Influence of grinding particle size on the non-targeted analysis of semivolatile organic pollutants in soil.

Abstract

Non-targeted analysis of soil pollutants requires effective sample pretreatment, yet current studies adopt inconsistent soil particle size conditions, with limited evaluation. This study aims to investigate how different soil particle size preparations influence the outcomes of non-targeted analysis of semi-volatile organic compounds (SVOCs). Soil samples collected from a maize field in Xingkai Lake Farm, Sanjiang Plain, Northeast China, were pretreated and sieved into seven particle size groups (60~140 mesh) and analyzed using GC × GC-MS. To verify data reliability, 16 priority PAHs and 53 pesticides were quantified by GC-MS. Total organic carbon (TOC) and compounds' physicochemical properties were analyzed for correlation with particle size. Additionally, these quantified compounds were grouped by log K_ow and vapor pressure, and their recovery rates were analyzed in relation to hydrophobicity and volatility. Results showed that different particle sizes led to variations in both the number and types of compounds detected, with the 60~80 mesh group achieving the highest detection count, compound diversity, and recovery rates. Significant correlations (p < 0.05) were observed between particle size, TOC, log K_ow, and vapor pressure. Overall, selecting soil particle sizes within the 60~80 mesh range provides an optimal balance between the number and diversity of compounds detected in non-targeted analysis, while the influence of hydrophobicity and volatility highlights the importance of considering compound properties during sample preparation. These findings offer experimental evidence for optimizing sample preparation conditions in non-targeted analysis and offer a scientific basis for enhancing the identification of SVOCs diversity in complex contaminated soils.