Zhang et al. (2021) measured amphibole and rutile in lower-crustal cumulates from the Kohistan and Gangdese arcs and showed that, although amphibole has high Nb/Ta, its absolute Nb and Ta contents are too low to be the principal sink, while small modal amounts of rutile can drive melt Nb/Ta from MORB-like to continental-crust values. They left open, first, whether rutile-bearing arc cumulates are volumetrically abundant enough across global arcs and geologic time to account for the continental crust's low Nb/Ta, and second, that the subchondritic Nb/Ta of the bulk silicate Earth requires additional mechanisms or reservoirs whose nature and location remain unidentified. As of August 2026 both cores remain open: the rutile-versus-amphibole (versus biotite) mechanism debate has intensified without a global budget, and the 25-year-old 'missing niobium' problem of the bulk silicate Earth has only gained new candidate reservoirs (metasomatized lithospheric mantle, recycled carbonate-rich sediments, eclogitic slabs, the core) with no quantitative closure. The problem is to determine, quantitatively and globally, the mineral-scale mechanisms and the reservoir(s) that reconcile the Nb–Ta budgets of both the continental crust and the bulk silicate Earth with their chondritic building blocks.

Background

Niobium (Nb) and tantalum (Ta) are refractory high-field-strength elements whose nearly identical valence and ionic radius keep their ratio, Nb/Ta, almost constant through most geological processes; they are considered 'geochemical twins'. Because both are refractory, the whole Earth is assumed to have the chondritic Nb/Ta of about 19.9 ± 0.6, yet high-precision measurements give a bulk silicate Earth (BSE, the mantle plus crust after core formation) Nb/Ta of only 14.0 ± 0.3 — the BSE is subchondritic, depleted in Nb relative to Ta (Münker et al., 2003). The continental crust is lower still, with Nb/Ta around 11–13 (Barth, McDonough & Rudnick, 2000), below both chondrites and the mid-ocean-ridge basalts (MORB, average Nb/Ta 15.4 ± 1.0; Tang et al., 2019) from which arc crust is ultimately built. Two coupled imbalances follow: the continental crust needs a complementary high-Nb/Ta reservoir among the products of basalt differentiation, and the silicate Earth as a whole needs a superchondritic reservoir hidden in the core or in unevaluated mantle domains — the long-standing 'missing niobium' problem (Wade & Wood, 2001; Münker et al., 2003; Pfänder et al., 2007). Candidate mineral controls include rutile (a TiO2_2 accessory mineral that is the chief host of Nb and Ta in mafic systems; cumulate rutiles are commonly reported to take Nb slightly more readily than Ta, although whether equilibrium rutile–melt partitioning alone can lower melt Nb/Ta is contested), amphibole (a hydrous calcic silicate that forms hornblendite cumulates in hydrous arc magmas and also prefers Nb over Ta), and biotite; diffusion during mineral growth can fractionate Nb from Ta even when equilibrium partitioning would not (Marschall et al., 2013). Delamination of garnet pyroxenite cumulates from arc roots — 'arclogites' — was proposed as the high-Nb/Ta complementary reservoir driven by rutile in those cumulates (Lee & Anderson, 2015). Zhang et al. (2021) tested the amphibole hypothesis directly by in-situ analysis of amphibole and rutile in magmatic cumulates from the Kohistan paleo-arc (an exposed Cretaceous island-arc crustal section in the Pakistani Himalaya) and the Gangdese arc of southern Tibet. They found that the amphiboles carry high Nb/Ta (averages 18.2 and 17.0) but contain far too little absolute Nb and Ta to strip enough of these elements from arc magmas, whereas rutile, even in small modal amounts, can shift melt Nb/Ta from MORB-like values to continental-crust values; they also emphasized that barometric uncertainties of ±2–4 kbar and poorly known melt water contents leave the pressure–temperature–water (PP–TT–H2OH_2O) windows of rutile versus amphibole control loosely defined.

Problem Statement

Determine the mechanisms and reservoirs that jointly close the terrestrial Nb–Ta mass balance, in two coupled parts. (1) Crustal balance: establish whether rutile-bearing and amphibole-rich arc lower-crust cumulates are volumetrically abundant and spatially pervasive enough across global arc systems and through geologic time to drive the Nb/Ta of basaltic parental melts down to continental-crust values — including whether the equilibrated cumulates are removed (e.g., by delamination) so that the residual crust inherits the low Nb/Ta — or whether additional crust-forming processes are required. Mineral-scale Nb and Ta concentrations, rutile and amphibole (and biotite) modes, and partitioning behavior over the relevant PP–TT–H2OH_2O (and CO2CO_2) range are the key controls bearing on this question. (2) Planetary balance: identify and quantify the reservoir or combination of reservoirs (e.g., metasomatized lithospheric mantle, recycled carbonate-rich or eclogitic slabs, hidden mantle domains, or the core) whose superchondritic Nb/Ta reconciles the subchondritic Nb/Ta of the bulk silicate Earth (about 14) with the chondritic value (about 19.9), together with the mechanism by which that reservoir acquired its Nb enrichment.

References

  1. Zhang, J., Wang, R., Hong, J., Tang, M., Zhu, D.-C. (2021). Nb-Ta systematics of Kohistan and Gangdese arc lower crust: Implications for continental crust formation. Ore Geology Reviews 133, 104131. DOI: 10.1016/j.oregeorev.2021.104131; https://doi.org/10.1016/j.oregeorev.2021.104131
  2. Münker, C., Pfänder, J. A., Weyer, S., Büchl, A., Kleine, T., Mezger, K. (2003). Evolution of Planetary Cores and the Earth-Moon System from Nb/Ta Systematics. Science 301, 84–87. DOI: 10.1126/science.1084662; https://doi.org/10.1126/science.1084662
  3. Wade, J., Wood, B. J. (2001). The Earth's 'missing' niobium may be in the core. Nature 409, 75–78. DOI: 10.1038/35051064; https://doi.org/10.1038/35051064
  4. Barth, M. G., McDonough, W. F., Rudnick, R. L. (2000). Tracking the budget of Nb and Ta in the continental crust. Chemical Geology 165, 197–213. DOI: 10.1016/S0009-2541(99)00173-4; https://doi.org/10.1016/S0009-2541(99)00173-4
  5. Pfänder, J. A., Münker, C., Stracke, A., Mezger, K. (2007). Nb/Ta and Zr/Hf in ocean island basalts — Implications for crust–mantle differentiation and the fate of Niobium. Earth and Planetary Science Letters 254, 158–172. DOI: 10.1016/j.epsl.2006.11.027; https://doi.org/10.1016/j.epsl.2006.11.027
  6. Lee, C.-T. A., Anderson, D. L. (2015). Continental crust formation at arcs, the arclogite 'delamination' cycle, and one origin for fertile melting anomalies in the mantle. Science Bulletin 60, 1141–1156. DOI: 10.1007/s11434-015-0828-6; https://doi.org/10.1007/s11434-015-0828-6
  7. Marschall, H. R., Dohmen, R., Ludwig, T. (2013). Diffusion-induced fractionation of niobium and tantalum during continental crust formation. Earth and Planetary Science Letters 375, 361–371. DOI: 10.1016/j.epsl.2013.05.055; https://doi.org/10.1016/j.epsl.2013.05.055
  8. Chen, W., Zhang, G., Ruan, M., Wang, S., Xiong, X. (2021). Genesis of Intermediate and Silicic Arc Magmas Constrained by Nb/Ta Fractionation. Journal of Geophysical Research: Solid Earth 126, e2020JB020708. DOI: 10.1029/2020JB020708; https://doi.org/10.1029/2020JB020708
  9. Tan, D., Xiao, Y., Li, D., Dai, L., Li, W., Hou, Z. (2022). Nb-Ta fractionation by amphibole and biotite during magmatic evolution: Implications for the low Nb/Ta ratios of continental crust. Lithos 434–435, 106941. DOI: 10.1016/j.lithos.2022.106941; https://doi.org/10.1016/j.lithos.2022.106941
  10. Gao, M., Xiong, X., Huang, F., Wang, J., Wei, C. (2023). Key Factors Controlling Biotite–Silicate Melt Nb and Ta Partitioning: Implications for Nb−Ta Enrichment and Fractionation in Granites. Journal of Geophysical Research: Solid Earth 128, e2022JB025889. DOI: 10.1029/2022JB025889; https://doi.org/10.1029/2022JB025889
  11. Bragagni, A., Avanzinelli, R., Münker, C., Mastroianni, F., Conticelli, S. (2024). Fractionation of Nb/Ta during subduction of carbonate-rich sediments. Geochemical Perspectives Letters 29, 38–42. DOI: 10.7185/geochemlet.2410; https://doi.org/10.7185/geochemlet.2410
  12. Yang, Y., Neubauer, F., Liu, Z., Wang, G.-C., Tan, S., He, X., Liang, C. (2024). Enriched lithospheric mantle storage of the Earth's missing niobium: New insights from alkaline rocks with superchondritic Nb/Ta ratios. Geosystems and Geoenvironment 3, 100289. DOI: 10.1016/j.geogeo.2024.100289; https://doi.org/10.1016/j.geogeo.2024.100289
  13. Chen, X., Schertl, H., Hart, E., Majka, J., Cambeses, A., Hernández-Uribe, D., Zheng, Y. (2022). Mobilization and fractionation of Ti-Nb-Ta during exhumation of deeply subducted continental crust. Geochimica et Cosmochimica Acta 319, 271–295. DOI: 10.1016/j.gca.2021.11.024; https://doi.org/10.1016/j.gca.2021.11.024
  14. Nicklas, R., Aulbach, S., Baxter, E. (2025). High field strength element systematics of rutile in cratonic eclogites: Evidence for Nb Ta disequilibrium. Lithos 508–509, 108080. DOI: 10.1016/j.lithos.2025.108080; https://doi.org/10.1016/j.lithos.2025.108080
  15. Manselle, P., Foley, S., Daczko, N. (2025). Origin of Amphibole Cumulates at the Base of an Exposed Arc Crustal Section: Perspectives From Fiordland, New Zealand. Island Arc 34, e70030. DOI: 10.1111/iar.70030; https://doi.org/10.1111/iar.70030
  16. Sahoo, S., Majumdar, A. S., Anand, R., Ray, D., Fuenlabrada, J. M. (2025). Petrogenesis of Mafic–Ultramafic Cumulates in the Mayudia Ophiolite Complex, NE Himalaya: Evidence of an Island Arc Root in Eastern Neo-Tethys. Minerals 15, 572. DOI: 10.3390/min15060572; https://doi.org/10.3390/min15060572
  17. Chen, W., Tiepolo, M., Zhang, G. (2025). Partitioning of Nb and Ta between rutile and carbonatitic melts: Implications for genesis of arc magmas and carbon cycle at subduction zones. Geochimica et Cosmochimica Acta 408, 71–83. DOI: 10.1016/j.gca.2025.09.037; https://doi.org/10.1016/j.gca.2025.09.037
  18. Tang, M., Lee, C.-T. A., Chen, K., Erdman, M., Costin, G., Jiang, H. (2019). Nb/Ta systematics in arc magma differentiation and the role of arclogites in continent formation. Nature Communications 10, 235. DOI: 10.1038/s41467-018-08198-3; https://doi.org/10.1038/s41467-018-08198-3