On-filter degradation of particle-bound benzo[a]pyrene by ozone during air sampling: A review of the experimental evidence of an artefact

On-filter loss of benzo[a]pyrene (BaP) during air sampling due to reaction with ozone was observed in several experimental studies. It was also of concern in the recent CEN method for BaP measurement in ambient air: a denuder-based sampling system minimising this loss is described in the method but, due to the scarcity of validated data, is not yet ready to be used for normative purposes. This paper presents a review of the available experimental results on this artefact, as reported in 16 published studies, with the aim of assessing the state of knowledge regarding its occurrence and extent. Almost all field studies showed that, regardless the type of sampling site, the on-filter BaP degradation by O3 constituted a not negligible artefact. When limiting the analysis of the results to the field sampling campaigns carried out under real-world conditions, mean BaP losses were typically in the 20–55% range (hence causing underestimates of mean BaP concentrations possibly in the order of 100%), with daily maxima up to 71%. The duration of the campaigns was typically of a few days; no experimental results are available on the actual O3 effect on the BaP annual mean. Various factors were investigated as potential contributors to the O3 effect, although their role has yet to be clarified: O3 concentration, relative humidity, distance from BaP sources, size distribution and source of particles, sampling duration. The use of denuders for O3 removal in the sampling system substantially eliminates this artefact, but their application and performance must still be validated.