Size and composition of particulate emissions from motor vehicles in the Kaisermühlen-Tunnel, Vienna

Abstract

Size segregated emissions of particle-phase species from on-road motor vehicles were investigated in the Kaisermühlen Tunnel (Vienna, Austria) during April and May 2005. Emission factors were calculated from concentration differences between tunnel inside and tunnel outside samples, the distance between tunnel entrance and sampling location, the ventilation rate and the number of vehicles passing the tunnel. For a mixed car fleet with an average contribution of 9.6% heavy duty vehicles (HDVs) mean particle mass emissions of 26±10 mg veh⁻¹ km⁻¹ in PM2.5, 62±18 mg veh⁻¹ km⁻¹ in PM10 and 129±45 mg veh⁻¹ km⁻¹ for total suspended particulates (TSPs) were observed. The released particles mainly consisted of elemental carbon (EC), organic carbon (OC) and the mineral components (MC=Si, Fe, Ca, Al, Mg). They accounted for 34.4% (EC), 30.3% (OC) and 18.2% (MC) of total PM10 emissions and 68.5%, 8.7% and 14.9% of PM2.5 emissions, respectively. Trace metal emissions (As, Ba, Cd, Co, Cr, Cu, Mn, Ni, Pb, Sb, Sn, Sr, Ti, V, Zn) contributed for less than 1% of total emissions in all size fractions. Emissions of coarse particles were found to be dominated by resuspended matter as well as by brake wear, whereas fine particles were mainly derived from combustion processes. On weekends for some components distinctly reduced emissions were observed which could be explained with changes in the driving conditions and/or fleet composition.

Introduction

Vehicle emissions make substantial contributions, both directly and indirectly, to atmospheric particle concentrations. Direct particulate emission sources from vehicles include their exhaust (Mulawa et al., 1997; Sagebiel et al., 1997), the mechanical wear of tires and brakes (Rogge et al., 1993; Garg et al., 2000), and the ejection of particles from the pavement (Kupiainen et al., 2005) and unpaved road shoulders (Moosmüller et al., 1998) by re-suspension processes (Nicholson et al., 1989; Sternbeck et al., 2002). Indirect contributions include the emission of reactive gases, both organic and inorganic, which form secondary particulate matter via atmospheric transformations.

Motor vehicle emission inventories normally used in air quality models are derived from tailpipe emissions measurements conducted on motor vehicles operated over simulated driving cycles on a chassis dynamometer (Cadle et al., 1997, Cadle et al., 1999; Kleeman et al., 2000). Conventional chassis dynamometer tests do not measure non-tailpipe emissions such as those from tire and brake wear. Usually the number of vehicles investigated in such tests is relatively small, since dynamometer studies are expensive. Thus information about the chemical composition and the size distribution of the emitted particles is limited.

An alternative to single vehicle emissions measurements is to measure the emissions from a large population of on-road vehicles as they are driven through a highway tunnel (El-Fadel and Hashisho, 2001). In the last years a range of studies focused on the characterization of PM vehicular emissions from on-road fleets, nevertheless mass and chemically speciated particulate emission rates from on-road, in-use vehicles remain relatively limited. Most of the studies reported so far were based on short term experiments conducted within several hours (Fraser et al., 1998; Allen et al., 2001; Gillies et al., 2001; Lough et al., 2005), which therefore include only data for relatively constant traffic and driving conditions. Just a few studies provided data derived from consecutive sampling intervals over longer periods, which allowed estimation of particulate vehicle emissions under varying traffic and driving conditions. Additionally it has to be considered that most of these studies, in particular where metals are investigated, have been performed in the US (Allen et al., 2001; Gillies et al., 2001; Lough et al., 2005, Fraser et al., 1998; Chellam et al., 2005), whereas for European fleets only data from a few studies are available (Sternbeck et al., 2002; Valiulis et al., 2002; Laschober et al., 2004). Thus newer and if possible size segregated emission data are required for inventory modeling and source apportionment studies, especially for European car fleets which are known to differ in fleet composition, engine type, fuel consumption and driving conditions from US fleets.

The purpose of this paper is to present size segregated (PM2.5 and PM10) and total suspended particulate (TSP) matter vehicular emission rates (24 h mean values) for particle mass (PM), total carbon (TC), organic carbon (OC), elemental carbon (EC), mineral components (MCs) and various trace metals for a mixed car fleet measured in an European highway tunnel (Kaisermühlen-tunnel, Vienna, Austria). The derived results are considered to be representative for central-European real-world conditions, since the entire sampling period (28 consecutive days) covered a wide field of different traffic conditions (in particular changes in traffic density and fleet composition) and traffic situations including free cruising conditions and stop-and-go traffic. Thus, the presented data set could be used to approximate particulate emissions of motor vehicles operated under real world conditions, which are known to be highly variable (e.g. braking and acceleration maneuvers) rather than well defined (e.g. constant speed).