Elsevier

Chemical Geology

Volume 524, 5 October 2019, Pages 136-157
Chemical Geology

Trace element analysis of high-Mg olivine by LA-ICP-MS – Characterization of natural olivine standards for matrix-matched calibration and application to mantle peridotites

https://doi.org/10.1016/j.chemgeo.2019.06.019Get rights and content

Highlights

  • LA-ICP-MS allows for quantification of trace elements in olivine down to ppb levels.

  • Elemental fractionation occurs at small laser spot sizes (< 100 μm) between olivine and typical silicate reference glasses.

  • Fractionation effects can be mitigated by using low laser fluence (≤ 4 J/cm2) and repetition rate (≤ 5 Hz).

  • Two in-house olivine standards (SC-GB and 355OL) are characterized.

  • These can be used for matrix-matched calibration of Li, Na, Al, P, Ca, Sc, Ti, V, Cr, Mn, Co, Ni, Cu, Zn in olivine.

Abstract

The trace element composition of olivine is becoming increasingly important in petrological studies due to the ubiquity of olivine in the Earth's upper mantle and in primitive magmatic rocks. The LA-ICP-MS method allows for the routine analysis of trace elements in olivine to sub-ppm levels, but a major drawback of this method is the lack of knowledge about possible downhole fractionation effects when non matrix-matched calibration is used. In this contribution, we show that matrix-matched (i.e., olivine-based) calibration is preferable for small laser spot sizes (<100 μm) due to significant laser-induced inter-element fractionation between olivine and commonly used silicate glass calibration materials, e.g., NIST SRM 612, GSD-1G and BHVO-2G. As a result, we present two Mg-rich natural olivine standards (355OL and SC-GB) that have been characterized by independent methods (EPMA, solution ICP-MS), and by LA-ICP-MS in four different laboratories. These natural olivines have been used 1) as primary standards for the matrix-matched calibration of olivine samples for most elements of interest (e.g., Li, Na, Al, P, Ca, Sc, Ti, V, Cr, Mn, Co, Ni, Cu, Zn), and 2) as secondary standards to assess the accuracy of results. Comparison of olivine- and silicate glass-calibrated results for natural peridotitic olivine reveals that matrix-matched calibration is essential when using small laser spot sizes (<100 μm) in order to mitigate downhole fractionation effects for certain elements, especially Na, P, Mn, Co, Ni and Zn. If matrix-matched calibration is not feasible, we recommend that spot sizes of ≥100 μm, laser fluence of ≤4.0 J/cm2, and total laser shot counts of ≤250 (e.g., 5 Hz repetition rate for 50 s) are used in order to minimize fractionation effects between olivine and silicate glass calibration materials. We demonstrate the applicability of matrix-matched calibration on olivine from a suite of different mantle peridotite xenoliths sampled by kimberlites and alkali basalts from on-craton and off-craton localities.

Introduction

Olivine represents the dominant mineral of the upper mantle (e.g., Ringwood, 1966), the most common silicate mineral inclusion in lithospheric diamonds (e.g., Meyer and Boyd, 1972; Stachel et al., 2005), and the most common sub-liquidus phase in primitive volcanic rocks (e.g., Sobolev et al., 2005; De Hoog et al., 2010; Foley et al., 2013). Previous studies on the minor and trace element chemistry of mantle olivine (e.g., De Hoog et al., 2010; Foley et al., 2013) showed that the incorporation of trace elements into olivine is limited by its simple crystal structure and major element composition which comprises >99 wt% of MgO, SiO2 and FeO. For olivine from mantle peridotites, the few additional elements that are incorporated can be divided into three groups following De Hoog et al. (2010): Group I elements (e.g., Ni, Mn, and Co) are the most compatible elements in olivine, being mostly divalent with ionic radii close to that of Mg; Group II elements (e.g., Cr, Al, V, Ca, and Na) are mainly controlled by equilibration temperature and pressure; Group III elements (e.g., Ti, Y, and Zr) show the largest concentration ranges in olivine and are strongly dependent on bulk rock contents and metasomatic overprinting.

The concentration of Al in olivine is increasingly of interest because it can be used as a geothermometer. Different calibrations exist for low-pressure magmatic olivine in equilibrium with spinel (Coogan et al., 2014; Wan et al., 2008) and for olivine in garnet peridotites (Bussweiler et al., 2017; De Hoog et al., 2010). Recent studies have applied Al-in-olivine thermometry to picrites (Trela et al., 2017) and komatiites (Waterton et al., 2017), mineral inclusions in diamonds (Korolev et al., 2018), and to the mantle olivine cargo in lamproites (e.g., Jaques and Foley, 2018; Shaikh et al., 2019) and kimberlites (Lawley et al., 2018). Moreover, the minor and trace element composition of magmatic olivine can be used as an indicator for different petrogenetic processes, including fingerprinting different magma sources (Howarth and Harris, 2017; Sobolev et al., 2005; Weiss et al., 2016; Zhang et al., 2016) or tracing its metasomatic history (e.g., Ammannati et al., 2016). Because of the wide petrological importance of olivine, it is critical to optimize analytical methods for probing its trace element composition with a high degree of precision and accuracy.

Laser ablation inductively coupled plasma mass spectrometry (LA-ICP-MS) allows the routine quantification of a wide range of elements in silicate minerals (e.g., Heinrich et al., 2003; Günther and Hattendorf, 2005). While some problems associated with LA-ICP-MS analysis of olivine, e.g., isobaric interferences, have been addressed in the literature (e.g., Foley et al., 2011; Bussweiler et al., 2015), a more extensive discussion of the analytical challenges and their effects on accuracy is required in order to fully understand the optimal analytical conditions.

A potentially major problem with analyzing trace elements in olivine by LA-ICP-MS is the lack of a matrix-matched olivine reference material. Such matrix-dependent elemental fractionation effects are well-documented, e.g., for reference glasses of different compositions (e.g., Czas et al., 2012; Hu et al., 2011). While matrix-matching is not always necessary when applying LA-ICP-MS to geological samples, especially when an internal standard (e.g., 29Si) is used (Jackson, 2008), this has not been tested for olivine, and the applicability of this approach depends on well-matched concentrations in the calibrant and the unknown. Routinely used calibration materials for olivine are silicate glasses, e.g., the NIST SRM 61X series (e.g., Kane, 1998). These certified reference material glasses have significantly higher SiO2, Na2O and CaO contents, but are much lower in MgO than olivine (Fig. 1). Alternatively, certified reference material glasses with basaltic compositions distributed by the United States Geological Survey (USGS), e.g., GSD-1G, BCR-2G, and BHVO-2G, can be used (e.g., Guillong et al., 2005). However, these glasses, in addition to being significantly darker in color than high-Mg olivine, have considerably higher Al2O3 contents which can produce background problems for analyzing Al in olivine due to memory effects (Fig. 1). A specific problem with the quantification of minor elements compatible in olivine (e.g., Ni and Mn) by LA-ICP-MS, is that they are present only at trace amounts in the commonly used calibration materials which can lead to large calibration errors.

The aim of this study is to optimize analytical protocols for the quantification of minor and trace element concentrations in olivine by LA-ICP-MS. We take a comparative approach by using different analytical methods, including electron probe microanalysis (EPMA), solution ICP-MS, and laser ablation (LA)-ICP-MS. The latter method was carried out in four different laboratories including the University of Alberta (UofA), the University of Melbourne (UofM), the Geological Survey of Canada (GSC), and the University of Münster (WWU). Two natural olivine grains, 355OL and SC-GB, are characterized for the use as in-house standards. This study shows that these standards are applicable as primary calibration materials, i.e., for the matrix-matched LA-ICP-MS analysis of olivine, and can also be used as secondary standards.

SC-GB is a >1 cm fragment from very coarse olivine grain from a spinel lherzolite from San Carlos, Arizona, USA (e.g., Jagoutz et al., 1979). Olivine grains from 355OL were extracted from a garnet harzburgite xenolith (XM1/355) entrained by the Bultfontein kimberlite, South Africa. Olivine in both samples is optically homogeneous and shows no zoning in major elements based on scanning electron microscope (SEM) imaging and EPMA.

The ‘unknown’ olivine grains from mantle peridotite xenoliths are derived from the on-craton Bultfontein kimberlite (South Africa) and off-craton alkali basalts from Bullenmerri and Mount Shadwell (Newer Volcanic Province, southeastern Australia; Table 1). All xenoliths show well-equilibrated granular textures and contain abundant fresh olivine together with variable amounts of orthopyroxene and other phases (e.g., clinopyroxene, spinel, garnet, phlogopite), with the exception of olivine megacryst sample BLFX-1 from Bultfontein, which is monomineralic.

Section snippets

Characterization of natural olivine standards by EPMA and solution ICP-MS

SC-GB was analyzed in multiple sessions by wavelength dispersive X-ray spectroscopy (WDS) using a JEOL 8900 electron probe microanalyzer (EPMA) at the University of Alberta. In addition to the major oxide components SiO2, MgO and FeO, the minor components NiO, MnO, CaO, Cr2O3, Al2O3, CoO were analyzed. An accelerating voltage of 20 kV was used with a beam size of 2 μm. Depending on element concentration, a beam current of 20 nA, 50 nA, or 100 nA was used on the natural and synthetic calibration

Limits of detection and suite of accessible elements

Fig. 2 shows all elements that could be analyzed in SC-GB and 355OL, sorted by concentration, along with their limits of detection (LODs) and limits of quantitation (LOQs) as measured by LA-ICP-MS at the UofA (130 μm laser spots, calibrated with NIST SRM 612). LODs were calculated using the formula by Pettke et al. (2012), which is based on Poisson statistics, and LOQs were estimated by multiplying LODs by a factor of three (Supplementary Table S3). Element concentrations down to ~0.004 ppm

Conclusions

Our investigation demonstrates that, other than for ultra-trace elements (e.g., REE, Y, Zr, Nb, Sr, Rb, Ba), matrix-matched calibration is preferable for the trace element analysis of olivine by LA-ICP-MS using small spot sizes (<100 μm) to minimize inaccuracies caused by calibration and fractionation effects. The fractionation effects depend on the employed laser spot size and become especially problematic at spot sizes of approximately <75 μm. They are caused by different ablation behavior of

Acknowledgments

Gerhard Brey is thanked for providing the San Carlos olivine (SC-GB). At the University of Alberta, Pedro Waterton is thanked for help with solution ICP-MS, Andrew Locock for assistance with EPMA, and Thomas Stachel for comments on an early version of the manuscript. At the University of Münster, Jasper Berndt-Gerdes and Beate Schmitte are thanked for assistance with LA-ICP-MS. At the University of Melbourne, Graham Hutchinson is thanked for performing EPMA and Emilie Lim for performing part of

References (55)

  • A. Giuliani et al.

    Constraints on kimberlite ascent mechanisms revealed by phlogopite compositions in kimberlites and mantle xenoliths

    Lithos

    (2016)
  • M. Guillong et al.

    Application of a particle separation device to reduce inductively coupled plasma-enhanced elemental fractionation in laser ablation-inductively coupled plasma-mass spectrometry

    Spectrochim. Acta - Part B At. Spectrosc.

    (2003)
  • D. Günther et al.

    Solid sample analysis using laser ablation inductively coupled plasma mass spectrometry

    TrAC - Trends Anal. Chem.

    (2005)
  • G.H. Howarth et al.

    Discriminating between pyroxenite and peridotite sources for continental flood basalts (CFB) in southern Africa using olivine chemistry

    Earth Planet. Sci. Lett.

    (2017)
  • C. Lawley et al.

    Trace metal and isotopic depth profiles through the Abitibi cratonic mantle

    Lithos

    (2018)
  • H.O.A. Meyer et al.

    Composition and origin of crystalline inclusions in natural diamonds

    Geochim. Cosmochim. Acta

    (1972)
  • T. Pettke et al.

    Recent developments in element concentration and isotope ratio analysis of individual fluid inclusions by laser ablation single and multiple collector ICP-MS

    Ore Geol. Rev.

    (2012)
  • A.M. Shaikh et al.

    Olivine trace element compositions in diamondiferous lamproites from India: proxies for magma origins and the nature of the lithospheric mantle beneath the Bastar and Dharwar cratons

    Lithos

    (2019)
  • P. Waterton et al.

    Age, origin, and thermal evolution of the ultra-fresh ~ 1.9 Ga Winnipegosis Komatiites, Manitoba, Canada

    Lithos

    (2017)
  • L.-Y. Zhang et al.

    Variation of olivine composition in the volcanic rocks in the Songliao basin, NE China: lithosphere control on the origin of the K-rich intraplate mafic lavas

    Lithos

    (2016)
  • J.T. Armstrong

    Quantitative analysis of silicate and oxide minerals: comparison of Monte Carlo, ZAF and phi-rho-z procedures

    Microbeam Anal

    (1988)
  • V.G. Batanova et al.

    New olivine reference material for in situ microanalysis

    Geostand. Geoanalytical Res. ggr

    (2019)
  • D. Chew et al.

    Ultrafast, >50 Hz LA-ICP-MS spot analysis applied to U–Pb dating of zircon and other U-bearing minerals

    Geostand. Geoanalytical Res.

    (2019)
  • J. Fietzke et al.

    Experimental evaluation of elemental behavior during LA-ICP-MS: influences of plasma conditions and limits of plasma robustness

    J. Anal. At. Spectrom.

    (2016)
  • S.F. Foley et al.

    Trace element variations in olivine phenocrysts from Ugandan potassic rocks as clues to the chemical characteristics of parental magmas

    Contrib. to Mineral. Petrol

    (2011)
  • B.J. Fryer et al.

    Design, operation and role of the laser-ablation microprobe coupled with an inductively-coupled plasma - mass-spectrometer (Lam-Icp-Ms) in the earth-sciences

    Can. Mineral.

    (1995)
  • M. Gaboardi et al.

    Elemental fractionation during LA-ICP-MS analysis of silicate glasses: implications for matrix-independent standardization

    J. Anal. At. Spectrom.

    (2009)
  • Cited by (39)

    • Olivine evidence for an ultramafic silicate precursor melt for the Jericho kimberlite (Slave Craton, Canada)

      2023, Lithos
      Citation Excerpt :

      Kimberlite olivine geochemistry therefore provides a window into the peridotite mantle sampled by the kimberlite magma and the magmatic processes that formed megacrysts and phenocrysts. Analytical advances have allowed new insights into the trace elemental geochemistry of olivine, which can provide information on melting and metasomatism (e.g. Bussweiler et al., 2015, 2017, 2019; De Hoog et al., 2010, 2019; Foley et al., 2013). Olivine being a liquidus phase in most mantle melts also presents the opportunity to probe the intensive variables involved in olivine crystallization, such as temperature and redox state (e.g. Beattie et al., 1991; Herzberg and O'Hara, 2002; Herzberg, 2011; Pilbeam et al., 2013; Moore, 2017).

    • Diamond formation beneath the Coromandel area, southwestern São Francisco Craton – The role of re-fertilization and subduction

      2022, Lithos
      Citation Excerpt :

      Al, Ca and Cr trace concentrations in olivine were measured using a CAMECA SX100 Electron Probe Microanalyzer (EPMA) with an accelerating voltage of 20 kV and a beam current of 300 nA. The detailed protocols of olivine trace element analysis using EPMA are given in Bussweiler et al. (2019). Our analyses of reference material SC-GB agree within uncertainty with the values stated in that study.

    • Olivine xenocrysts reveal carbonated mid-lithosphere in the northern Slave craton

      2022, Lithos
      Citation Excerpt :

      All data processing was completed with Iolite 3.0. All trace element data was checked for quality by comparing results from the EPMA and LA-ICP-MS for elements analyzed on both instruments and ensuring the secondary standard measurements (e.g., San Carlos olivine, PHN-1617B garnet, and USGS BCR-2G basaltic glass) were consistent with the accepted values (Bussweiler et al., 2019; Hardman et al., 2018). Analytical precision (e.g., average relative standard deviation for 48 elements = 5%) and accuracy (e.g., average difference between measured and target concentrations for 42 elements = 4%) for the secondary standard BCR-2G are also excellent, even for the unconventional suite of elements that are the focus of the present study.

    View all citing articles on Scopus
    View full text