Unconventional thinking.

Critical minerals, or critical metals, are often small markets, small production batches, byproducts, and difficult elements to extract. This includes elements like indium, gallium, germanium, scandium. Elements that are rare, don't follow the normal geological trajectories into sulphides, oxides, or silicates, and are poorly concentrated within the crust. Scandium (or scamdium as some refer to it) is a prime example: rare as hen's teeth, expensive as a unicorn horn, but forever on the cusp of making nickel laterite explorers rich as Creosote.

If we have an objective to develop secure supply of critical and rare metals, surely we should be exploring for these elements? That's often easier said than done.

Let’s say you were keen on gallium arsenide solar panels and you wanted to find a sulphide source not associated (as gallium often is) with bauxite, which is recovered via electrolysis of aluminium slag. Not the easiest metal to monetise, is it?

Have you ever considered when in Earth’s history was the time of greatest gallium flux? I haven’t until today, but there has been a lot of work on the temporal association of gold mineralisation, and people have written papers on it. There are very good reasons for these temporal phases of gold mineralisation, and the same might be true of other metals or mineralisation styles.

Armed with the GSWA WACHEM database, assigning the minimum age from the GSWA 1:500K geological interpretation polygons to the non-regolith geochemical data (excluding soils, alluvium, etc, is required to screen for hard rock sources and exclude unreasonably young associations), data can be assessed on gallium content vs age.

This shows enrichments of gallium above the average of 16ppm (n = 19,346) and above crustal abundance of ~18ppm, are clustered in rocks with a minimum age (via GSWA) of 2900-2600Ma, 1850-1600Ma, and 1150-1050Ma - in Western Australia. Already, this allows us to determine that gallium abundances increase in association with major episodes of magmatism, crustal growth and granitoid emplacement – as one would expect from its association with aluminium (being that crustal accretion occurs via agglomeration of sialic crust).

If you deep dive into the data this gallium is, however, hosted in interesting places. Chromitites in the period of 2950 to 2750Ma, for example, have gallium abundances on average twice the crustal abundance, up to 4 times (60ppm). For an element considered to correlate with aluminium, its presence in a chromite rock is intriguing.

Which leads me to investigate ultramafic rocks in general, and another rare element springs to mind: scamdium. Sorry, scandium. Plotting Sc vs Ga for 2950-250Ma rocks from Western Australia (basically, the work of Ivanic, Smithies and others in the Murchison and the basement of the Eastern Goldfields terranes) the following interesting graph can be shown;

Scandium and gallium correlate well in juvenile mafic material at 15ppm Ga and 30-40ppm Sc. There is a trend of lower Sc and Ga with primitive ultramafics (note Sc/Ni and Sc/Mg correlate well with pyroxene, so the tail here is olivine normative rocks). Granites are lower in Sc due to sequestration in pyroxenes (hence mafic rocks having equal Sc to ultramafic) and have the usual Ga.

We only extract gallium from bauxites because we have always done so. The fact that bauxites form from aluminous material has rguably led to gallium supply being correlated unfairly with granites; granites don't have undue amounts of gallium, we just don't extract it from ultramafic-rock based metallurgical processes.

Mineral exploration companies are already exploring for scandium resources in nickel-cobalt laterites, and scandium is being produced by Rio Tinto from bauxites (it seems to be a bit of a losing proposition according to this analysis). Ultramafic rocks have just as much gallium, and considerably more scandium, and we are seeking to exploit scandium from them, but are missing the gallium opportunity. This leads to the conclusion that ultramafic rocks with Sc:Ga ratios close to 1, and which form laterites, could present a new source of gallium. However, gallium – an essential metal for certain solar cell chemistries worth upwards of US$145 per kilogram, is not on the menu in nickel laterite processing.

For real, Roland?

Surely this is just hypothetical mumbo jumbo. But I argue it's not.

We have to only look at some of the biggest hydrochemical experiments in Australia to find a source rich in scandium and gallium, where there is a viable route to extraction. This is the High Pressure Acid Leach (HPAL) nickel laterite autoclave.

Within HPAL plants iron, aluminium, and potassium are stripped out of the pregnant leach liquor prior to Ni, Co sulphate precipitation, via a neat trick wherein jarosite and alunite group sulphates become insoluble at high pressures and temperatures (260 degrees is the trick). Keen observrs will note that Ga associates with aluminium, so if you drop jarosite-alunite out at high T and P, you ought to drop Ga out as well. Jarosite produced from HPAL nickel laterite processing ought to be a source of gallium from c. 2950 to 2750Ma ultramafic rocks; this possibly holds true for other major mafic-ultramafic Large Igneous Provinces such as the Warakurna LIP and the Kimberley events of c. 1800Ma. ?

Further thought needs to be put into chromitites - and associated sulphides - as a source of Ga, Sc hosted within the sulphides or chromite itself.

Just spitballing here, but there might be something to this kind of random thinking.