Evaluation and inter-comparison of oxygen-based OC-EC separation methods for radiocarbon analysis of ambient aerosol particle samples

Abstract

Radiocarbon analysis is a widely-used tool for source apportionment of aerosol particles. One of the big challenges of this method, addressed in this work, is to isolate elemental carbon (EC) for 14C analysis. In the first part of the study, we validate a two-step method (2stepCIO) to separate total carbon (TC) into organic carbon (OC) and EC against the EUSAAR_2 thermal-optical method regarding the recovered carbon concentrations. The 2stepCIO method is based on the combustion of OC in pure oxygen at two different temperature steps to isolate EC. It is normally used with a custom-built aerosol combustion system (ACS), but in this project, it was also implemented as a thermal protocol on a Sunset OC-EC analyzer. Results for the recovered EC mass concentration showed poor agreement between the 2stepCIO method on the ACS system and on the Sunset analyzer. This indicates that the EC recovery is sensitive not only to the temperature steps, but also to instrument-specific parameters, such as heating rates. We also found that the EUSAAR_2 protocol itself can underestimate the EC concentration on untreated samples compared to water-extracted samples. This is especially so for highly loaded filters, which are typical for 14C analysis. For untreated samples, the EC concentration on long-term filter samples (two to five days sampling time) was 20-45% lower than the sum of EC found on the corresponding 24-h filter samples. For water-extracted filter samples, there was no significant difference between long-term and the sum of daily filter samples. In the second part of this study, the 14C was measured on EC isolated by the 2stepCIO method and compared to methods from two other laboratories. The different methods agree well within their uncertainty estimates.