UC Davis

The Rotating Drum Impactor (RDI) is designed to collect samples of airborne particulate matter that are resolved by aerodynamic particle diameter as a function of time. The purpose of this study is to characterize the accuracy and precision of measurements for particulate matter less than 2.5 μm (PM2.5) from the latest generation of the RDI sampler for use in relatively low concentration environments. Nominal RDI size cuts were used to interpret the results; a laboratory calibration of the RDI was not undertaken. Airborne particulate matter was collected on the campus of the University of California at Davis during the winter and summer of 2013. Two RDI samplers operating with 1 h resolution, two IMPROVE samplers operating 21 h per day, and one MOUDI sampler operating 42 h every two days were deployed for both seasons to enable a comparison of co-located measurements. The RDI samples were scanned using synchrotron X-rays from the Advanced Light Source (ALS), Lawrence National Berkeley Laboratory, to obtain X-ray Fluorescence (XRF) spectra. IMPROVE and MOUDI samples were analyzed using Panalytical XRF instruments using the IMPROVE protocol, then analyzed using Inductively Coupled Plasma Mass Spectrometry (ICP-MS) using an Agilent 7500i. Repeated measurements from the same samplers and analytical techniques were compared to elucidate precision; inter-comparisons between measurements made with different sampler and analytical technique combinations were used to indirectly evaluate accuracy. The IMPROVE XRF analysis technique gave similar repeated measurements for 78% of the elements present above stated Minimum Detection Limits (MDLs) in winter and 42% of elements above MDL in summer. The MOUDI XRF analysis technique gave similar repeated measurements for 33% and 37% of the elements above stated MDLs in winter and summer, respectively. The MOUDI, ICP-MS analysis technique was the most repeatable method tested, yielding similar repeated measurements for 62% and 91% of elements above stated MDLs in winter and summer, respectively. The RDI, XRF analysis technique was the least repeatable method tested, yielding similar repeated measurements for 35–40% and 5–20% of elements above stated MDLs in winter and summer, respectively. PM2.5 measurements from IMPROVE samplers were in better agreement with integrated size-resolved measurements made by MOUDI samplers than RDI samplers in the summer season, however with higher concentrations in the winter season both instruments performed with similar accuracy when using the XRF analysis method. MOUDI samples analyzed by ICP-MS had the best agreement overall with IMPROVE samples analyzed by ICP-MS. These results suggest that MDLs should be reassessed for all methods and a longer collection time should be considered under low concentration conditions.