Persistent organic pollutant concentrations in air‐ and freeze‐dried compared to field‐fresh extracted soil samples of an eastern Slovak deposition gradient

Sample preparation affects the results of the determination of persistent organic pollutant (POP) concentrations in soils. We compared the extraction results of POPs from air‐ and freeze‐dried with field‐fresh samples. We determined the concentrations of 21 polycyclic aromatic hydrocarbons (PAHs) and 14 polychlorinated biphenyls (PCBs) in 35 horizons of forest soils (Lithosols and Cambisols) along a deposition gradient caused by a chemical factory in Strá?ske (eastern Slovakia). The organic C (C_org) concentrations of the studied samples ranged 14‐477 g kg^‐1, the sum of 21 PAH (Σ 21PAHs) concentrations 53‐6870 μg kg^‐1, and that of 14 PCBs (Σ 14PCBs) 0.12 96 μg kg^‐1. The benzofluoranthenes were the most abundant PAHs, and the hexa‐chlorinated PCBs 138 and 153 were the most abundant PCBs. The deposition gradient was reflected by decreasing PAH and PCB storages in the organic layer with increasing distance from the chemical factory (Σ 21PAHs: 82‐238, Σ 14PCBs: 0.34‐2.3 g ha^‐1). The PAH concentrations in air‐ and freeze‐dried samples were consistently lower than those in field‐fresh extracted samples. For naphthalene, this was mainly attributable to volatilization losses during drying. Naphthalene losses decreased with increasing C_org concentrations. For all other PAHs, drying reduced the extractability; the latter was correlated with the water content of the samples. The differences in most PCB concentrations among the sample preparation methods were small (within the determination error of ±15 %) and inconsistent. However, PCBs 8, 28, 52, and 209 had markedly higher concentrations in dried than in fieldfresh extracted samples. The increased recovery of low‐chlorinated PCBs in dried samples may be explained by redistribution of PCBs from higher to lower contaminated samples via the drying room atmosphere because there were no PCBs in analytical blanks. This assumption is supported by a close correlation between the octanol‐air distribution coefficient and the up to hexa‐chlorinated PCB concentrations (normalized to those in field‐fresh extracted samples) in air‐ (r = ‐0.90) and freeze‐dried (r = ‐0.86) samples. Our study shows that each sample preparation method has its specific advantages and disadvantages. Sample drying results in a standardization to a well‐defined water content and facilitates homogenization; therefore it increases the reproducibility of POP determinations. Extraction of field‐fresh samples reduces volatilization losses and contamination risks.