Answers are grounded in this paper and cite their sources.
Unconsolidated surface soil and dust samples of varying trace element (TE) content were collected from remote locations in central and south-eastern Australia. The finer grained fraction of the samples (<10 µm, PM10) was separated and geochemically compared to the parent particulate matter (PMPAR). TE are mostly hosted in phosphates and oxides/hydroxides or adsorbed to clay minerals, and are normally fractionated into the PM10, producing PM10/PMPAR ratios >1, especially in siliceous, TE-depleted dusts. In contrast, samples TE-enriched by primary silicate minerals eroded from igneous and metamorphic rocks can produce PM10/PMPAR <1 for more mobile elements such as K, Na, Ba, Rb, and Sr. K/Rb is normally lower in PM10 (unless the PMPAR is muscovite-rich) as is the light/heavy rare earth elements (LREE/HREE) ratio because both Rb and HREE are preferentially adsorbed by fine clay particles. Zr and Hf are mostly hosted by zircon crystals initially >10 µm but these diminish in size with time and sedimentological transport so that PM10 aerosol concentrations of these elements are typically telescoped into a narrower range than the PMPAR. Nb is strongly fractionated into PM10, with Nb/TiO2 ratios characteristic of the durable host mineral rutile in all but the most TE-enriched PM. TE content of PM10 in continental dusts is controlled by both physical and chemical processes. Fresh primary silicates suppress PM10/PMPAR ratios of TE with low ionic potential, whereas the opposite effect is induced by hydraulic sorting and/or physical attrition during surface transport, as well as clay absorbtion and fixation of TE in small, resistant accessory minerals.
No takes yet. Share an insight, caveat, or question.
Moreno et al. (2008) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: