Tungsten isotopes in Baffin Island lavas: Evidence of Iceland plume evolution

J. Kaare-Rasmussen^1,2,

¹The MIT-WHOI Joint Program in Oceanography/Applied Ocean Science and Engineering, Woods Hole, MA 02543, USA
²Department of Geology and Geophysics, Woods Hole Oceanographic Institution, Woods Hole, MA 02543, USA

D. Peters³,

³Department of Earth Sciences, Carleton University, Ottawa-Carleton Geoscience Centre, Ottawa, ON K1S 5B6, Canada

H. Rizo³,

³Department of Earth Sciences, Carleton University, Ottawa-Carleton Geoscience Centre, Ottawa, ON K1S 5B6, Canada

R.W. Carlson⁴,

⁴Earth & Planets Laboratory, Carnegie Institution for Science, Washington, DC 20015, USA

S.G. Nielsen^2,5,

²Department of Geology and Geophysics, Woods Hole Oceanographic Institution, Woods Hole, MA 02543, USA
⁵NIRVANA Laboratories, Woods Hole Oceanographic Institution, Woods Hole, MA 02543, USA

F. Horton²

²Department of Geology and Geophysics, Woods Hole Oceanographic Institution, Woods Hole, MA 02543, USA

Affiliations | Corresponding Author | Cite as | Funding information

Geochemical Perspectives Letters v28 | https://doi.org/10.7185/geochemlet.2337
Received 17 May 2023 | Accepted 13 October 2023 | Published 8 November 2023

Published by the European Association of Geochemistry
under Creative Commons License CC BY-NC-ND 4.0

Keywords: geochemical evolution of Earth’s mantle, core–mantle interactions, ¹⁸²W, ³He/⁴He



PDF	PDF+SI

Share this article
Article views:
569
Cumulative count of HTML views and PDF downloads.
Download Citation
Rights & Permissions

top

Abstract

Tungsten and helium isotope ratios in lavas derived from deeply rooted mantle plumes are tracers of lower mantle compositional heterogeneity or core–mantle exchange. We measured the tungsten isotopic compositions of lavas with exceptionally high-³He/⁴He ratios that erupted above the head of the Iceland plume on Baffin Island. These lavas have ¹⁸²W/¹⁸⁴W ratios that are indistinguishable from the convecting upper mantle, unlike younger lavas in Iceland that have lower ¹⁸²W/¹⁸⁴W ratios. This implies that only the Iceland plume tail was infused with low-¹⁸²W/¹⁸⁴W material, likely from the core. If high-³He/⁴He helium also comes from the core, then diffusion across the core–mantle boundary may stratify plume-source mantle domains, with elevated ³He/⁴He travelling farther into the lower mantle than ¹⁸²W/¹⁸⁴W anomalies. Over Earth history, tungsten diffusion from the core can explain the decline of ¹⁸²W/¹⁸⁴W in the convecting mantle. We speculate that the uneven pace of this decline corresponds with evolving lower mantle dynamics.

Figures and Tables

Figure 1 (a) μ¹⁸²W and (b) μ¹⁸³W for Alfa Aesar standard, NIST 3163 standard, and Baffin Island lavas.	Figure 2 Conceptual model of a stratified Iceland plume source region. (a) On Gyr timescales, diffusion across the CMB may lower μ¹⁸²W and raise ³He/⁴He in the lowermost mantle on different length scales. (b) In such a scenario, there would be isotopic gradients in the lowermost mantle from the core (μ¹⁸²W = −200, ³He/⁴He = 120 Ra) to ambient mantle (μ¹⁸²W = 0, ³He/⁴He = 8 Ra). Iceland and Baffin Island mantle sources may derive from within and beyond the region impacted by W isotopic diffusion, respectively. Calculations for the isotopic composition curves are in the Supplementary Information and assume a diffusion timescale of 1 Gyr.	Figure 3 (a) Modern mantle μ¹⁸²W may be a function of the mantle residence time, τ, at the CMB and the percentage of the core surface area, ξ, insulated by long-term stable structures. We assume that (i) the early mantle had μ¹⁸²W of the Moon (+27); (ii) basal mantle acquired core-like μ¹⁸²W corresponding to the characteristic length scale of diffusion (), where D is the diffusivity (D = 4.62 × 10⁻¹⁰ m² s⁻¹; Ferrick and Korenaga, 2023); and (iii) core-infused mantle parcels mixed efficiently into the bulk mantle. Diffusion can explain a μ¹⁸²W decline to zero if τ is short (<30 Myr) and the CMB is <20 % insulated. (b) A compilation of μ¹⁸²W data (Reimink et al., 2020; Nakanishi et al., 2023) and three potential mantle μ¹⁸²W trajectories: (i) constant τ = 35 Myr; (ii) fast Hadean decline (τ = 0.1 Myr) corresponding to rapid Hadean convection followed by slower decline (τ = 200 Myr); and (iii) fast early Paleoproterozoic decline (τ = 0.4 Myr) due to a change in the style of plate tectonics.	Table 1 Tungsten isotopic compositions of the Baffin Island samples. μ¹⁸²W and μ¹⁸³W are reported as deviations from the Alfa Aesar standard (¹⁸²W/¹⁸⁴W = 0.864888 ± 0.000006 and ¹⁸³W/¹⁸⁴W = 0.467151 ± 0.000004, 2 s.e., n = 8) and normalised to ¹⁸⁶W/¹⁸⁴W, denoted by subscript 6/4. 2 s.e. represents the internal run precision of each individual analyses.
Figure 1	Figure 2	Figure 3	Table 1

View all figures and tables

top

Introduction

Geochemical heterogeneities preserved in Earth since its formation place fundamental constraints on planetary accretion and long-term evolution. Mantle plumes that sample the deepest portions of the mantle contain isotopic evidence of ancient, >4.5 Gyr, geochemical reservoirs that survived the mixing caused by giant impacts during the final stages of planetary accretion and billions of years of mantle convection (Mundl-Petermeier et al., 2019

Mundl-Petermeier, A., Walker, R.J., Jackson, M.G., Blichert-Toft, J., Kurz, M.D., Halldórsson, S.A. (2019) Temporal evolution of primordial tungsten-182 and ³He/⁴He signatures in the Iceland mantle plume. Chemical Geology 525, 245–259. https://doi.org/10.1016/j.chemgeo.2019.07.026

). Two competing models have emerged that might explain the preservation of these ancient heterogeneities: (1) the preservation of ancient gas-rich mantle domains (e.g., Kurz et al., 1982

Kurz, M.D., Jenkins, W.J., Hart, S.R. (1982) Helium isotopic systematics of oceanic islands and mantle heterogeneity. Nature 297, 43–47. https://doi.org/10.1038/297043a0

) and (2) core–mantle chemical exchange since planetary accretion (e.g., Rizo et al., 2019

Rizo, H., Andrault, D., Bennett, N.R., Humayun, M., Brandon, A., Vlastelic, I., Moine, B., Poirier, A., Bouhifd, M.A., Murphy, D.T. (2019) ¹⁸²W evidence for core-mantle interaction in the source of mantle plumes. Geochemical Perspectives Letters 11, 6–11. https://doi.org/10.7185/geochemlet.1917

). To test these hypotheses, we measured the tungsten (W) isotopic composition of Baffin Island lavas erupted above the Iceland mantle plume, which contains high-³He/⁴He ratios that have been uniquely well preserved since planetary formation.

The ¹⁸²Hf-¹⁸²W isotope system is a sensitive tracer of core–mantle interaction. During the first ∼60 Myr of solar system history, ¹⁸²W was produced by the decay of the now extinct radionuclide ¹⁸²Hf (t_1/2 = 8.9 Myr; Vockenhuber et al., 2004

Vockenhuber, C., Bichler, M., Golser, R., Kutschera, W., Priller, A., Steier, P., Winkler, S. (2004) ¹⁸²Hf, a new isotope for AMS. Nuclear Instruments and Methods in Physics Research Section B: Beam Interactions with Materials and Atoms 223–224, 823–828. https://doi.org/10.1016/j.nimb.2004.04.152

). The upper terrestrial mantle has μ¹⁸²W ≈ 0 (where μ¹⁸²W = [(¹⁸²W/¹⁸⁴W)_sample/(¹⁸²W/¹⁸⁴W)_standard − 1] × 10⁶), which is substantially higher than the average μ¹⁸²W of chondrites of approximately −190 (Kleine et al., 2009

Kleine, T., Touboul, M., Bourdon, B., Nimmo, F., Mezger, K., Palme, H., Jacobsen, S.B., Yin, Q.-Z., Halliday, A.N. (2009) Hf–W chronology of the accretion and early evolution of asteroids and terrestrial planets. Geochimica et Cosmochimica Acta 73, 5150–5188. https://doi.org/10.1016/j.gca.2008.11.047

). Because Hf is lithophile while W is moderately siderophile under reducing conditions (e.g., Wade et al., 2013

Wade, J., Wood, B.J., Norris, C.A. (2013) The oxidation state of tungsten in silicate melt at high pressures and temperatures. Chemical Geology 335, 189–193. https://doi.org/10.1016/j.chemgeo.2012.10.011

), core formation increased the Hf/W of the mantle and left the metallic core with Hf/W near zero. The superchondritic μ¹⁸²W of the mantle thus most likely reflects core formation during the lifetime of ¹⁸²Hf, in which case the core has μ¹⁸²W lower than −190. Therefore, negative μ¹⁸²W values observed in some mantle plume-related magmas may be evidence of core–mantle exchange (e.g., Rizo et al., 2019

). Alternatively, the lowermost mantle could host ancient isotopic heterogeneities, either resulting from early silicate differentiation events (e.g., Touboul et al., 2012

Touboul, M., Puchtel, I.S., Walker, R.J. (2012) ¹⁸²W Evidence for Long-Term Preservation of Early Mantle Differentiation Products. Science 335, 1065–1069. https://doi.org/10.1126/science.1216351

) or introduced during the late accretion of chondritic material with low μ¹⁸²W relative to the terrestrial mantle (e.g., Willbold et al., 2011

Willbold, M., Elliott, T., Moorbath, S. (2011) The tungsten isotopic composition of the Earth’s mantle before the terminal bombardment. Nature 477, 195–198. https://doi.org/10.1038/nature10399

).

Intriguingly, some of the lowest μ¹⁸²W values have been measured in lavas with elevated ³He/⁴He compared to upper mantle values (greater than ∼8 Ra, where Ra is the atmospheric ratio; e.g., Mundl-Petermeier et al., 2020

Mundl-Petermeier, A., Walker, R.J., Fischer, R.A., Lekic, V., Jackson, M.G., Kurz, M.D. (2020) Anomalous ¹⁸²W in high ³He/⁴He ocean island basalts: Fingerprints of Earth’s core? Geochimica et Cosmochimica Acta 271, 194–211. https://doi.org/10.1016/j.gca.2019.12.020

). This suggests that μ¹⁸²W anomalies are associated with geochemical reservoirs that retain primordial ³He trapped during planetary accretion before nebular gases dispersed, or ³He that was accreted later from solar wind irradiated meteoritic material (Mukhopadhyay and Parai, 2019

Mukhopadhyay, S., Parai, R. (2019) Noble Gases: A Record of Earth’s Evolution and Mantle Dynamics. Annual Review of Earth and Planetary Sciences 47, 389–419. https://doi.org/10.1146/annurev-earth-053018-060238

). Traditionally, high ³He/⁴He has been attributed to the preservation of primordial mantle domains, either in the entire lower mantle (e.g., Kurz et al., 1982

Kurz, M.D., Jenkins, W.J., Hart, S.R. (1982) Helium isotopic systematics of oceanic islands and mantle heterogeneity. Nature 297, 43–47. https://doi.org/10.1038/297043a0

) or within certain regions in the lower mantle (e.g., Rizo et al., 2016

Rizo, H., Walker, R.J., Carlson, R.W., Horan, M.F., Mukhopadhyay, S., Manthos, V., Francis, D., Jackson, M.G. (2016) Preservation of Earth-forming events in the tungsten isotopic composition of modern flood basalts. Science 352, 809–812. https://doi.org/10.1126/science.aad8563

). Alternatively, high-³He/⁴He helium concentrated in the core might escape and become entrained in mantle plumes (Bouhifd et al., 2013

Bouhifd, M.A., Jephcoat, A.P., Heber, V.S., Kelley, S.P. (2013) Helium in Earth’s early core. Nature Geoscience 6, 982–986. https://doi.org/10.1038/ngeo1959

). If so, core–mantle exchange may explain the observation that high-³He/⁴He ratios correlate with low μ¹⁸²W in some mantle plumes (Mundl-Petermeier et al., 2019

, 2020

). As more data become available, however, the correlation between W and He isotopic compositions in modern ocean island basalts (OIBs) seems less universal.

Lavas erupted at c. 62 Ma on Baffin Island above the Iceland mantle plume have the highest ³He/⁴He of any measured terrestrial igneous rock (Horton et al., 2023

Horton, F., Asimow, P.D., Farley, K.A., Curtice, J., Kurz, M.D., Blusztajn, J., Biasi, J., Boyes, X.M. (2023) Highest terrestrial ³He/⁴He credibly from the core. Nature 623, 90–94. https://doi.org/10.1038/s41586-023-06590-8

) and therefore contain an unusually pure primordial helium component. Thus, if high ³He/⁴He is sourced from the core, it might reasonably be expected that these lavas also exhibit low μ¹⁸²W. In this study, we reassess μ¹⁸²W in high-³He/⁴He lavas from Baffin Island because previous attempts to measure their W isotopic compositions have produced inconsistent results (Rizo et al., 2016

; Jansen et al., 2022

Jansen, M.W., Tusch, J., Münker, C., Bragagni, A., Avanzinelli, R., Mastroianni, F., Stuart, F.M., Kurzweil, F. (2022) Upper mantle control on the W isotope record of shallow level plume and intraplate volcanic settings. Earth and Planetary Science Letters 585, 117507. https://doi.org/10.1016/j.epsl.2022.117507

).

top

Results

We analysed glass picked from five Baffin Island pillow lavas containing olivine phenocrysts. Tungsten concentrations (20.9–107.0 ng g⁻¹) correlate with other highly incompatible and immobile elements, such as Th, but do not vary systematically with Sr, Nd, or Hf isotopic compositions (Fig. S-4). Weighted average μ¹⁸²W and μ¹⁸³W are −2.7 ± 6.6 and +2.8 ± 6.6, respectively (2 s.d., n = 5; Table 1, Fig. 1). All μ¹⁸²W and μ¹⁸³W from individual samples are indistinguishable at the 2 s.e. confidence level from the Alfa Aesar and NIST SRM 3163 standards (Table S-1, Fig. S-5).

Table 1 Tungsten isotopic compositions of the Baffin Island samples. μ¹⁸²W and μ¹⁸³W are reported as deviations from the Alfa Aesar standard (¹⁸²W/¹⁸⁴W = 0.864888 ± 0.000006 and ¹⁸³W/¹⁸⁴W = 0.467151 ± 0.000004, 2 s.e., n = 8) and normalised to ¹⁸⁶W/¹⁸⁴W, denoted by subscript 6/4. 2 s.e. represents the internal run precision of each individual analyses.

Sample	μ¹⁸²W_6/4	2 s.e.	μ¹⁸³W_6/4	2 s.e.
PING18-H16	−7.3	5.0	−0.6	4.5
PING18-H2	−0.6	4.2	2.2	3.5
PING18-H20	0.5	3.8	6.9	3.3
DURB18-H11	−1.7	6.0	4.3	5.1
RB18-H3	−5.3	4.8	−0.8	3.9

**Figure 1 (a)** μ¹⁸²W and **(b)** μ¹⁸³W for Alfa Aesar standard, NIST 3163 standard, and Baffin Island lavas.

The lack of resolvable μ¹⁸²W anomalies in Baffin Island lavas agrees with the results published by Jansen et al. (2022)

and from the stratigraphically similar lavas from West Greenland (Mundl-Petermeier et al., 2019

) but differs from the positive μ¹⁸²W anomalies reported by Rizo et al. (2016)

. Improvements in N-TIMS techniques, including the ability to quantify the oxygen isotopic compositions during tungsten oxide measurements (see Supplementary Information), give us confidence that these results are more representative of the Baffin Island mantle source than results published in Rizo et al. (2016)

, which should be interpreted with caution. Also, the absence of μ¹⁸³W anomalies in data reported here alleviates concerns raised by the Jansen et al. (2022)

dataset about contamination and mass-independent fractionation.

The helium isotopic compositions of Baffin Island lavas are well characterised (e.g., Horton et al., 2023

; Table S-2) and have been reported for two samples analysed in this study: olivine crushing experiments for RB18-H3 and PING18-H2 imply minimum magmatic ³He/⁴He ratios of 36.8 ± 2.0 and 55.1 ± 1.3 Ra, respectively (Horton et al., 2023

). Olivine separates from the remaining three samples yielded insufficient helium for isotopic characterisation. These samples contained smaller olivines (<1 mm) than the helium-rich samples (>3 mm); we suspect that the low helium contents of the former reflect olivine growth after magma degassing. Nonetheless, the μ¹⁸²W results reported here are unambiguously associated with high-³He/⁴He lavas.

top

Discussion

The origins of the tungsten and helium in the Iceland plume. Our μ¹⁸²W results are unresolvable from the mantle and therefore do not require a core component in Baffin Island lavas source. Yet, the high-³He/⁴He helium and solar-like neon (Horton et al., 2023

) in these rocks and other lavas from the Iceland plume have presumably been preserved in Earth since the late stages of planetary accretion. On a global scale, rock samples from all 15 hotspots with anomalously low μ¹⁸²W also have anomalously high ³He/⁴He (Mundl-Petermeier et al., 2020

). This suggests a common origin of both elements in mantle plumes, such as primordial or ancient mantle, late accreted material, or the core.

Given the high W concentrations and positive μ¹⁸²W of Archean crust, small amounts of crustal assimilation could mask a core μ¹⁸²W signature. Assimilation modelling (see Supplementary Information) predicts correlations between μ¹⁸²W, trace elements, and long-lived radiogenic isotope ratios that are not observed in our data. This suggests that crustal assimilation is unlikely to have significantly influenced the W isotopic compositions of the Baffin Island lavas. Rather, the lack of μ¹⁸²W anomalies in Baffin Island high-³He/⁴He lavas indicates that helium and tungsten in plumes either (a) derive from a common source but are decoupled in the Iceland plume, or (b) have different origins. Either way, plumes appear to form in ways that produce systematic He-W correlations in many cases, but not universally.

The hypothesis that the high-³He/⁴He ratios are derived from primordial, non-degassed mantle in the Baffin Island lavas is inconsistent with other geochemical constraints, including our new μ¹⁸²W data. The primordial mantle likely had positive μ¹⁸²W, based on the positive μ¹⁸²W compositions of the Moon (Kruijer et al., 2012

Kruijer, T.S., Sprung, P., Kleine, T., Leya, I., Burkhardt, C., Wieler, R. (2012) Hf–W chronometry of core formation in planetesimals inferred from weakly irradiated iron meteorites. Geochimica et Cosmochimica Acta 99, 287–304. https://doi.org/10.1016/j.gca.2012.09.015

; Touboul et al., 2015

Touboul, M., Puchtel, I.S., Walker, R.J. (2015) Tungsten isotopic evidence for disproportional late accretion to the Earth and Moon. Nature 520, 530–533. https://doi.org/10.1038/nature14355

) and mantle-derived rocks in the Archean (e.g., Reimink et al., 2020

Reimink, J.R., Mundl-Petermeier, A., Carlson, R.W., Shirey, S.B., Walker, R.J., Pearson, D.G. (2020) Tungsten Isotope Composition of Archean Crustal Reservoirs and Implications for Terrestrial μ¹⁸²W Evolution. Geochemistry, Geophysics, Geosystems 21, e2020GC009155. https://doi.org/10.1029/2020GC009155

, and references therein). However, modern mantle plumes do not have positive μ¹⁸²W. Furthermore, the Baffin Island lavas have superchondritic ¹⁴³Nd/¹⁴⁴Nd and ¹⁷⁶Hf/¹⁷⁷Hf (Willhite et al., 2019

Willhite, L.N., Jackson, M.G., Blichert‐Toft, J., Bindeman, I., Kurz, M.D., Halldórsson, S.A., Harðardóttir, S., Gazel, E., Price, A.A., Byerly, B.L. (2019) Hot and Heterogenous High‐³He/⁴He Components: New Constraints From Proto‐Iceland Plume Lavas From Baffin Island. Geochemistry, Geophysics, Geosystems 20, 5939–5967. https://doi.org/10.1029/2019GC008654

), suggesting they are not derived from a primordial mantle component but a differentiated mantle reservoir. These observations, combined with correlations between high ³He/⁴He and low μ¹⁸²W in many hotspots globally, imply the latter likely derive from a common deep Earth reservoir that is not primordial mantle.

Ancient differentiated mantle reservoirs that formed while ¹⁸²Hf was extant are similarly difficult to reconcile with coupled He-W isotopic compositions. Magma ocean silicate cumulates generated in the aftermath of the Moon-forming giant impact should be depleted in incompatible trace elements and might host high-³He/⁴He helium (Coltice et al., 2011

Coltice, N., Moreira, M., Hernlund, J., Labrosse, S. (2011) Crystallization of a basal magma ocean recorded by Helium and Neon. Earth and Planetary Science Letters 308, 193–199. https://doi.org/10.1016/j.epsl.2011.05.045

). If formed while ¹⁸²Hf was extant, cumulates would presumably acquire a positive μ¹⁸²W composition because W is more incompatible than Hf in silicate minerals (Righter and Shearer, 2003

Righter, K., Shearer, C.K. (2003) Magmatic fractionation of Hf and W: constraints on the timing of core formation and differentiation in the Moon and Mars. Geochimica et Cosmochimica Acta 67, 2497–2507. https://doi.org/10.1016/S0016-7037(02)01349-2

). Therefore, an ancient depleted mantle with high ³He/⁴He would be expected to have higher μ¹⁸²W than primordial mantle. Furthermore, silicate differentiation would have fractionated Sm from Nd, thereby influencing the abundances of ¹⁴²Nd—the product of ¹⁴⁶Sm decay (t_1/2 ≈ 100 Myr)—in the segregates. However, like most post-Archean mantle-derived rocks, Baffin Island (de Leeuw et al., 2017

de Leeuw, G.A.M., Ellam, R.M., Stuart, F.M., Carlson, R.W. (2017) ¹⁴²Nd/¹⁴⁴Nd inferences on the nature and origin of the source of high ³He/⁴He magmas. Earth and Planetary Letters 472, 62–68. https://doi.org/10.1016/j.epsl.2017.05.005

) and Iceland lavas (Murphy et al., 2010

Murphy, D.T., Brandon, A.D., Debaille, V., Burgess, R., Ballentine, C. (2010) In search of a hidden long-term isolated sub-chondritic ¹⁴²Nd/¹⁴⁴Nd reservoir in the deep mantle: Implications for the Nd isotope systematics of the Earth. Geochimica et Cosmochimica Acta 74, 738–750. https://doi.org/10.1016/j.gca.2009.10.005

) lack ¹⁴²Nd/¹⁴⁴Nd anomalies. This indicates that the Iceland plume did not derive from mantle differentiated during the lifetimes of ¹⁴⁶Sm or ¹⁸²Hf.

Silicate differentiation within Hadean crust might have produced restite with low μ¹⁸²W decoupled from ¹⁴²Nd/¹⁴⁴Nd (Tusch et al., 2022

Tusch, J., Hoffmann, J.E., Hasenstab, E., Fischer-Gödde, M., Marien, C.S., Wilson, A.H., Münker, C. (2022) Long-term preservation of Hadean protocrust in Earth’s mantle. Proceedings of the National Academy of Sciences 119, e2120241119. https://doi.org/10.1073/pnas.2120241119

). However, the formation of, and subsequent magmatic differentiation within, Hadean crust would have caused extensive degassing of primordial gases. If so, Hadean crustal restites that foundered into the mantle would acquire low ³He/⁴He over time. Mantle plumes incorporating such material would acquire positively correlated ³He/⁴He and μ¹⁸²W, which is not observed.

Alternatively, hidden low-μ¹⁸²W mantle domains—perhaps formed during late accretion—are potential hosts of primordial ³He. The μ¹⁸²W of the convecting mantle may have decreased by ∼27 since Moon formation, based on lunar and Eoarchean terrestrial rock compositions (e.g., Willbold et al., 2011

Willbold, M., Elliott, T., Moorbath, S. (2011) The tungsten isotopic composition of the Earth’s mantle before the terminal bombardment. Nature 477, 195–198. https://doi.org/10.1038/nature10399

). About 0.5 wt. % of late accreting chondritic material with high W and highly siderophile element (HSE) concentrations but low μ¹⁸²W might explain the decline in mantle μ¹⁸²W (Willbold et al., 2011

Willbold, M., Elliott, T., Moorbath, S. (2011) The tungsten isotopic composition of the Earth’s mantle before the terminal bombardment. Nature 477, 195–198. https://doi.org/10.1038/nature10399

). Mantle domains that contain an above average amount of late accreting material have been proposed as alternative sources of negative μ¹⁸²W (Marchi et al., 2018

Marchi, S., Canup, R.M., Walker, R.J. (2018) Heterogeneous delivery of silicate and metal to the Earth by large planetesimals. Nature Geoscience 11, 77–81. https://doi.org/10.1038/s41561-017-0022-3

). However, late accretion seems an unlikely common source of W and He because most late accreting material is expected to have high ³He/⁴He and to be HSE-rich, yet high-³He/⁴He helium in late accreting materials would not necessarily enter the mantle, and HSE abundances do not correlate with μ¹⁸²W or ³He/⁴He in mantle plumes (e.g., Rizo et al., 2019

). Although estimating the mantle source HSE abundances from the HSE contents of erupted magmas is difficult, to our knowledge, no clear correlation between μ¹⁸²W and the HSE content has been demonstrated in spite of the wide range in μ¹⁸²W seen in OIBs.

Given the above discussion, combined with W isotope evidence from other mantle plumes, the negative μ¹⁸²W anomalies (as low as −12.9) in some Iceland high-³He/⁴He lavas might derive from the core (Mundl-Petermeier et al., 2019

). Even though the Baffin Island lavas do not have a μ¹⁸²W anomaly outside the analytical uncertainty, the external reproducibility of these samples still allow a small amount of bulk core material (∼1.08 %) mixed with ambient mid-ocean ridge basalt (MORB)-like depleted mantle (see Supplementary Information, Fig. S-6). However, bulk mixing of this much core into the mantle is mechanically improbable and inconsistent with the Os concentrations in Iceland and Baffin Island lavas (Rizo et al., 2016

; Mundl-Petermeier et al., 2019

), which limit the amount of bulk core contribution to <0.1 %, assuming 2.83 μg g⁻¹ Os in the core (Day, 2013

Day, J.M.D. (2013) Hotspot volcanism and highly siderophile elements. Chemical Geology 341, 50–74. https://doi.org/10.1016/j.chemgeo.2012.12.010

).

Importantly, ³He/⁴He appears decoupled from not only W but also from lithophile elements, HSEs, and heavier noble gases in the Iceland plume (e.g., Mundl-Petermeier et al., 2019

). This observation is consistent with helium diffusion into Iceland mantle from—rather than bulk mixing with—a ³He-rich reservoir. Theoretically, He concentration gradients (Horton et al., 2023

) and W isotopic gradients (Ferrick and Korenaga, 2023

Ferrick, A.L., Korenaga, J. (2023) Long-term core–mantle interaction explains W-He isotope heterogeneities. Proceedings of the National Academy of Sciences 120, e2215903120. https://doi.org/10.1073/pnas.2215903120

) exist across the core–mantle boundary (CMB) that could drive diffusion into the mantle (Fig. 2). Alternatively, He and W diffused into the Iceland plume source from different reservoirs, such as ancient differentiated mantle and the core, respectively.

**Figure 2** Conceptual model of a stratified Iceland plume source region. **(a)** On Gyr timescales, diffusion across the CMB may lower μ¹⁸²W and raise ³He/⁴He in the lowermost mantle on different length scales. **(b)** In such a scenario, there would be isotopic gradients in the lowermost mantle from the core (μ¹⁸²W = −200, ³He/⁴He = 120 Ra) to ambient mantle (μ¹⁸²W = 0, ³He/⁴He = 8 Ra). Iceland and Baffin Island mantle sources may derive from within and beyond the region impacted by W isotopic diffusion, respectively. Calculations for the isotopic composition curves are in the Supplementary Information and assume a diffusion timescale of 1 Gyr.

), the characteristic length scales of diffusion (

) for helium might be ∼40 km, if theoretical diffusion rates for the upper mantle (Wang et al., 2015

Wang, K., Brodholt, J., Lu, X. (2015) Helium diffusion in olivine from first principles calculations. Geochimica et Cosmochimica Acta 156, 145–153. https://doi.org/10.1016/j.gca.2015.01.023

) are extrapolated to lower mantle temperatures. For W, this length scale may be only 5–10 km (Ferrick and Korenaga, 2023

), suggesting that mantle domains farther from the CMB may be characterised by He but not W isotopic anomalies. If anomalous W and He in the Iceland plume diffused from the core, our results imply that the plume head originated farther from the CMB than the plume tail.

Baffin Island mantle may have originated from the periphery of a helium-infused zone in the lowermost mantle (10–40 km from the core), whereas low-μ¹⁸²W Iceland mantle might have resided nearer the core (Fig. 2). Negative μ¹⁸²W anomalies may only exist in the Iceland plume tail, which may have preferentially entrained denser material proximal to the core–mantle boundary (Jones et al., 2019

Jones, T.D., Davies, D.R., Sossi, P.A. (2019) Tungsten isotopes in mantle plumes: Heads it’s positive, tails it’s negative. Earth and Planetary Science Letters 506, 255–267. https://doi.org/10.1016/j.epsl.2018.11.008

). The plume head, from which Baffin Island lavas derived, may have instead entrained primarily portions of the lowermost mantle beyond the diffusion limit of W but still infused with ³He from the core. Thus, plume tails might be the most efficient conveyors of material from the CMB.

This model predicts that ³He/⁴He is highest in lavas with the most negative μ¹⁸²W, a trend observed for all high-³He/⁴He hotspots, except the Iceland plume (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., Beddam, 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 117, 30993–31001. https://doi.org/10.1073/pnas.2009663117

). Maximum ³He/⁴He in Iceland plume lavas apparently declined from >65 Ra at 62 Ma (Horton et al., 2023

) to <26 Ra in the neovolcanic zones of Iceland (Harðardóttir et al., 2018

Harðardóttir, S., Halldórsson, S.A., Hilton, D.R. (2018) Spatial distribution of helium isotopes in Icelandic geothermal fluids and volcanic materials with implications for location, upwelling and evolution of the Icelandic mantle plume. Chemical Geology 480, 12–27. https://doi.org/10.1016/j.chemgeo.2017.05.012

). This decline could be due to the incorporation of convecting upper mantle into the Iceland plume—enough to explain up to a 40 % MORB component in modern Iceland lavas—as a result of the ridge-centred plume position (Shorttle and Maclennan, 2011

Shorttle, O., Maclennan, J. (2011) Compositional trends of Icelandic basalts: Implications for short–length scale lithological heterogeneity in mantle plumes. Geochemistry, Geophysics, Geosystems 12, Q11008. https://doi.org/10.1029/2011GC003748

). The addition of this much material from the convecting upper mantle would have moderated ³He/⁴He and μ¹⁸²W, implying that the Iceland plume itself currently exhibits μ¹⁸²W closer to −21.

Implications for planetary accretion and convecting mantle evolution. By combining μ¹⁴²Nd and μ¹⁸²W constraints with our diffusion model, an early Earth chronology emerges. Some Eoarchean rocks have ¹⁴²Nd anomalies (e.g., Caro et al., 2003

Caro, G., Bourdon, B., Birck, J.-L., Moorbath, S. (2003) ¹⁴⁶Sm–¹⁴²Nd evidence from Isua metamorphosed sediments for early differentiation of the Earth’s mantle. Nature 423, 428–432. https://doi.org/10.1038/nature01668

) produced by igneous differentiation that fractionated Sm and Nd prior to 4 Ga. Such differentiation could be expected to also fractionate Hf from W, so any differentiation that occurred before ¹⁸²Hf became extinct would have produced μ¹⁸²W anomalies that would correlate with μ¹⁴²Nd variations (Touboul et al., 2012

). However, correlations between μ¹⁴²Nd and μ¹⁸²W are rarely observed. Instead, these observations seem consistent with: (i) a Moon-forming impact after ¹⁸²Hf was extinct (∼4.5 Ga) that homogenised the silicate portions of the Earth–Moon system, (ii) generation of ¹⁴²Nd/¹⁴⁴Nd anomalies in the mantle by differentiation events that post-dated the Moon-forming impact that were subsequently erased by mixing after the extinction of ¹⁴⁶Sm around 4 Ga, and (iii) μ¹⁸²W decline in the mantle until present caused by CMB diffusion.

The inferred mantle μ¹⁸²W decline since the Hadean requires that the average residence time (τ) of material diffused from the core at the CMB was ≤30 Myr (Fig. 3a). Due to the higher temperatures in early Earth, early mantle convection may have been rapid. Fast convection would also have efficiently homogenised the mantle and, hence, efficient W isotopic transfer across the CMB (Hadean rapid convection path, Fig. 3b). However, μ¹⁴²Nd and positive μ¹⁸²W heterogeneities throughout the mantle persisted at least until the end of the Archean, and potentially even for longer (Slowing mantle convection path, Fig. 3b). Perhaps the μ¹⁸²W of the mantle rapidly decreased during the late Archean to early Proterozoic, coinciding with development of continents and therefore a liminal stage of mantle dynamics. Alternatively, CMB cover by long-term stable structures may have increased in the late Archean (i.e. increasing ξ), inhibiting transfer of core-derived W to the convecting mantle. Either way, this transition suggests a link between continent formation and lower mantle dynamics during the initiation of modern plate tectonics.

**Figure 3 (a)** Modern mantle μ¹⁸²W may be a function of the mantle residence time, τ, at the CMB and the percentage of the core surface area, ξ, insulated by long-term stable structures. We assume that (i) the early mantle had μ¹⁸²W of the Moon (+27); (ii) basal mantle acquired core-like μ¹⁸²W corresponding to the characteristic length scale of diffusion (, where D is the diffusivity (D = 4.62 × 10⁻¹⁰ m² s⁻¹; Ferrick and Korenaga, 2023
Ferrick, A.L., Korenaga, J. (2023) Long-term core–mantle interaction explains W-He isotope heterogeneities. *Proceedings of the National Academy of Sciences* 120, e2215903120. https://doi.org/10.1073/pnas.2215903120
); and (iii) core-infused mantle parcels mixed efficiently into the bulk mantle. Diffusion can explain a μ¹⁸²W decline to zero if τ is short (<30 Myr) and the CMB is <20 % insulated. **(b)** A compilation of μ¹⁸²W data (Reimink *et al.*, 2020
Reimink, J.R., Mundl-Petermeier, A., Carlson, R.W., Shirey, S.B., Walker, R.J., Pearson, D.G. (2020) Tungsten Isotope Composition of Archean Crustal Reservoirs and Implications for Terrestrial μ¹⁸²W Evolution. *Geochemistry, Geophysics, Geosystems* 21, e2020GC009155. https://doi.org/10.1029/2020GC009155
; Nakanishi *et al.*, 2023
Nakanishi, N., Puchtel, I.S., Walker, R.J., Nabelek, P.I. (2023) Dissipation of Tungsten-182 Anomalies in the Archean Upper Mantle: Evidence from the Black Hills, South Dakota, USA. *Chemical Geology* 617, 121255. https://doi.org/10.1016/j.chemgeo.2022.121255
) and three potential mantle μ¹⁸²W trajectories: (i) constant τ = 35 Myr; (ii) fast Hadean decline (τ = 0.1 Myr) corresponding to rapid Hadean convection followed by slower decline (τ = 200 Myr); and (iii) fast early Paleoproterozoic decline (τ = 0.4 Myr) due to a change in the style of plate tectonics.

top

Acknowledgements

We thank M. Mahy of Parks Canada Nunavut Field Unit for assisting with fieldwork planning and Shuangquan Zhang at the IGGRC of Carleton University for technical support. This work was funded by a National Science Foundation grant awarded to F. Horton (NSF EAR-1911699), a Natural Sciences and Engineering Research Council of Canada Discovery Grant awarded to H. Rizo (RGPIN-477144-2015), and an Ontario Early Researcher Award received by H. Rizo. Additional support came from a National Science Foundation grant awarded to S.G. Nielsen (EAR-1829546), the Woods Hole Oceanographic Institution Andrew W. Mellon Foundation Endowed Fund for Innovative Research, and a National Geographic Society grant (CP4-144R-18), which supported fieldwork activities. Comments from two anonymous reviewers and editor Raul O.C. Fonseca improved this manuscript.

Editor: Raúl Fonseca

top

References

Bouhifd, M.A., Jephcoat, A.P., Heber, V.S., Kelley, S.P. (2013) Helium in Earth’s early core. Nature Geoscience 6, 982–986. https://doi.org/10.1038/ngeo1959

Show in context

Alternatively, high-³He/⁴He helium concentrated in the core might escape and become entrained in mantle plumes (Bouhifd et al., 2013).
View in article

Show in context

By combining μ¹⁴²Nd and μ¹⁸²W constraints with our diffusion model, an early Earth chronology emerges. Some Eoarchean rocks have ¹⁴²Nd anomalies (e.g., Caro et al., 2003) produced by igneous differentiation that fractionated Sm and Nd prior to 4 Ga.
View in article

Show in context

Magma ocean silicate cumulates generated in the aftermath of the Moon-forming giant impact should be depleted in incompatible trace elements and might host high-³He/⁴He helium (Coltice et al., 2011).
View in article

Day, J.M.D. (2013) Hotspot volcanism and highly siderophile elements. Chemical Geology 341, 50–74. https://doi.org/10.1016/j.chemgeo.2012.12.010

Show in context

However, bulk mixing of this much core into the mantle is mechanically improbable and inconsistent with the Os concentrations in Iceland and Baffin Island lavas (Rizo et al., 2016; Mundl-Petermeier et al., 2019), which limit the amount of bulk core contribution to <0.1 %, assuming 2.83 μg g⁻¹ Os in the core (Day, 2013).
View in article

Show in context

However, like most post-Archean mantle-derived rocks, Baffin Island (de Leeuw et al., 2017) and Iceland lavas (Murphy et al., 2010) lack ¹⁴²Nd/¹⁴⁴Nd anomalies. This indicates that the Iceland plume did not derive from mantle differentiated during the lifetimes of ¹⁴⁶Sm or ¹⁸²Hf.
View in article

Show in context

This observation is consistent with helium diffusion into Iceland mantle from—rather than bulk mixing with—a ³He-rich reservoir. Theoretically, He concentration gradients (Horton et al., 2023) and W isotopic gradients (Ferrick and Korenaga, 2023) exist across the core–mantle boundary (CMB) that could drive diffusion into the mantle (Fig. 2).
View in article
Helium may diffuse farther into mantle plume sources than W. Assuming the mantle plume source is stable on Gyr timescales (as expected for large low-shear wave velocity provinces, LLSVPs; e.g., Ferrick and Korenaga, 2023), the characteristic length scales of diffusion () for helium might be ∼40 km, if theoretical diffusion rates for the upper mantle (Wang et al., 2015) are extrapolated to lower mantle temperatures. For W, this length scale may be only 5–10 km (Ferrick and Korenaga, 2023), suggesting that mantle domains farther from the CMB may be characterised by He but not W isotopic anomalies.
View in article
We assume that (i) the early mantle had μ¹⁸²W of the Moon (+27); (ii) basal mantle acquired core-like μ¹⁸²W corresponding to the characteristic length scale of diffusion (, where D is the diffusivity (D = 4.62 × 10⁻¹⁰ m² s⁻¹; Ferrick and Korenaga, 2023); and (iii) core-infused mantle parcels mixed efficiently into the bulk mantle. Diffusion can explain a μ¹⁸²W decline to zero if τ is short (<30 Myr) and the CMB is <20 % insulated.
View in article

Show in context

Maximum ³He/⁴He in Iceland plume lavas apparently declined from >65 Ra at 62 Ma (Horton et al., 2023) to <26 Ra in the neovolcanic zones of Iceland (Harðardóttir et al., 2018).
View in article

Show in context

Lavas erupted at c. 62 Ma on Baffin Island above the Iceland mantle plume have the highest ³He/⁴He of any measured terrestrial igneous rock (Horton et al., 2023) and therefore contain an unusually pure primordial helium component.
View in article
The helium isotopic compositions of Baffin Island lavas are well characterised (e.g., Horton et al., 2023; Table S-2) and have been reported for two samples analysed in this study: olivine crushing experiments for RB18-H3 and PING18-H2 imply minimum magmatic ³He/⁴He ratios of 36.8 ± 2.0 and 55.1 ± 1.3 Ra, respectively (Horton et al., 2023).
View in article
Yet, the high-³He/⁴He helium and solar-like neon (Horton et al., 2023) in these rocks and other lavas from the Iceland plume have presumably been preserved in Earth since the late stages of planetary accretion. On a global scale, rock samples from all 15 hotspots with anomalously low μ¹⁸²W also have anomalously high ³He/⁴He (Mundl-Petermeier et al., 2020).
View in article
This observation is consistent with helium diffusion into Iceland mantle from—rather than bulk mixing with—a ³He-rich reservoir. Theoretically, He concentration gradients (Horton et al., 2023) and W isotopic gradients (Ferrick and Korenaga, 2023) exist across the core–mantle boundary (CMB) that could drive diffusion into the mantle (Fig. 2).
View in article
Maximum ³He/⁴He in Iceland plume lavas apparently declined from >65 Ra at 62 Ma (Horton et al., 2023) to <26 Ra in the neovolcanic zones of Iceland (Harðardóttir et al., 2018).
View in article

Show in context

This model predicts that ³He/⁴He is highest in lavas with the most negative μ¹⁸²W, a trend observed for all high-³He/⁴He hotspots, except the Iceland plume (Jackson et al., 2020).
View in article

Show in context

In this study, we reassess μ¹⁸²W in high-³He/⁴He lavas from Baffin Island because previous attempts to measure their W isotopic compositions have produced inconsistent results (Rizo et al., 2016; Jansen et al., 2022).
View in article
The lack of resolvable μ¹⁸²W anomalies in Baffin Island lavas agrees with the results published by Jansen et al. (2022) and from the stratigraphically similar lavas from West Greenland (Mundl-Petermeier et al., 2019) but differs from the positive μ¹⁸²W anomalies reported by Rizo et al. (2016).
View in article
Also, the absence of μ¹⁸³W anomalies in data reported here alleviates concerns raised by the Jansen et al. (2022) dataset about contamination and mass-independent fractionation.
View in article

Show in context

Negative μ¹⁸²W anomalies may only exist in the Iceland plume tail, which may have preferentially entrained denser material proximal to the core–mantle boundary (Jones et al., 2019).
View in article

Show in context

The upper terrestrial mantle has μ¹⁸²W ≈ 0 (where μ¹⁸²W = [(¹⁸²W/¹⁸⁴W)_sample/(¹⁸²W/¹⁸⁴W)_standard − 1] × 10⁶), which is substantially higher than the average μ¹⁸²W of chondrites of approximately −190 (Kleine et al., 2009).
View in article

Show in context

Kurz, M.D., Jenkins, W.J., Hart, S.R. (1982) Helium isotopic systematics of oceanic islands and mantle heterogeneity. Nature 297, 43–47. https://doi.org/10.1038/297043a0

Show in context

Mantle domains that contain an above average amount of late accreting material have been proposed as alternative sources of negative μ¹⁸²W (Marchi et al., 2018).
View in article

Show in context

This suggests that μ¹⁸²W anomalies are associated with geochemical reservoirs that retain primordial ³He trapped during planetary accretion before nebular gases dispersed, or ³He that was accreted later from solar wind irradiated meteoritic material (Mukhopadhyay and Parai, 2019).
View in article

Show in context

Mantle plumes that sample the deepest portions of the mantle contain isotopic evidence of ancient, >4.5 Gyr, geochemical reservoirs that survived the mixing caused by giant impacts during the final stages of planetary accretion and billions of years of mantle convection (Mundl-Petermeier et al., 2019).
View in article
If so, core–mantle exchange may explain the observation that high-³He/⁴He ratios correlate with low μ¹⁸²W in some mantle plumes (Mundl-Petermeier et al., 2019, 2020).
View in article
The lack of resolvable μ¹⁸²W anomalies in Baffin Island lavas agrees with the results published by Jansen et al. (2022) and from the stratigraphically similar lavas from West Greenland (Mundl-Petermeier et al., 2019) but differs from the positive μ¹⁸²W anomalies reported by Rizo et al. (2016).
View in article
Given the above discussion, combined with W isotope evidence from other mantle plumes, the negative μ¹⁸²W anomalies (as low as −12.9) in some Iceland high-³He/⁴He lavas might derive from the core (Mundl-Petermeier et al., 2019).
View in article
However, bulk mixing of this much core into the mantle is mechanically improbable and inconsistent with the Os concentrations in Iceland and Baffin Island lavas (Rizo et al., 2016; Mundl-Petermeier et al., 2019), which limit the amount of bulk core contribution to <0.1 %, assuming 2.83 μg g⁻¹ Os in the core (Day, 2013).
View in article
Importantly, ³He/⁴He appears decoupled from not only W but also from lithophile elements, HSEs, and heavier noble gases in the Iceland plume (e.g., Mundl-Petermeier et al., 2019).
View in article

Show in context

Intriguingly, some of the lowest μ¹⁸²W values have been measured in lavas with elevated ³He/⁴He compared to upper mantle values (greater than ∼8 Ra, where Ra is the atmospheric ratio; e.g., Mundl-Petermeier et al., 2020).
View in article
If so, core–mantle exchange may explain the observation that high-³He/⁴He ratios correlate with low μ¹⁸²W in some mantle plumes (Mundl-Petermeier et al., 2019, 2020).
View in article
Yet, the high-³He/⁴He helium and solar-like neon (Horton et al., 2023) in these rocks and other lavas from the Iceland plume have presumably been preserved in Earth since the late stages of planetary accretion. On a global scale, rock samples from all 15 hotspots with anomalously low μ¹⁸²W also have anomalously high ³He/⁴He (Mundl-Petermeier et al., 2020).
View in article

Show in context

Nakanishi, N., Puchtel, I.S., Walker, R.J., Nabelek, P.I. (2023) Dissipation of Tungsten-182 Anomalies in the Archean Upper Mantle: Evidence from the Black Hills, South Dakota, USA. Chemical Geology 617, 121255. https://doi.org/10.1016/j.chemgeo.2022.121255

Show in context

(b) A compilation of μ¹⁸²W data (Reimink et al., 2020; Nakanishi et al., 2023) and three potential mantle μ¹⁸²W trajectories: (i) constant τ = 35 Myr; (ii) fast Hadean decline (τ = 0.1 Myr) corresponding to rapid Hadean convection followed by slower decline (τ = 200 Myr); and (iii) fast early Paleoproterozoic decline (τ = 0.4 Myr) due to a change in the style of plate tectonics.
View in article

Show in context

The primordial mantle likely had positive μ¹⁸²W, based on the positive μ¹⁸²W compositions of the Moon (Kruijer et al., 2012; Touboul et al., 2015) and mantle-derived rocks in the Archean (e.g., Reimink et al., 2020, and references therein).
View in article
(b) A compilation of μ¹⁸²W data (Reimink et al., 2020; Nakanishi et al., 2023) and three potential mantle μ¹⁸²W trajectories: (i) constant τ = 35 Myr; (ii) fast Hadean decline (τ = 0.1 Myr) corresponding to rapid Hadean convection followed by slower decline (τ = 200 Myr); and (iii) fast early Paleoproterozoic decline (τ = 0.4 Myr) due to a change in the style of plate tectonics.
View in article

Show in context

If formed while ¹⁸²Hf was extant, cumulates would presumably acquire a positive μ¹⁸²W composition because W is more incompatible than Hf in silicate minerals (Righter and Shearer, 2003).
View in article

Show in context

Traditionally, high ³He/⁴He has been attributed to the preservation of primordial mantle domains, either in the entire lower mantle (e.g., Kurz et al., 1982) or within certain regions in the lower mantle (e.g., Rizo et al., 2016).
View in article
In this study, we reassess μ¹⁸²W in high-³He/⁴He lavas from Baffin Island because previous attempts to measure their W isotopic compositions have produced inconsistent results (Rizo et al., 2016; Jansen et al., 2022).
View in article
The lack of resolvable μ¹⁸²W anomalies in Baffin Island lavas agrees with the results published by Jansen et al. (2022) and from the stratigraphically similar lavas from West Greenland (Mundl-Petermeier et al., 2019) but differs from the positive μ¹⁸²W anomalies reported by Rizo et al. (2016).
View in article
Improvements in N-TIMS techniques, including the ability to quantify the oxygen isotopic compositions during tungsten oxide measurements (see Supplementary Information), give us confidence that these results are more representative of the Baffin Island mantle source than results published in Rizo et al. (2016), which should be interpreted with caution.
View in article
However, bulk mixing of this much core into the mantle is mechanically improbable and inconsistent with the Os concentrations in Iceland and Baffin Island lavas (Rizo et al., 2016; Mundl-Petermeier et al., 2019), which limit the amount of bulk core contribution to <0.1 %, assuming 2.83 μg g⁻¹ Os in the core (Day, 2013).
View in article

Show in context

Two competing models have emerged that might explain the preservation of these ancient heterogeneities: (1) the preservation of ancient gas-rich mantle domains (e.g., Kurz et al., 1982) and (2) core–mantle chemical exchange since planetary accretion (e.g., Rizo et al., 2019).
View in article
Therefore, negative μ¹⁸²W values observed in some mantle plume-related magmas may be evidence of core–mantle exchange (e.g., Rizo et al., 2019).
View in article
However, late accretion seems an unlikely common source of W and He because most late accreting material is expected to have high ³He/⁴He and to be HSE-rich, yet high-³He/⁴He helium in late accreting materials would not necessarily enter the mantle, and HSE abundances do not correlate with μ¹⁸²W or ³He/⁴He in mantle plumes (e.g., Rizo et al., 2019).
View in article

Show in context

This decline could be due to the incorporation of convecting upper mantle into the Iceland plume—enough to explain up to a 40 % MORB component in modern Iceland lavas—as a result of the ridge-centred plume position (Shorttle and Maclennan, 2011).
View in article

Show in context

Alternatively, the lowermost mantle could host ancient isotopic heterogeneities, either resulting from early silicate differentiation events (e.g., Touboul et al., 2012) or introduced during the late accretion of chondritic material with low μ¹⁸²W relative to the terrestrial mantle (e.g., Willbold et al., 2011).
View in article
Such differentiation could be expected to also fractionate Hf from W, so any differentiation that occurred before ¹⁸²Hf became extinct would have produced μ¹⁸²W anomalies that would correlate with μ¹⁴²Nd variations (Touboul et al., 2012).
View in article

Touboul, M., Puchtel, I.S., Walker, R.J. (2015) Tungsten isotopic evidence for disproportional late accretion to the Earth and Moon. Nature 520, 530–533. https://doi.org/10.1038/nature14355

Show in context

Silicate differentiation within Hadean crust might have produced restite with low μ¹⁸²W decoupled from ¹⁴²Nd/¹⁴⁴Nd (Tusch et al., 2022).
View in article

Show in context

During the first ∼60 Myr of solar system history, ¹⁸²W was produced by the decay of the now extinct radionuclide ¹⁸²Hf (t_1/2 = 8.9 Myr; Vockenhuber et al., 2004).
View in article

Show in context

Because Hf is lithophile while W is moderately siderophile under reducing conditions (e.g., Wade et al., 2013), core formation increased the Hf/W of the mantle and left the metallic core with Hf/W near zero.
View in article

Wang, K., Brodholt, J., Lu, X. (2015) Helium diffusion in olivine from first principles calculations. Geochimica et Cosmochimica Acta 156, 145–153. https://doi.org/10.1016/j.gca.2015.01.023

Show in context

Helium may diffuse farther into mantle plume sources than W. Assuming the mantle plume source is stable on Gyr timescales (as expected for large low-shear wave velocity provinces, LLSVPs; e.g., Ferrick and Korenaga, 2023), the characteristic length scales of diffusion () for helium might be ∼40 km, if theoretical diffusion rates for the upper mantle (Wang et al., 2015) are extrapolated to lower mantle temperatures. For W, this length scale may be only 5–10 km (Ferrick and Korenaga, 2023), suggesting that mantle domains farther from the CMB may be characterised by He but not W isotopic anomalies.
View in article

Willbold, M., Elliott, T., Moorbath, S. (2011) The tungsten isotopic composition of the Earth’s mantle before the terminal bombardment. Nature 477, 195–198. https://doi.org/10.1038/nature10399

Show in context

Alternatively, the lowermost mantle could host ancient isotopic heterogeneities, either resulting from early silicate differentiation events (e.g., Touboul et al., 2012) or introduced during the late accretion of chondritic material with low μ¹⁸²W relative to the terrestrial mantle (e.g., Willbold et al., 2011).
View in article
The μ¹⁸²W of the convecting mantle may have decreased by ∼27 since Moon formation, based on lunar and Eoarchean terrestrial rock compositions (e.g., Willbold et al., 2011).
View in article
About 0.5 wt. % of late accreting chondritic material with high W and highly siderophile element (HSE) concentrations but low μ¹⁸²W might explain the decline in mantle μ¹⁸²W (Willbold et al., 2011).
View in article

Show in context

However, modern mantle plumes do not have positive μ¹⁸²W. Furthermore, the Baffin Island lavas have superchondritic ¹⁴³Nd/¹⁴⁴Nd and ¹⁷⁶Hf/¹⁷⁷Hf (Willhite et al., 2019), suggesting they are not derived from a primordial mantle component but a differentiated mantle reservoir.
View in article

top