Ultrafine PM emissions from natural gas, oxidation-catalyst diesel, and particle-trap diesel heavy-duty transit buses

Abstract

This paper addresses how current technologies effective for reducing PM emissions of heavy-duty engines may affect the physical characteristics of the particles emitted. Three in-use transit bus configurations were compared in terms of submicron particle size distributions using simultaneous SMPS measurements under two dilution conditions, a minidiluter and the legislated constant volume sampler (CVS). The compressed natural gas (CNG)-fueled and diesel particulate filter (DPF)-equipped diesel configurations are two "green" alternatives to conventional diesel engines. The CNG bus in this study did not have an oxidation catalyst whereas the diesel configurations (with and without particulate filter) employed catalysts. The DPF was a continuously regenerating trap (CRT). Particle size distributions were collected between 6 and 237 nm using 2-minute SMPS scans during idle and 55 mph steady-state cruise operation. Average particle size distributions collected during idle operation of the diesel baseline bus operating on ultralow sulfur fuel showed evidence for nanoparticle growth under CVS dilution conditions relative to the minidiluter. The CRT effectively reduced both accumulation and nuclei mode concentrations by factors of 10-100 except under CVS dilution conditions where nuclei mode concentrations were measured during 55 mph steady-state cruise that exceeded baseline diesel concentrations. The CVS data suggest some variability in trap performance. The CNG bus had accumulation mode concentrations 10-100x lower than the diesel baseline but often displayed large nuclei modes, especially under CVS dilution conditions. Partly this may be explained by the lack of an oxidation catalyst on the CNG, but differences between the minidiluter and CVS size distributions suggest that dilution ratio, temperature-related wall interactions, and differences in tunnel background between the diluters contributed to creating nanoparticle concentrations that sometimes exceeded diesel baseline concentrations when driving under load. The results do not support use of CVS dilution methodology for ultrafine particle sampling, and, despite attention to collection of tunnel blanks in this study, results indicate that a protocol needs to be determined and prescribed for taking into account tunnel blank "emissions" to obtain meaningful comparisons between different technologies. Of critical importance is determining how temperature differences between tunnel blank and test cycle sampling compare in terms of background particle numbers. Total particle number concentrations for the minidiluter sampling point were not significantly different for the two alternative technologies when considering all the steady-cycle data collected. Concentrations ranged from 0.8 to 3 x 10(6) for the baseline bus operating on ultralow sulfur fuel, from 0.5 to 9 x 10(4) for the diesel bus equipped with the CRT filter, and from 1 to 8 x 10(4) particles/cc for the CNG bus.