Aerosol–computational fluid dynamics modeling of ultrafine and black carbon particle emission, dilution, and growth near roadways
Abstract. Many studies have shown that on-road vehicle emissions are the dominant source of ultrafine particles (UFPs; diameter < 100 nm) in urban areas and near-roadway environments. In order to advance our knowledge on the complex interactions and competition among atmospheric dilution, dispersion, and dynamics of UFPs, an aerosol dynamics–computational fluid dynamics (CFD) coupled model is developed and validated against field measurements. A unique approach of applying periodic boundary conditions is proposed to model pollutant dispersion and dynamics in one unified domain from the tailpipe level to the ambient near-road environment. This approach significantly reduces the size of the computational domain, and therefore allows fast simulation of multiple scenarios. The model is validated against measured turbulent kinetic energy (TKE) and horizontal gradient of pollution concentrations perpendicular to a major highway. Through a model sensitivity analysis, the relative importance of individual aerosol dynamical processes on the total particle number concentration (N) and particle number–size distribution (PSD) near a highway is investigated. The results demonstrate that (1) coagulation has a negligible effect on N and particle growth, (2) binary homogeneous nucleation (BHN) of H2SO4–H2O is likely responsible for elevated N closest to the road, and (3) N and particle growth are very sensitive to the condensation of semi-volatile organics (SVOCs), particle dry deposition, and the interaction between these processes. The results also indicate that, without the proper treatment of the atmospheric boundary layer (i.e., its wind profile and turbulence quantities), the nucleation rate would be underestimated by a factor of 5 in the vehicle wake region due to overestimated dilution. Therefore, introducing atmospheric boundary layer (ABL) conditions to activity-based emission models may potentially improve their performance in estimating UFP traffic emissions.
