. 7E and F). Chondrite-normalized REE patterns display strong negative Eu anomalies(Eu* = 0.003–0.069) and variable Ce anomalies (Ce* = 0.9–31.5)(Fig. 7G). In the Ce* vs (Sm/La)N and (Sm/La)N vs La diagrams, type2 zircon grains plot in the transitional area between magmatic andhydrothermal zircon, but close to the magmatic field (Hoskin, 2005)(Fig. 7H and I).

Type 3 zircon grains are similar in morphology to Type 2 varieties,but high porosity and abundant mineral inclusions distinguish themfrom Type 2 grains (Fig. 6E). These zircon grains have high Hf (3.3–6.3 wt.%), P (978–1890 ppm), Ca (1360–4400 ppm), Ti (15.9–45.3 ppm), Nb (225–852 ppm) and Ta (63.8–70.4 ppm)with Nb/Ta ra-tios of 3.2–13.5 (Table 3, Fig. 7A–C). They contain unusually high Th andU, ranging from10,160 to 24,070 ppmand 12,880–41,720 ppm, respec-tively (Tables 3 and 4), corresponding to Th/U ratios of 0.6–0.9 (Fig. 7D).Type 3 zircon grains have relative high LREE/HREE, LREE/MREE, andMREE/HREE ratios, ranging from 0.17 to 0.86, 0.19 to 0.69 and 0.90 to1.25, respectively (Fig. 7E and F). Chondrite-normalized REE patternsshow significant negative Eu anomalies (Eu* = 0.015–0.046) andweak positive Ce anomalies Ce (Ce* = 1.05–1.49) (Fig. 7G). In the Ce*vs (Sm/La)N and (Sm/La)N vs La diagrams, type 3 zircon grains plotvery close to the hydrothermal zircon field (Fig. 7H and I).

5.3. U–Pb ages of columbite-(Mn) and zircon

The U–Pb data of zircon and columbite-(Mn) are summarized inTable 4 and graphically illustrated in Fig. 8. Ten spot analyses from10 columbite-(Mn) grains are concordant or nearly concordant(Table 4; Fig. 8A), and have a weighted mean 206Pb/238U age of143 ± 1 Ma (2σ, MSWD = 0.31). A total of 17 analyses were madeon 15 zircon grains, all yielding concordant or nearly concordant U–Pb ages (Table 4; Fig. 8B). Five analyses on cores and rims of Type 1

fied from Černý, 1989); (B) (Nb + Ta)–(Fe + Mn)–(Ti + Sn + U + Zr + Sc) (atomicdefined by the substitution: (Fe,Mn)2+ + 2(Nb,Ta)5+ = 3(Ti + Sn + U + Zr + Sc)4+

Fig. 5. Plots showing the geochemical characteristics of columbite-(Mn) from the G2 pegmatite dike. (A–D) positive correlations of Ti vs Sn, U, Sc, and Zr; (E) tetravalent cations(Ti + Sn + U + Zr + Sc) vs Ta / (Ta + Nb) diagram showing a general negative correlation; (F) a positive correlation between W and Ta; (G) well positive correlation betweenU and Th; (H) positive correlation between U and REEs; (I) chondrite-normalized REE patterns of columbite-(Mn).

6 X.-D. Deng et al. / Chemical Geology 344 (2013) 1–11

zircon yielded similar 207Pb/206Pb ages ranging from 1873 ± 50to 1883 ± 44 Ma, with a weighted mean of 1879 ± 19 Ma (2σ,MSWD = 0.04). Nine Type 2 zircon grains have 206Pb/238U ages of142 ± 3 to 145 ± 11 Ma, with a weighted mean 206Pb/238U age of

Table 2Trace element analyses (ppm) of columbite by the LA-ICP-MS.

Spot no. WY08-1 WY08-2 WY08-3 WY08-4 WY08-

La 0.6 1.3 6.9 3.1 3.7Ce 11.8 16.0 35.2 62.8 113Pr 5.2 6.9 12.7 24.8 12.6Nd 57.8 70.9 125 272 139Sm 106 133 155 352 173Eu 0.1 0.2 0.4 0.8 0.4Gd 211 257 273 594 310Tb 49.2 58.8 56.2 114 63.5Dy 271 324 323 605 385Ho 37.6 45.6 51.7 92.7 68.7Er 83.6 102 138 232 180Tm 13.6 16.1 23.8 37.4 32.3Yb 105 125 201 318 278Lu 13.6 16.8 32.6 51.5 46.6Hf 461 644 571 805 673Th 50.6 79.2 863 1190 514Eu* 0.002 0.003 0.006 0.006 0.00Ce* 1.70 1.30 0.92 1.77 4.07

Note: Eu* = EuN / sqrt (Sm ∗ Gd) N; Ce* = CeN / sqrt(La ∗ Pr) N; abbreviations N = chond

143 ± 1 Ma (2σ, MSWD = 0.16). The remaining 3 Type 3 grainsare also concordant or marginally concordant and yield a reproduc-ible 206Pb/238U age of 125 ± 5 to 128 ± 4 Ma, with a weightedmean of 127 ± 3 Ma (2σ, MSWD = 0.44).

5 WY08-6 WY08-7 WY08-8 WY08-9 WY08-10

7.1 9.7 0.9 1.2 3.260.3 114 10.8 12.4 27.320.4 17.1 4.6 4.8 10.0

213 164 44.9 49.1 109243 204 91.9 88.1 154

0.7 0.5 0.2 0.2 0.5431 352 169 158 27185.6 73.9 41.0 36.6 57.3

503 439 216 196 33995.2 80.0 28.3 25.8 57.5

271 227 59.7 56.0 15147.7 40.7 9.8 8.8 26.9

410 348 71.7 69.2 22670.8 60.1 9.4 9.1 35.5

818 842 364 340 508661 755 49 37 365

6 0.007 0.005 0.004 0.006 0.0081.23 2.17 1.30 1.25 1.19

rite normalized.

Fig. 6. BSE (A, C, E) and CL (B, D) images showing themorphological and textural features of zircon grains. (A, B)Homogeneous Type 1 zirconwith oscillatory zoning; (C, D) homogeneousand inclusion-free Type 2 zircon with dark CL image; (E) typical Type 3 zircon with abundant micrometer-sized pores and mineral inclusions, thorite and uranium oxides (UOX) in thiscase.

7X.-D. Deng et al. / Chemical Geology 344 (2013) 1–11

6. Discussion

6.1. Interpretation of columbite-(Mn) and zircon U–Pb ages

Columbite-(Mn)within the pegmatite is commonly included in andtexturally equilibrated with albite, quartz, and garnet (Fig. 3A–C), indi-cating cogenetic growth of these minerals from the same parental gra-nitic magma. This view is confirmed by the high Nd, REE and Uabundances in columbite-(Mn) (Figs. 4 and 5; Ercit, 1994). In addition,chondrite-normalized REE patterns of columbite-(Mn) are character-ized by HREE b MREE and prominent negative Eu anomalies (Fig. 5I),features consistentwith the crystallization of columbite-groupmineralsfrom highly fractionated magma (Graupner et al., 2010). The petro-graphic and geochemical data, therefore, suggest a magmatic originfor the columbite-(Mn), and thus its U–Pb age (143 ± 1 Ma) can be re-liably considered as the time of emplacement of the host pegmatite.

In contrast, zircon grains extracted from the same pegmatite dikehave much more complicated U–Pb age patterns. Type 1 zircon has aweighted mean 207Pb/206Pb age of 1879 ± 19 Ma. They have darkcores and irregular oscillatory zones (Fig. 6B) and Th/U ratios (0.16–0.64), typical of magmatic zircon. Their REE patterns (Fig. 7G) arealso consistent with a magmatic origin (Hoskin and Schaltegger,2003). In the Ce* vs (Sm/La)N and (Sm/La)N vs La diagrams (Fig. 7Hand I), Type 1 zircon all plots in or proximal to the magmatic zirconfield (Hoskin, 2005). Their 207Pb/206Pb age is comparable with thatof zircon grains from many Paleoproterozoic diabase and pegmatitedikes in the Xiaoqinling district (1.9–1.8 Ga, Li et al., 2012a). Taken to-gether, Type 1 zircon grains are interpreted as inherited components.

Type 2 zircon grains have abnormally high Hf contents (up to20.8 wt.%), which is consistent with a highly differentiated pegmatiticsource (Pupin, 2000). They are characterized by the enrichment ofMREE relative to HREE (MREE/HREE = 0.35–2.68; Table 3), indicat-ing a magmatic source with garnet as a residue or separation ofgarnet during the evolution of the magma (Rubatto, 2002); the pres-ence of abundant garnet in the pegmatite (Fig. 2) favors the secondpossibility. Type 2 zircon grains display strong negative Eu anomalies(Eu* = 0.003–0.069) that resemble the whole-rock samples of thepegmatite dike (Eu* = 0.013–0.030, Table S.2), implying a close rela-tionship between zircon and the pegmatite. In the Ce* vs (Sm/La)Nand (Sm/La)N vs La diagrams (Fig. 7H and I), Type 2 zircon grainsplot between the hydrothermal and magmatic zircon fields, butclose to the magmatic field, confirming that they grew from a hydroussiliceous melt (London, 2005). It is thus concluded that Type 2 zirconcrystallized from volatile-rich pegmatitic magmas. Consequently, theweighted mean 206Pb/238U age (143 ± 1 Ma) of the Type 2 zircon isinterpreted as the emplacement age of the pegmatite.

Type 3 zircon grains contain abundant mineral inclusions (e.g. tho-rite, uranium oxides) and have high concentrations of Ca, P, Nd, Ta,

and LREE (Table 3; Fig. 7A–D); features commonly observed in hydro-thermal zircon grains (Geisler et al., 2007; Kusiak et al., 2009). In theCe* vs (Sm/La)N and (Sm/La)N vs La diagrams (Fig. 7H and I), Type 3zircon grains plot in the vicinity of the hydrothermal zircon field, dis-tinctly different from the Type 2 crystals, indicating that they aremost likely of hydrothermal origin. It is noted that Type 3 zircongrains are similar in morphology to the Type 2 varieties and havehigh Hf contents (up to 6.3 wt.%), implying that the former possiblyformed by the hydrothermal alteration of the latter. In addition,Type 3 grains have significantly higher Nb/Ta and Th/U ratios comparedto Type 2 crystals, confirming that they may have resulted fromre-equilibration of the Type 2 zircons with a post-crystallization Nb-and Th-rich fluid. The sericitic alteration of albite is consistent withsuch hydrothermal processes. The ages of Type 3 zircon (125 ± 5 to128 ± 4 Ma) overlap the ages of pervasive hydrothermal alterationand goldmineralization throughout the Xiaoqinling district that peakedat 130–125 Ma (Li et al., 2012a, 2012b). This indicates that some zircongrains from the pegmatite have been affected by subsequent hydrother-mal activities, presumably related to the district-wide hydrothermalflow and gold mineralization. Collectively, columbite-(Mn) U–Pb agesprovide direct and reliable constraints on the timing of G2 pegmatiteformation in the Xiaoqinling district, whereas some zircon grains fromthe same dike have been affected by later hydrothermal activities.

6.2. Implications for geochronology of pegmatite and associated oredeposits

The U–Pb zircon geochronometer is a powerful tool for datingigneous rocks including pegmatite bodies (e.g., Tapponnier et al.,1990; Schärer et al., 1994). It has been shown that pegmatite bodiesmay contain significant amounts of inherited zircon (Ghosh et al.,2008; Marsh et al., 2012), which would complicate the age distribu-tion and result in erroneous interpretations of the U–Pb data(Romer and Wright, 1992). In many cases, zircon in pegmatite mayhave unusually high U and Th concentrations, up to 7.1 wt.% U and2.4 wt.% Th in the studied G2 pegmatite (Tables 3 and 4). The ex-tremely high U and Th may cause intensive metamictic damage tothe crystal lattice of zircon, and consequently the alteration andre-crystallization of the mineral by later hydrothermal fluids(Geisler et al., 2007). In this case, U–Pb ages of hydrothermally al-tered zircon constrain the timing of post-pegmatite fluid processesrather than the emplacement/crystallization of the pegmatite(Soman et al., 2010). The U–Pb ages of Type 3 zircon grains fromthe G2 pegmatite provide a good example illustrating how high-U–Th zircon crystals can be affected by post-magmatic hydrothermalalteration.

In contrast, the clearly unimodal columbite-(Mn) U–Pb ages indi-cate no evidence of inheritance, metamictic damage, or hydrothermal

Table 3Trace element analyses (ppm) of zircon by LA-ICP-MS.

AnalysisNo

WY08-101 WY08-102 WY08-103 WY08-104 WY08-105 WY08-106 WY08-107 WY08-108 WY08-109 WY08-110 WY08-111 WY08-112 WY08-113 WY08-114 WY08-115 WY08-116 WY08-117

Type 1 zircons Type 2 zircons Type 3 Zircons

Li 97.9 121 86.9 60.5 73.9 53.0 73.9 98.1 59.1 40.5 164 126 37.6 78.4 67.4 96.8 51.8P 325 860 144 355 430 1260 1030 538 284 1150 315 449 284 727 978 1610 1890Ca 36.9 b.d.l. b.d.l. 80.5 71.3 155 196 45.4 15.5 703 82.1 b.d.l. 64.6 544 4400 1540 1360Ti 1.1 3.2 1.9 2.5 3.3 4.7 2.1 7.0 5.6 9.2 1.9 5.4 2.8 5.0 15.9 36.5 45.3Y 1120 1870 383 1480 1900 11,220 8870 3290 1660 7700 1600 3620 477 1310 5010 6540 3210Nb 25.4 3.2 1.2 4.1 5.3 128 87.9 202 37.1 129 32.0 127 50.9 106 225 404 852La 0.0 0.0 0.0 0.1 0.1 1.9 0.9 2.5 1.0 11.9 0.7 2.4 1.5 3.9 210 35.0 41.0Ce 6.9 15.2 9.1 23.4 31.5 161 126 78.5 23.7 100 18.2 79.5 5.6 12.5 530 178 214Pr 0.0 0.0 0.0 0.2 0.2 1.9 1.1 1.6 1.1 4.8 0.6 1.8 1.2 3.2 73.6 27.9 30.5Nd 0.3 0.7 0.3 3.8 5.2 35.8 21.6 16.8 11.9 39.9 7.4 18.3 7.8 22.2 409 174 180Sm 1.5 3.0 0.9 8.8 10.5 230 170 72.7 53.2 150 40.5 92.8 7.3 20.6 234 162 132Eu 0.2 0.4 0.3 0.6 1.1 0.6 0.4 0.2 0.2 0.4 0.3 0.3 0.3 0.3 1.7 4.3 1.4Gd 14.0 25.4 5.2 37.2 48.2 1290 1020 377 228 839 195 473 20.2 63.0 515 486 273Tb 6.2 10.5 2.3 11.3 15.4 384 310 112 86.3 240 78.7 143 7.3 23.8 111 142 72.7Dy 86.9 144 29.4 129 173 2500 2000 719 557 1570 534 913 68.3 238 747 1180 572Ho 36.8 60.6 12.3 49.5 63.7 398 311 105 91.2 259 86.0 133 18.7 61.3 153 255 116Er 183 289 61.3 230 288 902 677 221 237 600 224 284 73.1 209 455 764 338Tm 41.7 69.8 14.1 50.8 64.2 140 104 32.7 44.7 94.3 44.9 44.6 19.2 46.1 85.0 151 62.8Yb 420 715 145 519 648 1040 749 237 378 692 388 332 213 435 752 1350 535Lu 79.9 136.7 29.8 100 121 122 84.5 26.5 50.4 84.7 55.1 37.9 40.0 63.1 122 198 70.2Hf 17,700 17,100 10,800 11,300 13,600 94,400 96,200 92,500 204,000 44,400 208,000 119,000 41,400 48,000 32,700 62,700 35,400Ta 9.8 2.9 1.2 3.3 3.3 71.3 51.5 137 35.1 71.7 50.0 89.9 28.4 81.8 70.4 63.8 63.3Th 261 301 67.4 311 399 6480 5140 4900 1950 15,800 1790 4770 261 1000 23,620 24,070 10,160LREE/MREE 0.05 0.07 0.19 0.12 0.12 0.04 0.04 0.07 0.04 0.05 0.03 0.06 0.13 0.10 0.69 0.19 0.40LREE/HREE 0.01 0.01 0.04 0.03 0.03 0.09 0.09 0.19 0.05 0.11 0.04 0.15 0.05 0.06 0.86 0.17 0.46MREE/HREE 0.20 0.20 0.20 0.26 0.28 2.18 2.37 2.68 1.43 2.08 1.31 2.51 0.35 0.54 1.25 0.90 1.16Nb/Ta 2.6 1.1 1.0 1.3 1.6 1.8 1.7 1.5 1.1 1.8 0.6 1.4 1.8 1.3 3.2 6.3 13.5(Sm/La)N 169.9 462.7 37.2 227.0 295.5 191.7 300.8 44.8 85.9 19.5 93.5 60.8 7.3 8.2 1.7 7.2 5.0Ce* 99.5 277.2 135.3 56.0 67.0 20.7 31.5 9.7 5.7 3.3 6.9 9.3 1.0 0.9 1.0 1.4 1.5Eu* 0.111 0.147 0.376 0.103 0.144 0.003 0.003 0.003 0.006 0.003 0.010 0.004 0.069 0.028 0.015 0.046 0.022

Note: LREE = La + Ce + Pr + Nd; MREE = Sm + Eu + Gd + Tb + Dy + Ho; HREE = Er + Tm + Yb + Lu.(Sm/La)N = SmN / LaN.b.d.l. = below detection limit.

8X.-D

.Deng

etal./

ChemicalG

eology344

(2013)1–11

Fig. 7. Plots showing the geochemical characteristics of the three types of zircon from the G2 pegmatite dike. (A–D) correlation between Ca and P, Ti, Nb/Ta, and Th/U showing anincrease of the “non-formula” elements in zircon due to the alteration; (E–F) correlation of LREE/HREE vs LREE/MREE (E) and LREE/HREE vs MREE/HREE (F) showing the distribu-tions of HREE, MREE, and LREE in zircon; (G) chondrite-normalized REE patterns of zircon; (H, I) cerium anomaly (Ce*) vs (Sm/La)N and (Sm/La)N vs La diagrams. The referenceareas of magmatic and hydrothermal zircon are after Hoskin (2005).

9X.-D. Deng et al. / Chemical Geology 344 (2013) 1–11

overprinting (Fig. 8). The textural and geochemical data indicatethat they are unequivocally of magmatic origin (Figs. 3 and 5). Thecolumbite-(Mn) U–Pb age (143 ± 1 Ma) is consistent with thevalue of the Type 2 zircon, providing a reliable age constraint on thecrystallization of the pegmatite dike. This age is within analytical un-certainty of the zircon U–Pb age of the Wenyu monzogranite pluton(141 ± 2 Ma; Li et al., 2012a). An implication of this age compatibil-ity is that the G2 pegmatites and the Wenyu pluton may have beenevolved from a common granitic magma chamber, or that the pegma-tites were derived from a parental magma represented by the Wenyupluton.

Columbite-groupminerals are important oreminerals in pegmatite-related Nb and Ta deposits (e.g., Černý and Ercit, 1989; Beurlen et al.,2008; Linnen et al., 2012). Columbite-groupminerals also occurswidelyin a range of hydrothermal deposits genetically linked to pegmatite orgranite, such as Li, Be, B, and Cs deposits (e.g. Černý and Lenton, 1995;Selway et al., 2005). They are also common in Sn and W deposits (e.g.Lerouge et al., 2007), and REE deposits as well (e.g. Möller, 1989).Thus, dating columbite-groupminerals may provide significant insightsinto the timing of the mineralization of the aforementioned deposittypes. Romer and Lehmann (1995) and Melcher et al. (2008) havedemonstrated that U–Pb dating of columbite-(Mn, Fe) and tantalite-(Mn, Fe) can yield useful chronological constrains on Ta–Nb-(Sn–Li)

mineralization in Africa. Our results add an additional weight of evi-dence that columbite-group minerals are an ideal target for U–Pb geo-chronology study to precisely constrain the timing and history ofpegmatite-related ore deposits.

7. Conclusions

Columbite-(Mn) grains from a pegmatite dike in the Xiaoqinlingdistrict, North China Craton yield concordant U–Pb ages with aweighted mean 206Pb/238U age of 143 ± 1 Ma. Textures and geo-chemical data confirm that the columbite-(Mn) was crystallizedfrom a pegmatitic magma, and its U–Pb age provides a good con-straint on the time of pegmatite formation. Zircon grains from thesame dike consist of inherited (Type 1), syn-magmatic (Type 2),and hydrothermally altered (Type 3) varieties, which yield distinctU–Pb ages of 1879 ± 19 Ma, 143 ± 1 Ma, and 127 ± 3 Ma, respec-tively. Morphological, textural, and geochemical data indicate thatType 2 zircon grains are magmatic minerals formed during pegmatitecrystallization, whereas Type 3 zircon reflects post-pegmatite hydro-thermal alteration presumably related to the pervasive gold mineral-ization in the Xiaoqinling district. The age of Type 2 zircon isconsistent with that of columbite-(Mn) and considered as theemplacement age of the pegmatite dike. This study shows that

Table 4LA-ICP-MS U–Pb isotope data for columbite and zircon from the Wenyu G2 pegmatite.

Analysis no U Pb Th/U 207Pb/206Pb 2σ 207Pb/235U 2σ 206Pb/238U 2σ 207Pb/206Pb 2σ 207Pb/235U 2σ 206Pb/238U 2σ

ppm ppm Ma Ma Ma

ColumbiteWY08-1 2561 63.8 0.02 0.0510 0.0028 0.1574 0.0118 0.0224 0.0006 239 121 148 5 143 2WY08-2 3852 122 0.02 0.0545 0.0150 0.1659 0.0485 0.0221 0.0004 391 485 156 42 141 3WY08-3 4988 224 0.17 0.0379 0.0047 0.1235 0.0198 0.0226 0.0009 -413 251 118 18 144 6WY08-4 19,310 490 0.06 0.0492 0.0014 0.1523 0.0064 0.0225 0.0004 157 52 144 6 143 3WY08-5 7356 294 0.07 0.0478 0.0114 0.1496 0.0383 0.0227 0.0004 87 371 142 34 144 3WY08-6 13,630 690 0.05 0.0376 0.0187 0.1518 0.0805 0.0224 0.0008 -431 738 143 71 143 5WY08-7 10,300 424 0.07 0.0468 0.0096 0.1504 0.0313 0.0225 0.0007 40 316 142 28 143 4WY08-8 1763 58.4 0.03 0.0472 0.0178 0.1501 0.0603 0.0227 0.0006 57 578 142 53 144 4WY08-9 1713 62.3 0.02 0.0454 0.0204 0.1791 0.0968 0.0225 0.0023 -1 677 167 83 143 15WY08-10 4759 176 0.08 0.0486 0.0086 0.1504 0.0300 0.0223 0.0006 130 266 142 26 142 4

ZirconWY08-101 964 381 3.85 0.1152 0.0044 5.4094 0.2094 0.3384 0.0047 1883 46 1886 33 1879 22WY08-102 1366 526 4.75 0.1150 0.0041 5.3644 0.1915 0.3361 0.0044 1880 44 1879 31 1868 21WY08-103 411 159 6.22 0.1147 0.0034 5.3985 0.1670 0.3386 0.0044 1876 36 1885 27 1880 21WY08-104 523 224 1.72 0.1146 0.0049 5.3208 0.2301 0.3346 0.0052 1873 50 1872 37 1861 25WY08-105 618 259 1.73 0.1152 0.0051 5.3492 0.2399 0.3346 0.0058 1884 45 1877 38 1861 28WY08-106 65,040 1552 0.10 0.0482 0.0020 0.1515 0.0072 0.0226 0.0005 109 60 143 6 144 3WY08-107 57,640 1379 0.09 0.0489 0.0021 0.1532 0.0071 0.0225 0.0004 142 69 145 6 143 3WY08-108 40,700 1211 0.10 0.0488 0.0092 0.1595 0.0308 0.0225 0.0009 139 291 150 27 144 6WY08-109 34,080 723 0.08 0.0303 0.0309 0.1517 0.1553 0.0227 0.0018 -271 1003 143 137 145 11WY08-110 71,180 1670 0.21 0.0530 0.0057 0.1679 0.0184 0.0225 0.0006 328 197 158 16 144 4WY08-111 30,560 667 0.05 0.0465 0.0177 0.2058 0.0792 0.0224 0.0012 25 512 190 67 143 7WY08-112 52,560 1297 0.08 0.0493 0.0013 0.1529 0.0068 0.0225 0.0005 162 62 144 6 143 3WY08-113 3270 100 0.07 0.0467 0.0089 0.1502 0.0311 0.0223 0.0004 34 322 142 27 142 3WY08-114 6396 596 0.09 0.0061 0.0064 0.1508 0.1598 0.0224 0.0014 -1584 435 143 141 143 9WY08-115 41,720 1010 0.52 0.0470 0.0031 0.1301 0.0103 0.0195 0.0009 49 83 124 9 125 5WY08-116 26,790 766 0.81 0.0522 0.0039 0.1446 0.0115 0.0200 0.0006 293 114 137 10 128 4WY08-117 12,880 1779 0.28 0.0534 0.0219 0.1621 0.0665 0.0198 0.0006 346 656 153 58 127 4

10 X.-D. Deng et al. / Chemical Geology 344 (2013) 1–11

columbite-group minerals can be an ideal U–Pb geochronometer forprecisely dating pegmatites and associated ore deposits. The resultspresented here also demonstrate that trace element geochemistry ofcolumbite-group minerals and zircon may provide information usefulin distinguishing their origins and the interpretation of the U–Pb dataof these minerals.

Acknowledgments

We thank Zhang Suxin, Yang Meijun, and Zheng Shu for their helpin SEM, EMP, and CL analysis, respectively. Thorough and constructivereviews by two anonymous reviewers and editorial suggestions of

Fig. 8. LA-ICP-MS U–Pb concordia plots of columbite-(Mn) (A) and zircon (B) from the G2 p(2σ).

Prof. KLaus Mezger, Editor of Chemical Geology, have been very help-ful in our revision, which are gratefully acknowledged. This researchwas supported by the Natural Science Foundation of China (grant90814004), the Fundamental Research Funds for the Central Universities(CUG120102), and the MOST special fund from the GPMR lab(MSFGPMR201205).

Appendix A. Supplementary data

Supplementary data to this article can be found online at http://dx.doi.org/10.1016/j.chemgeo.2013.02.014.

egmatite dike under investigation. Age uncertainties are quoted as 95% confidence level

11X.-D. Deng et al. / Chemical Geology 344 (2013) 1–11

References

Beurlen, H., Da Silva, M., Thomas, R., Soares, D., Olivier, P., 2008. Nb–Ta-(Ti–Sn) oxidemineral chemistry as tracer of rare-element granitic pegmatite fractionation in theBorborema Province, Northeastern Brazil. Mineralium Deposita 43 (2), 207–228.

Černý, P., 1989. Characteristics of pegmatite deposits of tantalum. In: Möller, P., Černý, F.,Saupé, F. (Eds.), Lanthanides, Tantalum and Niobium. Springer, Berlin, pp. 195–239.

Černý, P., Ercit, T.S., 1989. Mineralogy of niobium and tantalum; crystal chemical rela-tionships, paragenetic aspects and their economic implications. In: Möller, P.,Černý, F., Saupé, F. (Eds.), Lanthanides, Tantalum and Niobium. Springer, Berlin,pp. 27–79.

Černý, P., Ercit, T.S., 2005. The classification of granitic pegmatites revisited. The CanadianMineralogist 43, 2005–2026.

Černý, P., Lenton, P.G., 1995. The Buck and Pegli lithium deposits, southeasternManitoba:the problem of updip fractionation in subhorizontal pegmatite sheets. EconomicGeology 90, 658–675.

Černý, P., Roberts, W.L., Ercit, T.S., Chapman, R., 1985. Wodginite and associated oxideminerals from the Peerless pegmatite, Pennington County, South Dakota. AmericanMineralogist 70 (9–10), 1044–1049.

Černý, P., Goad, B.E., Hawthorne, F.C., Chapman, R., 1986. Fractionation trends of theNb- and Ta-bearing oxide minerals in the Greer Lake pegmatitic granite and itspegmatite aureole, southeastern Manitoba. American Mineralogist 71 (3–4),501–517.

Ercit, T.S., 1994. The geochemistry and crystal chemistry of columbite-group mineralsfrom granitic pegmatites, southwestern Grenville Province, Canadian Shield. TheCanadian Mineralogist 32 (2), 421–438.

Geisler, T., Pidgeon, R.T., van Bronswijk, W., Kurtz, R., 2002. Transport of uranium,thorium, and lead in metamictic zircon under low-temperature hydrothermal con-ditions. Chemical Geology 191 (1–3), 141–154.

Geisler, T., Schaltegger, U., Tomaschek, F., 2007. Re-equilibration of zircon in aqueousfluids and melts. Elements 3 (1), 43–50.

Ghosh, A.K., Frei, R., Whitehouse, M.J., 2008. U–Pb geochronologic study of magmaticzircon in Paleoproterozoic granitic pegmatite and associated metapelites, BlackHills, South Dakota: Implications for gold petrogenesis and sedimentary prove-nance. Geological Society of America Abstracts with Programs 40 (6), 145.

Graupner, T., Melcher, F., Gäbler, H.E., Sitnikova, M., Brätz, H., Bahr, A., 2010. Rare earthelement geochemistry of columbite-group minerals: LA-ICP-MS data. MineralogicalMagazine 74 (4), 691–713.

Hoskin, P.W.O., 2005. Trace-element composition of hydrothermal zircon and thealteration of Hadean zircon from the Jack Hills, Australia. Geochimica etCosmochimica Acta 69 (3), 637–648.

Hoskin, P.W.O., Schaltegger, U., 2003. The composition of zircon and igneous and meta-morphic petrogenesis. Reviews in Mineralogy and Geochemistry 53, 27–62.

Hu, S.X., Lin, Q.L., 1989. Geology and Prospecting on the Collided Zone between NorthChina Paleo-plate and South China Paleo-plate. Nanjing University Press, Nanjing,pp. 29–35 (in Chinese).

Hu, Z.C., Gao, S., Liu, Y.S., Hu, S.H., Chen, H.H., Yuan, H.L., 2008. Signal enhancement inlaser ablation ICP–MS by addition of nitrogen in the central channel gas. Journal ofAnalytical Atomic Spectrometry 23 (8), 1093–1101.

Jackson, S.E., Pearson, N.J., Griffin, W.L., Belousova, E.A., 2004. The application of laserablation inductively coupled plasma-mass spectrometry to in situ U–Pb zircongeochronology. Chemical Geology 211 (1–2), 47–69.

Keppler, H., Wyllie, P.J., 1990. Role of fluids in transport and fractionation of uraniumand thorium in magmatic processes. Nature 348, 531–533.

Kusiak, M.A., Dunkley, D.J., Słaby, E., Martin, H., Budzyń, B., 2009. Sensitive high-resolution ion microprobe analysis of zircon reequilibrated by late magmatic fluidsin a hybridized pluton. Geology 37 (12), 1063–1066.

Lerouge, C., Deschamps, Y., Piantone, P., Gilles, C., Breton, J., 2007. Metal-carrier accessoryminerals associated with W ± Sn mineralization, La Châtaigneraie tungsten ore dis-trict, Massif Central, France. The Canadian Mineralogist 45 (4), 875–889.

Li, S.M., Qu, L.Q., Su, Z.B., 1996. Geology of Gold Deposits in the Xiaoqinling District andMetallogenic Prognosis. Geological Publishing House, Beijing (250 pp. (in Chinese)).

Li, H.M., Chen, Y.C., Wang, D.H., Ye, H.S., Wang, Y.B., Zhang, C.Q., Dai, J.Z., 2007. SHRIMPU–Pb zircon ages of metamorphic rocks and veins in the Xiaoqinling area, Henan,and their geological significance. Acta Petrologica Sinica 23 (10), 2504–2512 (inChinese with English abstract).

Li, J.W., Deng, X.D., Zhou, M.F., Liu, Y.S., Zhao, X.F., Guo, J.L., 2010. Laser ablation ICP–MStitanite U–Th–Pb dating of hydrothermal ore deposits: a case study of theTonglushan Cu–Fe–Au skarn deposit, SE Hubei Province, China. Chemical Geology270 (1–2), 56–67.

Li, J.W., Li, Z.K., Zhou, M.F., Chen, L., Bi, S.J., Deng, X.D., Qiu, H.N., Cohen, B., Selby, D.,Zhao, X.F., 2012a. The early Cretaceous Yangzhaiyu lode gold deposit, NorthChina Craton: a link between craton reactivation and gold veining. Economic Geology107 (1), 43–79.

Li, J.W., Bi, S.J., Selby, D., Chen, L., Vasconcelos, P., Thiede, D., Zhou, M.F., Zhao, X.F., Li,Z.K., Qiu, H.N., 2012b. Giant Mesozoic gold provinces related to the destruction ofthe North China craton. Earth and Planetary Science Letters 349–350, 26–37.

Linnen, R.L., Keppler, H., 1997. Columbite solubility in granitic melts; consequences forthe enrichment and fractionation of Nb and Ta in the Earth's crust. Contributions toMineralogy and Petrology 128 (2–3), 213–227.

Linnen, R.L., Van Lichtervelde, M., Černý, P., 2012. Granitic pegmatites: granitic pegma-tites as sources of strategic metals. Elements 8, 275–280.

Liu, Y.S., Hu, Z.C., Gao, S., Günther, D., Xu, J., Gao, C.G., Chen, H.H., 2008. In situ analysisof major and trace elements of anhydrous minerals by LA-ICP-MS without applyingan internal standard. Chemical Geology 257 (1–2), 34–43.

Liu, Y., Hu, Z., Zong, K., Gao, C., Gao, S., Xu, J., Chen, H., 2010a. Reappraisement and re-finement of zircon U–Pb isotope and trace element analyses by LA-ICP-MS. ChineseScience Bulletin 55 (15), 1535–1546.

Liu, Y.S., Gao, S., Hu, Z.C., Gao, C.G., Zong, K.Q., Wang, D.B., 2010b. Continental and oce-anic crust recycling-induced melt-peridotite interactions in the Trans-North ChinaOrogen: U–Pb Dating, Hf isotopes and trace elements in zircons from mantle xeno-liths. Journal of Petrology 51 (1–2), 537–571.

London, D., 2005. Granitic pegmatites; an assessment of current concepts and direc-tions for the future. Lithos 80 (1), 281–303.

Ludwig, K.R., 2003. ISOPLOT 3.00: A Geochronological Toolkit for Microsoft Excel.Berkeley Geochronology Center, California, Berkeley.

Mao, J.W., Xie, G.Q., Pirajno, F., Ye, H.S., Wang, Y.B., Li, Y.F., Xiang, J.F., Zhao, H.J., 2010.Late Jurassic–Early Cretaceous granitoid magmatism in Eastern Qinling, central-eastern China: SHRIMP zircon U–Pb ages and tectonic implications. AustralianJournal of Earth Sciences 57 (1), 101–112.

Marsh, J.H., Gerbi, C.C., Culshaw, N.G., Johnson, S.E., Wooden, J.L., Clark, C., 2012. Usingzircon U–Pb ages and trace element chemistry to constrain the timing of metamor-phic events, pegmatite dike emplacement, and shearing in the southern ParrySound domain, Grenville Province, Canada. Precambrian Research 192–195,142–165.

Melcher, F., Sitnikova, M.A., Graupner, T., Martin, N., Oberthür, T., Henjes-Kunst, F.,Gäbler, E., Gerdes, A., Brätz, H., Davis, D.W., Dewaele, S., 2008. Fingerprinting ofconflict minerals: columbite–tantalite (“coltan”) ores. SGA News 23, 1–14.

Möller, P., 1989. REE(Y), Nb, and Ta enrichment in pegmatites and carbonatite-alkalicrock complexes. In: Möller, P., Černý, F., Saupé, F. (Eds.), Lanthanides, Tantalumand Niobium. Springer, Berlin, pp. 103–144.

Nie, F.J., Jiang, S.H., Zhao, Y.M., 2001. Lead and sulfur isotopic studies of the Wenyu andthe Dongchuang quartz vein type gold deposits in the Xiaoqinling area, Henan andShaanxi Provinces, Central China. Mineral Deposits 20, 163–173 (in Chinese withEnglish abstract).

Pupin, J.P., 2000. Granite genesis related to geodynamics from Hf–Y in zircon. Transac-tions of the Royal Society of Edinburgh: Earth Sciences 91 (1–2), 245–256.

Romer, R.L., Lehmann, B., 1995. U–Pb columbite age of Neoproterozoic Ta–Nb mineral-ization in Burundi. Economic Geology 90 (8), 2303–2309.

Romer, R.L., Smeds, S.A., 1994. Implications of U–Pb ages of columbite-tantalites fromgranitic pegmatites for the Palaeoproterozoic accretion of 1.90–1.85 Ga magmaticarcs to the Baltic Shield. Precambrian Research 67 (1–2), 141–158.

Romer, R.L., Smeds, S.A., 1996. U–Pb columbite ages of pegmatites from Sveconorwegianterranes in southwestern Sweden. Precambrian Research 76 (1–2), 15–30.

Romer, R.L., Smeds, S.A., 1997. U–Pb columbite chronology of post-kinematicPalaeoproterozoic pegmatites in Sweden. Precambrian Research 82 (1–2), 85–99.

Romer, R.L., Wright, J.E., 1992. U–Pb dating of columbites; a geochronologic tool to datemagmatism and ore deposits. Geochimica et Cosmochimica Acta 56 (5), 2137–2142.

Rubatto, D., 2002. Zircon trace element geochemistry: partitioning with garnet and thelink between U–Pb ages and metamorphism. Chemical Geology 184 (1–2), 123–138.

Schärer, U., Zhang, L.S., Tapponnier, P., 1994. Duration of strike-slip movements in largeshear zones: the Red River belt, China. Earth and Planetary Science Letters 126 (4),379–397.

Selway, J.B., Breaks, F.W., Tindle, A.J., 2005. A review of rare-element (Li–Cs–Ta) peg-matite exploration techniques for the Superior Province, Canada, and large world-wide tantalum deposits. Exploration and Mining Geology 14 (1–4), 1–30.

Simmons, W.B., Pezzotta, F., Shigley, J.E., Beurlen, H., 2012. Granitic pegmatites: granit-ic pegmatites as sources of colored gemstones. Elements 8, 281–287.

Smith, S.R., Foster, G.L., Romer, R.L., Tindle, A.G., Kelley, S.P., Noble, S.R., Horstwood, M.,Breaks, F.W., 2004. U–Pb columbite–tantalite chronology of rare-element pegmatitesusing TIMS and laser ablation-multi collector-ICP–MS. Contributions to Mineralogyand Petrology 147 (5), 549–564.

Soman, A., Geisler, T., Tomaschek, F., Grange, M., Berndt, J., 2010. Alteration of crystal-line zircon solid solutions: a case study on zircon from an alkaline pegmatite fromZomba–Malosa, Malawi. Contributions to Mineralogy and Petrology 160 (6),909–930.

Stacey, J.S., Kramers, J.D., 1975. Approximation of terrestrial lead isotope evolution by atwo-stage model. Earth and Planetary Science Letters 26 (2), 207–221.

Storey, C.D., Jeffries, T.E., Smith, M., 2006. Common lead-corrected laser ablation ICP–MS U–Pb systematics and geochronology of titanite. Chemical Geology 227(1–2), 37–52.

Tapponnier, P., Lacassin, R., Leloup, P.H., Schärer, U., Zhong, D.L., Wu, H.W., Liu, X.H., Ji,S.C., Zhong, J.Y., 1990. The Ailao Shan/Red River metamorphic belt; tertiary left-lateral shear between Indochina and South China. Nature 343, 431–437.

Wiedenbeck, M., Allé, P., Corfu, F., Griffin, W.L., Meier, M., Oberli, F., Quadt, A.V.,Roddick, J.C., Spiegel, W., 1995. Three natural zircon standards for U–Th–Pb, Lu–Hf,trace element and REE analyses. Geostandards Newsletter 19 (1), 1–23.

Xu, X.S., Griffin, W.L., Ma, X., O'Reilly, S.Y., He, Z.Y., Zhang, C.L., 2009. The Taihua groupon the southern margin of the North China Craton: further insights from U–Pb agesand Hf isotope compositions of zircons. Mineralogy and Petrology 97 (1-2), 43–59.