Addressing analytical uncertainties in the determination of trichloroacetic acid in soil
Dickey, Catherine A
Heal, Kate V
Heal, Mathew R
Soil is an important compartment in the environmental cycling of trichloroacetic acid (TCA), but soil TCA concentration is a methodologically defined quantity; analytical methods either quantify TCA in an aqueous extract of the soil, or thermally decarboxylate TCA to chloroform in the whole soil sample. The former may underestimate the total soil TCA, whereas the latter may overestimate TCA if other soil components (e.g. humic material) liberate chloroform under the decarboxylation conditions. The aim of this work was to show that extraction and decarboxylation methods yield different TCA concentrations because the decarboxylation method can also determine bound TCA. Experiments with commercial humic acid solutions showed there was no additional chloroform formation under decarboxylation conditions, and that all TCA in a TCA–humic acid mixture could be quantitatively determined (108 ± 13%). Anion exchange resin was used as a provider of solid-phase TCA binding; only 5 ± 1% of a TCA solution mixed with the resin was present in the aqueous extract subsequently separated from the resin, yet the decarboxylation method yielded mass balance (123 ± 22%) with TCA remaining in the resin. In aqueous extraction of a range of soil samples (with or without added TCA spike), the decarboxylation method was able to satisfactorily account for TCA in the extractant + residue post-extraction, compared with whole-soil TCA (+ spike) pre-extraction: e.g. mass balances for unspiked soil from Sikta spruce and larch forest were 99 ± 8% and 93 ± 6%, respectively, and for TCA-spiked forest and agricultural soils were 114 ± 13% and 102 ± 2%. In each case recovery of TCA in the extractant was substantially less than 100% (<20% for unspiked soils, <55% for spiked soils). Extraction efficiencies were generally lower in more organic soils. The results suggest that analytical methods which utilise aqueous extraction may underestimate whole-soil TCA concentrations. Application of both methodologies together may enhance insight into TCA behaviour in soil.