Slopes and intercepts from log-log correlations of gas/particle quotient and octanol-air partition coefficient (vapor-pressure) for semi-volatile organic compounds: II. Theoretical predictions vs. monitoring

• Published in: Chemosphere (Oxford) 2021-06, Vol.273, p.128860, Article 128860
• Main Authors: Qiao, Li-Na, Ma, Wan-Li, Zhang, Zi-Feng, Liu, Li-Yan, Song, Wei-Wei, Jia, Hong-Liang, Zhu, Ning-Zheng, Li, Wen-Long, Macdonald, Robie W., Nikolaev, Anatoly, Li, Yi-Fan
• Format: Article
• Language: English
• Subjects: Air Pollutants - analysis; Dibenzofurans; Environmental Monitoring; Gas/particle partition; Halogenated Diphenyl Ethers - analysis; Intercept; Octanols; Polychlorinated Dibenzodioxins; Slope; Steady state; SVOCs; Vapor Pressure; Volatile Organic Compounds - analysis
• ISSN: 0045-6535; 1879-1298
• DOI: 10.1016/j.chemosphere.2020.128860

The logarithm of gas/particle (G/P) partition quotient (logKP) has been found to have a linear relationship with logKOA (octanol-air partition coefficient) with slope mo and intercept bo and logPL (subcooled liquid vapor pressure) with slope mp and intercept bp. In the sister paper of the present work, analytical equations to predict the slope mo and intercept bo based on logKOA and predict the slope mp and intercept bp based on logPL are developed using steady state theory. In this work, the equations are evaluated using world-wide monitoring data (262 pairs for mo and bo values and 292 pairs for mp and bp values produced from more than 10,000 monitiring data worldwide) for selected seven groups of semi-volatile organic compounds (SVOCs), including polybrominated diphenyl ethers (PBDEs), polychlorinated dibenzo-p-dioxins and polychorinated dibenzofurans (PCDD/Fs), polyclorinated biphenyl (PCBs), polycyclic aromatic hydrocarbons (PAHs), polychlorinated naphthalenes (PCNs), organochlorinated pesticides (OCPs), novel brominated flame retardants (NBFRs), and other selected halogenated flame retardants. The slopes and intercepts predicted by the steady state equations reproduce the trends observed in monitoring regression results for the seven SVOC groups, with 44.4% of the variation of monitoring mo values accounted for by logKOA and 48.2% of the variation of monitoring mp values accounted for by logPL. Theoretically, the values of mo can be any value between 0 and 1 dependent on the values of KOA, and are not constrained to 1 as in equilibrium theory. Likewise, the values of mp can be any value between 0 and -1 dependent on the values of PL, and not constrained to −1 predicted by the equilibrium theory. The influence of sampling artifacts on the G/P partitioning of SVOCs has most likely been overemphasized by the equilibrium theory. Thus, the equilibrium approach should be abandoned in favor of the steady state approach for calculating the G/P partition quotients for SVOCs with high KOA values (>1011.38) or low PL values (