Comparison of different approaches to quantify substituted polycyclic aromatic compounds

Abstract

Unlike native polycyclic aromatic hydrocarbons (PAHs), quantitation of substituted polycyclic aromatic compounds (PACs) has been a challenge in the environmental industry. The challenge can be attributed in part to the large number of theoretically possible isomers and the lack of authentic standards for quantitation. In addition, the lack of a unified approach to the quantitation of these compounds has led to poor interlaboratory accuracy. Because these compounds are often used for toxicology studies or to delineate sources and fingerprinting, it is vital that a standardized approach to quantify them is established. This study evaluated different quantitation approaches to quantify both 16 individual PACs and 32 groups/clusters of substituted PACs in three standard reference materials (SRM 1944 - New York / New Jersey waterway sediments, SRM 1597 - a coal tar sample and SRM 2779 - Gulf of Mexico crude oil). The methods employed include: (1) external calibration taking into account recovery correction factor for each analyte, (2) an average relative response factor (ARRF) of PACs obtained with a recovery correction, (3) ARRF of PACs obtained using uncorrected peak areas (i.e., no recovery correction), (4) ARRF of PACs calculated by normalization to deuterated PAHs and (5) ARRF of native PAHs to quantify substituted PACs. The evaluation of concentrations of individually substituted PACs from the different quantitative approaches compared to the certified/reference values showed that methods 1, 2 and 3 performed best. The average percentage of compounds that fell within our acceptable limit (±30%) using methods 1, 2 and 3 for SRM-1944, -1597a and -2779 was 87, 75 and 100%, respectively. Using native PAHs to quantify their substituted analogs resulted in data of the poorest quality. Irrespective of the approach used, there were significant systematic errors in measurements on clusters/groups PACs most notably C1 and C2-benzanthracenes/ chrysenes/triplenylenes, and C2- and C3-dibenzothiophenes being consistently greater than 100% of the stated value. Commerical availability of more substituted PACs will mitigate the biases associated with the quanititation of PAC clusters/groups.