Figures

top

Introduction

The chlorine isotope ratio (δ³⁷Cl value) of Earth’s mantle has important implications for both the extent of volatile exchange between Earth’s surface and mantle and the source of Earth’s volatiles during early accretion. The δ³⁷Cl values of Earth’s major surface reservoirs (e.g., hydrosphere, evaporites, sediments) have ranges that are largely agreed upon (e.g., Barnes and Sharp, 2017

Barnes, J.D., Sharp, Z.D. (2017) Chlorine Isotope Geochemistry. Reviews in Mineralogy and Geochemistry 82, 345–378. https://doi.org/10.1515/9783110545630-010

), but the δ³⁷Cl value of Earth’s mantle is more contentious. Previous studies have examined the δ³⁷Cl values of mid-ocean ridge basalt (MORB) glasses because Cl isotopes do not significantly fractionate at high temperatures (e.g., Balan et al., 2019

Balan, E., Créon, L., Sanloup, C., Aléon, J., Blanchard, M., Paulatto, L., Bureau, H. (2019) First-principles modeling of chlorine isotope fractionation between chloride bearing molecules and minerals. Chemical Geology 525, 424–434. https://doi.org/10.1016/j.chemgeo.2019.07.032

) during partial melting or fractional crystallisation, so MORB samples should preserve the δ³⁷Cl values of their mantle source if there is minimal shallow assimilation. Studies of MORB glasses and other mantle-derived samples have come to contrasting conclusions about the δ³⁷Cl values of the depleted MORB-source mantle (DMM), with recent estimates ranging from −3 ‰ to +0.9 ‰ (Sharp et al., 2007

Sharp, Z.D., Barnes, J.D., Brearley, A.J., Chaussidon, M., Fischer, T.P., Kamenetsky, V.S. (2007) Chlorine isotope homogeneity of the mantle, crust, and carbonaceous chondrites. Nature 446, 1062–1065. https://doi.org/10.1038/nature05748

, 2013

Sharp, Z.D., Mercer, J.A., Jones, R.H., Brearley, A.J., Selverstone, J., Bekker, A., Stachel, T. (2013) The chlorine isotope composition of chondrites and Earth. Geochimica et Cosmochimica Acta 107, 189–204. https://doi.org/10.1016/j.gca.2013.01.003

; Bonifacie et al., 2008

Bonifacie, M., Jendrzejewski, N., Agrinier, P., Humler, E., Coleman, M., Javoy, M. (2008) The chlorine isotope composition of Earth’s mantle. Science 319, 1518–1520. https://doi.org/10.1126/science.1150988

; Layne et al., 2009

Layne, G.D., Kent, A.J.R., Bach, W. (2009) δ³⁷Cl systematics of a backarc spreading system: the Lau Basin. Geology 37, 427–430. https://doi.org/10.1130/G25520A.1

; Pinti et al., 2020

Pinti, D.L., Shouakar-Stash, O., Castro, M.C., Lopez-Hernández, A., Hall, C.M., Rocher, O., Shibata, T., Ramírez-Montes, M. (2020) The bromine and chlorine isotopic composition of the mantle as revealed by deep geothermal fluids. Geochimica et Cosmochimica Acta 276, 14–30. https://doi.org/10.1016/j.gca.2020.02.028

). These DMM estimates span a large portion of the natural δ³⁷Cl range of predominant Cl hosts on Earth’s surface (−2 ‰ to +2 ‰; Barnes and Sharp, 2017

Barnes, J.D., Sharp, Z.D. (2017) Chlorine Isotope Geochemistry. Reviews in Mineralogy and Geochemistry 82, 345–378. https://doi.org/10.1515/9783110545630-010

). Each estimate has differing implications for the cosmochemical source of Earth’s volatile elements and the extent of volatile exchange between Earth’s mantle and surface. It is therefore imperative to better constrain the δ³⁷Cl composition of Earth’s mantle. To do that, we analysed the δ³⁷Cl values of MORB glasses spanning a range of trace element and radiogenic isotope compositions. We combine our new data with previously published MORB glass δ³⁷Cl values to examine the effects of shallow assimilation and mantle heterogeneity and more tightly constrain the δ³⁷Cl value of Earth’s mantle.

top

Samples and Geologic Context

Samples analysed in this study are MORB glasses from the Mid-Atlantic Ridge (MAR); the East Pacific Rise (EPR); the Galapagos Spreading Centre (GSC); the Southwest Indian Ridge (SWIR); and the Central Indian Ridge (CIR) (Fig. S-3a). These localities range from ultrafast to ultraslow spreading centres. Samples were dredged from depths of 1305–5150 metres below sea level. Selected samples have MgO contents of 5.6–10.4 wt. % and have trace element (e.g., K₂O/TiO₂, Nb/Zr, Th/La) and radiogenic isotope (e.g., ¹⁴³Nd/¹⁴⁴Nd) compositions (Fig. S-3b) that include D-MORB, N-MORB, and E-MORB. Samples also span a wide range of Cl/K (0.010–0.819; Fig. S-3b). Details of sample locations and compositions are provided in Table S-3. Measurements of δ³⁷Cl values of pristine MORB glasses were conducted via isotope ratio mass spectrometry at the University of Texas at Austin and via secondary ionisation mass spectrometry at the University of Lausanne, Switzerland. For details of preparation procedures and analytical methods see the Supplementary Information S-1 and S-2.

top

Results

Chlorine isotope ratios of MORB glasses measured in this study range from −1.5 ‰ to +1.0 ‰, slightly larger than the range of previous studies (−1.9 ‰ to +0.4 ‰; Sharp et al., 2007

Sharp, Z.D., Barnes, J.D., Brearley, A.J., Chaussidon, M., Fischer, T.P., Kamenetsky, V.S. (2007) Chlorine isotope homogeneity of the mantle, crust, and carbonaceous chondrites. Nature 446, 1062–1065. https://doi.org/10.1038/nature05748

; Bonifacie et al., 2008

Bonifacie, M., Jendrzejewski, N., Agrinier, P., Humler, E., Coleman, M., Javoy, M. (2008) The chlorine isotope composition of Earth’s mantle. Science 319, 1518–1520. https://doi.org/10.1126/science.1150988

). After integrating MORB δ³⁷Cl data from previous studies filtered for potential analytical artifacts (see Supplementary Information S-3), δ³⁷Cl values do not correlate with indices of seawater/brine assimilation such as Cl/K when all data are considered together (Fig. 1). The most depleted sample from this study (RC2806 1D-1; ¹⁴³Nd/¹⁴⁴Nd = 0.513234, ⁸⁷Sr/⁸⁶Sr = 0.702125; Le Voyer et al., 2015

Le Voyer, M., Cottrell, E., Kelley, K.A., Brounce, M., Hauri, E.H. (2015) The effect of primary versus secondary processes on the volatile content of MORB glasses: An example from the equatorial Mid-Atlantic Ridge (5°N–3°S). Journal of Geophysical Research Solid Earth 120, 125–144. https://doi.org/10.1002/2014JB011160

) has a δ³⁷Cl value of −0.5 ‰. For all localities, samples with higher K₂O/TiO₂ and lower ¹⁴³Nd/¹⁴⁴Nd extend to both positive and negative δ³⁷Cl values (Fig. 2). See Table S-3 for all collected and compiled δ³⁷Cl values and other geochemical data used in this study.

Full size image

Full size image

top

Discussion

Effects of seawater/brine assimilation. Shallow assimilation of seawater or brine has been shown to significantly increase the Cl content of MORB glasses (e.g., Michael and Schilling, 1989

Michael, P.J., Schilling, J.-G. (1989) Chlorine in mid-ocean ridge magmas: Evidence for assimilation of seawater-influenced components. Geochimica et Cosmochimica Acta 53, 3131–3143. https://doi.org/10.1016/0016-7037(89)90094-X

). Previous studies have used Cl/K as a filter for seawater and high salinity brine assimilation (e.g., Michael and Cornell, 1998

Michael, P.J., Cornell, W.C. (1998) Influence of spreading rate and magma supply on crystallization and assimilation beneath mid-ocean ridges: Evidence from chlorine and major element chemistry of mid-ocean ridge basalts. Journal of Geophysical Research 103, 18325–18356. https://doi.org/10.1029/98JB00791

; Le Roux et al., 2006

Le Roux, P.J., Shirey, S.B., Hauri, E.H., Perfit, M.R., Bender, J.F. (2006) The effects of variable sources, processes and contaminants on the composition of northern EPR MORB (8–10°N and 12–14°N): Evidence from volatiles (H2O, CO2, S) and halogens (F, Cl). Earth and Planetary Science Letters 251, 209–231. https://doi.org/10.1016/j.epsl.2006.09.012

). Although Michael and Cornell (1998)

Michael, P.J., Cornell, W.C. (1998) Influence of spreading rate and magma supply on crystallization and assimilation beneath mid-ocean ridges: Evidence from chlorine and major element chemistry of mid-ocean ridge basalts. Journal of Geophysical Research 103, 18325–18356. https://doi.org/10.1029/98JB00791

proposed that any sample with Cl/K > 0.08 has been affected by brine assimilation, subsequent studies have shown higher Cl/K values up to 0.09–0.14 can be primary mantle signatures (Shimizu et al., 2016

Shimizu, K., Saal, A.E., Myers, C.E., Nagle, A.N., Hauri, E.H., Forsyth, D.W., Kamenetsky, V.S., Niu, Y. (2016) Two-component mantle melting-mixing model for the generation of mid-ocean ridge basalts: Implications for the volatile content of the Pacific upper mantle. Geochimica et Cosmochimica Acta 176, 44–80. https://doi.org/10.1016/j.gca.2015.10.033

; Kendrick et al., 2017

Kendrick, M.A., Hémond, C., Kamenetsky, V.S., Danyushevsky, L., Devey, C.W., Rodermann, T., Jackson, M.G., Perfit, M.R. (2017) Seawater cycled throughout Earth’s mantle in partially serpentinized lithosphere. Nature Geoscience 10, 222–227. https://doi.org/10.1038/ngeo2902

). Samples with Cl/K in this range may contain primary or assimilated Cl. In this study, we filter samples based on Cl/K thresholds from Shimizu et al. (2016)

Shimizu, K., Saal, A.E., Myers, C.E., Nagle, A.N., Hauri, E.H., Forsyth, D.W., Kamenetsky, V.S., Niu, Y. (2016) Two-component mantle melting-mixing model for the generation of mid-ocean ridge basalts: Implications for the volatile content of the Pacific upper mantle. Geochimica et Cosmochimica Acta 176, 44–80. https://doi.org/10.1016/j.gca.2015.10.033

(see Supplementary Information S-3). Filtered samples may contain assimilated Cl and so are not considered in the discussion of mantle δ³⁷Cl values.

Bonifacie et al. (2008)

Bonifacie, M., Jendrzejewski, N., Agrinier, P., Humler, E., Coleman, M., Javoy, M. (2008) The chlorine isotope composition of Earth’s mantle. Science 319, 1518–1520. https://doi.org/10.1126/science.1150988

interpreted correlations between MORB glass δ³⁷Cl values and both Cl concentration and Cl/K as the result of seawater/brine assimilation driving MORB towards seawater-like δ³⁷Cl values (∼0 ‰) at higher Cl/K. However, when data from this study and filtered literature (see Supplementary Information S-3) are considered together, there is no clear correlation between δ³⁷Cl and Cl/K (Fig. 1a). When samples are considered by locality, samples with unambiguous signs of brine assimilation (Cl/K > 0.2) tend to have lower δ³⁷Cl values than other samples from the same locality, extending to δ³⁷Cl values lower than −0.6 ‰ to −1.5 ‰ (Fig. 1b). This indicates that shallow assimilation does not drive MORB glasses to seawater-like δ³⁷Cl values of 0.0 ‰, as suggested by Bonifacie et al. (2008)

Bonifacie, M., Jendrzejewski, N., Agrinier, P., Humler, E., Coleman, M., Javoy, M. (2008) The chlorine isotope composition of Earth’s mantle. Science 319, 1518–1520. https://doi.org/10.1126/science.1150988

, but rather towards δ³⁷Cl values lower than seawater. An exception to this is the GSC suite because the samples with the highest Cl/K do extend towards seawater-like δ³⁷Cl values (Fig. 1b). However, all GSC samples measured in this study and by Bonifacie et al. (2008)

Bonifacie, M., Jendrzejewski, N., Agrinier, P., Humler, E., Coleman, M., Javoy, M. (2008) The chlorine isotope composition of Earth’s mantle. Science 319, 1518–1520. https://doi.org/10.1126/science.1150988

have Cl/K of 0.70–1.04, and so have likely all assimilated significant Cl. The generally negative δ³⁷Cl value of assimilated components indicates that assimilation of seawater-derived Cl from a high salinity brine or hydrothermal fluid will not necessarily result in seawater-like δ³⁷Cl values in MORB, as has been previously suggested.

Variations in MORB δ³⁷Cl values related to mantle heterogeneity. Previous studies of MORB δ³⁷Cl values assumed that the DMM source of MORB was largely a single reservoir (Sharp et al., 2007

Sharp, Z.D., Barnes, J.D., Brearley, A.J., Chaussidon, M., Fischer, T.P., Kamenetsky, V.S. (2007) Chlorine isotope homogeneity of the mantle, crust, and carbonaceous chondrites. Nature 446, 1062–1065. https://doi.org/10.1038/nature05748

) or that it was divided into an E-MORB and N-MORB source (Bonifacie et al., 2008

Bonifacie, M., Jendrzejewski, N., Agrinier, P., Humler, E., Coleman, M., Javoy, M. (2008) The chlorine isotope composition of Earth’s mantle. Science 319, 1518–1520. https://doi.org/10.1126/science.1150988

). However, the mantle is not composed of just one or two reservoirs, but a large range of chemically heterogenous materials including at least three end members (EM-I, EM-II, HIMU) in addition to DMM (e.g., Stracke, 2012

Stracke, A. (2012) Earth’s heterogeneous mantle: A product of convection-driven interaction between crust and mantle. Chemical Geology 330-331, 274–299. https://doi.org/10.1016/j.chemgeo.2012.08.007

). These end members likely result from incorporation of subducted oceanic crust and sediment into the mantle source of basalts, resulting in higher K₂O/TiO₂ (e.g., Jackson and Dasgupta, 2008

Jackson, M.G., Dasgupta, R. (2008) Compositions of HIMU, EM1, and EM2 from global trends between radiogenic isotopes and major elements in ocean island basalts. Earth and Planetary Sciences Letters 276, 175–186. https://doi.org/10.1016/j.epsl.2008.09.023

) and lower ¹⁴³Nd/¹⁴⁴Nd (e.g., White and Hofmann, 1982

White, W.M., Hofmann, A.W. (1982) Mantle heterogeneity and isotopes in oceanic basalts. Nature 295, 363–364. https://doi.org/10.1038/295363a0

). Although these components are more abundant in ocean island basalts (OIBs), they can also be introduced at mid-ocean ridges either from interaction with a mantle plume (e.g., Le Voyer et al., 2015

Le Voyer, M., Cottrell, E., Kelley, K.A., Brounce, M., Hauri, E.H. (2015) The effect of primary versus secondary processes on the volatile content of MORB glasses: An example from the equatorial Mid-Atlantic Ridge (5°N–3°S). Journal of Geophysical Research Solid Earth 120, 125–144. https://doi.org/10.1002/2014JB011160

) or from distributed components in the upper mantle (e.g., Castillo et al., 1998

Castillo, P.R., Natland, J.H., Niu, Y., Lonsdale, P.F. (1998) Sr, Nd and Pb isotopic variation along the Pacific–Antarctic risecrest, 53–578S: Implications for the composition and dynamics of the South Pacific upper mantle. Earth and Planetary Science Letters 154, 109–125. https://doi.org/10.1016/S0012-821X(97)00172-6

). Some of the disagreement on the DMM δ³⁷Cl values may arise from incorporation of these subduction-derived enriched components in some of the previously analysed samples. The possibility of MORB Cl isotope heterogeneity was raised by both Sharp et al. (2007)

Sharp, Z.D., Barnes, J.D., Brearley, A.J., Chaussidon, M., Fischer, T.P., Kamenetsky, V.S. (2007) Chlorine isotope homogeneity of the mantle, crust, and carbonaceous chondrites. Nature 446, 1062–1065. https://doi.org/10.1038/nature05748

and Bonifacie et al. (2008)

Bonifacie, M., Jendrzejewski, N., Agrinier, P., Humler, E., Coleman, M., Javoy, M. (2008) The chlorine isotope composition of Earth’s mantle. Science 319, 1518–1520. https://doi.org/10.1126/science.1150988

but was never explored.

When MORB data filtered for shallow assimilation from this study and previous studies are combined, the most depleted samples (lowest K₂O/TiO₂; highest ¹⁴³Nd/¹⁴⁴Nd) from several localities have δ³⁷Cl values of −0.2 ‰ to −0.5 ‰, and other MORB samples diverge towards lower and higher δ³⁷Cl values at greater degrees of enrichment (higher K₂O/TiO₂, lower ¹⁴³Nd/¹⁴⁴Nd; Fig. 2). Previously published OIB data (John et al., 2010

John, T., Layne, G.D., Haase, K.M., Barnes, J.D. (2010) Chlorine isotope evidence for crustal recycling into the Earth’s mantle. Earth and Planetary Sciences Letters 298, 175–182. https://doi.org/10.1016/j.epsl.2010.07.039

; Halldórsson et al., 2016

Halldórsson, S.A., Barnes, J.D., Stefánsson, A., Hilton, D.R., Hauri, E.H., Marshall, E.W. (2016) Subducted lithosphere controls halogen enrichments in the Iceland mantle plume source. Geology 44, 679–682. https://doi.org/10.1130/G37924.1

) also follow these general trends (Fig. 2). We calculated the standard deviation in δ³⁷Cl values of subsets of MORB and OIB samples grouped by increasing K₂O/TiO₂ and ¹⁴³Nd/¹⁴⁴Nd ranges (see Fig. 2a,b). The standard deviation of the sample subsets increases with increasing subduction influence (Fig. 2a,b), consistent with addition of subducted altered oceanic crust and sediment with heterogenous δ³⁷Cl values (−3.0 to +3.0 ‰; Barnes and Sharp, 2017

Barnes, J.D., Sharp, Z.D. (2017) Chlorine Isotope Geochemistry. Reviews in Mineralogy and Geochemistry 82, 345–378. https://doi.org/10.1515/9783110545630-010

and references therein; Fig. 2) into the DMM. The δ³⁷Cl values higher or lower than DMM are not uniquely associated with specific mantle components. Samples with HIMU/FOZO influence (Fig. S-3b) extend to both higher and lower δ³⁷Cl values (Fig. S-4a), as do samples influenced by EM-I/EM-II components (Figs. S-3b, S-4b). Although the addition of subducted crust and sediment in the upper mantle influences MORB δ³⁷Cl values, there is currently no clear differentiation between the δ³⁷Cl values of different subducted materials.

To quantify the relationship between changes in δ³⁷Cl values and incorporation of subducted components, we calculate the absolute deviation from the DMM end member δ³⁷Cl value for each sample. Here the DMM δ³⁷Cl value selected is −0.5 ‰ based on the value of the most depleted sample (RC2806 1D-1). We then compare these absolute deviations to the D^Pb/Nd/Sr values for each sample (method modified from Jackson et al., 2020

Jackson, M.G., Blichert-Toft, J., Halldórsson, S.A., Mundl-Petermeier, A., Bizimis, M., Kurz, M.D., Price, A.A., Harðardóttir, S., Willhite, L.N., Breddam, K., Becker, T.W., Fischer, R.A. (2020) Ancient helium and tungsten isotopic signatures preserved in mantle domains least modified by crustal recycling. Proceedings of the National Academy of Sciences 202009663. https://doi.org/10.1073/pnas.2009663117

; Supplementary Information S-4). The D^Pb/Nd/Sr value describes the distance in three dimensional radiogenic isotope space that a sample is from the DMM end member. As the samples extend farther from the DMM component in radiogenic isotope space (higher D^Pb/Nd/Sr), the δ³⁷Cl values also deviate more strongly from the DMM value (Fig. 3). This indicates much of the observed variability in the MORB δ³⁷Cl values results from incorporation of subduction-derived components with heterogeneous δ³⁷Cl values into an ambient DMM reservoir with a δ³⁷Cl value of ∼ −0.5 ‰. Although this trend is largely defined by the HIMU-influenced Equatorial MAR samples (Fig. 3), the observed correlation is still statistically significant if the Equatorial MAR samples are not considered (non-directional p-value = 0.008). When including OIB data from John et al. (2010)

John, T., Layne, G.D., Haase, K.M., Barnes, J.D. (2010) Chlorine isotope evidence for crustal recycling into the Earth’s mantle. Earth and Planetary Sciences Letters 298, 175–182. https://doi.org/10.1016/j.epsl.2010.07.039

, the correlation remains significant (non-directional p-value < 0.0001). Therefore, after accounting for shallow assimilation, mantle heterogeneity, and potential analytical artifacts in previous studies (see Supplementary Information S-3), contradictory estimates of the DMM δ³⁷Cl value from studies in the past two decades can largely be reconciled.

Full size image

Estimate of the Bulk Silicate Earth δ³⁷Cl value and similarity to chondrites. Previous studies have suggested multiple sources of volatile elements to Earth, including nebular ingassing, comet addition, chondrite addition, or a mixture of these three (e.g., Sharp, 2017

Sharp, Z.D. (2017) Nebular ingassing as a source of volatiles to the Terrestrial planets. Chemical Geology 448, 137–150. https://doi.org/10.1016/j.chemgeo.2016.11.018

). The δ³⁷Cl value of the Bulk Silicate Earth (BSE) can shed light on the source of Earth’s volatile elements. We calculated the BSE δ³⁷Cl value by adding all surface material into the processed mantle with their respective Cl concentrations and δ³⁷Cl values. The methods and parameters for this procedure are given in the Supplementary Information S-5. Because seawater and evaporites both have near-zero δ³⁷Cl value and account for a combined 44 ± 6 % of the BSE Cl (Table S-2), whereas the processed mantle contains only 9 ± 2 % (Table S-2), the BSE δ³⁷Cl value must be close to 0 ‰. Indeed, we calculated that the BSE δ³⁷Cl value is −0.04 ± 0.13 ‰ (2 s.e.), statistically indistinguishable from bulk δ³⁷Cl values of all chondrites considered together (−0.12 ± 0.26 ‰; 2 s.e.), or carbonaceous (−0.2 ± 0.4 ‰; 2 s.e.), ordinary (−0.4 ±0. 6‰; 2 s.e.), or enstatite (+0.2 ± 0.4 ‰; 2 s.e.) chondrites considered individually (Sharp et al., 2013

Sharp, Z.D., Mercer, J.A., Jones, R.H., Brearley, A.J., Selverstone, J., Bekker, A., Stachel, T. (2013) The chlorine isotope composition of chondrites and Earth. Geochimica et Cosmochimica Acta 107, 189–204. https://doi.org/10.1016/j.gca.2013.01.003

). In contrast, the BSE δ³⁷Cl value is dissimilar to δ³⁷Cl estimates of the solar nebula (−7 ‰; Gargano and Sharp, 2019

Gargano, A., Sharp, Z.D. (2019) The chlorine isotope composition of iron meteorites: Evidence for the Cl isotope composition of the solar nebula and implications for extensive devolatilization during planet formation. Meteoritics and Planetary Science 54, 1619–1631. https://doi.org/10.1111/maps.13303

).

Previous authors have noted the general similarity between the δ³⁷Cl values of Earth’s various reservoirs and chondrites and interpreted that as indicative of a chondritic source of Cl on Earth (Sharp et al., 2013

Sharp, Z.D., Mercer, J.A., Jones, R.H., Brearley, A.J., Selverstone, J., Bekker, A., Stachel, T. (2013) The chlorine isotope composition of chondrites and Earth. Geochimica et Cosmochimica Acta 107, 189–204. https://doi.org/10.1016/j.gca.2013.01.003

). However, later studies suggested the δ³⁷Cl value of the proto-Earth was originally similar to the solar nebula (−7 ‰; Gargano and Sharp, 2019

Gargano, A., Sharp, Z.D. (2019) The chlorine isotope composition of iron meteorites: Evidence for the Cl isotope composition of the solar nebula and implications for extensive devolatilization during planet formation. Meteoritics and Planetary Science 54, 1619–1631. https://doi.org/10.1111/maps.13303

) and obtained its current δ³⁷Cl value through kinetic fractionation associated with volatile loss during degassing or impact erosion (Sharp et al., 2016

Sharp, Z.D., Williams, J., Shearer, C., Agee, C., McKeegan, K. (2016) The chlorine isotope composition of Martian meteorites 2. Implications for the early solar system and the formation of Mars. Meteoritics and Planetary Science 51, 2111–2126. https://doi.org/10.1111/maps.12591

; Gargano and Sharp, 2019

Gargano, A., Sharp, Z.D. (2019) The chlorine isotope composition of iron meteorites: Evidence for the Cl isotope composition of the solar nebula and implications for extensive devolatilization during planet formation. Meteoritics and Planetary Science 54, 1619–1631. https://doi.org/10.1111/maps.13303

). Other studies suggest the δ³⁷Cl values of chondrites are derived through equilibrium isotope fractionation of nebular gases at low temperature (Sharp et al., 2013

Sharp, Z.D., Mercer, J.A., Jones, R.H., Brearley, A.J., Selverstone, J., Bekker, A., Stachel, T. (2013) The chlorine isotope composition of chondrites and Earth. Geochimica et Cosmochimica Acta 107, 189–204. https://doi.org/10.1016/j.gca.2013.01.003

, 2016

Sharp, Z.D., Williams, J., Shearer, C., Agee, C., McKeegan, K. (2016) The chlorine isotope composition of Martian meteorites 2. Implications for the early solar system and the formation of Mars. Meteoritics and Planetary Science 51, 2111–2126. https://doi.org/10.1111/maps.12591

). These studies imply that the δ³⁷Cl values of Earth and chondrites are not genetically related and any similarity between the two is coincidental. The total range of δ³⁷Cl values on differentiated bodies that have undergone extensive Cl loss spans >46 ‰ (Sarafian et al., 2017

Sarafian, A.R., John, T., Roszjar, J., Whitehouse, M.J. (2017) Chlorine and hydrogen degassing in Vesta’s magma ocean. Earth and Planetary Science Letters 459, 311–319. https://doi.org/10.1016/j.epsl.2016.10.029

; Gargano and Sharp., 2019

Gargano, A., Sharp, Z.D. (2019) The chlorine isotope composition of iron meteorites: Evidence for the Cl isotope composition of the solar nebula and implications for extensive devolatilization during planet formation. Meteoritics and Planetary Science 54, 1619–1631. https://doi.org/10.1111/maps.13303

; Barrett et al., 2019

Barrett, T.J., Barnes, J.J., Anand, M., Franchi, I.A., Greenwood, R.C., Charlier, B.L.A., Zhao, X., Moynier, F., Grady, M.M. (2019) Investigating magmatic processes in the early Solar System using the Cl isotopic systematics of eucrites. Geochimica et Cosmochimica Acta 266, 582–597. https://doi.org/10.1016/j.gca.2019.06.024

). Given this large range, it appears unlikely that varying degrees of planetesimal degassing, impact erosion, and/or subsequent ingassing would result in nearly identical δ³⁷Cl values in the BSE and in distinct classes of chondrites. We do not argue that such a mechanism is impossible, but rather that a more parsimonious model is that Earth’s chondrite-like average δ³⁷Cl value reflects Cl derivation from a chondrite source.

The chondrite-like BSE δ³⁷Cl value requires that little to no Cl was added to Earth from ingassing of the solar nebula. Previous studies have suggested ingassing of the solar nebula into a terrestrial magma ocean as a potential major source of volatiles to the early Earth (e.g., Sharp 2017

Sharp, Z.D. (2017) Nebular ingassing as a source of volatiles to the Terrestrial planets. Chemical Geology 448, 137–150. https://doi.org/10.1016/j.chemgeo.2016.11.018

; Olson and Sharp, 2019

Olson, P.L., Sharp, Z.D. (2019) Nebular atmosphere to magma ocean: A model for volatile capture during Earth accretion. Physics of Earth and Planetary Interiors 294, 106294. https://doi.org/10.1016/j.pepi.2019.106294

). Such incorporation would also impact Cl, which existed as HCl gas in the proto-solar nebula (Fegley et al., 2020

Fegley, B., Lodders, K., Jacobson, N.S. (2020) Volatile element chemistry during accretion of the earth. Geochemistry 80, 125594. https://doi.org/10.1016/j.chemer.2019.125594

). Given the chondritic δ³⁷Cl value of Earth, a maximum of 1–9 % of Earth’s Cl can be derived from ingassing of the solar nebula. Chlorine abundance in the early solar nebula was 10–1000 times lower than C, N, Ne, and Ar (Lodders et al., 2009

Lodders, K., Palme H., Gail, H.P. (2009) Abundances of the elements in the solar system. In: Trümper J.E. (Ed.), Landolt Börnstein, New Series, Springer-Verlag, Berlin, Heidelberg, New York, 560–630. https://doi.org/10.1007/978-3-540-88055-4_34

), but HCl is 100–100,000 times more soluble than CH₄, CO₂, N₂, Ne, and Ar in silicate melts at high temperature/low pressure conditions (Fegley et al., 2020

Fegley, B., Lodders, K., Jacobson, N.S. (2020) Volatile element chemistry during accretion of the earth. Geochemistry 80, 125594. https://doi.org/10.1016/j.chemer.2019.125594

and references therein). Therefore, ingassing of these volatiles will be similar to or lower than for Cl, so Earth’s low nebular Cl fraction requires relatively little ingassing of these other volatiles as well. This is consistent with the estimate made by Marty (2012)

Marty, B. (2012) The origins and concentrations of water, carbon, nitrogen and noble gases on Earth. Earth and Planetary Science Letters 313–314, 56–66. https://doi.org/10.1016/j.epsl.2011.10.040

that ≤10 % of Earth’s volatile elements are derived from nebular ingassing based on C, N, H, and noble gas isotope compositions. Therefore, the estimates of Earth’s δ³⁷Cl value as well as other volatile element isotope ratios are consistent with minimal nebular ingassing contribution to Earth’s volatile content.

top

Conclusions

This study examines δ³⁷Cl values of MORB glasses to better constrain the δ³⁷Cl value of Earth’s mantle and the source of volatiles on Earth. The unmodified DMM has a δ³⁷Cl value ≈ −0.5 ‰, and deviations from that value are either due to incorporation of subducted material with heterogeneous δ³⁷Cl values or assimilation of seawater/brine. In the context of a heterogeneous mantle, the previous disparate mantle δ³⁷Cl estimates are not contradictory, but rather sample different subduction-related components in the upper mantle. Shallow assimilation of hydrothermal brines results in MORB δ³⁷Cl values that extend lower than −0.6 ‰ to −1.5 ‰ and does not necessarily result in seawater-like δ³⁷Cl values. We calculate that the BSE has a δ³⁷Cl value ≈ −0.04 ± 0.13 ‰, consistent with a chondritic source of Cl on Earth with a maximum of 1–9 % contribution from early ingassing of the solar nebula.

top

Acknowledgements

This work was supported by funding by NSF-EAR-1850749 to JDB and JCL, GSA Graduate Research Grant to GSW, and the Jackson School of Geoscience student support to GSW. Thanks go to the Smithsonian NMNH Mineral Sciences Department and the Marine Geological Samples Laboratory, University of Rhode Island for access to samples. NSF grant OCE-2116199 provides support for the curation and distribution of geological samples at the Marine Geological Samples Laboratory, University of Rhode Island. Our thanks go Dr. Jeff Cullen for assistance with IRMS measurements and to Clémence Le Lay for assistance with SIMS measurements. Additional thanks go to undergraduate students Julie Hammons and Enrique Morales for their many hours separating clean glass chips. This manuscript benefitted from constructive comments from Peter Michael, one anonymous reviewer, and editor Helen Williams.

Editor: Helen Williams

top

References

top

Supplementary Information

The Supplementary Information includes:

• IRMS Methods
• SIMS Procedure
• Filtering for Data Quality in Literature and Filtering All Samples for Shallow Assimilation
• Calculating the Distance from the DMM Endmember in Pb, Sr, Nd Multi-Isotope Space
• δ³⁷Cl Values of Terrestrial Reservoirs and the Calculation of the Bulk Silicate Earth δ³⁷Cl value
• Tables S-1 to S-3
• Figures S-1 to S-5
• Supplementary Data Table
• Supplementary Information References

Download the Supplementary Information (PDF)

Download Table S-3 (.xlsx)