Aviation and the Global Atmosphere

	Other reports in this collection

7.5.3. Production of Engine Emissions

Pollutant formation in combustion, regulated aircraft emissions during landing/take-off, and currently unregulated cruise emissions are covered in some depth by Brasseur et al. (1998). For convenience, we briefly describe the principal points here, with somewhat greater detail given to more recent findings about soot and particulate emissions.

Under ideal conditions, combustion of kerosene-type fuels produces carbon dioxide CO2) and water vapor (H2O), the proportions of which depend on the specific fuel carbon to hydrogen ratio. Figure 7-14 shows the ideal and "real" combustion processes. Figure 7-14 also illustrates the scale of the combustion products by showing that at cruise conditions they constitute only about 8.5% of total mass flow emerging from the engine. Of these combustion products, only a very small volume (about 0.4%) of residual products arise from non-ideal combustion processes (soot, HC, and CO) and the oxidation of nitrogen (NOx).

Table 7-2 gives typical emission levels for various operating regimes. The emission values are quoted as an emission index (EI) in units of grams of emittant species per kilogram of fuel burned.

Table 7-2 clearly illustrates the constancy of emissions indices for CO2, H2O, and SOxO throughout the flight cycle. These emissions are directly related to the fuel consumption of the engine in its various flight phases. In contrast, emissions such as NOx, CO, HC, and soot are strongly influenced by a wide range of variables but particularly engine power setting and ambient engine inlet conditions. CO and HC are products of incomplete combustion. They are highest at low power settings, when the temperature of the air is relatively low and fuel atomization and mixing processes least efficient. This problem area is proving responsive to improvements linked to detailed studies of basic fuel/air mixing processes. NOx and soot (not shown in the table), on the other hand, are highest at high power settings.

The majority of NOx emissions are generated in the highest temperature regions of the combustor-usually in the primary combustion zone, before the products are diluted. The fundamental processes of NOx formation are well known and documented (reviewed and described in detail by Bowman, 1992). They are best expressed as a function of local combustion temperature, pressure, and time. Combustion zone temperature depends on combustor inlet air temperatures and pressure, as well as the fuel/air mass ratio. The dependence of NOx on fuel/air ratio is shown in Figure 7-15. As illustrated, peak NOx formation coincides with peak temperature, which occurs close to the stoichiometric fuel/air ratio (or equivalence ratio = 1). In current gas turbine engine combustors, there are always some regions of the flame that burn stoichiometrically, so NOx formation is very strongly linked to combustor inlet temperature.

"Soot" generally refers to particulates in emissions. These particles are composed primarily of carbonaceous material, the sum of graphite carbon and primary organics resulting from incomplete combustion of carbonaceous material (Novakov, 1982; Chang and Novakov, 1983). "Smoke" refers to combustion emissions particulates that contribute to a visible plume. The formation of soot and its partial oxidation in gas turbine combustors are very complex processes. Soot is produced mainly in the fuel-rich primary zone of the combustor, then oxidized in the high-temperature regions of the dilution and intermediate zone. A simplified description of the soot production mechanism is provided in Figure 7-16 (Mullins, 1988).

Estimates of the fleet averaged emission index of soot is 0.04g/kg fuel burned (D�pelheuer, 1997). However, the derivation of this estimate carries considerable uncertainty and at best is only within a factor of 2. The formation of soot in gas turbine combustors and the precise design features and conditions that can influence the process continue to be subjects of important ongoing research. Since 1990, there has been a marked improvement in the engine soot and aerosol emissions database through skilled use of the most modern aerosol measurement techniques (Hagen et al., 1992; Rickey, 1995; Pueschel et al., 1997; Petzold and Schr�der, 1998). State-of-the-art instrumentation permits detection of particles as small as 3 nm in diameter (Alofs et al., 1995). These data also show that jet engines emit soot particles, which have log normal-type size distributions peaking in the 20-30 nm range. Concentrations range between 10⁶ and 10⁷ particles cm^-3 (Hagen et al., 1992; Whitefield and Hagen, 1995; Schumann et al., 1996; Pueschel et al., 1997; Anderson et al., 1998; Petzold and Schr�der, 1998). Number-based emission indices fall within the range of 10¹² soot particles per kg of fuel burned for current advanced combustors to 10¹⁵ for older type engines (Howard et al., 1996; Whitefield et al., 1996). A general range for most engines operating in the current commercial fleet is 10¹⁴-10¹⁵ particles per kg fuel burned (Hagen et al., 1996; D�pelheuer, 1997; Anderson et al., 1998; Petzold and Schr�der, 1998; Petzold et al., 1999).