Contents lists available at ScienceDirect Chemical Geology journal homepage: www.elsevier.com/locate/chemgeo Role of fractionation correction in accurate determination of 142 Nd/ 144 Nd by TIMS: A case study of 1.48 Ga alkaline rocks from Khariar, India Ikshu Gautam a,b,⁎ , Jyotiranjan S. Ray a , Rajneesh Bhutani c , S. Balakrishnan c , J.K. Dash c a Physical Research Laboratory, Ahmedabad 380 009, India b Department of Geology, The Maharaja Sayaji Rao University of Baroda, Vadodara 390002, India c Department of Earth Sciences, Pondicherry University, Puducherry 605 014, India ARTICLE INFO Keywords: 142Nd/144Nd analyses TIMS 1.48 Ga Khariar rocks ABSTRACT The short-lived isotopic systematics of 146 Sm- 142 Nd is a tracer of early silicate Earth differentiation events. Evidence for these events comes from anomalous 142 Nd/ 144 Nd, defined in terms of μ 142 Nd (μ 142 Nd = [{( 142 Nd/ 144 Nd) sample /( 142 Nd/ 144 Nd) standard } - 1] × 10 6 ) with respect to a terrestrial standard representing the modern accessible mantle. This requires measurement of accurate and highly precise 142 Nd/ 144 Nd, which is carried out by Thermal Ionisation Mass Spectrometry (TIMS). Since multiple factors affect the accuracy of the final results, we carried out a detailed investigation on the effect of various data acquisition, fractionation correction and normalization methods on the accuracy of 142 Nd/ 144 Nd determinations. Based on the analyses of Ames Nd standard using various combinations of the most commonly employed methods we observed that for a multi-dynamic mode of data acquisition, the power-normalised exponential law is the most appropriate method for mass fractionation correction. The time delays between successive sequences in a multi-dynamic mode had little effect on the final value of 142 Nd/ 144 Nd. The different standards have different 142 Nd/ 144 Nd ratios and therefore, their uses yield different μ 142 Nd values for the same sample. We extended this information to understand the two contradicting results from 1.48 Ga alkaline rocks from Khariar, India, carried out on the same sample aliquots (Upadhyay et al. 2009; Roth et al. 2014b). A confirmation of 142 Nd anomalies in such younger rocks is important because it could establish the longevity of early silicate differentiation signatures beyond Archean. Our ex- periment on freshly collected samples from the same outcrops, using identical analytical procedures, could not reproduce the results of Upadhyay et al. (2009). We did, however, observe slightly negative μ 142 Nd values with respect to Ames Nd, which became normal with respect to JNdi-1. 1. Introduction The near absence of rock record from the first 500 million years of the Earth's history makes it difficult to understand the earliest differ- entiation processes that caused the separation of its various reservoirs. Short lived radionuclides and their decay products have been useful in such studies as they provide critical information about these processes from meteorites and ancient magmatic systems. 146 Sm- 142 Nd (t 1/ 2 = 103 Ma; (Marks et al., 2014) or 68 Ma; Kinoshita et al., 2012) is one such systematics which has been widely utilized to decode the early silicate Earth differentiation. Anomalous abundances of 142 Nd with respect to terrestrial standards, expressed as μ 142 Nd (μ 142 Nd = [{( 142 Nd/ 144 Nd) sample /( 142 Nd/ 144 Nd) standard } - 1] × 10 6 ), provide clues to fractionation of Sm/Nd during the differentiation events that took place during the first 500 million years of the Earth's formation, when 146 Sm was extant. Considering the highly dynamic nature of the Earth's earliest mantle and the time elapsed since its formation, it is extremely difficult to encounter μ 142 Nd anomalies in rocks younger than Archean. Also, because of the small magnitude of these anomalies and isobaric interferences of Sm and Ce on various Nd isotopes in- cluding 142 Ce (11.4% abundance) on 142 Nd (27.2%), their detection through mass spectrometry is analytically challenging. All of the ac- cepted discoveries of μ 142 Nd anomalies come from the Hadean and Archean rocks. The positive anomalies possibly represent the earliest Large Ion Lithophile Element (LILE) depleted source (Bennett et al., 2007; Boyet and Carlson, 2006; Boyet et al., 2003; Caro et al., 2006, 2003; Rizo et al., 2011). Only four examples of negative μ 142 Nd anomalies are known today (O'Neil et al., 2008; Rizo et al., 2012; Roth et al., 2014a; Roth et al., 2013; Upadhyay et al., 2009),which are be- lieved to be vestiges of a Hadean LILE enriched reservoir, possibly re- presenting the earliest crust (Rizo et al., 2012; Roth et al., 2014a) and/ or non-convecting lithospheric mantle (Upadhyay et al., 2009). http://dx.doi.org/10.1016/j.chemgeo.2017.06.036 Received 11 April 2017; Received in revised form 18 June 2017; Accepted 26 June 2017 ⁎ Corresponding author. E-mail address: ikshu@prl.res.in (I. Gautam). Chemical Geology 466 (2017) 479–490 Available online 01 July 2017 0009-2541/ © 2017 Elsevier B.V. All rights reserved. MARK