Ca-Sr isotope and chemical evidence for distinct sources of carbonatite and silicate mantle metasomatism
Introduction
Studies of chemical and mineralogical composition of mantle rocks suggested that they were affected by two types of metasomatism, ‘silicate’ and ‘carbonatite’, the latter characterized by strong enrichments in light rare earth elements (LREE) at low high-field-strength elements (HFSE) (Ionov et al., 1993, Rudnick et al., 1993, Yaxley et al., 1991). These results implied an important, widespread role for carbonatite melts as agents of mantle metasomatism, despite the rarity of these melts at the Earth’s surface (Hauri et al., 1993). However, other studies on mantle xenoliths and peridotite massifs inferred that enrichment patterns similar to those attributed to carbonate-rich melts could be produced by reactive percolation of volatile-bearing silicate melts. Numerical modeling of this process (Navon and Stolper, 1987) obtained a range of chemical patterns from a single parental liquid at different distances from its source as a result of chromatographic fractionation (due to different element compatibility), with ‘carbonatite’ signatures at the percolation front (e.g. Bodinier et al., 1990, Ionov et al., 2002a).
The existence of distinct, carbonatite and silicate, agents of mantle metasomatism can be ascertained using isotope compositions, which may differ in samples affected by media generated from different sources, but are similar in samples reworked by derivatives of a single initial agent. Here we present new Ca-Sr-Nd isotope and trace element data for a suite of mantle peridotites from the island of Spitsbergen that contain carbonate-bearing pockets of mantle origin replacing earlier minerals (Ionov et al., 1993, Ionov et al., 1996). Calcium and strontium are typical chemical components of carbonatites. The main objective of this study is to distinguish the Ca-Sr isotope inputs of the carbonates in the rocks using leaching with acetic acid that dissolves only carbonates. The data are used to evaluate directly the effects of carbonate-rich liquids on mantle metasomatism as well as to establish whether different types and stages of metasomatism were produced by distinct types and sources of media.
Calcium is a major lithophile element in the Earth’s mantle and crust (e.g. Hofmann, 1988). It has six naturally occurring stable isotopes (40Ca, 42Ca, 43Ca, 44Ca, 46Ca and 48Ca) with ~ 10% mass difference (Δm/m) between 40Ca and 44Ca (DePaolo, 2004). Calcium isotopic compositions are usually expressed as delta notation, e.g. δ44/40Ca relative to NIST SRM 915a: [(44Ca/40Ca)sample/(44Ca/40Ca)NIST-SRM-915a – 1] × 1000.
The application of Ca stable isotopic composition in geosciences has rapidly developed in the last two decades because Ca isotopes show important fractionation at low temperatures near the Earth’s surface, and are considered important indices of a wide range of biological and chemical processes in sedimentology, environmental geochemistry, weathering etc. (e.g. Blättler and Higgins, 2017, Depaolo, 2004, Fantle and Tipper, 2014, Farkaš et al., 2007, Skulan et al., 1997). Ca isotopes may also give new insights into global reservoirs and high-temperature processes in the Earth’s interior and extraterrestrial samples (see Antonelli and Simon (2020) for a review).
Recent studies revealed significant Ca isotope variations in mantle-derived mafic igneous rocks (Huang et al., 2011, Liu et al., 2017a, Zhu et al., 2020, Zhu et al., 2018b) and carbonatites (Amsellem et al., 2020, Banerjee and Chakrabarti, 2019, Banerjee et al., 2021, Sun et al., 2021). Ca isotope variations were reported also for the lithospheric mantle and ascribed to additions of recycled crustal materials (Chen et al., 2018, Huang et al., 2011, Liu et al., 2017a, Zhu et al., 2020, Zhu et al., 2018b) and/or equilibrium and disequilibrium isotope fractionation at high temperatures during: (a) melt extraction (Chen et al., 2019, Kang et al., 2017), (b) reaction with silicate and carbonate-rich melts or fluids (Ionov et al., 2019, Kang et al., 2017, Kang et al., 2019, Kang et al., 2016, Zhao et al., 2017). This opens up new prospects in mantle studies, but also creates new challenges, in particular how to distinguish Ca isotope variations in mantle rocks produced by intra-mantle fractionation from those caused by mixing with recycled crustal materials.
An important notion in mantle studies is that of the Bulk Silicate Earth (BSE), sometimes called “primitive mantle” (PM), defined as silicate portion of the Earth as it existed after separation of the core but before it was differentiated to crust and heterogeneous mantle (Hofmann, 1997). Present-day mantle reservoirs were formed by melt extraction from the BSE (Workman and Hart, 2005), metasomatism and recycling of surface materials. The BSE composition may be evaluated using fertile mantle peridotites that experienced no melt extraction or metasomatism (Palme and O'Neill, 2014, Mcdonough and Sun, 1995). So far, the only robust δ44/40Ca estimate for the BSE (0.94 ± 0.05‰, 2sd) has been given by Kang et al. (2017) based on 14 xenoliths of non-metasomatized fertile lherzolites with Al2O3 (3.7–4.7 wt.%) and CaO (3.2–4.0 wt.%) overlapping the PM estimates (~4.5 and 3.6%, respectively).
Kang et al. (2017) obtained a higher than BSE average δ44/40Ca (1.06 ± 0.06‰, 2sd) for strongly melt-depleted (1.3–1.7 wt.% Al2O3), non-metasomatised (LREE-depleted) peridotites and suggested that high degrees of melt extraction produce slightly higher (by ~ 0.12‰) δ44/40Ca in residues. A similar δ44/40Ca range (and δ44/40Ca evolution trend during melting) was found by Chen et al. (2019) in melt-depleted lherzolites from Alpine massifs. Kang et al. (2017) also reported usually lower δ44/40Ca (0.25–0.96‰) in metasomatized (LREE-enriched) peridotites, in particular for low-CaO (≤1 wt.%) rocks with Ca budget dominated by metasomatic inputs.
Strontium, like Ca, is an alkali earth element, but its BSE abundance is three orders of magnitude lower (20 vs. 25400 μg/g). Strontium is highly incompatible during peridotite melting because cation Sr2+ is much larger than Ca2+ such that Sr partition coefficients between clinopyroxene (major Sr host in common lherzolites) and mafic melts are well below unity (https://earthref.org/KDD/e:38/). Using the Rb-Sr radiogenic isotope system for dating mantle materials is problematic due to high mobility of these elements (e.g. Pearson et al., 2014), but the 87Sr/86Sr ratios may indicate the sources of mantle-derived rocks and magmas (Hofmann, 2003), as well as recycling of crustal materials (e.g. Guo et al., 2020, Xu, 2002). The 87Sr/86Sr ratios in mantle rocks are affected both by their age and the evolution of the Rb/Sr ratio that may be hard to disentangle to identify sources of metasomatism. By comparison, Ca isotope compositions do not evolve with time, but may reveal mantle sources, mixing or, alternatively, mass-dependent fractionation during mantle processes. Overall, the evidence from Sr and Ca isotopes may be complementary.
Previous studies found carbonates usually associated with (Na,Al)-rich silicate glass in some Spitsbergen xenoliths (Ionov et al., 1993, Ionov et al., 1996). The carbonates do not represent quenched carbonatite liquids but are crystal cumulates from carbonate-rich silicate melts reacting with host peridotites (Ionov and Harmer, 2002). Trace element patterns estimated for such liquids (Ionov, 1998) are consistent with many features of ‘carbonatite’ metasomatism (enrichments in LREE, Sr, Ba and negative HFSE anomalies). However, similar element patterns were found also in peridotites that contain no carbonates and attributed to reactive percolation and extreme fractionation of volatile-bearing silicate melts (Ionov et al., 2002a).
Overall, it remains uncertain whether the carbonates in mantle xenoliths from Spitsbergen and other worldwide localities (e.g. Dautria et al., 1992, Ionov et al., 2018, Laurora et al., 2001) originate from particular carbonate-rich liquids or are late-stage derivatives of evolving volatile-bearing silicate melts. In a more general sense, the ambiguity concerns the origin of chemical signatures usually seen as evidence for ‘carbonatite’ mantle metasomatism (e.g. Rivalenti et al., 2004, Tappe et al., 2017).
This uncertainty is addressed here using Ca-Sr isotope and chemical data on carbonate-bearing Spitsbergen xenoliths, notably by treating WR samples with acetic acid that dissolves only carbonates. Analyses of the leachates, residues and bulk-rocks discern metasomatic inputs of late-stage carbonates (extracted by leaching) from those of earlier events (hosted by residues). We show that trace element patterns and Ca-Sr isotope compositions of the leachates, hence carbonates, are distinct from those in the silicate minerals of the host peridotites. We infer that the carbonate pockets were formed from specific carbonate-rich liquids responsible for ‘carbonatite’ metasomatism. We further discuss the application of Ca-Sr isotope data to gain insights into the sources and mechanisms of metasomatism in the mantle.
Section snippets
Geological setting and samples
Spitsbergen is the largest island of the Svalbard archipelago, the northwestern (NW) edge of the Eurasian continent. It was connected to Greenland until the opening of the North Atlantic Ocean, and is located now ~ 200 km east of the mid-ocean ridge (e.g. Vagnes and Amundsen, 1993), between the North Atlantic and the Arctic Oceans at similar distances from Norway, Greenland and the North Pole.
Mantle xenoliths reported here were collected in alkali basaltic rocks at three Quaternary eruption
Sample treatment and analytical methods
The xenoliths analyzed in this study are listed in Table 1, along with a summary of information on available data from this and previous studies.
Results
Major and trace element compositions of leachates, residues and WR xenoliths in this study are listed in S-Table 1, Table 2 of Electronic Appendix 1 (EA1) along with relevant literature data (Ionov et al., 2002a, Ionov et al., 1993, Ionov et al., 1996). The Ca-Sr-Nd isotope compositions obtained in this study are given in Table 3, Table 4. Acid leaching was used here to separate the metasomatic inputs of late-stage carbonates (extracted by leaching) from those of earlier melt extraction and
Major element composition and origin of leached carbonate materials
Electron probe analyses of carbonates in our samples (Ionov et al., 1996) show that the dolomite contains 28–40 wt.% CaO, 13–24 wt.% MgO and 0.3–5.0 wt.% FeO while the Mg-calcite contains 44–53 wt.% CaO, 3–11 wt.% MgO and 0.3–2.7 wt.% FeO. The leachates from three xenoliths that host large mosaic carbonate aggregates (4-36-90, 4-90-9 and 21-6) contain 21–33 wt.% CaO, 12–20% MgO and 2.2–3.4% FeO, with higher CaO than for MgO in each case. The leachates from these xenoliths appear to be derived
Summary of conclusions
Leaching of peridotite xenoliths from Spitsbergen with acetic acid extracted carbonates formed in the latest metasomatic event and thus allowed to distinguish chemical and Ca-Sr-isotope imprints of their parental carbonate-rich melts from the combined effects of melt extraction and silicate metasomatism in leaching residues.
PM-normalized trace element patterns of the leached carbonates, unlike silicate residues, show distinctive indices (LREE-enrichments, HFSE-depletions) of carbonatite mantle
Declaration of Competing Interest
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 are grateful to Yajun An, Fang Liu, Jinting Kang, Jianghao Bai, Wei Wu and Zebin Luo for help with analyses and discussions. This study was supported by grants from the National Natural Science Foundation of China (No. 41773009 and 41873002) and the Programme National de Planétologie (PNP) of CNRS/INSU/CNES (grants to DAI in 2018-2019). DAI acknowledges Chinese Academy of Sciences President’s International Fellowship Initiative (PIFI) for Visiting Scientists (Grant No. 2017VCA0009). We
References (99)
- et al.
Metasomatism In Lithospheric Mantle Roots: Constraints From Whole-Rock And Mineral Chemical Composition Of Deformed Peridotite Xenoliths From Kimberlite Pipe Udachnaya
Lithos
(2013) - et al.
Calcium Isotopes In High-Temperature Terrestrial Processes
Chem. Geol.
(2020) - et al.
A Geochemical And Nd, Sr And Stable Ca Isotopic Study Of Carbonatites And Associated Silicate Rocks From The ~65 Ma Old Ambadongar Carbonatite Complex And The Phenai Mata Igneous Complex, Gujarat, India: Implications For Crustal Contamination, Carbonate Recycling, Hydrothermal Alteration And Source-Mantle Mineralogy
Lithos
(2019) - et al.
Temporal Evolution Of Δ44/40ca And 87sr/86sr Of Carbonatites: Implications For Crustal Recycling Through Time
Geochim. Cosmochim. Acta
(2021) - et al.
Testing Urey's Carbonate-Silicate Cycle Using The Calcium Isotopic Composition Of Sedimentary Carbonates
Earth Planet Sci Lett
(2017) - et al.
Pyroxene-Carbonate Reactions In The Upper Mantle
Earth Planet Sci Lett
(1983) - et al.
Compositional Characteristics Of The Morb Mantle And Bulk Silicate Earth Based On Spinel Peridotites From The Tariat Region
Mongolia. Geochim. Cosmochim. Acta
(2019) - et al.
Methods For The Microsampling And High-Precision Analysis Of Strontium And Rubidium Isotopes At Single Crystal Scale For Petrological And Geochronological Applications
Chem. Geol.
(2006) - et al.
Calcium Isotopic Compositions Of Oceanic Crust At Various Spreading Rates
Geochim. Cosmochim. Acta
(2020) - et al.
Calcium Isotope Fractionation During Magmatic Processes In The Upper Mantle
Geochim. Cosmochim. Acta
(2019)