Core-mantle chemical interaction via convection within thermochemical piles
Affiliations | Corresponding Author | Cite as | Funding information
Published by the European Association of Geochemistry
under Creative Commons License CC BY-NC-ND 4.0
Keywords: core-mantle interaction, mantle geochemistry, mantle convection
- Share this article
-
Article views:318Cumulative count of HTML views and PDF downloads.
- Download Citation
- Rights & Permissions
top
Abstract
Figures
 Figure 1 (a–f) Snapshots of the composition field for the case starting with a laterally homogeneous, 750 km thick metal-impregnated layer with the total metal of 1022 kg, and with Ra = 3 × 107, non-dimensional internal heating of 25, and the Frank-Kamenetskii parameter θ = 4.61. See Figure S-1 for the corresponding temperature field. (g) Summary of vertical trajectories of dense tracers, with three sample trajectories highlighted. |  Figure 2 Histograms of (a) transit time scale and (b) minimum distance to CMB, measured from the trajectories of dense tracers for the case shown in Figure 1. Also shown in (b) are the corresponding cumulative distribution function (CDF) in red, a fitted exponential CDF in red-dashed, and a reference uniform CDF in dotted. |  Figure 3 Summary of convection diagnostics for all model runs: (a) transit time scale and (b) amplification factor. Solid and open symbols for Ra = 107 and 3 × 107, respectively, and stars, squares, and triangles denote the cases with the Frank-Kamenetskii parameter θ of 4.61, 5.70, and 6.91, respectively. |
| Figure 1 | Figure 2 | Figure 3 |
top
Introduction
The silicate mantle and the metallic core interact both thermally and chemically at the core-mantle boundary (CMB). Thermal interaction manifests as core heat flux, which has long been studied by numerous authors because of its importance for plume dynamics (e.g., Sleep, 1990
Sleep, N.H. (1990) Hotspots and mantle plumes: Some phenomenology. Journal of Geophysical Research: Solid Earth 95, 6715–6736. https://doi.org/10.1029/JB095iB05p06715
), the geodynamo (e.g., Nimmo, 2015Nimmo, F. (2015) 8.02 - Energetics of the Core. In: Schubert, G. (Ed.) Treatise on Geophysics. Second Edition, Elsevier, Amsterdam, 8, 27–55. https://doi.org/10.1016/B978-0-444-53802-4.00139-1
), and Earth’s thermal evolution (e.g., Korenaga, 2008Korenaga, J. (2008) Urey ratio and the structure and evolution of Earth’s mantle. Reviews of Geophysics 46, RG2007. https://doi.org/10.1029/2007RG000241
). By contrast, less attention has traditionally been paid to chemical interactions, largely because chemical diffusion can be slower than thermal diffusion by many orders of magnitude and other possible transfer mechanisms are not well understood (e.g., Porcelli and Halliday, 2001Porcelli, D., Halliday, A.N. (2001) The core as a possible source of mantle helium. Earth and Planetary Science Letters 192, 45–56. https://doi.org/10.1016/S0012-821X(01)00418-6
; Walker and Walker, 2005Walker, R.J., Walker, D. (2005) Does the core leak? EOS 86, 237–242. https://doi.org/10.1029/2005EO250001
). However, recent years have witnessed a growing number of observational, experimental, and theoretical studies that consider the possibility of core-mantle chemical interaction to explain the geochemistry of ocean island basalts (OIBs) (e.g., Mundl et al., 2017Mundl, A., Touboul, M., Jackson, M.G., Day, J.M.D., Kurz, M.D., Lekic, V., Helz, R.T., Walker, R.J. (2017) Tungsten-182 heterogeneity in modern ocean island basalts. Science 356, 66–69. https://doi.org/10.1126/science.aal4179
; Rizo et al., 2019Rizo, 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
; Yoshino et al., 2020Yoshino, T., Makino, Y., Suzuki, T., Hirata, T. (2020) Grain boundary diffusion of W in lower mantle phase with implications for isotopic heterogeneity in oceanic island basalts by core-mantle interactions. Earth and Planetary Science Letters 530, 115887. https://doi.org/10.1016/j.epsl.2019.115887
; Olson and Sharp, 2022Olson, P.L., Sharp, Z.D. (2022) Primordial Helium-3 Exchange Between Earth’s Core and Mantle. Geochemistry, Geophysics, Geosystems 23, e2021GC009985. https://doi.org/10.1029/2021GC009985
; Ferrick and Korenaga, 2023bFerrick, A.L., Korenaga, J. (2023b) 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
; Deng and Du, 2023Deng, J., Du, Z. (2023) Primordial helium extracted from the Earth’s core through magnesium oxide exsolution. Nature Geoscience 16, 541–545. https://doi.org/10.1038/s41561-023-01182-7
; Horton et al., 2023Horton, F., Asimow, P.D., Farley, K.A., Curtice, J., Kurz, M.D., Blusztajn, J., Biasi, J.A., 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
). For example, negative μ182W anomalies reported for OIB (e.g., Mundl et al., 2017Mundl, A., Touboul, M., Jackson, M.G., Day, J.M.D., Kurz, M.D., Lekic, V., Helz, R.T., Walker, R.J. (2017) Tungsten-182 heterogeneity in modern ocean island basalts. Science 356, 66–69. https://doi.org/10.1126/science.aal4179
), especially in conjunction with their negative correlation with 3He/4He, are difficult to explain without some interactions with the core, which is considered to have a strongly negative μ182W value.The CMB region is physically and chemically heterogeneous, as indicated by the presence of large low shear-velocity provinces (LLSVPs) and ultra-low velocity zones (ULVZs) (e.g., Garnero et al., 2016
Garnero, E.J., McNamara, A.K., Shim, S.-H. (2016) Continent-sized anomalous zones with low seismic velocity at the base of Earth’s mantle. Nature Geoscience 9, 481–489. https://doi.org/10.1038/ngeo2733
). The LLSVPs appear to be the source regions for most large igneous provinces (Burke and Torsvik, 2004Burke, K., Torsvik, T.H. (2004) Derivation of Large Igneous Provinces of the past 200 million years from long-term heterogeneities in the deep mantle. Earth and Planetary Science Letters 227, 531–538. https://doi.org/10.1016/j.epsl.2004.09.015
), which in turn often serve as the precursors of hot spot magmatism (Richards et al., 1989Richards, M.A., Duncan, R.A., Courtillot, V.E. (1989) Flood Basalts and Hot-Spot Tracks: Plume Heads and Tails. Science 246, 103–107. https://doi.org/10.1126/science.246.4926.103
). Thus, the nature of core-mantle chemical interaction at the LLSVPs becomes important when considering how possible core signatures can be imparted to the OIB source mantle. Based on their apparent longevity (Torsvik et al., 2014Torsvik, T.H., van der Voo, R., Doubrovine, P.V., Burke, K., Steinberger, B., Ashwal, L.D., Trønnes, R.G., Webb, S.J., Bull, A.L. (2014) Deep mantle structure as a reference frame for movements in and on the Earth. Proceedings of the National Academy of Sciences 111, 8735–8740. https://doi.org/10.1073/pnas.1318135111
), the LLSVPs have been suggested to represent thermochemical piles compositionally denser than the ambient mantle. Whereas this presumed intrinsic density anomaly can stabilise thermochemical piles against the ambient mantle, convection can still take place within the piles themselves. The purpose of this paper is to investigate the potential role of this convection within a thermochemical pile, or “inter-pile convection”, in core-mantle chemical interaction. ULVZs may contribute to the geochemistry of the OIB source mantle, but given their global distribution and highly variable characteristics (e.g., Hansen et al., 2023Hansen, S.E., Garnero, E.J., Li, M., Shim, S.-H., Rost, S. (2023) Globally distributed subducted materials along the Earth’s core-mantle boundary: Implications for ultralow velocity zones. Science Advances 9, eadd4838. https://doi.org/10.1126/sciadv.add4838
), their role in hot spot magmatism is not very clear, and it is beyond the scope of this paper. Core-mantle interaction affects both the ambient mantle and thermochemical piles, but core signatures imparted to the former would be quickly diluted by convective mixing with the bulk mantle. By contrast, dense piles could potentially reside on the CMB for billions of years and acquire strong core signatures, and despite its limited surface coverage on the CMB (∼30 %), they could play a disproportionate role in creating chemical heterogeneities in the mantle. Because of inter-pile convection, however, their longevity does not necessarily translate to strong core signals in the geochemistry of OIB source mantle.Inter-pile convection has two competing effects. Without it, core-mantle chemical interactions would affect only the lowermost portion of the piles, most of which would remain inaccessible. With convection, the base of the piles can be brought directly to the top of the piles, facilitating sampling by mantle plumes. At the same time, convective mixing within piles dilutes the effect of core-mantle chemical interactions. Imparting core signatures on the OIB source mantle, therefore, depends on how directly inter-pile convection can bring materials from the bottom to the top and side boundaries, where mixing with the ambient mantle allows the entrainment of dense pile materials. Quantifying the efficiency of such convective transfer is, however, not straightforward because the volume of thermochemical piles can be time dependent; they could have been smaller if they represent accumulated subducted materials (e.g., Christensen and Hofmann, 1994
Christensen, U.R., Hofmann, A.W. (1994) Segregation of subducted oceanic crust in the convecting mantle. Journal of Geophysical Research: Solid Earth 99, 19867–19884. https://doi.org/10.1029/93JB03403
) or larger if they are primordial anomalies (e.g., Korenaga and Marchi, 2023Korenaga, J., Marchi, S. (2023) Vestiges of impact-driven three-phase mixing in the chemistry and structure of Earth’s mantle. Proceedings of the National Academy of Sciences 120, e2309181120. https://doi.org/10.1073/pnas.2309181120
). This fundamentally transient nature of inter-pile convection makes it difficult to build a generic model. In this study, therefore, we opt to present a case study by utilising our earlier numerical model of thermochemical piles (Korenaga and Marchi, 2023Korenaga, J., Marchi, S. (2023) Vestiges of impact-driven three-phase mixing in the chemistry and structure of Earth’s mantle. Proceedings of the National Academy of Sciences 120, e2309181120. https://doi.org/10.1073/pnas.2309181120
). Even though this model is based on a particular hypothesis for pile formation, our approach of characterising the nature of inter-pile convection is sufficiently general, allowing us to discuss core-mantle chemical interaction in the presence of inter-pile convection.top
Convection Diagnostics with Dense Tracers
Our convection model was designed to study the fate of a metal-impregnated layer formed in the mid-mantle resulting from a large late accretion impact (Korenaga and Marchi, 2023
Korenaga, J., Marchi, S. (2023) Vestiges of impact-driven three-phase mixing in the chemistry and structure of Earth’s mantle. Proceedings of the National Academy of Sciences 120, e2309181120. https://doi.org/10.1073/pnas.2309181120
) (Fig. 1a). The metal-impregnated layer contains ∼1 × 1022 kg of metal (corresponding to ∼1–2 % density anomalies), and it first sinks to the CMB by the Rayleigh-Taylor instability (Fig. 1b). The layer is then gradually entrained by mantle convection, increasing the abundance of highly siderophile elements (HSE) in the ambient mantle (Fig. 1c–f). Our hypothesis is that the LLSVPs represent the residue of this metal-impregnated layer. The details of our convection modelling are given in the Supplementary Information. Even though this convection model was originally built to test the feasibility of our hypothesis, our modelling results can also be utilised to discuss inter-pile convection in a more general context.Figure 1 (a–f) Snapshots of the composition field for the case starting with a laterally homogeneous, 750 km thick metal-impregnated layer with the total metal of 1022 kg, and with Ra = 3 × 107, non-dimensional internal heating of 25, and the Frank-Kamenetskii parameter θ = 4.61. See Figure S-1 for the corresponding temperature field. (g) Summary of vertical trajectories of dense tracers, with three sample trajectories highlighted.
The fate of the metal-impregnated layer is tracked by dense tracers that represent metallic components, and their trajectories contain valuable information on the nature of inter-pile convection (Fig. 1g). As the metal-impregnated layer initially descends to the CMB, most of dense tracers become part of thermochemical piles, and they circulate within the piles before being entrained by the ambient mantle. We focus on the dense tracers that initially sank to the bottom 10 % of the mantle and monitor their trajectories up to the time of their entrainment, te (defined as the time when a tracer enters the upper half of the mantle). The criterion of the bottom 10 % is chosen to easily identify dense particles that constitute long lived piles. For each trajectory, we measure the number of extrema (Nx) of its vertical coordinate as well as the minimum distance to the CMB. The transit time scale, τT, or the average period of particle motion within convection, is estimated as 2te/(Nx) (Fig. 2a).
Figure 2 Histograms of (a) transit time scale and (b) minimum distance to CMB, measured from the trajectories of dense tracers for the case shown in Figure 1. Also shown in (b) are the corresponding cumulative distribution function (CDF) in red, a fitted exponential CDF in red-dashed, and a reference uniform CDF in dotted.
To quantify how directly inter-pile convection can bring core signatures to its boundaries, we need to know the fate of mantle parcels that directly touch the CMB, and our inference is based on the statistics of the minimum distance to the CMB for dense tracers (Fig. 2b). The distribution of the minimum distance is usually similar to the exponential distribution. In thermal convection, the delamination of the bottom boundary layer is the source of upwelling, and simply bringing the bottom boundary layer to the top of the pile would result in a relatively uniform distribution of the minimum distance, within the length scale corresponding to the thickness of the boundary layer. We define the amplification factor as a measure of deviations from this expected uniform distribution (see the Supplementary Information for details). The amplification factor of three, for example, means that entraining the mantle parcels that have directly touched the CMB is three times more likely than expected from the simple upwelling of a bottom boundary layer.
As expected, the transit time scale is basically a function of the Rayleigh number (Fig. 3a), and a higher Rayleigh number (i.e. lower viscosity) results in a shorter transit time scale. The transit time scale does not vary much as a function of the entrainment ratio, which measures the fraction of entrained tracers among the dense tracers that initially sank to the bottom 10 % of the mantle. The statistics of the amplification factor are more variable (Fig. 3b). Considerable scatters are seen at low entrainment ratios because of the overpopulation of dense tracers in the piles. In these cases, dense tracers located near the CMB are unlikely to ever be entrained, and thus the distribution of the minimum distance for the entrained tracers does not necessarily take its maximum at zero distance. For relatively high entrainment ratios (>0.5), the amplification factor is consistently ∼3 to ∼4.
Figure 3 Summary of convection diagnostics for all model runs: (a) transit time scale and (b) amplification factor. Solid and open symbols for Ra = 107 and 3 × 107, respectively, and stars, squares, and triangles denote the cases with the Frank-Kamenetskii parameter θ of 4.61, 5.70, and 6.91, respectively.
top
Discussion: Origins of Thermochemical Piles
The characterisation of inter-pile convection allows us to evaluate the strengths and the weaknesses of the existing hypotheses put forward for the origin of thermochemical piles. The diagnostics of inter-pile convection given in the previous section are based on one scenario for pile formation, but the main points of our discussion should remain valid as long as inter-pile convection is concerned. Here we focus on whether a given hypothesis can explain the correlation between the He and W isotopes observed in OIBs (Mundl et al., 2017
Mundl, A., Touboul, M., Jackson, M.G., Day, J.M.D., Kurz, M.D., Lekic, V., Helz, R.T., Walker, R.J. (2017) Tungsten-182 heterogeneity in modern ocean island basalts. Science 356, 66–69. https://doi.org/10.1126/science.aal4179
; Mundl-Petermeier et al., 2020Mundl-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
). A recent modelling study suggests that both He and W isotope signatures can be explained by core-mantle chemical interaction (Ferrick and Korenaga, 2023bFerrick, A.L., Korenaga, J. (2023b) 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
), but this model treats the OIB source mantle (or equivalently, thermochemical piles) as a static reservoir lying on the CMB. Based on experiments on W diffusivity through lower mantle minerals (Yoshino et al., 2020Yoshino, T., Makino, Y., Suzuki, T., Hirata, T. (2020) Grain boundary diffusion of W in lower mantle phase with implications for isotopic heterogeneity in oceanic island basalts by core-mantle interactions. Earth and Planetary Science Letters 530, 115887. https://doi.org/10.1016/j.epsl.2019.115887
), the W isotopic signature of the core can diffuse into the bottom ∼20 km of the mantle over four billion years, but this time scale for chemical interaction is much longer than the transit time scale (Fig. 3a), which is a more appropriate diffusion time scale in the presence of inter-pile convection. With a residence time of only 200 Myr, for example, the diffusion length would be reduced to ∼4.5 km. At the same time, inter-pile convection can efficiently deliver this thin core-affected layer to the OIB source mantle, so it is still possible to reproduce the observed isotope correlation by core-mantle interaction. In this case, scatters in the observed correlation are likely caused by the stochastic nature of convective mass transfer, instead of differences in the sampled depth as originally proposed (Ferrick and Korenaga, 2023bFerrick, A.L., Korenaga, J. (2023b) 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
).Inter-pile convection can transport core signatures over a few hundred kilometres, but whether such mass transfer would be quantitatively sufficient still depends on the efficiency of diffusive interaction across the CMB. Consider, for example, the Hawaiian plume, which is the most volumetrically significant plume with its plume buoyancy flux estimated to be ∼5 × 103 kg s−1 (Sleep, 1990
Sleep, N.H. (1990) Hotspots and mantle plumes: Some phenomenology. Journal of Geophysical Research: Solid Earth 95, 6715–6736. https://doi.org/10.1029/JB095iB05p06715
; King and Adam, 2014King, S.D., Adam, C. (2014) Hotspot swells revisited. Physics of the Earth and Planetary Interiors 235, 66–83. https://doi.org/10.1016/j.pepi.2014.07.006
). By dividing the buoyancy flux with the product of thermal expansivity (3 × 10−5 K−1) and a temperature contrast (200 K), we obtain the corresponding mass flux of ∼8 × 105 kg s−1. If the average μ182W anomaly of Hawaiian basalts is −5 (Willhite et al., 2024Willhite, L.N., Finlayson, V.A., Walker, R.J. (2024) Evolution of tungsten isotope systematics in the Mauna Kea volcano provides new constraints on anomalous μ182W and high 3He/4He in the mantle. Earth and Planetary Science Letters 640, 118795. https://doi.org/10.1016/j.epsl.2024.118795
) and the core μ182W anomaly is −220, ∼2.3 % of the source mantle must diffusively interact with the core. To provide such a source mantle for the duration of 20 Myr, for example, the mass of core-affected material needs to be ∼1011 kg. This translates to the thickness of core-affected layer of ∼670 m, with the density of the lowermost mantle (5500 kg m−3) and the radius of the plume stem area of 500 km. For the residence time along the CMB of 200 Myr, this diffusion length scale requires W diffusivity to be greater than 1.7 × 10−11 m2 s−1. Higher W diffusivities have been suggested for the CMB conditions (∼8 × 10−8 m2 s−1) (Yoshino et al., 2020Yoshino, T., Makino, Y., Suzuki, T., Hirata, T. (2020) Grain boundary diffusion of W in lower mantle phase with implications for isotopic heterogeneity in oceanic island basalts by core-mantle interactions. Earth and Planetary Science Letters 530, 115887. https://doi.org/10.1016/j.epsl.2019.115887
), but this involves considerable extrapolations from the experimental conditions of 1600–1900 K and 25 GPa to the CMB conditions of ∼4000 K and 130 GPa. Thus, if the actual W diffusivity at the CMB conditions were too low, core-mantle interaction would be insufficient to explain the observed W isotopic signatures. The same consideration applies to the He isotopic signatures, but the situation is different from imparting core W isotopic signatures, which involves only isotope exchange across the CMB with no net W flux. On the other hand, the transfer of core He isotopic signatures depends on the net diffusional flux of He across the CMB, which is proportional not only to the He diffusivity but also to the He concentration in the core, both of which are poorly constrained (Olson and Sharp, 2022Olson, P.L., Sharp, Z.D. (2022) Primordial Helium-3 Exchange Between Earth’s Core and Mantle. Geochemistry, Geophysics, Geosystems 23, e2021GC009985. https://doi.org/10.1029/2021GC009985
).Thus, W diffusivity at the CMB conditions, along with the transit time scale of inter-pile convection, may be the key to distinguishing different formation hypotheses for thermochemical piles. If the W diffusivity is too low to explain the magnitude of hot spot magmatism with W isotope anomalies, pile formation mechanisms without involving metallic components (e.g., Christensen and Hofmann, 1994
Christensen, U.R., Hofmann, A.W. (1994) Segregation of subducted oceanic crust in the convecting mantle. Journal of Geophysical Research: Solid Earth 99, 19867–19884. https://doi.org/10.1029/93JB03403
; Yuan et al., 2023Yuan, Q., Li, M., Desch, S.J., Ko, B., Deng, H., Garnero, E.J., Gabriel, T.S.J., Kegerreis, J.A., Miyazaki, Y., Eke, V., Asimow, P.D. (2023) Moon-forming impactor as a source of Earth’s basal mantle anomalies. Nature 623, 95–99. https://doi.org/10.1038/s41586-023-06589-1
) would be difficult to explain the observed W isotope signatures. By contrast, W diffusivity becomes irrelevant for pile formation by late accretion (Korenaga and Marchi, 2023Korenaga, J., Marchi, S. (2023) Vestiges of impact-driven three-phase mixing in the chemistry and structure of Earth’s mantle. Proceedings of the National Academy of Sciences 120, e2309181120. https://doi.org/10.1073/pnas.2309181120
) because the core-like W isotope signatures (μ182W of −220) are already embedded in the piles from the outset. Note that, contrary to our earlier conjecture (Korenaga and Marchi, 2023Korenaga, J., Marchi, S. (2023) Vestiges of impact-driven three-phase mixing in the chemistry and structure of Earth’s mantle. Proceedings of the National Academy of Sciences 120, e2309181120. https://doi.org/10.1073/pnas.2309181120
), pile formation by late accretion can still satisfy the observed absence of correlation between μ182W and HSE concentrations (Mundl et al., 2017Mundl, A., Touboul, M., Jackson, M.G., Day, J.M.D., Kurz, M.D., Lekic, V., Helz, R.T., Walker, R.J. (2017) Tungsten-182 heterogeneity in modern ocean island basalts. Science 356, 66–69. https://doi.org/10.1126/science.aal4179
). To be entrained and become part of the OIB source mantle, the density anomaly of pile materials must be compensated by the thermal buoyancy of hot ambient mantle, and as a result, pile materials can constitute only ∼10 % of the OIB source mantle (Ferrick and Korenaga, 2023bFerrick, A.L., Korenaga, J. (2023b) 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
). Mixing with the ambient mantle during entrainment thus sufficiently dilutes the effect of metallic components. A major challenge for the late accretion-based hypothesis is instead how to explain the correlation between He and W isotope anomalies. Because W isotope signatures would be sufficiently strong everywhere within the piles, the magnitude of He isotope signatures need to be similarly ubiquitous, requiring efficient 3He transport from the core to the mantle. The slopes of the correlations between μ182W and 3He/4He vary from plume to plume (or even within a single plume), and some plume-derived systems (e.g., Baffin Bay) are characterised by very high 3He/4He but no W anomaly (e.g., Mundl-Petermeier et al., 2020Mundl-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
; Willhite et al., 2024Willhite, L.N., Finlayson, V.A., Walker, R.J. (2024) Evolution of tungsten isotope systematics in the Mauna Kea volcano provides new constraints on anomalous μ182W and high 3He/4He in the mantle. Earth and Planetary Science Letters 640, 118795. https://doi.org/10.1016/j.epsl.2024.118795
). It is to be seen whether such complexity of observations can be explained by the combination of core-mantle interaction and inter-pile convection.The threshold diffusivity for W is suggested to be on the order of 10−11 m2 s−1 in the above, but there are a few important assumptions, the validity of which requires more careful analyses of hot spot magmatism. First, we need to accurately estimate the spatial and temporal average of a W isotope anomaly for a given mantle plume. If very negative μ182W signatures are spatially or temporally localised features, the required volume of core-affected materials could be substantially reduced. Second, plume buoyancy fluxes themselves are subject to non-trivial uncertainties (e.g., King and Adam, 2014
King, S.D., Adam, C. (2014) Hotspot swells revisited. Physics of the Earth and Planetary Interiors 235, 66–83. https://doi.org/10.1016/j.pepi.2014.07.006
). It is also important to consider various types of diffusive interactions. For example, the present day core-side temperature of the CMB is close to the mantle solidus, so the lowermost part of the mantle is likely to have been partially molten when the core was hotter in the past. This tendency is even more important for the piles because they are generally hotter than the ambient mantle. Diffusive transport of core signatures would be rapid through a partially molten mantle. Though it seems difficult to maintain a partially molten state as melt is denser than the coexisting solid phase at the CMB conditions (e.g., Nomura et al., 2011Nomura, R., Ozawa, H., Tateno, S., Hirose, K., Hernlund, J., Muto, S., Ishii, H., Hiraoka, N. (2011) Spin crossover and iron-rich silicate melt in the Earth’s deep mantle. Nature 473, 199–202. https://doi.org/10.1038/nature09940
) and likely drains downward, it may be possible to avoid downward percolation by convective stirring (Hernlund and Jellinek, 2010Hernlund, J.W., Jellinek, A.M. (2010) Dynamics and structure of a stirred partially molten ultralow-velocity zone. Earth and Planetary Science Letters 296, 1–8. https://doi.org/10.1016/j.epsl.2010.04.027
).Direct mass transfer from the core can also be treated as a diffusive process by defining an ‘effective’ diffusion coefficient. Whereas the penetration of molten iron into the lower mantle by morphological instability may be unlikely (Yoshino, 2019
Yoshino, T. (2019) Penetration of molten iron alloy into the lower mantle phase. Comptes Rendus Géoscience 351, 171–181. https://doi.org/10.1016/j.crte.2018.06.013
), MgO exsolution from a cooling core can still happen if the core was initially saturated with Mg (Deng and Du, 2023Deng, J., Du, Z. (2023) Primordial helium extracted from the Earth’s core through magnesium oxide exsolution. Nature Geoscience 16, 541–545. https://doi.org/10.1038/s41561-023-01182-7
). This mechanism is important for our discussion because efficient exsolution could result in a high effective W diffusivity, but its efficacy is still uncertain. One important point is that a substantial fraction of an impactor’s core would directly merge with the target’s core (Marchi et al., 2018Marchi, 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
), i.e. without chemically equilibrating with the target’s mantle. Thus, the early core is not necessarily saturated with Mg, and the weak temperature dependence of Mg solubility in the core (Du et al., 2019Du, Z., Boujibar, A., Driscoll, P., Fei, Y. (2019) Experimental Constraints on an MgO Exsolution-Driven Geodynamo. Geophysical Research Letters 46, 7379–7385. https://doi.org/10.1029/2019GL083017
) further reduces the likelihood of MgO exsolution. MgO exsolution was originally proposed to explain the early geodynamo (O’Rourke and Stevenson, 2016O’Rourke, J.G., Stevenson, D.J. (2016) Powering Earth’s dynamo with magnetism precipitation from the core. Nature 529, 387–389. https://doi.org/10.1038/nature16495
), but it is also possible to explain the early geodynamo without invoking MgO exsolution (Ferrick and Korenaga, 2023aFerrick, A.L., Korenaga, J. (2023a) Defining Earth’s elusive thermal budget in the presence of a hidden reservoir. Earth and Planetary Science Letters 601, 117893. https://doi.org/10.1016/j.epsl.2022.117893
).Delineating the origin of thermochemical piles using mantle geochemistry will thus require further observational, experimental, and theoretical efforts. It is hoped that our exploration of inter-pile convection will serve as a useful theoretical reference when assimilating the outcomes of future research.
top
Acknowledgements
This research was supported in part by the U.S. National Science Foundation EAR-2102777 and EAR-2102571. The authors thank Richard Walker and an anonymous reviewer for their constructive comments.
Editor: Anat Shahar
top
References
Burke, K., Torsvik, T.H. (2004) Derivation of Large Igneous Provinces of the past 200 million years from long-term heterogeneities in the deep mantle. Earth and Planetary Science Letters 227, 531–538. https://doi.org/10.1016/j.epsl.2004.09.015
Show in context The LLSVPs appear to be the source regions for most large igneous provinces (Burke and Torsvik, 2004), which in turn often serve as the precursors of hot spot magmatism (Richards et al., 1989).
View in article
Christensen, U.R., Hofmann, A.W. (1994) Segregation of subducted oceanic crust in the convecting mantle. Journal of Geophysical Research: Solid Earth 99, 19867–19884. https://doi.org/10.1029/93JB03403
Show in context Quantifying the efficiency of such convective transfer is, however, not straightforward because the volume of thermochemical piles can be time dependent; they could have been smaller if they represent accumulated subducted materials (e.g., Christensen and Hofmann, 1994) or larger if they are primordial anomalies (e.g., Korenaga and Marchi, 2023).
View in article
If the W diffusivity is too low to explain the magnitude of hot spot magmatism with W isotope anomalies, pile formation mechanisms without involving metallic components (e.g., Christensen and Hofmann, 1994; Yuan et al., 2023) would be difficult to explain the observed W isotope signatures.
View in article
Deng, J., Du, Z. (2023) Primordial helium extracted from the Earth’s core through magnesium oxide exsolution. Nature Geoscience 16, 541–545. https://doi.org/10.1038/s41561-023-01182-7
Show in context However, recent years have witnessed a growing number of observational, experimental, and theoretical studies that consider the possibility of core-mantle chemical interaction to explain the geochemistry of ocean island basalts (OIBs) (e.g., Mundl et al., 2017; Rizo et al., 2019; Yoshino et al., 2020; Olson and Sharp, 2022; Ferrick and Korenaga, 2023b; Deng and Du, 2023; Horton et al., 2023).
View in article
Whereas the penetration of molten iron into the lower mantle by morphological instability may be unlikely (Yoshino, 2019), MgO exsolution from a cooling core can still happen if the core was initially saturated with Mg (Deng and Du, 2023).
View in article
Du, Z., Boujibar, A., Driscoll, P., Fei, Y. (2019) Experimental Constraints on an MgO Exsolution-Driven Geodynamo. Geophysical Research Letters 46, 7379–7385. https://doi.org/10.1029/2019GL083017
Show in context Thus, the early core is not necessarily saturated with Mg, and the weak temperature dependence of Mg solubility in the core (Du et al., 2019) further reduces the likelihood of MgO exsolution. MgO exsolution was originally proposed to explain the early geodynamo (O’Rourke and Stevenson, 2016), but it is also possible to explain the early geodynamo without invoking MgO exsolution (Ferrick and Korenaga, 2023a).
View in article
Ferrick, A.L., Korenaga, J. (2023a) Defining Earth’s elusive thermal budget in the presence of a hidden reservoir. Earth and Planetary Science Letters 601, 117893. https://doi.org/10.1016/j.epsl.2022.117893
Show in context Thus, the early core is not necessarily saturated with Mg, and the weak temperature dependence of Mg solubility in the core (Du et al., 2019) further reduces the likelihood of MgO exsolution. MgO exsolution was originally proposed to explain the early geodynamo (O’Rourke and Stevenson, 2016), but it is also possible to explain the early geodynamo without invoking MgO exsolution (Ferrick and Korenaga, 2023a).
View in article
Ferrick, A.L., Korenaga, J. (2023b) 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
Show in context However, recent years have witnessed a growing number of observational, experimental, and theoretical studies that consider the possibility of core-mantle chemical interaction to explain the geochemistry of ocean island basalts (OIBs) (e.g., Mundl et al., 2017; Rizo et al., 2019; Yoshino et al., 2020; Olson and Sharp, 2022; Ferrick and Korenaga, 2023b; Deng and Du, 2023; Horton et al., 2023).
View in article
A recent modelling study suggests that both He and W isotope signatures can be explained by core-mantle chemical interaction (Ferrick and Korenaga, 2023b), but this model treats the OIB source mantle (or equivalently, thermochemical piles) as a static reservoir lying on the CMB.
View in article
In this case, scatters in the observed correlation are likely caused by the stochastic nature of convective mass transfer, instead of differences in the sampled depth as originally proposed (Ferrick and Korenaga, 2023b).
View in article
To be entrained and become part of the OIB source mantle, the density anomaly of pile materials must be compensated by the thermal buoyancy of hot ambient mantle, and as a result, pile materials can constitute only ∼10 % of the OIB source mantle (Ferrick and Korenaga, 2023b).
View in article
Garnero, E.J., McNamara, A.K., Shim, S.-H. (2016) Continent-sized anomalous zones with low seismic velocity at the base of Earth’s mantle. Nature Geoscience 9, 481–489. https://doi.org/10.1038/ngeo2733
Show in context The CMB region is physically and chemically heterogeneous, as indicated by the presence of large low shear-velocity provinces (LLSVPs) and ultra-low velocity zones (ULVZs) (e.g., Garnero et al., 2016).
View in article
Hansen, S.E., Garnero, E.J., Li, M., Shim, S.-H., Rost, S. (2023) Globally distributed subducted materials along the Earth’s core-mantle boundary: Implications for ultralow velocity zones. Science Advances 9, eadd4838. https://doi.org/10.1126/sciadv.add4838
Show in context ULVZs may contribute to the geochemistry of the OIB source mantle, but given their global distribution and highly variable characteristics (e.g., Hansen et al., 2023), their role in hot spot magmatism is not very clear, and it is beyond the scope of this paper.
View in article
Hernlund, J.W., Jellinek, A.M. (2010) Dynamics and structure of a stirred partially molten ultralow-velocity zone. Earth and Planetary Science Letters 296, 1–8. https://doi.org/10.1016/j.epsl.2010.04.027
Show in context Though it seems difficult to maintain a partially molten state as melt is denser than the coexisting solid phase at the CMB conditions (e.g., Nomura et al., 2011) and likely drains downward, it may be possible to avoid downward percolation by convective stirring (Hernlund and Jellinek, 2010).
View in article
Horton, F., Asimow, P.D., Farley, K.A., Curtice, J., Kurz, M.D., Blusztajn, J., Biasi, J.A., 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
Show in context However, recent years have witnessed a growing number of observational, experimental, and theoretical studies that consider the possibility of core-mantle chemical interaction to explain the geochemistry of ocean island basalts (OIBs) (e.g., Mundl et al., 2017; Rizo et al., 2019; Yoshino et al., 2020; Olson and Sharp, 2022; Ferrick and Korenaga, 2023b; Deng and Du, 2023; Horton et al., 2023).
View in article
King, S.D., Adam, C. (2014) Hotspot swells revisited. Physics of the Earth and Planetary Interiors 235, 66–83. https://doi.org/10.1016/j.pepi.2014.07.006
Show in context Consider, for example, the Hawaiian plume, which is the most volumetrically significant plume with its plume buoyancy flux estimated to be ∼5 × 103 kg s−1 (Sleep, 1990; King and Adam, 2014).
View in article
Second, plume buoyancy fluxes themselves are subject to non-trivial uncertainties (e.g., King and Adam, 2014).
View in article
Korenaga, J. (2008) Urey ratio and the structure and evolution of Earth’s mantle. Reviews of Geophysics 46, RG2007. https://doi.org/10.1029/2007RG000241
Show in context Thermal interaction manifests as core heat flux, which has long been studied by numerous authors because of its importance for plume dynamics (e.g., Sleep, 1990), the geodynamo (e.g., Nimmo, 2015), and Earth’s thermal evolution (e.g., Korenaga, 2008).
View in article
Korenaga, J., Marchi, S. (2023) Vestiges of impact-driven three-phase mixing in the chemistry and structure of Earth’s mantle. Proceedings of the National Academy of Sciences 120, e2309181120. https://doi.org/10.1073/pnas.2309181120
Show in context Quantifying the efficiency of such convective transfer is, however, not straightforward because the volume of thermochemical piles can be time dependent; they could have been smaller if they represent accumulated subducted materials (e.g., Christensen and Hofmann, 1994) or larger if they are primordial anomalies (e.g., Korenaga and Marchi, 2023).
View in article
In this study, therefore, we opt to present a case study by utilising our earlier numerical model of thermochemical piles (Korenaga and Marchi, 2023).
View in article
Our convection model was designed to study the fate of a metal-impregnated layer formed in the mid-mantle resulting from a large late accretion impact (Korenaga and Marchi, 2023) (Fig. 1a).
View in article
By contrast, W diffusivity becomes irrelevant for pile formation by late accretion (Korenaga and Marchi, 2023) because the core-like W isotope signatures (μ182W of −220) are already embedded in the piles from the outset.
View in article
Note that, contrary to our earlier conjecture (Korenaga and Marchi, 2023), pile formation by late accretion can still satisfy the observed absence of correlation between μ182W and HSE concentrations (Mundl et al., 2017).
View in article
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
Show in context One important point is that a substantial fraction of an impactor’s core would directly merge with the target’s core (Marchi et al., 2018), i.e. without chemically equilibrating with the target’s mantle.
View in article
Mundl, A., Touboul, M., Jackson, M.G., Day, J.M.D., Kurz, M.D., Lekic, V., Helz, R.T., Walker, R.J. (2017) Tungsten-182 heterogeneity in modern ocean island basalts. Science 356, 66–69. https://doi.org/10.1126/science.aal4179
Show in context However, recent years have witnessed a growing number of observational, experimental, and theoretical studies that consider the possibility of core-mantle chemical interaction to explain the geochemistry of ocean island basalts (OIBs) (e.g., Mundl et al., 2017; Rizo et al., 2019; Yoshino et al., 2020; Olson and Sharp, 2022; Ferrick and Korenaga, 2023b; Deng and Du, 2023; Horton et al., 2023).
View in article
For example, negative μ182W anomalies reported for OIB (e.g., Mundl et al., 2017), especially in conjunction with their negative correlation with 3He/4He, are difficult to explain without some interactions with the core, which is considered to have a strongly negative μ182W value.
View in article
Here we focus on whether a given hypothesis can explain the correlation between the He and W isotopes observed in OIBs (Mundl et al., 2017; Mundl-Petermeier et al., 2020).
View in article
Note that, contrary to our earlier conjecture (Korenaga and Marchi, 2023), pile formation by late accretion can still satisfy the observed absence of correlation between μ182W and HSE concentrations (Mundl et al., 2017).
View in article
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
Show in context Here we focus on whether a given hypothesis can explain the correlation between the He and W isotopes observed in OIBs (Mundl et al., 2017; Mundl-Petermeier et al., 2020).
View in article
The slopes of the correlations between μ182W and 3He/4He vary from plume to plume (or even within a single plume), and some plume-derived systems (e.g., Baffin Bay) are characterised by very high 3He/4He but no W anomaly (e.g., Mundl-Petermeier et al., 2020; Willhite et al., 2024).
View in article
Nimmo, F. (2015) 8.02 - Energetics of the Core. In: Schubert, G. (Ed.) Treatise on Geophysics. Second Edition, Elsevier, Amsterdam, 8, 27–55. https://doi.org/10.1016/B978-0-444-53802-4.00139-1
Show in context Thermal interaction manifests as core heat flux, which has long been studied by numerous authors because of its importance for plume dynamics (e.g., Sleep, 1990), the geodynamo (e.g., Nimmo, 2015), and Earth’s thermal evolution (e.g., Korenaga, 2008).
View in article
Nomura, R., Ozawa, H., Tateno, S., Hirose, K., Hernlund, J., Muto, S., Ishii, H., Hiraoka, N. (2011) Spin crossover and iron-rich silicate melt in the Earth’s deep mantle. Nature 473, 199–202. https://doi.org/10.1038/nature09940
Show in context Though it seems difficult to maintain a partially molten state as melt is denser than the coexisting solid phase at the CMB conditions (e.g., Nomura et al., 2011) and likely drains downward, it may be possible to avoid downward percolation by convective stirring (Hernlund and Jellinek, 2010).
View in article
Olson, P.L., Sharp, Z.D. (2022) Primordial Helium-3 Exchange Between Earth’s Core and Mantle. Geochemistry, Geophysics, Geosystems 23, e2021GC009985. https://doi.org/10.1029/2021GC009985
Show in context However, recent years have witnessed a growing number of observational, experimental, and theoretical studies that consider the possibility of core-mantle chemical interaction to explain the geochemistry of ocean island basalts (OIBs) (e.g., Mundl et al., 2017; Rizo et al., 2019; Yoshino et al., 2020; Olson and Sharp, 2022; Ferrick and Korenaga, 2023b; Deng and Du, 2023; Horton et al., 2023).
View in article
On the other hand, the transfer of core He isotopic signatures depends on the net diffusional flux of He across the CMB, which is proportional not only to the He diffusivity but also to the He concentration in the core, both of which are poorly constrained (Olson and Sharp, 2022).
View in article
O’Rourke, J.G., Stevenson, D.J. (2016) Powering Earth’s dynamo with magnetism precipitation from the core. Nature 529, 387–389. https://doi.org/10.1038/nature16495
Show in context Thus, the early core is not necessarily saturated with Mg, and the weak temperature dependence of Mg solubility in the core (Du et al., 2019) further reduces the likelihood of MgO exsolution. MgO exsolution was originally proposed to explain the early geodynamo (O’Rourke and Stevenson, 2016), but it is also possible to explain the early geodynamo without invoking MgO exsolution (Ferrick and Korenaga, 2023a).
View in article
Porcelli, D., Halliday, A.N. (2001) The core as a possible source of mantle helium. Earth and Planetary Science Letters 192, 45–56. https://doi.org/10.1016/S0012-821X(01)00418-6
Show in context By contrast, less attention has traditionally been paid to chemical interactions, largely because chemical diffusion can be slower than thermal diffusion by many orders of magnitude and other possible transfer mechanisms are not well understood (e.g., Porcelli and Halliday, 2001; Walker and Walker, 2005).
View in article
Richards, M.A., Duncan, R.A., Courtillot, V.E. (1989) Flood Basalts and Hot-Spot Tracks: Plume Heads and Tails. Science 246, 103–107. https://doi.org/10.1126/science.246.4926.103
Show in context The LLSVPs appear to be the source regions for most large igneous provinces (Burke and Torsvik, 2004), which in turn often serve as the precursors of hot spot magmatism (Richards et al., 1989).
View in article
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
Show in context However, recent years have witnessed a growing number of observational, experimental, and theoretical studies that consider the possibility of core-mantle chemical interaction to explain the geochemistry of ocean island basalts (OIBs) (e.g., Mundl et al., 2017; Rizo et al., 2019; Yoshino et al., 2020; Olson and Sharp, 2022; Ferrick and Korenaga, 2023b; Deng and Du, 2023; Horton et al., 2023).
View in article
Sleep, N.H. (1990) Hotspots and mantle plumes: Some phenomenology. Journal of Geophysical Research: Solid Earth 95, 6715–6736. https://doi.org/10.1029/JB095iB05p06715
Show in context Thermal interaction manifests as core heat flux, which has long been studied by numerous authors because of its importance for plume dynamics (e.g., Sleep, 1990), the geodynamo (e.g., Nimmo, 2015), and Earth’s thermal evolution (e.g., Korenaga, 2008).
View in article
Consider, for example, the Hawaiian plume, which is the most volumetrically significant plume with its plume buoyancy flux estimated to be ∼5 × 103 kg s−1 (Sleep, 1990; King and Adam, 2014).
View in article
Torsvik, T.H., van der Voo, R., Doubrovine, P.V., Burke, K., Steinberger, B., Ashwal, L.D., Trønnes, R.G., Webb, S.J., Bull, A.L. (2014) Deep mantle structure as a reference frame for movements in and on the Earth. Proceedings of the National Academy of Sciences 111, 8735–8740. https://doi.org/10.1073/pnas.1318135111
Show in context Based on their apparent longevity (Torsvik et al., 2014), the LLSVPs have been suggested to represent thermochemical piles compositionally denser than the ambient mantle.
View in article
Walker, R.J., Walker, D. (2005) Does the core leak? EOS 86, 237–242. https://doi.org/10.1029/2005EO250001
Show in context By contrast, less attention has traditionally been paid to chemical interactions, largely because chemical diffusion can be slower than thermal diffusion by many orders of magnitude and other possible transfer mechanisms are not well understood (e.g., Porcelli and Halliday, 2001; Walker and Walker, 2005).
View in article
Willhite, L.N., Finlayson, V.A., Walker, R.J. (2024) Evolution of tungsten isotope systematics in the Mauna Kea volcano provides new constraints on anomalous μ182W and high 3He/4He in the mantle. Earth and Planetary Science Letters 640, 118795. https://doi.org/10.1016/j.epsl.2024.118795
Show in context If the average μ182W anomaly of Hawaiian basalts is −5 (Willhite et al., 2024) and the core μ182W anomaly is −220, ∼2.3 % of the source mantle must diffusively interact with the core.
View in article
The slopes of the correlations between μ182W and 3He/4He vary from plume to plume (or even within a single plume), and some plume-derived systems (e.g., Baffin Bay) are characterised by very high 3He/4He but no W anomaly (e.g., Mundl-Petermeier et al., 2020; Willhite et al., 2024).
View in article
Yoshino, T. (2019) Penetration of molten iron alloy into the lower mantle phase. Comptes Rendus Géoscience 351, 171–181. https://doi.org/10.1016/j.crte.2018.06.013
Show in context Whereas the penetration of molten iron into the lower mantle by morphological instability may be unlikely (Yoshino, 2019), MgO exsolution from a cooling core can still happen if the core was initially saturated with Mg (Deng and Du, 2023).
View in article
Yoshino, T., Makino, Y., Suzuki, T., Hirata, T. (2020) Grain boundary diffusion of W in lower mantle phase with implications for isotopic heterogeneity in oceanic island basalts by core-mantle interactions. Earth and Planetary Science Letters 530, 115887. https://doi.org/10.1016/j.epsl.2019.115887
Show in context However, recent years have witnessed a growing number of observational, experimental, and theoretical studies that consider the possibility of core-mantle chemical interaction to explain the geochemistry of ocean island basalts (OIBs) (e.g., Mundl et al., 2017; Rizo et al., 2019; Yoshino et al., 2020; Olson and Sharp, 2022; Ferrick and Korenaga, 2023b; Deng and Du, 2023; Horton et al., 2023).
View in article
Based on experiments on W diffusivity through lower mantle minerals (Yoshino et al., 2020), the W isotopic signature of the core can diffuse into the bottom ∼20 km of the mantle over four billion years, but this time scale for chemical interaction is much longer than the transit time scale (Fig. 3a), which is a more appropriate diffusion time scale in the presence of inter-pile convection.
View in article
Higher W diffusivities have been suggested for the CMB conditions (∼8 × 10−8 m2 s−1) (Yoshino et al., 2020), but this involves considerable extrapolations from the experimental conditions of 1600–1900 K and 25 GPa to the CMB conditions of ∼4000 K and 130 GPa.
View in article
Yuan, Q., Li, M., Desch, S.J., Ko, B., Deng, H., Garnero, E.J., Gabriel, T.S.J., Kegerreis, J.A., Miyazaki, Y., Eke, V., Asimow, P.D. (2023) Moon-forming impactor as a source of Earth’s basal mantle anomalies. Nature 623, 95–99. https://doi.org/10.1038/s41586-023-06589-1
Show in context If the W diffusivity is too low to explain the magnitude of hot spot magmatism with W isotope anomalies, pile formation mechanisms without involving metallic components (e.g., Christensen and Hofmann, 1994; Yuan et al., 2023) would be difficult to explain the observed W isotope signatures.
View in article
top
Supplementary Information
The Supplementary Information includes:
Download the Supplementary Information (PDF)
Figures
Figure 1 (a–f) Snapshots of the composition field for the case starting with a laterally homogeneous, 750 km thick metal-impregnated layer with the total metal of 1022 kg, and with Ra = 3 × 107, non-dimensional internal heating of 25, and the Frank-Kamenetskii parameter θ = 4.61. See Figure S-1 for the corresponding temperature field. (g) Summary of vertical trajectories of dense tracers, with three sample trajectories highlighted.
Figure 2 Histograms of (a) transit time scale and (b) minimum distance to CMB, measured from the trajectories of dense tracers for the case shown in Figure 1. Also shown in (b) are the corresponding cumulative distribution function (CDF) in red, a fitted exponential CDF in red-dashed, and a reference uniform CDF in dotted.
Figure 3 Summary of convection diagnostics for all model runs: (a) transit time scale and (b) amplification factor. Solid and open symbols for Ra = 107 and 3 × 107, respectively, and stars, squares, and triangles denote the cases with the Frank-Kamenetskii parameter θ of 4.61, 5.70, and 6.91, respectively.