Heavy-mineral concentrates record sediment provenance, but their standard quantification by optical grain counting is slow, operator-dependent, and blind to opaque grains. Mounteney et al. (2018) showed that complete-digestion lithium-metaborate (LiBO₂) flash fusion followed by ICP-AES elemental analysis of heavy-mineral concentrates separates United Arab Emirates (UAE) sediment populations at roughly ten-fold higher throughput, and that the geochemical index CrZri reproduces the optically counted chrome-spinel:zircon index - while stating precisely what the method still lacks: element-mineral calibrations fail for elements carried by several minerals (Ti: R2=0.22R^2 = 0.22), systematic biases (grain volume, mineral stoichiometry, excluded opaque grains, selective dissolution) make optical counts and element concentrations disagree, and the protocol has been tested only on UAE assemblages. Later work resolved the UAE interpretation itself (Farrant et al. 2019) and advanced neighboring techniques (interlaboratory comparability of counts, pXRF, SEM-BEX mapping) but left the methodological core untouched. The open problem is to make geochemical heavy-mineral-concentrate analysis a quantitative, transferable provenance tool: validated element-mineral transfer functions including multi-carrier elements, quantified or corrected count-versus-chemistry biases, and demonstrated provenance discrimination across source lithologies and grain-size fractions beyond the UAE, with optimal index formulations for mixed-source basins.

Background

Heavy minerals are detrital grains denser than about 2.8–2.9 g/cm³ (e.g. zircon, rutile, chrome-spinel, amphibole, pyroxene) that typically make up about 1% of a siliciclastic sediment. Because they are refractory and source-rock specific, their abundances and proportions are a standard basis for sedimentary provenance analysis - reconstructing where sediment came from - used in palaeogeographic reconstruction, mineral reconnaissance and petroleum reservoir correlation. The traditional quantification is optical: an analyst identifies 300-500 non-opaque grains per sample by petrographic microscope ribbon counting (a few samples per day), and provenance is read from mineral indices, i.e. ratios of mineral pairs with similar size, density and alteration resistance such as the chrome-spinel:zircon index (CZi) or rutile:zircon index (RZi), or from ternary plots of mineral proportions. This is slow, operator-dependent, and excludes opaque and optically isotropic grains. Mounteney et al. (2018) proposed a bulk-geochemical alternative: separate the heavy-mineral concentrate (HMC) by dense liquid, mill a 0.1 g portion to <20 µm, flash-fuse it with 0.9 g of lithium metaborate (LiBO₂) flux at 1050 °C for 15 min - a digestion that dissolves even resistate minerals such as zircon and chrome-spinel that acid digestion leaves incomplete - and measure major and trace elements by inductively coupled plasma atomic emission spectrometry (ICP-AES, a solution multi-element technique). Element-to-mineral transfer was calibrated on artificial heavy mineral concentrates (AHMCs), gravimetric mixtures of pure mineral separates: measured Cr and Zr reproduced known mineral proportions with R2=0.99R^2 = 0.99 (P: 0.96), but Ti correlated poorly (R2=0.22R^2 = 0.22) because Ti is carried by several heavy minerals (rutile, anatase, brookite, titanite, ilmenite) and also occurs as a minor component in silicates such as augite. On 118 UAE desert-sand HMCs, the geochemical index CrZri = 100 Cr/(Cr+Zr) agreed with the optically counted CZi (R2=0.63R^2 = 0.63; a Mann-Whitney test, a non-parametric two-sample rank test whose Z-score must stay below the 1.96 critical value for the populations to be indistinguishable, gave Z = 0.32), and Zr-Ti-Cr and Ca-Fe-Mg ternary plots of the ICP-AES data separated the Miocene, Hili, Ghayathi and modern-dune populations at least as discretely as the mineralogical plots. The paper itself identified what keeps this a screening demonstration rather than a calibrated, transferable method. First, the calibration problem: element-mineral transfer works only for elements with an essentially unique carrier mineral (Cr - chrome-spinel, Zr - zircon) and fails for multi-carrier elements such as Ti. Second, systematic, unquantified differences between grain counts and element concentrations: optical counting weights every grain equally regardless of volume (a slender prismatic zircon of about 49 mm³ and an equant one of about 1020 mm³ count the same but contribute very different Zr); counts exclude opaque, optically isotropic, authigenic and carbonate grains that the total digestion includes; chrome-spinel stoichiometry (its Cr:Fe ratio) varies with mode of formation, so equal mineral counts do not mean equal Cr; and selective dissolution of unstable pyroxenes and amphiboles (etched 'hacksaw' grain terminations seen near the Arabian Gulf coast) alters the chemical signal of some samples. Third, generality: the protocol was tested only on UAE assemblages; the paper cautions that sedimentary basins may be infilled from several sources, in which case bulk geochemistry may not necessarily be diagnostic of a specific geological terrain though it may still yield valuable data, notes that optical identification of HMC is still required for minerals lacking a unique geochemical signature, advocates the method as a powerful screening tool for larger-scale provenance studies, and defers full interpretation of the UAE dataset to a companion paper. Its discrimination limits, the optimal geochemical index formulations for mixed-source basins, and best-practice integration of geochemical with optical and mineral-chemical datasets across other source lithologies and grain-size fractions therefore remain untested.

Problem Statement

Establish bulk geochemical analysis of heavy-mineral concentrates by complete-digestion LiBO₂ fusion and ICP-AES as a quantitative, transferable sediment-provenance tool, by resolving three linked strands that jointly serve this one objective: (1) derive and validate element-mineral calibration or transfer functions - including for elements carried by more than one heavy mineral, notably Ti - that hold beyond artificial-mixture calibrations and beyond the UAE assemblages; (2) quantify, correct, or rigorously bound the systematic differences between optical grain counts and bulk element concentrations (grain-volume/mass bias from equal counting of unequal grains, variable mineral stoichiometry such as Cr:Fe in chrome-spinel, exclusion of opaque/optically isotropic/authigenic grains from counts, and selective dissolution of unstable minerals), so that geochemical and optical indices can be compared or combined on a calibrated footing; and (3) validate the provenance-discrimination power of geochemical HMC data and the optimal geochemical index formulations (CrZri-type ratios and multi-element discrimination schemes) in mixed-source basins across a range of source-rock lithologies and grain-size fractions beyond the UAE case, together with best-practice integration of geochemical, optical and mineral-chemical datasets. Resolution requires all three strands to be advanced jointly: the count-versus-chemistry differences of strand (2) are resolved when they are quantified and either corrected or bounded, the 'optimal' index formulations of strand (3) are those that best discriminate source lithologies and grain-size fractions under an explicitly stated metric, and the best-practice integration is closed by a reproducible procedure for combining the geochemical, optical and mineral-chemical datasets.

References

  1. Mounteney, I., Burton, A.K., Farrant, A.R., Watts, M.J., Kemp, S.J. & Cook, J.M. (2018). Heavy mineral analysis by ICP-AES a tool to aid sediment provenancing. Journal of Geochemical Exploration 184, 1-10. DOI: 10.1016/j.gexplo.2017.10.007. https://doi.org/10.1016/j.gexplo.2017.10.007
  2. Farrant, A.R., Mounteney, I., Burton, A., Thomas, R.J., Roberts, N.M.W., Knox, R.W.O. & Bide, T. (2019). Gone with the wind: dune provenance and sediment recycling in the northern Rub' al-Khali, United Arab Emirates. Journal of the Geological Society 176(2), 269-283. DOI: 10.1144/jgs2017-044. https://doi.org/10.1144/jgs2017-044
  3. Dunkl, I., von Eynatten, H., Ando, S., Lunsdorf, K., Morton, A., Alexander, B., Aradi, L., Augustsson, C., Bahlburg, H., Barbarano, M., Benedictus, A., Berndt, J., Bitz, I., Boekhout, F., Breitfeld, T., Cascalho, J., Costa, P.J.M., Ekwenye, O., Feher, K., Flores-Aqueveque, V., Fuhring, P., Giannini, P., Goetz, W., Guedes, C., Gyurica, G., Hennig-Breitfeld, J., Hulscher, J., Jafarzadeh, M., Jagodzinski, R., Jozsa, S., Kelemen, P., Keulen, N., Kovacic, M., Liebermann, C., Limonta, M., Luzar-Oberiter, B., Markovic, F., Melcher, F., Miklos, D.G., Moghalu, O., Mounteney, I., Nascimento, D., Novakovic, T., Obbagy, G., Oehlke, M., Omma, J., Onuk, P., Passchier, S., Pfaff, K., Pinto Linconir, L., Power, M., Razum, I., Resentini, A., Sagi, T., Salata, D., Salgueiro, R., Schoenig, J., Sitnikova, M., Sternal, B., Szakmany, G., Szokaluk, M., Thamo-Bozso, E., Toth, A., Tremblay, J., Verhaegen, J., Villasenor, T., Wagreich, M., Wolf, A. & Yoshida, K. (2020). Comparability of heavy mineral data - The first interlaboratory round robin test. Earth-Science Reviews 211, 103210. DOI: 10.1016/j.earscirev.2020.103210. https://doi.org/10.1016/j.earscirev.2020.103210
  4. Gallhofer, D. & Lottermoser, B.G. (2020). The application of pXRF for the chemical and mineralogical characterization of heavy mineral sands. Geochemistry: Exploration, Environment, Analysis 20(1), 99-111. DOI: 10.1144/geochem2019-015. https://doi.org/10.1144/geochem2019-015
  5. Buckman, J., Gough, A. & Webb, M. (2024). Fast Elemental Analysis of Heavy Mineral Suites by Scanning Electron Microscopy (SEM-Unity BEX). Minerals 14(9), 950. DOI: 10.3390/min14090950. https://doi.org/10.3390/min14090950
  6. Zhang, C., Li, Z., Chen, Q., Dong, S., Yu, X. & Yu, Q. (2020). Provenance of eolian sands in the Ulan Buh Desert, northwestern China, revealed by heavy mineral assemblages. Catena 193, 104624. DOI: 10.1016/j.catena.2020.104624. https://doi.org/10.1016/j.catena.2020.104624