Environmental context. Charcoal is widespread in soils and may be a major component of soil organic matter. Trace metal ions in soils are predominantly associated with solid phase materials, including charcoal, and the identity of the solid phase and the mechanisms of association influence the geochemical behaviour of metals. Metals associated with soil mineral phases are estimated using techniques such as selective sequential extraction, and the sorption reactions of metal ions are well understood. Much less is known about the associations of trace metals with natural charcoal, and metals associated with charcoal in soils are likely to be misidentified in sequential extraction procedures.

Abstract. Given that up to 50% of the soil carbon store can consist of charcoal, it is possible that trace elements can become immobilised through their interaction with natural charcoal. Hence, natural charcoal may be a significant sink that has yet to be accounted for in trace element biogeochemical cycles. Testing this hypothesis becomes problematic considering the typically small size (<53 µm) of charcoal particles that occur naturally in Australian soils, making isolation and analysis of natural soil charcoal difficult. Therefore, in this study, we test the robustness of a typical sequential extraction technique by applying it to naturally occurring charcoal that had been spiked with five different concentrations of metal ions (Al³⁺, Cr³⁺, Cu²⁺, Ni²⁺, Zn²⁺, Cd²⁺, Ag⁺, Pb²⁺). The method was then applied to contrasting soils mixed with this spiked charcoal. The sequential extraction scheme consisted of the following five extractions the in order: (1) sodium acetate (targeting the adsorbed-exchangeable-carbonate fraction), (2) sodium pyrophosphate (organic fraction), (3) ammonium oxalate (amorphous iron/manganese oxides), (4) hydroxylamine hydrochloride (crystalline iron/manganese oxides) and (5) residual (aqua regia digest). The majority of metals added to the charcoal were extracted in the fractions targeting both the amorphous and crystalline iron and manganese oxides, at low additions of metal ions. At higher additions of metals, the metals were mostly extracted from charcoal in the adsorbed-exchangeable-carbonate fraction. When the spiked charcoal was added to soils, a trend similar to the charcoal-only experiment was observed in the sequential extraction data. Higher concentrations of metals (compared with the control) were extracted for the charcoal-amended soils, in the same fractions as in the charcoal-only extractions. Since the concentration of metals extracted in the various extractants changed with increasing metal loads on charcoal, sequential extractions cannot be used to identify the contribution of metals from the charcoal pool. Therefore, a potentially large pool of trace elements could be misrepresented when sequential extraction techniques are applied, particularly for soils in which there is a large concentration of charcoal. Hence, there is still a large gap in knowledge with regard to the significance of charcoal in ‘real’ soils, particularly with respect to the role of charcoal as a trace element sink.