Elsevier

Talanta

Volume 219, 1 November 2020, 121244
Talanta

A new chemical separation procedure for the determination of rare earth elements and yttrium abundances in carbonates by ICP-MS

https://doi.org/10.1016/j.talanta.2020.121244Get rights and content

Highlights

  • A novel cation exchange method was developed for the analysis of REE+Y in carbonates by ICP-MS.

  • Measurements of REEs were free of Ba interferences.

  • Low blanks and high enrichment-factor for REEs+Y are achieved, and allow the analysis of ultradepleted samples.

  • Precision better than 3 % RSD was achieved for most elements analysed.

Abstract

The determination of rare earth elements (REEs) and Y in carbonates can be complicated by low REE abundances and the presence of significant amounts of Ba resulting in problematic interferences when analysed by ICP-MS. We describe here a novel ion-exchange method using the DGA resin (TODGA), combined with addition of a Tm spike, which allows the separation of the REEs+Y as a whole prior to analysis using an Element XR ICP-MS. This method was validated with results obtained on three different reference carbonate materials (CAL-S, JLs-1 and BEAN, an in-house standard), yielding reproducibility levels better than 3% (RSD) in most cases. This new separation scheme is particularly well suited for carbonate samples having very low REE contents, but could be equally applied to various rock types and organic-rich sample matrices whenever quantitative Ba removal is required.

Introduction

The lanthanides and other rare earth elements (REEs) are amongst the most studied chemical elements in geochemistry, being of prime importance for addressing a wide range of issues in earth and planetary sciences, such as e.g. the condensation of the first solids in the Solar System, the formation of magmas, the reconstruction of present and past ocean circulation patterns [e.g., 1]. The usefulness of REE as unique tracers of geochemical processes comes first from their overall very consistent behavior in nature, which enables the modelling of their abundance in geochemical reservoirs. Most REEs are trivalent and, as a consequence, cannot be easily fractionated from each other during petrogenetic processes. However, two REEs, Ce and Eu, can also exist in different valence states (Ce4+ and Eu2+, respectively), which implies that they can be decoupled from the other REEs, resulting in quantifiable abundance anomalies, which, in turn, can be used to provide constraints on rock formation processes and other redox-sensitive bio-geochemical reactions. Another important field of application of REE geochemistry is based on the use of particular radioactive and radiogenic isotopes for geochronological purposes (i.e. the La/Ce, Sm/Nd, and Lu/Hf isotope systematics).

Over the recent years, an increasing interest has been placed in the application of REEs to both biogenic and chemical carbonates, as chemical tracers of the composition of seawater and other natural waters. For instance, the abundances of REEs in various carbonate material (e.g. stromatolites, shells, corals, stalagmites, methane-derived carbonates, hydrothermal carbonates) can provide unique constraints on the chemistry of oceans [e.g., [2], [3], [4], to characterize the fluids from which these mineral phases were precipitated [e.g., 5], to reconstruct past climatic changes [e.g., 6], or to track pollution linked to medical or industrial uses [[7], [8], [9]].

Since the 1990s, ICP-MS has become the method of choice for determining trace element concentrations in rocks, minerals and waters. A multitude of protocols have been described to date, and successfully applied to a wide range of materials [e.g., [10], [11], [12], [13], [14], [15], [16], [17]. The very high sensitivity of ICP-MS instruments makes it possible to determine very low trace element abundances in solution, with dilution factors (=solution weight/sample weight) up to several tens of thousands. Compared to previous state-of-the-art techniques, such as isotope dilution-thermo-ionization mass spectrometry (ID-TIMS) or instrumental neutron activation analyses (INAA), the use of ICP-MS allows high sample throughput at comparatively low analytical costs. The dissolution of biogenic and chemical carbonate samples is generally relatively straightforward, so that high quality trace element data can be obtained in the vast majority of case studies [e.g., 2, 4, 6, 7 among many others]. One potential difficulty in analyzing carbonates is that they are typically characterized by much lower REE abundances compared to other commonly studied rocks, such as basalts, granites and sediments. Using ICP-MS techniques, this difficulty can be generally successfully overcome by simply analyzing less diluted solutions. Another important issue when measuring REE concentrations in carbonates is that they can contain substantial amounts of Ba. During the course of ICP-MS analysis, a fraction of Ba atoms present in the plasma forms oxides, which can generate isobaric interferences with Eu isotopes (e.g. 135Ba16O+ and 134BaOH+ with 151Eu, 137Ba16O and 136BaOH+with 153Eu [e.g., 14, [18], [19]). Uncorrected isobaric interferences cannot be neglected because they typically result in the occurrence of non-natural positive Eu anomalies in studied samples. In most cases, the presence of isobaric interferences on Eu can be successfully corrected by monitoring the oxide formation rate during an analytical ICP-MS session, using a mono-elemental solution of Ba. Many rocks, such as basalts, granites, terrigenous sediments or peridotites commonly display Ba/Eu ratios < 1000. The contribution of Ba oxides to the 151 or 153 Eu masses is not dominant here, and the correction is generally very satisfactory (Fig. 1). However, many marine or hydrothermal carbonates have much higher Ba/Eu ratios, frequently >10000. For these samples, the interference contribution to measured 151Eu and 153Eu signals can become dominant. The corrections that can be made using estimates of oxide formation rates commonly yield Eu abundances with poor accuracy and/or associated with a relatively high uncertainty. The use of collision/reaction cells and/or of ICP-MS operated in high-resolution modes can provide efficient means for eliminating the formation of Ba oxides and allowing quantitative separation of Eu+ and BaO+ peaks, respectively. However, these options significantly reduce the signal intensities, and hence are generally not adapted to low-level samples such as carbonates. One alternative to resolve the specific problem raised by the potential occurrence of isobaric interferences upon carbonate analysis is to quantitatively separate REEs from Ba prior to ICP-MS measurements. Another advantage of separating REEs is that it allows one to analyse less diluted solutions during ICP-MS measurements, hence improving the quality of data acquisition.

Several ion-exchange and diverse Fe–Mg hydroxide coprecipitation techniques have been developed for application to samples having low REE abundances [e.g., 10, 12, [20], [21], [22], [23], [24], [25], [26], [27], [28]. Over the past twenty-five years, we have used one of these previously developed chromatographic methods, based on the use cation-exchange resins (e.g. 50WX12 or 50WX8, as first described by Strelow [29]), combined with the addition of a Tm spike in order to overcome the potential problem of any sample loss during handling and throughout the ion chromatography process [30]. It was initially designed for silicate rocks, and over time has proven particularly useful for analysing various minerals and rocks with very low REE abundances [e.g., [31], [32], [33]. The utility of this technique for measuring REE abundances in carbonates has been already demonstrated [9,34]. However, this procedure also had some drawbacks, which required further improvements. First, the columns are small (1.6 ml resin), which restricts their use to small sample size only (<30 mg), hence being problematic when processing REE-depleted materials such as biogenic carbonates, for which larger amounts of material would be ideally required. Second, the procedure based on cation-exchange chromatography did not yield quantitative Ba removal, resulting in eluted solutions that still contain non-negligible amounts of this element. Finally, this procedure was also accompanied with poor recovery of Y, hence leading to non-reproducible determination of Y abundances.

Novel ion exchange resins have been developed and commercialized in recent years, including the DGA, normal resin. It is an extraction chromatographic resin based on N,N,N’,N’-tetra-n-octyldiglycolamide extractant, commonly referred to as either DN Resin or TODGA in the literature. It has been previously used for the preconcentration of actinides and lanthanides from various samples for radioactive waste management [35,36]. A complete set of partition coefficients has been published for this resin [37], and its capability for the preconcentration of REEs for difficult geological samples has already been demonstrated [38,39]. In this study, we used the remarkable properties of this resin to separate the REEs from carbonates. We describe here a new procedure aimed at quantitatively separating REE from other elements in carbonate samples, and illustrate its utility using two international (JLs-1and CAL-S) and one in-house standard from hereafter referred to as BEAN for Brest carbonate and available on request.

Section snippets

Sample preparation and REE separation

All sample preparations were conducted in a Class 1000 (ISO 6) clean laboratory. Deionized water purified with a Milli-Q system (Millipore®) at 18.2 MΩ (from hereafter referred to as ultrapure water) was used for material cleaning and acid dilutions. Nitric and hydrochloric acids were purified using sub-boiling systems.

Three carbonate reference materials displaying low REE abundances were selected for this study:

-JLs-1, a marine limestone prepared by the Geological Survey of Japan, which is one

Results and discussion

The results for the three carbonate samples investigated in this study are given in Table 2, while corresponding REE patterns normalised to Post Archaean Australian Shale (PAAS [38]) are shown in Fig. 3. Concentrations obtained from samples with or without separation are analytically indistinguishable, and in excellent agreement with literature values. This confirms that the recovery of rare earths and yttrium with our procedure is complete. For the case of JLs-1, i.e. the studied carbonate

Conclusions

We report on a novel analytical protocol for rapid and efficient separation of REEs and Y from carbonate samples, which results in quantitative removal of Ba and other major alkaline earth matrix elements. This procedure was validated using a suite of three carbonate reference materials (CAL-S, JLs-1; BEAN) analysed by Element XR ICP-MS, providing precise and accurate REE data even for depleted carbonate material characterized by high Ba abundances, for which efficient Ba removal prior to

Author contribution statement

Jean-Alix Barrat: Conceptualization, Supervision, Methodology, Investigation, Validation, Writing - original draft, Germain Bayon: Conceptualization, Supervision, Methodology, Investigation, Validation, Writing - original draft. Xudong Wang: Investigation, Validation, Samuel Le Goff: Investigation, Validation, Marie-Laure Rouget: Investigation, Validation, Bleuenn Gueguen: Investigation, Validation. Douraied Ben Salem: Conceptualization.

Declaration of competing interests

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

We thank Jean-Michel Kauffmann for the editorial handling, and the two anonymous reviewers for constructive comments. We thank Stefan Lalonde for providing the Acros Organics® carbonate. This work was supported by the “Laboratoire d’Excellence” LabexMER (ANR-10-LABX-19) and funded by grants from the French Government under the program “Investissements d’Avenir”.

References (43)

  • M. Bau et al.

    Anthropogenic origin of positive gadolinium anomalies in river waters

    Earth Planet Sci. Lett.

    (1996)
  • M. Bau et al.

    Rare earth elements in mussel shells of the Mytilidae family as tracers for hidden and fossil high-temperature hydrothermal systems

    Earth Planet Sci. Lett.

    (2010)
  • N. Freslon et al.

    Determination of rare earth elements and other trace elements (Y, Mn, Co, Cr) in seawater using Tm addition and Mg(OH)2 co-precipitation

    Talanta

    (2011)
  • F.W.E. Strelow

    Separation of trivalent rare earths plus Sc(III) from Al, Ga, In, Tl, Fe, Ti, U and other elements by cation-exchange chromatography

    Anal. Chim. Acta

    (1966)
  • J.A. Barrat et al.

    Evidence from Tm anomalies for non-CI refractory lithophile element proportions in terrestrial planets and achondrites

    Geochem. Cosmochim. Acta

    (2016)
  • J.A. Barrat et al.

    The origin of aubrites : evidence from lithophile trace element abundances and oxygen isotope compositions

    Geochem. Cosmochim. Acta

    (2016)
  • J.A. Barrat et al.

    Melting of a C-rich asteroid: lithophile trace elements in ureilites

    Geochem. Cosmochim. Acta

    (2016)
  • J.A. Barrat et al.

    Strontium isotopes and rare earth element geochemistry of hydrothermal carbonate deposits from the Tanganyika Lake, East Africa

    Geochem. Cosmochim. Acta

    (2000)
  • H. Hoshi et al.

    Group separation of trivalent minor actinides and lanthanides by TODGA extraction chromatography for radioactive

    Waste Manag.

    (2004)
  • A. Pourmand et al.

    Distribution coefficients of 60 elements on TODGA resin: application to Ca, Lu, Hf, U and Th isotope geochemistry

    Talanta

    (2010)
  • A. Pourmand et al.

    A novel extraction chromatography and MC-ICP-MS technique for rapid analysis of REE, Sc and Y: revising CI-chondrite and Post-Archean Australian Shale (PAAS) abundances

    Chem. Geol.

    (2012)
  • Cited by (19)

    • Mixed carbonate-siliceous hydrothermal chimneys ahead of the Asal propagating rift (SE Afar Rift, Republic of Djibouti)

      2023, Journal of African Earth Sciences
      Citation Excerpt :

      In-run uncertainties on measurements were generally better than 4% for all elements, and invariably <10%. The analysis of in-house carbonate standard (BE-AN) yielded REE abundances in general agreement (<13%) with recommended values of Barrat et al. (2020), except for Gd (>16%). REE abundances are reported using the Post-Archean Australian shale (PAAS) reference values (Pourmand et al., 2012).

    • Rare earth elements and yttrium in suspension-feeding bivalves (dog cockle, Glycymeris glycymeris L.): Accumulation, vital effects and pollution

      2022, Geochimica et Cosmochimica Acta
      Citation Excerpt :

      The reader is referred to previous papers where our routine sequence for carbonates (Barrat et al., 2020), calibration, isobaric interference corrections and calculations of concentrations with the Tm spike have been extensively described, and results for many international standards were reported (e.g., Barrat et al., 1996, 2012,2016,2020; Charles et al., 2021). We use for the normalisation of the concentrations, the Post Archean Australian Shale (PAAS) average obtained by Pourmand et al. (2012), adjusted to standard results obtained in our laboratory (Barrat et al., 2020). The La, Ce, and Gd anomalies are calculated using the La/La*, Ce/Ce*, Gd/Gd* ratios, where X* is the extrapolated concentrations for a smooth PAAS-normalised REE pattern and Xsn is the concentration of element X normalised to PAAS: Lasn* = Prsn3/Ndsn2, Cesn* = Prsn2/Ndsn, Gdsn* = Tbsn2/Dysn.

    View all citing articles on Scopus
    View full text