Experimental observations on noble metal nanonuggets and Fe-Ti oxides, and the transport of platinum group elements in silicate melts

Abstract

Platinum group element (PGE) nanonuggets are a nuisance in experimental studies designed to measure solubility or partitioning of noble metals in silicate melts. Instead of treating nanonuggets as experimental artifacts, we studied their behaviour motivated by recent discoveries of PGE nanonuggets in a variety of natural settings. We used an experimental setup consisting of AgPd, Pt or AuPd capsules and Fe(-Ti) oxide-saturated hydrous peralkaline silicate melts to maximise nanonugget production. TABS (Te, As, Bi, Sb, Sn) commonly occur in PGM (platinum group minerals), prompting addition of Bi to our experiments to investigate its properties as well. Three-dimensional optical examination by 100× objective and immersion oil reveals variable colour which correlates with nanonugget size and shape due to plasmon resonance effects. We observe two textural types: (1) intermediate-sized nanonuggets dispersed in the glass and adhering to oxides, and (2) abundant fine nanonuggets dispersed in the glass with coarse euhedral crystals in contact with oxides. Slow cooling removes dispersed nanonuggets and greatly coarsens existing oxide-associated metal crystals. Nanonugget-free halos are commonly observed around oxide grains. All metal phases are composed of major (Ag, Pd) and trace (Pt, Ir, Au) capsule material.Our results show reduction processes, imposed by growing oxides, causing local metal saturation in the oxide rich zones with preferential nucleation on smaller oxide grains. The redox gradient then blocks additional metals from diffusing into oxide rich zones, forming halos. As the entire experimental charge is reduced throughout the run, nanonuggets form in the distal glass. Bismuth contents of metal phases do not depend on Bi2O3 amounts dissolved in the melt. Further PGM crystallisation consumes nanonuggets as feedstock. We conclude that the appearance of metallic PGE phases happens in two stages: first as nanonuggets and then as larger PGM. Once formed, nanonuggets cannot be removed by oxidation or dissolution, but only by PGM coarsening.Sulfur-poor PGE ore deposits commonly contain more PGE than permitted by existing estimates of equilibrium solubility in silicate melts. This is commonly explained by initial scavenging of PGE by sulfide liquids followed by S-loss, but evidence for S-loss is not conclusive. We suggest that nanonuggets may be a means to transport PGE from source regions to form ore deposits, followed by direct PGM crystallisation from silicate melt without intermediate concentration by sulfide liquids.