Particle number emission factors and volatile fraction of particles emitted from on-road gasoline direct injection passenger vehicles A. Momenimovahed, D. Handford, M.D. Checkel, J.S. Olfert * Mechanical Engineering Department, University of Alberta, Edmonton, T6G 2G8, Canada highlights  Particle emission rates are proposed for on-road gasoline direct injection vehicles.  More semi-volatile particles are produced during initial stages of the acceleration.  The emission factor is higher in urban driving cycles compared to highway cycles.  Emission rates increase as tractive power increases for the entire range of speed. article info Article history: Received 8 July 2014 Received in revised form 18 November 2014 Accepted 21 November 2014 Available online 22 November 2014 Keywords: Gasoline direct injection Particle number emission factor Emission model Volatility abstract Particle number emission factors, and the volatility of the particles, are measured on-road for five gas- oline direct injection vehicles over a wide range of operating conditions including urban and highway driving conditions. Two condensation particle counters (CPC) were used to measure nascent and non- volatile (thermodenuded) particle concentrations for transient urban and highway tests. To measure the non-volatile concentration and also the volatility of the particles, a thermodenuder was employed to remove the semi-volatile material from the aerosol sample. Rapid accelerations were also studied in more detail by measuring the particle size distributions in real-time using a differential mobility spec- trometer (DMS). The ratio of semi-volatile particles to total particle number is generally higher during acceleration followed by the idle operating mode. The number emission factors (for particles larger than 2.5 nm) ranged between 5.46  10 11 e3.50  10 12 /km for freshly emitted (nascent) particles and between 2.87  10 11 e3.31  10 12 /km for non-volatile (thermodenuded) particles. More particles per kilometer are produced during acceleration compared to cruise conditions where the non-volatile particle number emission factor for acceleration is 2.3 and 1.8 times higher than vehicle cruise for urban and highway driving cycles, respectively. Particle number emission factor models are also presented in terms of particle emission rate as a function of vehicle tractive power and also as a function of vehicle specific power as defined for the US Environmental Protection Agency's MOVES model. © 2014 Elsevier Ltd. All rights reserved. 1. Introduction In recent years, gasoline direct injection (GDI) engines have been widely used on passenger vehicles and trucks. GDI engines have better fuel economy and higher power output compared to port fuel injection (PFI) gasoline engines, however, they produce more particulate emissions in terms of both number and mass (Zhao et al., 1999). Concerns about the health effects of the particles emitted from these vehicles have resulted in particle mass emissions limits, and more recently, particle number emission limits defined in the Euro 6 standard for GDI vehicles (Commission Regulation (EC) No 459/2012). According to the standard, emission factors are measured on a chassis dynamometer using standard driving cycles. Only non-volatile particles larger than 23 nm are included in the particle number limit according to the particle measurement programme (PMP). Several studies have been done on chassis dynamometers to examine the effect of air-fuel mixing method (Choi et al., 2012), gasoline particulate filters (Chan et al., 2012; Mamakos et al., 2013), fuel volatility (Khalek et al., 2010; Liang et al., 2013) and ambient temperature (Chan et al., 2013; Mamakos et al., 2013) on particulate emissions from GDI vehicles. However, it has been shown that * Corresponding author. E-mail address: jolfert@ualberta.ca (J.S. Olfert). Contents lists available at ScienceDirect Atmospheric Environment journal homepage: www.elsevier.com/locate/atmosenv http://dx.doi.org/10.1016/j.atmosenv.2014.11.045 1352-2310/© 2014 Elsevier Ltd. All rights reserved. Atmospheric Environment 102 (2015) 105e111