PubMed2004
The relationships between transient vehicle operation and ultrafine particle emissions are not well-known, especially for low-emission alternative bus technologies such as compressed natural gas (CNG) and diesel buses equipped with particulate filters/traps (TRAP). In this study, real-time particle number concentrations measured on a nominal 5 s average basis using an electrical low pressure impactor (ELPI) for these two bus technologies are compared to that of a baseline catalyst-equipped diesel bus operated on ultralow sulfur fuel (BASE) using dynamometer testing. Particle emissions were consistently 2 orders of magnitude lower for the CNG and TRAP compared to BASE on all driving cycles. Time-resolved total particle numbers were examined in terms of sampling factors identified as affecting the ability of ELPI to quantify the particulate matter number emissions for low-emitting vehicles such as CNG and TRAP as a function of vehicle driving mode. Key factors were instrument sensitivity and dilution ratio, alignment of particle and vehicle operating data, sampling train background particles, and cycle-to-cycle variability due to vehicle, engine, after-treatment, or driver behavior. In-cycle variability on the central business district (CBD) cycle was highest for the TRAP configuration, but this could not be attributed to the ELPI sensitivity issues observed for TRAP-IDLE measurements. Elevated TRAP emissions coincided with low exhaust temperature, suggesting on-road real-world particulate filter performance can be evaluated by monitoring exhaust temperature. Nonunique particle emission maps indicate that measures other than vehicle speed and acceleration are necessary to model disaggregated real-time particle emissions. Further testing on a wide variety of test cycles is needed to evaluate the relative importance of the time history of vehicle operation and the hysteresis of the sampling train/dilution tunnel on ultrafine particle emissions. Future studies should monitor particle emissions with high-resolution real-time instruments and account for the operating regime of the vehicle using time-series analysis to develop predictive number emissions models.