Comparison of gas with liquid chromatography for the determination of benzenepolycarboxylic acids as molecular tracers of black carbon

Existing methods for black carbon (BC) quantification measure different parts of the BC continuum, which complicates the calculation of a global BC budget. Benzenepolycarboxylic acids (BPCA) are used as molecular markers to quantify and characterize BC in soils and sediments using gas chromatography for BPCA separation (GC-BPCA). Recently, this method was refined for BC analysis in seawater using high performance liquid chromatography (LC-BPCA), which omits the cleaning steps and derivatization necessary in GC analysis. As yet it is not clear whether the two analytical methods yield similar results. Here we apply both methods to a suite of laboratory produced charcoals derived from wood and grass. We found systematically lower total BPCA-C contents and larger analytical variability for all tested charcoals when using GC-BPCA compared to LC-BPCA, the latter giving 1.5 ± 0.3 times higher yields for the charcoal samples formed at 275-700 °C. At lower and higher pyrolysis temperatures the differences between the two analytical methods were larger. The main reason for the differences between the two methods is the loss of BPCA during sample preparation for GC analysis. We propose a correction factor of 1.5 to account for at least part of these losses. No qualitative biases, i.e. towards more or less functionalized BPCAs, were observed between the two methods. The relative contribution of mellitic acid C to total BPCA-C, a measure for the degree of condensation of BC, was the same in the two analytical techniques. Qualitative differences between wood and grass charcoals as detected by both methods were small.