In situ perovskite Sr–Nd isotopic constraints on the petrogenesis of the Ordovician Mengyin kimberlites in the North China Craton Yue-Heng Yang a , Fu-Yuan Wu a, ⁎, Simon A. Wilde b , Xiao-Ming Liu c , Yan-Bin Zhang a , Lie-Wen Xie a , Jin-Hui Yang a a State Key Laboratory of Lithospheric Evolution, Institute of Geology and Geophysics, Chinese Academy of Sciences, P. O. Box 9825, Beijing 100029, China b Department of Applied Geology, Curtin University of Technology, P. O. Box U1987, Perth, Western Australia 6845, Australia c State Key Laboratory of Continental Dynamics, Department of Geology, Northwest University, Xi'an 750069, China abstract article info Article history: Received 5 October 2008 Received in revised form 17 February 2009 Accepted 18 February 2009 Editor: R.L. Rudnick Keywords: Laser ablation analyses Sr–Nd–Hf isotopes Perovskite Mengyin kimberlite North China Craton Precise Sr, Nd and Hf isotopic data are important indicators in deciphering the evolution of magmatic rocks and their source. However, such data are difficult to obtain for kimberlite, since these rocks are commonly contaminated by both crustal and mantle materials and also show extensive alteration and weathering following emplacement. In this study, numerous kimberlitic samples from Mengyin in northeast China were selected for U–Pb dating and Sr–Nd–Hf isotopic analysis of perovskite. In situ U–Pb analyses of fresh perovskite yield an age of 470±4 Ma, which is considered the emplacement age of the Mengyin kimberlite. However, the altered perovskite shows Pb loss and yields Paleozoic-Mesozoic ages, indicating that perovskite is not as strongly resistant to isotopic modification as previously thought. In situ Sr–Nd isotopic analyses by laser ablation of perovskite grains collected from the main Mengyin kimberlite record uniform Sr and Nd isotopic compositions with an average initial 87 Sr/ 86 Sr ratio of 0.70371±12 and ε Nd (t) value of 0.13±0.22, which are identical, within uncertainties, to the values obtained by solution analyses. However, they are significantly different from comparable data obtained from whole rock samples, indicating that initial Sr–Nd isotopic ratios calculated from whole rock measurements on kimberlites are likely to record mixed isotopic signatures, due to crustal contamination and/or subsequent alteration. This conclusion is also supported by the Hf isotopic data for perovskite. Meaningful isotopic ratios for kimberlite can therefore only be obtained from single minerals such as pervoskite. The Mengyin samples investigated in this study have isotopic compositions similar to the Group I kimberlite, as defined in Southern Africa, and are interpreted to reflect derivation from a primitive mantle source. Combined with the fact that the Mengyin kimberlites were emplaced coevally with regional lithospheric uplift in the Ordovician at ∼ 470 Ma, it is proposed that a mantle plume triggered kimberlite magmatism in this part of the North China Craton. © 2009 Elsevier B.V. All rights reserved. 1. Introduction Kimberlite is a unique ultramafic rock derived from either the lithosphere or sub-lithospheric mantle (Mitchell, 1986, 1995; Woolley et al., 1996) and can provide invaluable information about the composition and structure of that lithosphere (Mitchell, 1986). Based on mineralogical and petrological characteristics and Sr–Nd isotopic data, kimberlites can be divided into two main types; Group I (non-micaceous, with low initial Sr and high initial Nd isotope signatures) and Group II (phlogopite-bearing, with high initial Sr and low initial Nd isotope signatures) (Smith, 1983); a transitional type has also been reported (Becker and Le Roex, 2006). However, two important problems hamper the determination of kimberlite genesis. Firstly, kimberlite, by definition, contains abundant, variably-sized crustal and mantle xenoliths (Mitchell, 1986), which makes it difficult to determine the initial isotopic composition of the kimberlitic magma. Secondly, kimberlite is highly susceptible to alteration during both emplacement and by later weathering, due to its high content of volatiles and mantle minerals. Therefore, the combined effects of contamination and post-emplacement alteration limit the reliability of whole-rock isotopic analyses (Mitchell, 1986), and make these largely ineffective when trying to determine mantle sources (Heaman, 1989). Fortunately, perovskite (CaTiO 3 ) can be used to circumvent many of the problems outlined above since it occurs mainly in the kimberlite groundmass and crystallized early in the magmatic history, along with ilmenite, rutile and magnesian chromite, and prior to crystallization of monticellite, phlogopite, serpentine and calcite (Mitchell, 1972, 1986; Thy et al., 1987). Therefore, perovskite has the potential to record the primary geochemical and isotopic signature of the magma, prior to any contamination and/or weathering. Furthermore, although perovskite is unstable in residual magmatic fluids (H 2 O and CO 2 ) and is commonly resorbed and/or mantled by thin rims of rutile, it is Chemical Geology 264 (2009) 24–42 ⁎ Corresponding author. Tel.: +86 10 82998217; fax: +86 10 62010846. E-mail address: wufuyuan@mail.igcas.ac.cn (F.-Y. Wu). 0009-2541/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.chemgeo.2009.02.011 Contents lists available at ScienceDirect Chemical Geology journal homepage: www.elsevier.com/locate/chemgeo