A high-efficiency, low-bias method for extracting particulate matter from filter and impactor substrates

•Novel PM extraction methods are presented that:•Maximize extraction efficiency and minimize artifacts and compositional biases.•Achieve 10-40% efficiency increase for ultrafine PM extracted from afterfilters.•Achieve 20-50% efficiency increase for submicron fine PM extracted from PUF.•Demonstrate that extraction efficiencies are compositionally, or source, specific.

Atmospheric particles are frequently collected onto filter and impactor substrates for studies related to the composition, health effects and climate impact of ambient particulate matter (PM). Many of these studies require extraction of that PM from the substrates but available methods have low extraction efficiencies that may lead to compositional and thus toxicity bias. Here, novel PM extraction methods are presented that (a) maximize extraction efficiency, (b) minimize compositional biases in extracted PM, relative to sampled PM and (c) minimize extraction artifacts. Method development was based upon strengths and weaknesses of existing SOPs and current requirements in the field of aerosol health effects research. Extraction objectives were accomplished using a combination of sonication in solvents of varying polarity, selective filtration, liquid-liquid extraction of water-based extracts, solvent removal and final reconstitution of the total extracted PM. Relying largely on intensive gravimetric analyses and comparison to existing SOPs, the new technique has been fully validated on nearly 40 different size-segregated, source-oriented samples collected during two separate seasons in Fresno, CA. Compared to existing methods, and depending on the source, compositionally-specific increases in extraction efficiencies of 10-40% and 20-50% were obtained for the ultrafine and submicron fine PM fractions, respectively, indicating significant increases in total extraction efficiency and significant decreases in compositional bias.

Graphical abstractDownload high-res image (233KB)Download full-size image