Working Group III: Mitigation

Other reports in this collection Direct Injection Gasoline and Diesel Engines

Direct injection lean-burn gasoline engines have already been introduced in Japan and Europe, but have been restricted in North America by a combination of tight emission standards and high sulphur content in gasoline. Fuel sulphur levels will be drastically reduced in Europe and North America over the next 10 years. The US EPA, for example, has proposed regulations that would set caps on sulphur content of 30 ppm for gasoline and 15 ppm for diesel fuel (Walsh, 2000). While planned reductions in the sulphur content of fuels to the range of 10 to 30 ppm will allow direct injection gasoline engines to be introduced, it is not yet clear that the full fuel efficiency benefits can be retained at lower NOx levels. Preliminary evaluations suggest that benefits may be in the 12% to 15% range rather than the 16% to 20% range available in Japan and Europe, but even this assumes some advances in after treatment technology. Engine costs, however, seem quite moderate, in the range of US$200 to US$300 more than a conventional engine.

Direct injection (DI) diesel engines have long been available for heavy trucks, but recently have become more competitive for automobiles and light trucks as noise and emission problems have been resolved. These new engines attain about 35% greater fuel economy than conventional gasoline engines and produce about 25% less carbon emissions over the fuel cycle. In light-duty applications, DI diesels may cost US$500 to US$1000 more than a comparable gasoline engine. Tightening of NOx and particulate emissions standards presents a challenge to the viability of both diesel and gasoline lean-burn engines, but one that it may be possible to overcome with advanced emissions controls and cleaner fuels (e.g., Martin et al., 1997; Gerini and Montagne, 1997; Mark and Morey, 1999; Greene, 1999). Further improvements in diesel technology also offer substantial promise in heavy-duty applications, especially heavy trucks but also including marine and rail applications. Current research programmes are aiming to achieve maximum thermal efficiencies of 55% in heavy-duty diesels (compared to current peak efficiencies of about 40%-45%), with low emissions. Automotive Fuel Cells

Fuel cells, which have the potential to achieve twice the energy conversion efficiency of conventional internal combustion engines with essentially zero pollutant emissions, have received considerable attention recently, with most major manufacturers announcing their intentions to introduce such vehicles by the 2005 model year. The recent optimism about the fuel cell has been driven by strong advances in technology performance, including rapid increases in specific power that now allow a fuel cell powertrain to fit into a conventional vehicle without sacrificing its passenger or cargo capacities. While fuel cell costs have been reduced by approximately an order of magnitude, they are still nearly 10 times as expensive per kW as spark ignition engines. Recent analyses project that costs below US$40/kW for complete fuel cell drivetrains powered by hydrogen can be achieved over the next ten years (Thomas et al., 1998). Hydrogen is clearly the cleanest and most efficient fuel choice for fuel cells, but there is no hydrogen infrastructure and on-board storage still presents technical and economic challenges. Gasoline, methanol or ethanol are possible alternatives, but require on-board reforming with consequent cost and efficiency penalties. Mid-size fuel cell passenger cars using hydrogen could achieve fuel consumption rates of 2.5 gasoline equivalent l/100 km in vehicles with lightweight, low drag bodies; comparable estimates for methanol or gasoline-powered fuel cell vehicles would be 3.2 and 4.0 l/100 km (gasoline equivalent), respectively. While gasoline is relatively more difficult to reform, it has the benefit of an in-place refuelling infrastructure, and progress has been made in reformer technology (NRC, 1999a).

The fuel economy of hydrogen fuel cell vehicles is projected to be 75% to 250% greater than that of conventional gasoline internal combusiton engine (ICE) vehicles, depending on the drive cycle (Thomas et al., 1998). Primarily as a result of energy losses in reforming, comparable estimates of the fuel economy benefit of methanol-powered fuel cells range from 25% to 125%. The GHG reduction potential of hydrogen or methanol fuel cells, however, requires a “well-to-wheels” analysis to measure the full fuel cycle impacts. Both sources cited here include emissions of all significant greenhouse gases produced in the respective processes. Assuming hydrogen produced by local reforming of natural gas, Thomas et al. (1998, Figure 8) estimated roughly a 40% reduction in well-to-wheels GHG emissions for a direct hydrogen fuel cell vehicle versus a conventional gasoline ICE vehicle getting 7.8 l/100km (about 150 g CO2 equivalent per km, versus 250). Wang (1999a, p. 4) concluded that direct hydrogen fuel cell vehicles, with hydrogen produced at the refuelling station by reforming natural gas, would reduce full fuel cycle GHG emissions by 55% to 60% versus a comparably sized 9.8 l/100km gasoline vehicle. Hydrogen could also be produced from methane in large-scale centralized facilities. This could create opportunities for sequestering carbon but would also require an infrastructure for hydrogen transport. Hydrogen produced via electrolysis was estimated to produce 50% to 100% more full fuel cycle GHG emissions, depending on the energy sources used to generate electricity. Methanol produced from natural gas was estimated to give a 50% reduction in full fuel cycle GHG emissions. Wang (1999b, Table 4.4) projected direct hydrogen fuel cell vehicles to be 180% to 215% more energy efficient, and methanol fuel cell vehicles to be 110% to 150% more efficient. These analyses attempt to hold other vehicle characteristics constant but, of course, that is never entirely possible. Fuel Cycle Emissions

In considering the impacts of advanced technologies and alternative fuels on emissions of greenhouse gases, it is important to include the full fuel cycle, since emissions in feedstock and fuel production can vary substantially. The same fuel can be produced from several feedstocks, and this too has important implications for greenhouse gas emissions. Finally, as Ishitani et al. (2000) have demonstrated, the use of different drive cycles as a basis for comparison can also change the ranking of various advanced technologies. Hybrid vehicles, for example, will perform relatively better under congested, low-speed driving conditions. Table 3.9 shows a sample of results obtained by Wang (1999a) based on US assumptions for passenger car technologies expected to be available in the year 2010. In all cases, carbon dioxide is the predominant GWP-weighted greenhouse gas. Advanced direct injection gasoline engines appear to achieve nearly the same greenhouse gas emissions reductions as spark-ignition engine vehicles fuelled by propane or compressed natural gas. Direct-injection diesel vehicles show a reduction of one-third over advanced gasoline vehicles. The gasoline hybrid achieves almost a 50% reduction, while the grid-connected hybrid does no better because of the large share of coal in the US electricity generation mix. The dependence of electric vehicle (EV) emissions on the power generation sector is illustrated by the very large difference between EVs using California versus US average electricity. Fuel cell vehicles using gasoline are estimated by Wang (1999a) to achieve a 50% reduction in emissions, but hybrid vehicles fuelled by compressed natural gas (CNG) do slightly better. Fuel cells powered by hydrogen produced by reforming natural gas locally at refuelling outlets are estimated to reduce fuel cycle greenhouse gas emissions by almost two thirds, while those using hydrogen produced from solar energy achieve more than a 90% reduction. Clearly, Wang’s (1999b) estimates differ substantially from those of Thomas et al. (1998) as noted above. Such differences are common, as a result of differences in the many assumptions that must be made in fuel cycle analyses.

Table 3.9: GHG emissions from advanced automotive technologies and alternative fuels
(Wang, 1999,
App. B-II).
  CO2-equivalent grams per km
  Fuel cycle stage
  Greenhouse gas
  Feedstock Fuel Operation   CO2 N2O CH4 Total
Gasoline (reformulated) 15.6 52.7 228.9   282.2 5.7 9.4 295.6
Gasoline direct injection (DI) 12.6 42.1 184.3   225.6 5.7 7.7 237.6
Propane (from natural gas) 19.0 13.6 197.6   217.5 5.5 7.3 228.9
Compressed natural gas (CNG) 30.7 21.3 174.6   206.2 3.1 17.3 225.3
Diesel DI 10.6 27.2 161.6   191.7 3.3 4.5 198.4
20% biodiesel DI 11.7 32.7 132.7   169.1 3.7 4.3 176.1
Grid-Hybrid (RFG) 9.8 63.5 88.8   152.7 4.1 5.3 161.2
Hybrid (RFG) 8.6 27.5 123.3   148.3 5.7 5.4 158.5
Electric vehicle (EV, US mix) 12.3 145.2 0.0   152.1 0.6 4.8 156.6
Fuel Cell (Gasoline) 7.8 26.1 112.6   140.8 1.4 4.4 145.7
Hybrid. CNG 19.1 13.2 110.5   127.6 2.7 12.5 142.0
Fuel cell (methanol. NG) 8.1 17.9 83.1   105.0 1.2 3.0 108.5
Fuel cell (H2 from CH4)) 11.0 97.3 0.0   103.1 0.2 5.0 107.62
EV (CA mix) 10.4 51.1 0.0   58.5 0.2 2.8 61.1
Fuel cell (solar) 0.0 20.3 0.0   18.9 0.2 1.2 20,2
  100-year global warming potentials          
  CO2 N2O CH4          
  1 310 21          

Other reports in this collection

IPCC Homepage