Vehicular emissions

It is well established that vehicle exhaust contains N2O (Smith and Carey, 1982; Metz, 1984; Prigent and De Soete, 1989; Dasch, 1992; Berges et al, 1993; Sjödin et al, 1995). Furthermore, it is widely accepted that new vehicles equipped with three-way catalysts generally emit more N2O than older vehicles without catalysts (Dasch, 1992; Berges et al, 1993; Hupa and Matinlinna, 1994; Sjödin et al, 1995; Siegl et al, 1996). In the early 1990s concern was expressed that N2O emissions would rise substantially as the global fleet of old vehicles was replaced with modern vehicles equipped with three-way catalysts to reduce urban air pollution (Berges et al, 1993).

In 1992, Berges et al (1993) studied N2O emissions from motor traffic in tunnels in Stockholm, Sweden, and Hamburg, Germany, and concluded that catalyst-equipped vehicles emit 106mg N2O km-1 (170mg N2O mile-1).

Extrapolation to the global car fleet led Berges et al (1993) to predict that global N2O emissions from vehicles could reach 6-32 per cent of the atmos -pheric growth rate. Sjodin et al (1995) studied N2O emissions from motor traffic passing though a different tunnel in Sweden in 1992 (in Gothenburg) and reported an average emission rate of 25mg N2O km-1. Sjodin et al (1995) estimated that the traffic passing through the tunnel consisted of approximately 10 per cent heavy-duty vehicles, 45 per cent catalyst-equipped and 45 per cent non-catalyst-equipped cars. If one takes the extreme assumption that N2O emission from heavy-duty vehicles and non-catalyst equipped cars is zero, then an EF of approximately 56mg N2O km-1 (90mg N2O mile-1) for catalyst-equipped vehicles could be derived, which is significantly less than that reported by Berges et al (1993).

To better understand the environmental impact of vehicle exhaust, Becker et al (1999) conducted field measurements of traffic passing through the Kiesbergtunnel in Wuppertal, Germany, and this was followed by laboratory studies at the Ford Motor Company using a chassis dynamometer facility to measure N2O emissions from vehicles (Becker et al, 2000b). Figure 9.4 shows as an example of N2O and CO2 mixing ratios from the measurements in the Kiesbergtunnel and their good correlation, which indicates that high concentrations of N2O measured in the tunnel resulted from vehicle exhaust. The combined database from the laboratory studies at the Ford Motor Company provided emission rates for non-methane hydrocarbons (NMHC), CO, CO2, NOx, NO, NO2, N2O and NH3 from 26 different vehicles powered by six different fuels and represents a comprehensive survey of the emissions of N-containing compounds from individual modern vehicles. The EF (g N2O g-1 CO2) measured in the 'real world' of the tunnel was 4.1 ± 1.2 X 10-5 while that measured in the 'laboratory' was 4.3 ± 1.2 X 10-5. The consistency between these EFs was rather surprising. Typically, 'real world' studies report emission rates that are significantly higher (perhaps by a factor of two to three) than those observed in the laboratory. This has been ascribed to the fact that a small fraction of the 'real world' vehicle fleet is poorly maintained and contributes disproportionately to the observed pollutant levels. The consistency between the 'real world' and 'laboratory' results reported in the papers by Becker et al (1999, 2000b) may reflect either: (1) the contribution of heavy-duty and non-catalyst vehicles in the tunnel that are believed to emit less N2O than catalyst-equipped vehicles; or (2) that N2O emissions are not very sensitive to the maintenance condition of the vehicle.

In the light of the results of Jiménez et al (1997, 2000), the former explanation seems more likely. In any case, the results suggest that a reasonable estimate of the EF is (4 ± 2) X 10-5 when calculating traffic-related N2O emissions for emission inventory purposes. This range is consistent with the overall EF measured by Berges et al (1993) in 1992 in a tunnel study in Germany of (6 ± 3) X 10-5 but is substantially lower than the range of (1.4 ± 0.9) X 10-4 derived from a similar study in Sweden (Berges et al, 1993).

It is useful to place these emission rates into perspective in terms of vehicle contributions to the global N2O budget and to radiative forcing of climate change. The contribution of global vehicular traffic to the global N2O budget can be estimated using two different approaches. First, one could assume that the N2O observed in the tunnel studies is attributable solely to catalyst-equipped passenger cars, calculate the EF for such vehicles, and combine this result with an estimate of their global fuel consumption. Second, one could assume that the vehicle mix travelling through the Wuppertal tunnel is representative of the global vehicle population and simply multiply the measured EF by the global vehicle fuel consumption. For simplicity, and because the tunnel measurements do not support the assumption that the N2O is attributable solely to catalyst-equipped passenger cars, Becker et al (1999) adopted the second approach. Using values of 964Tg for the annual global vehicle fuel consumption in 1995 (637Tg of gasoline, 327Tg of automotive diesel (Associated Octel, 1996), 0.855 for the average carbon content of gasoline by mass (Marland and Rotty, 1984), 44 for the molecular weights of N2O and CO2, 12 for the atomic weight of carbon, and 4 ± 2 X 10-5 for the EF of N2O, the authors estimated the contribution of vehicular traffic to the global N2O budget to be 964 X 0.855 X (44/12) X (4 ± 2) X 10-5 = 0.12 ±

0.06Tg. Atmospheric levels of N2O are increasing at a rate of 4.7 ± 0.9Tg yr-(3.0 ± 0.6Tg N yr-1) (Berges et al, 1993). Hence, emissions from the global vehicle fleet represent approximately 1-4 per cent of the atmospheric growth rate of N2O. The global warming potential of N2O is 298 times that of CO2 (IPCC, 2007). Using an emission factor (g N2O g-1 CO2) of (4 ± 2) X 10-5 it follows that N2O emissions from vehicles have a global warming impact of 12 per cent of that of the CO2 emitted from vehicles.

Figure 9.4 Correlation between CO2 and N2O emitted from vehicular traffic in Wuppertal Kiesbergtunnel

Source:After Becker et al (2000b)

Figure 9.4 Correlation between CO2 and N2O emitted from vehicular traffic in Wuppertal Kiesbergtunnel

Source:After Becker et al (2000b)

Becker et al (1999) concluded that N2O emissions from vehicles make minor (though non-negligible) contributions to the global atmospheric N2O budget and to anthropogenic radiative forcing of global climate change. These findings are in disagreement with the conclusions of the previous study by Berges et al (1993) described above, i.e. that if the entire fleet of cars were to be equipped with catalysts the global N2O emissions from vehicles would double and reach 6-32 per cent of the atmospheric growth rate. This work of Berges et al (1993) took place in 1992, when catalyst-equipped passenger cars represented only a small fraction (20-30 per cent) of the total traffic. In contrast, in 1997 catalyst-equipped passenger cars dominated the traffic volume. Berges et al (1993) assumed that N2O emissions from non-catalyst cars and trucks were zero and derived EFs by attributing all of the measured N2O to the relatively small number of catalyst-equipped cars. The results from the work of Becker et al (1999, 2000b) indicate that this assumption is not valid and that N2O emissions from catalyst cars operating under 'real world' conditions are substantially lower than previously reported. Becker et al (1999, 2000b) recommended the use of an EF of (4 ± 2) X 10-5 in emission inventory calculations for the modern vehicle fleet.

Since 2000, several publications have appeared reporting N2O emissions from various type of vehicles (for example Durbin et al, 2003; Huai et al, 2003, 2004; Behrentz et al, 2004; Karlson, 2004; Winer and Behrentz, 2005). Data from these publications were added to the body of data reported in a review (US EPA, 2004) by the United States Environmental Protection Agency (US EPA).

It is nowadays commonly accepted that N2O emissions from light-duty gasoline-powered vehicles are very dependent on the type of pollution control technology and the age of this technology. Emissions also vary with the vehicle in question and with operating conditions such as fuel sulphur level, driving cycle, ambient temperature and catalyst operating temperature (Jobson et al, 1994; Laurikko and Aakko, 1995; Michaels, 1998; Odaka et al, 1998, 2000, 2002; Koike et al, 1999; Baronick et al, 2000; Meffert et al, 2000). The most up-to-date EFs for light-duty gasoline- and diesel-powered vehicles have been published very recently by Graham et al (2009), who investigated the N2O emission from 200 of those vehicles between April 2004 and June 2007. This work was mainly triggered by the fact that major changes in both vehicle and fuel specifications had occurred during the previous ten years. Taken together with data from the same group going back to 2004, information on a test fleet of 467 vehicles is available, representing a wide range of emission standards.

The data analysis has shown that the distinction between light-duty auto -mobiles and light-duty trucks within a given emission standard is not significant, whereas the distinction between new and aged catalysts remains. The N2O emission rates appeared to be well correlated with the numerical NMHC or non-methane organic gases (NMOG) emission standard to which the vehicles were certified, but less so with NOx or CO emission standards. Of the vehicles investigated, those with aged catalysts showed a greater rate of increase of N2O emissions with NMHC or NMOG emission standard than did new vehicles.

To what extent new NOx reduction techniques such as the AdBlue® may alter the N2O emission in particular from heavy-duty vehicles is still an open question and needs further clarification. The basic principle of AdBlue®, which is used in heavy-duty vehicles, is very similar to the thermal DeNOx process used in stationary combustion; an aqueous solution of urea is injected into a SCR catalyst, leading to almost 90 per cent NOx reduction in the exhaust.

0 0

Post a comment