Personal Cars and China (2003)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

4 Present and Future Automotive Technologies T he tenth five-year plan envisions a future for China in which cars will be widely available to Chinese families and in which the Chi- nese automotive industry will grow into a “pillar industry” of China’s economy. Technology will play an important role in facilitating these goals. This chapter describes the automotive technology options that are available today and some that may become available in the longer term, and it comments on their applicability to the development of China’s au- tomotive industry and road transportation fleet. The choice of automotive technology is closely related to the kinds of materials and fuels selected, the economic impacts of the technology, and infrastructure requirements. The development of substantial transportation infrastructure, which re- quires major investments in land, influences in turn how land is devel- oped. These issues are complex and variable, strongly depending on local values and conditions (see Chapter 6 for more discussion about the effects of motorization on land use). Because the car is part of a much larger system, the overall costs and benefits of each technology choice must be assessed on a system basis. Such an assessment may reveal that the optimization of a single compo- nent may not be the best choice when the total system is considered. This life cycle analysis has been used in technology assessment studies world- wide and is appropriate for strategic decision making. Although this chapter focuses on automotive technologies, many of the issues considered elsewhere in this report—automotive industry is- sues (Chapter 3), energy/fuels issues (Chapter 5), societal change (Chap- 61 

62 PERSONAL CARS AND CHINA ter 6), environmental and health concerns (Chapter 7), and government policies (Chapter 8)—impinge on the choices of technology. China has already decided to require Chinese cars to meet European Emission Stan- dard I (Euro I) now and Euro II standards by 2004–2005, based on state environmental protection regulations. The European Union1 has been en- acting increasingly stringent emissions standards and now requires Euro III standards for its member nations, with Euro IV standards planned for implementation in Europe in 2005. State plans call for Chinese cars to attain the current Euro emissions standards by 2010. To meet the “wide availability” objective (that is, be affordable), the “China car” would have to cost about RMB60,000–80,000 ($7,200–9,600). Below this price range, the basic requirements for emissions control, per- formance, and safety cannot realistically be met. To make the automotive industry into a pillar of the economy, China will have to build its domestic capabilities for supplying materials, manu- facturing components, and providing the design, assembly, maintenance, sales, credit, and other services that in turn will contribute to an increase in China’s gross domestic product (GDP). As discussed in Chapter 3, con- solidation of the Chinese automotive industry and improvements in effi- ciency and cost competitiveness will take on new importance. Imported cars and components create unfavorable trade balances for China, whereas exports will expand opportunities for its automotive industry. Where technology choices bear on the future competitiveness of the in- dustry, these factors must be considered carefully. Likewise, the China car should be highly fuel-efficient, because China’s domestic supply of petroleum fuel is limited at present and higher fuel demand can be met in the near future only through increased im- ports. This situation will affect China’s balance of trade over the next de- cade and perhaps longer. The China car also should be highly “clean” in emissions, because the air quality in many Chinese large cities ranks among the worst in the world. The situation is expected to worsen if ve- hicle tailpipe emissions are not controlled rigorously. The buyer of the China car will have a heavy influence on automobile design and sales. Successful auto companies select product attributes that maximize markets. Consumers typically want the most performance and convenience they can afford. Initially, most Chinese drivers will be lo- cated in urban areas or on rural and intercity roads that are not designed for high speeds. The performance requirements and fuel efficiency of cars are heavily influenced by the “driving cycles” used in designing and op- 1 Through rulemaking proposed by the European Union Council of Environmental Minis- ters and approved by the European Parliament. 

PRESENT AND FUTURE AUTOMOTIVE TECHNOLOGIES 63 erating the vehicle. In the United States, where motorists engage in a lot of high-speed highway driving, many cars are designed with large engines to enable passing and hill-climbing capability at high speed. But these cars have poor fuel efficiency at low speeds and continue to burn fuel at a high rate when idling in urban traffic jams. Cars designed specifically for urban driving cycles could be much more fuel-efficient under these con- ditions. For example, new technologies based on greater electrification of the car are being developed to switch the engine off during vehicle stops. This feature has the double benefit of reducing emissions and wasted fuel. The envisioned China car also will have to be highly reliable, easily ser- viceable, and rugged to accommodate China’s present road and mainte- nance infrastructure. Over the century of development of automobiles in member coun- tries of the Organisation for Economic Co-operation and Development (OECD), governments and automakers continually responded to prob- lems and applied the lessons learned to the development path. For ex- ample, pollution problems led to increasingly stringent emissions regula- tions; safety problems led to improved roadway designs, crashworthy vehicle designs, and safety features such as seat belts and air bags; and operational problems led to driver training and vehicle inspection pro- grams. China, facing an anticipated period of rapid investment and growth for automobiles and the associated infrastructure of a decade or two rather than a century, will benefit from those lessons as well. The technology choices that must be made in China are constrained by the marketplace and by various government policies and regulations. With that in mind, this chapter will focus on technologies that seem to best fit the rapidly growing automotive industry in China over the next five years or so, and will examine them within the context of both their short- and long-term implications for the industry and consumers. The discussion of near-term technologies will be placed in the context of cur- rent expectations of government policies. The discussion of longer-term technologies will address those that could be important whether current government policies remain as they are or not. FLEET ISSUES FOR VEHICLE TECHNOLOGIES Emissions Control In view of the projected growth of China’s vehicle fleet, air quality issues in major urban areas are of great importance. As just noted, Chi- nese cars produced in 2004–2005 must meet Chinese emissions standards equivalent to the Euro II standards. China’s Technical Policy on Preven- tion and Control of Motor Vehicle Pollution states that emissions levels 

64 PERSONAL CARS AND CHINA Miles per hour Miles per hour Miles per hour Miles per hour Seconds Seconds FIGURE 4-1 Driving cycles for measuring emissions. NOTE: 20 miles per hour (mph) = 32 kilometers per hour (kph); 40 mph = 64 kph; 60 mph = 97 kph; 80 mph = 129 kph. should approach the international level of control around the year 2010 (Chinese State Environmental Protection Agency, 1999). Table 4-1 sum- marizes the principal elements of the existing U.S., Japanese, and Euro- pean standards. Emissions control technologies to meet these standards are discussed later in this chapter. Vehicle emissions are the products of the incomplete combustion of propulsion fuels. Propulsion system efficiencies vary with output energy requirements, and different propulsion technologies achieve optimum ef- ficiency under different conditions. In urban driving, conventional en- gines continue to run, using energy and producing emissions, even if a vehicle is stopped or moving slowly in traffic. Because any comparison of vehicle options must take into account the typical local driving condi- tions, when emissions standards are set they are based on a specified “driving cycle.” Figure 4-1 presents four examples of driving cycles speci- fied by regulatory groups in the United States, Japan, and Europe. 

TABLE 4-1 Examples of Emissions Standards in the United States, Japan, and Europe Date of Implementation CO HC NOx PM U.S. Standards SI car 1997 3.4 g/mi 0.41 g/mi 0.40 g/mi N.S. (2.1 g/km) (0.26 g/km) (0.25 g/km) SI and diesel cars 2007 2.1 g/mi 0.09 g/mi 0.07 g/mi 1.01 g/mi (1.3 g/km) (0.06 g/km) (0.04 g/km) (0.006 g/km) Japanese Standards SI car 2000 0.67 g/km 0.08 g/km 0.08 g/km N.S. Diesel car 2002 0.63 g/km 0.12 g/km 0.30 g/km 0.056 g/km European Standards Euro II SI car 1996 3.28 g/km 0.34 g/km 0.25 g/km N.S. Euro II diesel car 1996 1.06 g/km 0.19 g/km 0.73 g/km 0.10 g/km Euro III SI car 2000 2.30 g/km 0.20 g/km 0.15 g/km N.S. Euro III diesel car 2000 0.64 g/km 0.06 g/km 0.50 g/km 0.05 g/km Euro IV SI car 2005 1.00 g/km 0.10 g/km 0.08 g/km N.S. Euro IV diesel car 2005 0.50 g/km 0.05 g/km 0.25 g/km 0.025 g/km NOTE: CO = carbon monoxide; HC = hydrocarbons; NOx = nitrogen oxides; PM = particulate matter; SI = spark ignition; g/mi = grams per mile; g/km = grams per kilometer; N.S. = no specification. SOURCE: Yasuhira Daisho, Waseda University. 65 

66 PERSONAL CARS AND CHINA In the United States, emissions are set on a per mile basis using a combined urban-highway cycle. Japanese and European driving cycles are more similar to the U.S. urban driving cycle than to the U.S. highway driving cycle. A comparison of the urban driving cycles used in Japan (Figure 4-1c), Europe (Figure 4-1d), and the United States (Figure 4-1a) to measure emissions indicates that in the Japanese and European cycles more time is spent with the vehicle stopped, engine idling, than in the U.S. cycle, which has an higher average speed and less idling time. Figure 4-1b depicts the U.S. highway cycle that was used in combination with the U.S. city cycle to determine vehicle fuel economy (45 percent share of the highway cycle and 55 percent of the city cycle). When vehicles are operated only in urban environments where traffic constrains their rate of acceleration and maximum speed, users have mod- est expectations of vehicular performance in terms of acceleration and top speed. If users are able to drive vehicles on improved roads at high speeds with little competing vehicular traffic, their expectations for engine power and high-speed capability will rise. Because China has chosen the Euro emissions standards, it will have to measure emissions performance for the driving cycle that is established for the Euro system. In the future, however, China may choose to implement new standards, with a driving cycle and emissions standards suited to its specific environment. Energy Use and Fuel Economy The energy use of a vehicle fleet depends on the size, weight, type, and efficiency of vehicles in the fleet and on the driving conditions en- countered in their use. Energy use includes not only the fuel consumed in operating the vehicle, but also the energy consumed in making the ve- hicle, producing and processing the fuel, and disposing of the vehicle at the end of its life. Life Cycle Assessment Anyone weighing the automotive technology options discussed later in this chapter would benefit from comparing alternative automotive tech- nology systems over their full life cycle (see Figure 4-2, which depicts the steps in the life cycle of automotive technology from the production of the raw materials used to make the fuel and the vehicle through the vehicle’s useful life to its final disposition). Such an assessment would allow one to track the key parameters involved through each of these life cycle stages and assess the overall results as part of the technology selection process. For the vehicle, the major parameters of interest are: cost, performance, local emissions of air pollutants, greenhouse gas (GHG) emissions, and 

PRESENT AND FUTURE AUTOMOTIVE TECHNOLOGIES 67 Sources Sources of of raw primary materials energy Government Vehicle Fuel at all levels manufacturer manufacturer Vehicle distribution, Fuel repair, and distribution Vehicle purchaser or user maintenance Scrapping and recycling FIGURE 4-2 Steps in the life cycle of automotive technology. SOURCE: Weiss et al. (2000). energy use. The vehicle also must have attributes that make it attractive to its purchasers and users over its lifetime, and it must meet established regulatory standards. Various recent studies have compared the technologies for clean, fuel- efficient cars (Office of Technology Assessment, 1995; Automotive Engi- neering, 1996; Sierra Research, 1997; Höhlein et al., 1998; Singh et al., 1998; Ogden et al., 1999; NRC, 2000; Pembina Institute, 2000; Weiss et al., 2000; GMC, 2001). Not all of these studies treat the full life cycle of comparable fuel/vehicle systems. For example, the U.S. Partnership for a New Gen- eration of Vehicles (PNGV) program does not consider the fuel cycle in its performance goals or comparisons of vehicles (NRC, 2000). And the Gen- eral Motors study does not provide information on the production costs of new vehicles, nor does it state detailed design assumptions about the vehicles evaluated. The results of the study by Weiss et al. (2000) are sum- marized here to illustrate the importance of a life cycle review. Beyond the organizations directly involved in producing fuels and vehicles are the vehicle purchasers and various levels of government. Those purchasing vehicles make their choices based largely on affordability, convenience, comfort, availability, and appearance. Local governments impose local zoning and safety codes; subnational govern- ments issue planning, tax, and regional environmental regulations; and ultimately the national government is responsible for central investments in infrastructure, for national tax policies, and for the national trade, envi- 

68 PERSONAL CARS AND CHINA ronmental, safety, and other requirements that are applied to fuels and vehicles. In the life cycle assessment used here for illustrative purposes (Weiss et al., 2000), the starting point is a vehicle similar to a 1996 Toyota Camry. It is assumed that this vehicle, the baseline vehicle for the study, evolves forward to the year 2020 into a car of similar performance and capacity. The analysis is based on a U.S. Environmental Protection Agency (U.S. EPA) combined city-highway driving cycle. Various choices in technol- ogy are then considered, resulting in the following set of study cases: • Baseline vehicle—the 2020 version of a car similar to a 1996 Toyota Camry, with a gasoline engine, 600 kilometer (km) refueling range, and some body lightweighting • Advanced body vehicle—a similar car that is about 10 percent lighter than the baseline because of changes in materials and that costs about 10 percent more • Advanced body, diesel (with both petroleum-based fuel and Fischer-Tropsch synthetic diesel fuel made from natural gas) • Advanced body, hybrid—gasoline, diesel, or compressed natural gas (CNG) • Advanced body, fuel cell hybrid—dependent on reforming gaso- line, reforming methanol, or utilizing high-pressure hydrogen gas made from natural gas • Advanced body, electric (requiring recharging every 400 km be- cause of battery limitations). The comparisons that follow are based on these vehicle designs. But because forecasting technology some 20 years in the future involves some uncertainty, the forecasts here related to evolutionary and advanced body combustion engine cars are subject to underestimation or overestimation by about 10 percent; the hybrids by about 20 percent, and the fuel cell and electric vehicles by about 30 percent. The uncertainties are greater for the rapidly evolving technologies because of the possibilities of technological breakthroughs or the identification of unanticipated barriers. The study by Weiss et al. (2000) assumes a vehicle lifetime of 15 years and an annual distance traveled of 20,000 km. Life Cycle Energy Consumption Figure 4-3 shows the life cycle energy consumption, on a megajoule (MJ) per kilometer basis, of the various technology combinations evalu- ated. The top portion of each bar represents the energy equivalent of the fuel used by the vehicle in propulsion, the middle portion the energy used 

PRESENT AND FUTURE AUTOMOTIVE TECHNOLOGIES 69 Vehicle Operation Fuel Cycle Embodied 4 3 Energy (MJ/km) 2 1 0 e r e el el id id id id id d id c ca tri lin lin 2 F bri es es br br br br br br ec so so e hy hy hy hy hy hy hy di di lin el ga ga AB T e el el G FC FC C se AB F- lin N es es ry AB ba C so e H am di di AB lin H eO AB ga 20 AB T AB so C F- M 20 AB 96 ga AB AB 19 AB FIGURE 4-3 Comparisons of life cycle energy use. NOTE: MJ/km = megajoule per kilometer; AB = advanced body; F-T = Fischer-Tropsch; CNG = compressed natural gas; FC = fuel cell; MeOH = methanol; H2 = hydrogen. SOURCE: Weiss et al. (2000). in producing that amount of fuel, and the bottom portion the energy used in the actual manufacture of the vehicle (embodied energy). For the cases studied, the energy involved in vehicle manufacture is a relatively small part of the life cycle energy use. For much smaller, more efficient vehicles, the embodied energy becomes a more significant factor. For petroleum fuels and natural gas, the energy required for fuel pro- duction is a relatively small part of total energy requirements (Figure 4-3). However, the energy associated with producing a synthetic fuel by a Fischer-Tropsch process adds substantially to life cycle energy use, which is related to GHG emissions as well. The energy production requirement for electric vehicles is based on the mix of energy sources and power gen- eration efficiencies of the U.S. electricity supply. On that basis, the fuel cycle is the predominant energy requirement for the electric vehicle. Life Cycle Costs and Emissions Figure 4-4 presents similar comparative data, but it looks at total an- nual operating costs, new vehicle costs, and total GHG emissions. It is assumed that the vehicles would have to incorporate technology to meet 

70 PERSONAL CARS AND CHINA Annual operating cost (new) × $0.1/km Vehicle cost (new) × $10,000 8 Carbon × 10 gC/km 7 6 5 4 3 2 1 0 r e el el rid rid rid rid rid rid rid c e ca tri lin es es lin b b b b b b b ec so e hy hy hy hy hy hy hy so di di lin el ga ga AB T e el el G FC FC FC se AB F- lin es es N AB ry ba 2 C so e H AB di di H am lin eO AB ga 20 AB T AB so F- C M 20 AB ga 96 AB AB AB 19 FIGURE 4-4 Life cycle comparisons of costs and carbon emissions. NOTE: An- nual operating costs include amortized new vehicle costs and running costs as shown in Table 4-2. AB = advanced body; F-T = Fischer-Tropsch; CNG = com- pressed natural gas; FC = fuel cell; MeOH = methanol; H2 = hydrogen; gC/km = grams carbon per kilometer. SOURCE: Weiss et al. (2000). the 2020 U.S. emissions standards for local pollutants. The pattern of GHG emissions follows that of energy use for the petroleum fuel-based sys- tems, but emissions are somewhat reduced for the CNG-fueled hybrid. GHG emissions for the hydrogen fuel cell and for the electric car are slightly higher than those for the gasoline hybrid, based on the U.S. elec- tric sector average GHG emissions. Table 4-2 presents estimates of the annual operating costs for new U.S. vehicles based on fuel cost averages and vehicle fuel consumption. A flat fuel tax of $0.0033 (RMB0.027) per megajoule of fuel ($0.40 per gallon of gasoline equivalent) is used across all the fuel sources (this assumption is made so that tax policy does not affect relative results; taxation is a policy tool that may be used to influence the economic choice between technologies). A constant maintenance or other charge of $0.036 (RMB0.30) per kilometer for the various technologies is assumed to avoid introduc- ing an additional bias. Total fixed costs are based on known fixed annual 

TABLE 4-2 Vehicle Costs per Kilometer for Selected New Vehicle Options, 2020 (1997 U.S. dollars per kilometer) AB AB Meth- AB Evolved AB AB Gasoline anol Hydro- Body AB AB Gasoline Diesel AB, CNG FC FC gen FC AB Gasoline Gasoline Diesel Hybrid Hybrid Hybrid Hybrid Hybrid Hybrid Electric Total 0.056 0.053 0.047 0.049 0.044 0.049 0.056 0.050 0.054 0.045 running costs Fuel ex tax 0.014 0.012 0.007 0.009 0.005 0.010 0.014 0.010 0.015 0.007 (percent of (5%) (4%) (2%) (3%) (1%) (3%) (4%) (3%) (4%) (2%) total) FT=0.009 FT=0.006 Fuel tax 0.006 0.005 0.004 0.004 0.003 0.003 0.006 0.004 0.003 0.002 Other (oil, 0.036 0.036 0.036 0.036 0.036 0.036 0.036 0.036 0.036 0.036 tires, main- tenance) Total fixed 0.250 0.268 0.281 0.292 0.304 0.297 0.317 0.315 0.303 0.363 costs Insurance 0.050 0.052 0.053 0.056 0.057 0.056 0.057 0.057 0.057 0.063 License, 0.020 0.022 0.023 0.024 0.025 0.024 0.026 0.026 0.025 0.030 excise tax, registration Capital costs 0.180 0.194 0.205 0.212 0.222 0.217 0.234 0.232 0.221 0.270 Total costs 0.306 0.321 0.328 0.341 0.348 0.346 0.373 0.365 0.357 0.408 NOTE: AB = advanced body; FC = fuel cell; CNG = compressed natural gas; FT = Fischer-Tropsch. 71 SOURCE: Weiss et al. (2000). 

72 PERSONAL CARS AND CHINA costs (license, registration, and insurance) and on 20 percent per year of the new vehicle cost. Running costs are based on 20,000 km per year of travel. Fees for license and registration of $0.02 (RMB0.17) per kilometer (scaled by new vehicle cost relative to the baseline to represent some ex- cise tax and other costs) are incorporated into the calculations, as well as insurance costs of $0.05 (RMB0.4) per kilometer, with half of the cost scaled by the purchase price. These assumptions are consistent with cur- rent U.S. analysis (Davis, 1999). The annual operating costs reflect the assumption that the new cars would be sold in the United States in 2020. As cars age, their capital value decreases, and fuel and maintenance costs become a larger fraction of the decreasing total annual operating cost. Likewise, in countries where cer- tain fuels are heavily taxed, the ratio of capital to running costs would be less for vehicles using those more expensive fuels. All these costs are sub- ject to uncertainties inherent in the assumptions made in this analysis. The cost difference between the baseline vehicle and the most expensive option is 22 percent. Like today, the total annual cost for a new U.S. vehicle in 2020 is domi- nated by the capital cost, which is tied to the vehicle cost. Estimates indi- cate that the more efficient vehicles from an energy consumption stand- point are the more expensive ones, and the charges associated with increased price more than offset any fuel savings at current U.S. tax rates. For fuel cell vehicles, the total operating costs vary from the baseline of about 0.30 per kilometer to about $0.37 (RMB2.5–3.1) per kilometer. This difference reflects the roughly 30 percent higher estimated purchase price for the fuel cell vehicles. The $0.41 (RMB3.4) per kilometer cost of the electric vehicle is mostly attributable to the increased capital cost associ- ated with the storage batteries. Overall, only large differences in fuel costs or fuel taxes are likely to have a significant influence on the annual oper- ating costs of new cars. For example, at a UK tax rate of $3.53 per gallon of gasoline (8.8 times higher than the U.S. rate), the baseline vehicle fuel tax would increase to $0.044 (RMB0.36) per kilometer and the total new baseline vehicle operating cost would rise to $0.343 (RMB2.80), about 13 percent higher than in the United States. Overall Life Cycle Comparisons The following general comments are based on the cases evaluated: • Reducing vehicle weight improves life cycle efficiency. In the cases studied, a 10 percent weight reduction produced about a 10 percent re- duction in energy but resulted in about a 10 percent increase in vehicle cost. A smaller, lighter “China car” could be considerably more efficient than the typical U.S.-size car studied. 

PRESENT AND FUTURE AUTOMOTIVE TECHNOLOGIES 73 • Hybrid technologies offer significant energy savings, particularly in urban driving cycles. Typically, these cars are more expensive because of the more complex drive trains. A partial hybrid that shuts the engine on and off in stop-and-go traffic can achieve some of the emissions reduc- tions and energy savings of a true hybrid, with only a modest impact on cost. • The diesel engine offers some improvements in efficiency at a somewhat higher cost, but challenges remain as to whether it can meet emerging emissions standards for nitrogen oxides and particulates. • A hydrogen fuel cell car does appear to offer some advantages. It eliminates harmful vehicle emissions (although it can produce other emis- sions, depending on how the fuel is produced), but because of the cost and uncertainty about the development of the technology, it is not clearly a winner in the near term. Major investments in research and develop- ment (R&D) and in infrastructure will be required to move this technol- ogy into a significant market share over the next decade or two. The uncertainty bounds allow for the possibility of revolutionary im- provements in any of the technologies. Where these happen, a predomi- nant new technology may emerge. Because the vehicle system has so many components, breakthroughs are possible in many different areas— perhaps in control systems, batteries, engine or fuel cell technology, or fuels. VEHICLE COMPONENT TECHNOLOGIES Vehicle Weight and Body The dependence of fuel economy on vehicle weight is shown sche- matically in Figure 4-5 (Horton and Compton, 1984). In the past, the main incentive for substituting lighter materials has been to meet requirements for decreasing emissions below critical levels while minimizing the eco- nomic impact. In the United States, where fuel costs are not of major con- cern to vehicle purchasers at present, average vehicle weights have been increasing as sport-utility vehicle (SUV) sales increase. However, this find- ing masks the fact that vehicles of a given size are becoming increasingly efficient. Lighter materials, including high-strength steels, aluminum, and plastics, are replacing heavier carbon steels. The potential for weight re- duction is considerable, but the present barrier is the higher costs of many of the lighter materials. Automakers are working hard on ways to bring down these costs. The high-strength steels now being used extensively in new vehicles can provide, with minimal additional cost, equivalent strength at less 

74 PERSONAL CARS AND CHINA 60 Fuel economy, city and highway driving Improved (miles per gallon) power train efficiency W 40 ei gh t red uc tio n 20 2,000 3,000 4,000 Pounds FIGURE 4-5 Typical fuel economy improvements in new vehicle options result- ing from vehicle weight reduction for a typical power train efficiency. NOTE: Improvements in power train efficiency also increase fuel economy. SOURCE: Horton and Compton (1984). weight. About a 15 percent reduction in vehicle fuel use can be achieved by substituting high-strength steel for regular steel. Additional weight reduction can be achieved through the increased use of aluminum or fi- ber-reinforced plastic composites. Substitution of these two light materi- als for steel must not, however, compromise vehicle safety. The costs and availability of virgin materials, as well as the costs of disposing of scrapped vehicles, will influence decisions about the level of recycling that is appropriate. Significant energy savings and some cost savings are associated with the recycling of metals, especially aluminum. However, aluminum alloys vary in properties as a function of composition, and so a mixed recycled alloy is unlikely to match the desired physical properties for particular car components. The car market is therefore still largely de- pendent on virgin aluminum. Some research is under way on developing an all-purpose alloy that would meet structural needs and yet be suitable for recycling. Other approaches are to design for a few alloys that would separate easily on disassembly. This mixing problem occurs with other materials as well and should be considered in the design for recycling. Several European countries and Japan are enacting strict requirements for recycling or for manufacturers to be responsible for vehicles at the end 

PRESENT AND FUTURE AUTOMOTIVE TECHNOLOGIES 75 of their useful life. A large vehicle fleet generates a considerable waste stream. In countries such as the United States where land is available for inexpensive disposal sites, this problem does not appear to be a priority one. But in congested countries where landfill disposal is not available and more expensive incineration is used for wastes, more aggressive mea- sures seem appropriate. For metals, recycling is usually cost-effective, but other materials are more difficult to separate and are associated with only a marginal cost incentive (or a cost penalty) for recycling. Cars tradition- ally have not been designed for end-of-life waste minimization. However, some manufacturers are now modifying designs to facilitate disassembly, reuse of some components, and recycling of many materials. Mixed plas- tics and fiber materials can even be burned to generate process heat, for example. This new approach to design also has suggested ways in which repairs can be modularized for simpler servicing. The Chinese government has examined domestic material resources for use in the manufacture of today’s vehicles; where new materials are needed for the next generation of vehicles, similar analyses should be un- dertaken. Furthermore, the government will have to provide for the waste streams generated by scrapped vehicles, which it may wish to begin do- ing now in the requirements it imposes on industry—whether domestic or foreign. Vehicle design comprises a complex set of decisions about aerody- namics, stability, interior space, and safety. The characteristics of a vehicle’s propulsion system, transmission, and fuel storage systems are related to its weight and performance. Safety depends both on intrinsic performance capabilities and on crashworthiness. Today’s new collision warning systems and other devices also may increase the safety of smaller vehicles. Vehicle design that seeks to protect the passengers in an accident requires a sophisticated combination of understanding the crush behav- ior of structural elements and the performance of passenger restraint and protection systems, such as air bags. It is certainly possible to design a very safe small car, but it still will be less safe than a larger car of similar design if it crashes into an object larger in mass. Air drag and tire resistance provide opportunities for designers to reduce energy losses. Vehicle drag coefficients have fallen from the typi- cal level of 0.3 in the 1990s to about 0.25 today—and even below for some of the PNGV concept cars (but at significant cost and with the elimination of outside mirrors and other items). Drag reduction is most important during high-speed driving conditions. Improved tires also are providing some benefits, such as traction, reduced noise, and ride comfort. But ben- efits can be lost if drivers do not maintain properly inflated tires. Cost is a factor, although improvements in tire efficiency affect safety as well. 

76 PERSONAL CARS AND CHINA Propulsion Systems Conventional Propulsion Systems Today’s vehicles typically use gasoline or diesel engines as their pri- mary power source. These engine configurations have been developed over the years to provide reliable and easy-to-operate sources of vehicu- lar mechanical power. It is expected that these technologies will continue to evolve and improve over the next decade as well. The gasoline engine serves most personal transportation vehicles worldwide. Gasoline internal combustion engines (ICEs) burn air-fuel mixtures using spark ignition (SI) to initiate combustion. They are capable of operating over a broad speed range, from several hundred revolutions per minute (rpm) to as high as 7,000 rpm, and of starting rapidly over ambient temperatures ranging from –35°C to well over 38°C. Much of the improvement in engine efficiency over the last two de- cades in the United States has resulted indirectly from increasing the engine’s specific power, measured in kilowatts (kW) per liter (rated power per liter of engine displacement). This achievement has enabled engine downsizing of some 58 percent and a 26 percent reduction in the average 0–60 miles per hour (mph) acceleration time (An et al., 2001a; DeCicco et al., 2001). The prospects for further increases in specific power are excel- lent (Jost, 2002; NRC, 2002), which translates into further engine down- sizing while maintaining vehicle performance. Engine downsizing implies reduced engine friction and weight. Specific power has been increased by the addition of valves, fuel injection, improved fuel/air controls, low-fric- tion and lightweight materials, higher engine speed, turbo charging, ap- plication of numerical analysis techniques to optimize engine processes, precision manufacturing, and greatly improved quality control. Today, most production gasoline engines rely on a homogeneous stoichiometric fuel/air mixture for the internal combustion process. The choice of this combustion process for production engines stems from its flexibility for operation over broad speed ranges, combined with catalyst exhaust after- treatment technology to control effectively vehicle exhaust emissions. Al- though there is little room to improve thermal efficiency significantly for this type of engine, some potential exists to improve part-load engine effi- ciency by reducing engine friction and pumping losses, resulting in an enlarged high-efficiency area on the engine performance map. This im- provement can be achieved by reducing throttling losses through various valve train control technologies. Examples of such technologies include variable valve timing, variable valve lift, and throttleless “valvetronic” engine technology introduced by BMW in its 7-series sedan. Other tech- nologies, such as cylinder deactivation and the ability to vary the com- 

PRESENT AND FUTURE AUTOMOTIVE TECHNOLOGIES 77 pression ratio (see Flynn et al., 1999; Jost, 2002), also improve engine effi- ciency significantly. Recently, many engine manufacturers have initiated efforts to develop gasoline direct-injection (GDI) stratified lean combustion processes to improve both engine thermal and part-load efficiency. Japanese manufac- turers have introduced this technology to Asian markets. The GDI strati- fied lean-burn technology presents several new problems for engine de- signers. Similar to diesel direct-injection engines, the process is initiated with the injection of liquid fuel directly into the combustion chamber, forming regions of stratified rich and lean fuel/air ratios in the combus- tion space. The rich zones yield carbon-based particulates, which must be trapped with particulate filters. The lean zones yield nitrogen oxide (NOx) emissions combined with available oxygen in the exhaust stream. This combination of NOx emissions and available oxygen render three-way catalyst processes ineffective in removing nitrogen oxides; such systems require the use of new catalyst processes that effectively remove nitrogen oxides from an environment that includes free oxygen. Diesel engines that have been developed for broad use in passenger and commercial vehicles operate over a somewhat narrower speed range than gasoline engines (Flynn et al., 1999). Today’s diesel engines with high-quality fuel systems operate at speeds of between 500 and 4,000 rpm. In these en- gines, which operate at higher pressures than spark ignition engines, cylin- der combustion is initiated by injecting fuel into high-temperature com- pressed air, causing compression ignition (CI) (Naber and Siebers, 1996; Dec, 1997; Siebers, 1999). Compared with gasoline engines, diesel engines are more difficult to start rapidly under cold ambient conditions. The minimum start- ing temperature for diesel engines without special starting aids is typically 0°C. Although diesel engines were once prone to produce more noise and vibration than gasoline engines similar in size, recent design developments have produced smooth running, quiet diesel engine configurations that are barely distinguishable from gasoline engines (Flynn, 2000). New diesel engines incorporate a wide variety of technologies that improve performance and fuel economy and reduce emissions. Most new diesel engines apply high-injection pressure, which is enabled by a com- mon-rail unit injection system with advanced injection timing manage- ment, turbocharging, aftercooling, and an integrated exhaust gas recircu- lation (EGR) manifold system. Diesel cars have significantly penetrated markets in Europe and else- where, but future emissions standards are likely to challenge the ability of diesels to meet NOx requirements (this is discussed more specifically later in the section Diesel Engine Emissions). At a somewhat higher cost than spark ignition engines, diesels offer improved efficiency and are the tech- nology of choice for hauling heavy loads in freight transport where the fuel savings outweighs the initial capital cost investment. 

78 PERSONAL CARS AND CHINA Figure 4-6 illustrates the performance differences between the three types of propulsion technologies (fuel cell, diesel, and spark ignition) for a vehicle similar to the Volkswagen Golf, using the European driving cycle and based on the present state of technology (Wengel and Schirrmeister, 2000). The potential for improvement exists in each of the technologies. The figure is based on a prototype fuel cell design utilizing hydrogen. Fuel cell propulsion offers advantages in efficiency, especially for low- speed operation and for idling conditions in which the fuel cell output goes to charge batteries or in which the system is shut off (hybrid vehicles also offer similar advantages that are described later in this chapter). The efficiency curves shown in Figure 4-6 are for a particular vehicle configuration and driving cycle. The relationship between performance and efficiency under different driving conditions is much more complex. Vehicle designers use “performance maps,” a plot of normalized torque— expressed as brake mean effective pressure (BMEP)—from the engine as a 0.4 Fuel cell propulsion system Diesel propulsion system 0.2 Spark ignition propulsion system 0 0 10 20 30 40 50 60 Power (kilowatts) FIGURE 4-6 Comparisons of power train efficiency of combustion engine and fuel cell systems. NOTE: Information based on car similar to a Volkwagen Golf. SOURCE: Wengel and Schirrmeister (2000). 

PRESENT AND FUTURE AUTOMOTIVE TECHNOLOGIES 79 1,200 1,000 Break mean effective pressure (kilopascals) 300 275 5 27 260 800 27 5 260 250 0 26 250 250 600 27 260 5 0 260 30 30 0 275 275 400 300 300 325 325 325 350 350 350 375 200 375 375 400 400 400 450 60 450 450 500 0 500 500 600 0 500 1,000 1,500 2,000 2,500 3,000 3,500 4,000 4,500 5,000 Engine speed (revolutions per minute) FIGURE 4-7 Typical performance map for a spark ignition engine. NOTE: Brake- specific fuel consumption (BSFC) contours are shown in grams per kilowatt-hour. Performance map based on a 3-liter, 155-brake horsepower (bhp) engine. SOURCE: An et al. (2001a). function of engine speed in revolutions per minute. Figure 4-7 shows a typical performance map for a spark ignition engine. Performance maps depict how efficiency varies throughout the range of operation associated with particular driving cycles, with various pat- terns of requirements for power and acceleration. A shaded region of “maximum efficiency” is indicated in Figure 4-7 (contours are indicated in grams of fuel per kilowatt-hour). As engine torque and speed combina- tions move outside this region, fuel efficiency drops. The characteristics of the combustion engine are closely linked to the design of the transmission system to provide the desired performance characteristics for the vehicle. Trade-offs between performance and effi- ciency occur throughout the driving cycles. For U.S. cars, where fuel economy has been of less concern to the consumer than performance, the trade-off usually favors performance. European and Asian car manufac- turers, who are seeking sales where fuel efficiency is valued more than high-speed acceleration and other performance factors, make a different 

80 PERSONAL CARS AND CHINA trade-off in their designs. In the future, however, it appears the stronger emphasis will be on fuel economy because of concerns about foreign oil dependence and GHG emissions. Emerging Propulsion Technologies The homogeneous charge compression ignition (HCCI) diesel engine differs from both the typical gasoline and diesel combustion processes in that the energy release does not take place in a flame front. Fuel oxidation is accomplished by inducting a lean premixed charge of fuel and air into the engine cylinder. The lean premixed charge is subsequently com- pressed to higher temperatures and pressures until reactions similar to those encountered in gasoline engine knock phenomena occur (Westbrook et al., 1991). These reactions take place at temperatures considerably lower than those occurring in flame propagation, but still can be completed in the time allowed for piston engine combustion. If the combination of fuel/ air ratio and starting temperatures and pressures are controlled so that peak combustion temperatures do not exceed 1850 K, the combustion pro- cess can proceed to completion without the formation of any particulate or NOx emissions. The temperatures and pressures at which such pro- cesses take place are determined by the ignition characteristics of the fuel used. If high-octane fuels such as natural gas are used as the primary fuel for HCCI systems, the indicated efficiency of the process can approach that of the diesel engine. If low-octane fuels such as diesel are used, the pressures and temperatures at which these knock-like reactions take place are too low to allow compression and expansion ratios of above 8:1. Be- cause these expansion ratios are similar to those in gasoline engines, the fuel economy of the cycle using diesel as its main fuel would be similar to that in gasoline engine operation, but without the output of the usual pollutants. Power output also would be low because of the very lean fuel/ air ratio. Researchers are exploring the HCCI ignition process for use in both light-duty vehicular applications and heavy-duty engine applications. Research presently in the feasibility study phase indicates that significant improvements in light-load efficiency can be achieved when using this process in light-duty gasoline engines. The process is very difficult to con- trol, however, because it depends on reaching specific combinations of pressure and temperature for its initiation. The lack of a spark or injection event to control the initiation of combustion makes it more difficult to coordinate variables such as intake temperature, fuel octane number, or intake pressure, which must be controlled on a cycle-by-cycle basis to manage the process. Because small variations in temperature cause large differences in the times of combustion within the cylinder, it is unclear 

PRESENT AND FUTURE AUTOMOTIVE TECHNOLOGIES 81 whether intake processes can be managed closely enough to control HCCI combustion. Presently, funding of research on the HCCI combustion pro- cess is quite large. But it remains to be seen whether viable systems can be developed to incorporate such an approach into the production engine. Electric propulsion for cars has generated great interest because of its potential for “zero emissions” during use. Emissions may be generated in the production of the electricity, but when the generating plant is located outside of the urban air shed, such emissions may be of less concern to those residing in polluted urban areas, and yet they may be transported hundred of miles to affect the air quality of others. Electric power genera- tion from fossil fuels does produce GHG emissions, however. The main barrier to the use of electric energy in transportation ve- hicles is the difficulty in storing electricity. At present, batteries are heavy, cumbersome, and expensive. And charging times are long relative to liq- uid fueling times for a vehicle. Researchers are seeking a more efficient battery storage system, but a major breakthrough is needed if electric cars are to compete in price, convenience, and range with today’s liquid fuel vehicles. Although applications for electric vehicles do exist, especially for short travel distances, they are unlikely to be a major competitor with conventional vehicle technologies in the next decade or two. Fuel cell propulsion systems offer an alternative way to produce elec- tricity for propulsion from onboard fuels. Most fuel cells, especially those for transportation applications, operate with hydrogen fuel that can be either stored on board or chemically reformed from gasoline or other liq- uid hydrocarbon fuels. Liquid methanol can be used as a direct fuel for a fuel cell, but the technology is still far behind the hydrogen fuel cell tech- nology. A fuel cell is an electrochemical device that produces electricity by separating the hydrogen fuel into electrons and protons (hydrogen ions) via a catalyst. Because the fuel is converted directly to electricity, a fuel cell can operate at higher efficiencies than internal combustion en- gines, extracting more electricity from the same amount of fuel. The fuel cell itself has no moving parts, making it a quiet, reliable source of power. Fuel cell technologies are presently being developed for a variety of applications (see the appendix to this chapter for a brief description of the alternatives). The most promising fuel cell technology choice for trans- portation applications is the proton exchange membrane (PEM) fuel cell. These cells operate at relatively low temperatures (about 200°F or 95ºC) and have high power density. They can vary their output quickly to meet shifts in power demand and are suited for applications—such as in auto- mobiles—where quick start-up is required. PEM fuel cells are the primary candidates for light-duty vehicles, for buildings, and potentially for much smaller applications such as replacements for rechargeable batteries. The proton exchange membrane is a thin plastic sheet that allows passage of 

82 PERSONAL CARS AND CHINA hydrogen ions. The membrane is coated on both sides with highly dis- persed metal alloy particles (mostly platinum) that are active catalysts. Hydrogen is fed to the anode side of the fuel cell where the catalyst en- courages the hydrogen atoms to release electrons and become hydrogen ions (protons). The electrons travel in the form of an electric current that can be utilized before it returns to the cathode side of the fuel cell where oxygen has been fed. At the same time, the protons diffuse through the membrane to the cathode, where the hydrogen atom is recombined and reacted with oxygen from the air to produce water, thus completing the overall process. Comparing Fuel Cell Systems with Gasoline and Diesel Engine Systems Figure 4-6 compared the typical system efficiencies of a gasoline in- ternal combustion engine, a diesel engine, and a fuel cell for current tech- nologies. Although both gasoline and diesel engines have very low part- load efficiencies, the efficiency of a fuel cell system peaks around 20 percent of full load. Figure 4-8 shows how typical gasoline engine effi- ciency varies with percentage of maximum engine power, along with the operating modes of a U.S. car in an urban driving environment. For U.S. driving behaviors, average engine power demands occur at about 10 per- cent of maximum engine power. This level is well below peak efficiency for the gasoline and diesel engines, but it is where the fuel cell is most efficient. Thus potentially a fuel cell vehicle can be much more efficient than its internal combustion engine counterparts. Fuel cell vehicle developers must overcome many technological and economic challenges if they hope to match the performance and cost of today’s conventional technology vehicles. Fundamental problems with fuel cell technology are fuel selection, generation, distribution, and stor- age. The only truly zero emissions vehicle fuel cell is the direct hydrogen fuel cell. However, hydrogen infrastructure and onboard storage pose a huge challenge. Gasoline infrastructure and onboard storage to provide fuel for fuel cells already exist, but unfortunately a reformer that can con- vert gasoline to hydrogen adds more weight and technical complexity to a car, and in situ reforming technology is still far off for lower-tempera- ture automotive fuel cell technologies. Furthermore, “reforming” liquid fuels to make hydrogen still generates GHG emissions and deteriorates vehicle start-up and transient performance. For direct hydrogen fuel cell vehicles, one critical technological issue is onboard hydrogen storage. Hydrogen can be stored on board vehicles in many forms, including as compressed gas, as liquid, or within metal hydride alloys. Table 4-3 compares some competing hydrogen storage technologies and compares those technologies with gasoline and other 

83 Engine efficiency (%) Average engine power Maximum engine in U.S. c ity cycle power in U.S. cycles 37 27 Average engine power 20 in U.S. highway cycle Load (hp) 13.3 16.8 51.5 155 8.6% 11% 33% 100% FIGURE 4-8 Typical engine efficiency and average driving cycle operating modes for U.S. cars. Basis: Approximate efficiency versus horsepower (hp) curve for a 155 hp engine. SOURCE: Calculations by Feng An. energy storage media. Both the volume and weight of the hydrogen stor- age are based on the equivalent performance of 15 gal of gasoline fuel used in a typical U.S. car giving a 400-mile range. It is estimated that about 6 kg of hydrogen are needed to drive a fuel cell vehicle (that is smaller and more efficient) for 400 miles (Ashley, 2001). The comparison in Table 4-3 assumes that the hydrogen fuel cell vehicles are about twice as efficient as diesel vehicles. The last four columns of Table 4-3 show simply that, compared with gasoline technology, today’s hydrogen storage technology faces chal- lenges in both volumetric efficiency and weight penalties. Furthermore, a hydrogen infrastructure would have to be developed and its costs would be affected by the low density of hydrogen, which would require high- pressure transmission and distribution lines. Because hydrogen also has very wide flammability limits, safety will be a concern, especially during fueling because consumers are accustomed to fueling cars with liquid fuels. Overall, the infrastructure costs will likely run into the billions of dollars. Beyond technological challenges, the costs associated with fuel cell 

TABLE 4-3 Typical Energy Content and Storage Requirements for Automotive Energy Sources 84 Fuel Storage Quantities Equivalent to 15 Gal Gasoline or Percent of 1,800 MJ (ex tank) Gasoline Energy MJ/kg MJ/liter Energy Volume, Volume, Weight, Weight, Type (LHV) (LHV) Density Liters Gallons Kilograms Pounds Gasoline 32.0 043.7 100 56.4 15.0 40.9 90 Diesel 36.0 041.8 112 50.1 13.3 42.8 94 Methanol 16.0 020.1 50 112.8 30.0 88.9 195 Methane (STP) 00.036 050.0 0.11 50,133 13,300 35.7 78.6 Fuel Storage Quantities Equivalent to 6 Kg of Hydrogen or 750 MJ (ex tank) Hydrogen (STP) 00.011 120 0.03 66,700 Gaseous 6 13 Hydrogen 03.25 120 9.3 256 80.2 120 267 (gas, 350 atm) Hydrogen (liquid, 05.05 072 14.4 82.7 36.7 28 61 20 K, 1 atm) Metal hydride 11.9 001.4 34 60 16 500 1,100 Mg-based metal 06 0v2.4 17 120 32 300 660 hydride (ECD) NiMH battery 00.25 USABC comm. 00.54 goal 2000 NOTE: Energy equivalent to 15 gallons (gal) of gasoline; hydrogen based on 6 kilograms (kg) of fuel stored. LHV = lower heating value; STP = at standard temperature and pressure; ECD = Energy Conversion Devices Inc.; Mg = magnesium; NiMH = nickel metal hydride; USABC = U.S. Advanced Battery Consortium; MJ = megajoule; atm = atmosphere. SOURCES: Stodolsky et al. (1999); Wang (1999); Wang and Huang (1999). 

PRESENT AND FUTURE AUTOMOTIVE TECHNOLOGIES 85 systems, including fuel cell stacks, system accessories, and the onboard reformer, remain a major barrier to the commercialization of fuel cell tech- nology. Today, gasoline power trains cost around $25 (RMB207) per kilo- watt, diesel power trains about $50 (RMB415) per kilowatt, and limited production PEM fuel cells (e.g., the Ballard Model Mark 900), about $500 (RMB4,100) per kilowatt (Ashley, 2001). Fuel cell costs are expected to decrease in the future as technology advances and production grows. The U.S. Department of Energy (2001) estimates that if PEM fuel cells had been mass-produced (500,000 units per year), the cost would have been about $200 (RMB1,700) per kilowatt in 2001 and could be reduced to $125 (RMB1,000) per kilowatt by 2005—the technical target for the proposed U.S. FreedomCAR (Cooperative Automotive Research). Emissions Control Systems Gasoline Spark Ignition Emissions For gasoline internal combustion engines, three-way catalyst systems are highly effective (99+ percent) in achieving Euro II emissions standards. Figure 4-9 portrays the engine out and tailpipe out emissions of a typical three-way catalyst-equipped spark ignition engine. The catalysts, which are composed of mixtures of noble metals supported on a ceramic sub- strate, are used to eliminate carbon monoxide, nitrogen oxides, and hy- drocarbons from gasoline engine exhaust. But they are expensive and can be poisoned by impurities in the fuel—notably lead, which is now being phased out of gasoline in China. The effective operation of these catalyst systems depends on maintaining a stoichiometric fuel/air ratio in the engine’s combustion chamber. This stoichiometric operation provides the catalyst with an oxygen-depleted exhaust stream containing an appropri- ate level of hydrocarbon emissions so that, in the presence of the catalyst, the reducing atmosphere liberates oxygen from the nitrogen oxides and supplies them to the unburned hydrocarbons for oxidation. The catalyst combinations required to complete the joint reduction and oxidation processes have been developed after many years of empiri- cal work by engine manufacturers and catalyst suppliers. Because a three- way catalyst is not efficient under cold operating conditions, recent work has focused on reducing tailpipe emissions during a vehicle cold-start, which contributes more than 75 percent of the emissions of modern U.S. cars. Solving the problem requires development of quick “light-off” sys- tems. Cooperation between the various suppliers of the three-way cata- lysts and the manufacturers of their associated electronic control systems is needed to manage performance during start-up and other transient con- ditions. Although Chinese industry can supply these components, it has 

86 PERSONAL CARS AND CHINA 2 5 Engine out UHC 3 4 Engine out NOx Engine out CO NOx (1,000 ppm) Range of high UHC (100 ppm) CO (volume %) catalyst effectiveness 3 2 1 2 Catalyst out NOx 1 Catalyst out CO 1 Catalyst out UHC 0 0 0 14.25 14.50 14.75 Fuel/air ratio FIGURE 4-9 Three-way catalyst system behavior versus fuel/air ratio. NOTE: UHC = unburned hydrocarbons; NOx = nitrogen oxides; CO = carbon monoxide; ppm = parts per million. SOURCES: Data replotted from Sher (1998: Figure 6.4); catalyst effectiveness: Kummer (1981). not yet developed an independent capability for designing and optimiz- ing overall emissions control systems. Meanwhile, a promising new catalyst technology under development offers much the same efficiency at a substantially lower cost by replacing some of the noble metals currently used with rare earth elements. Use of these less expensive catalysts depends on reducing the sulfur in the fuel to levels significantly below those currently being planned (Zhan et al., 2001). Diesel Engine Emissions The diesel combustion process yields emissions of particulates and nitrogen oxides. Just as for the gasoline engine, meeting the most rigorous of the emissions standards proposed today will require applying exhaust aftertreatment devices to diesel engines to bring their emissions of nitro- gen oxides, unburned hydrocarbons, and particulates to the desired low levels. Because the exhaust of a diesel engine is cooler and contains more oxygen than that of a gasoline engine operating under stoichiometric con- ditions, removing the pollutants from the exhaust of a diesel requires a somewhat different technology. The particulates in the diesel exhaust pose an additional problem. 

PRESENT AND FUTURE AUTOMOTIVE TECHNOLOGIES 87 The NOx emissions from diesel engines can be controlled by cooling the diffusion flames within the engine in one of a variety of ways (De Witt and Wan, 2000). The most common is to add diluents, in the form of water-fuel emulsions, inert gases, or recirculated cooled exhaust gas, to the combustion process. All these methods provide additional thermal mass near the diffusion flame and thus limit the overall rise in the flame temperature. The addition of diluents (exhaust gas recirculation or EGR) is the most effective way to control in-cylinder diesel NOx emissions. As shown in Figure 4-10, the lowest NOx emissions level from diesel engine combustion is 5.5 g per kilogram of fuel burned for engines oper- ating at 1,500 rpm (Flynn et al., 2000). These results at 1,500 rpm translate to lower numbers at higher engine speeds, because NOx conversion is directly proportional to combustion residence times. These levels of NOx production are significantly above the minimum NOx levels produced by gasoline engines operating with effective three-way catalyst systems. Therefore meeting newly legislated NOx emissions targets will require additional aftertreatment of the exhaust gas stream with selective cata- lytic reduction (SCR) or other techniques to remove nitrogen oxides. Diesel engines also have particulate emissions. The rich combustion process converts a significant portion of the carbon mass in the fuel to particulate precursors. Those particulate precursors that avoid going through the vigorous diffusion flame are left as tailpipe emissions. The California Air Resources Board (CARB) has labeled diesel particulates as toxic emissions, and such emissions have been associated with degrada- tion in lung function by a variety of epidemiological studies. Particulates also degrade visibility as they accumulate in the atmosphere. As such, to satisfy future particulate emission legislation in the United States the manufacturers of diesel engines probably will have to add exhaust par- ticulate filters to their products. These filters must trap the particulates and then, by managing temperatures, provide an opportunity for the oxi- dation of the particulates on the trap. Under favorable operational and ambient conditions, this combination of trapping and oxidation can be completed without additional management or manipulation of trap tem- peratures. If ambient conditions or operational constraints prevent ex- haust temperatures (and thus trap temperatures) from rising to the level needed to oxidize the particulates, trap temperatures may require active management through the injection of additional fuel or the modification of engine operation to raise exhaust temperatures. Fuel Cell Emissions A fuel cell operating on hydrogen emits only water as a waste stream. However, when fuel cells operate through reforming a fuel such as gaso- 

88 PERSONAL CARS AND CHINA fsNOx (grams per kilogram fuel) 30 30 Complete Diesel combustion fsNOx (g/kg fuel) 20 20 NOx level Incomplete at minimum combustion viable flame (high CO, temperature 10 UHC, soot) 10 5.5 g/kg fuel at 1,500 rpm 0 0 0.29 0.30 0.31 0.32 0.33 2000 2200 2400 2600 2800 GISFC ( lbm/hp-hr) Peak flame temperature (K) FIGURE 4-10 Limits of diesel combustion at 1,500 revolutions per minute (rpm). NOTE: GISFC is the fuel consumption per net power produced in the compres- sion and expansion strokes of engine operation and thus disregards the gas ex- change strokes of engine operation; fsNOx is the NOx emissions on a fuel-specific basis. EGR = exhaust gas recirculation; UHC = unburned hydrocarbons; NOx = nitrogen oxides; g/kg = grams per kilogram; lbm/hp-hr = pounds mass per horse- power-hour. SOURCE: Flynn et al. (2000). line to produce hydrogen, GHG emissions may be produced by the re- former. Because the reformer operates at a lower temperature than a com- bustion engine, NOx emissions are negligible. A clean fuel (low sulfur) is required to maintain satisfactory performance of the system. This higher- efficiency propulsion system also provides a benefit in reducing emis- sions per kilometer. Meeting Future Emissions Standards For diesel engines, both the particulate and NOx aftertreatment sys- tems are likely to require active management of the exhaust system tem- perature and the fuel/air ratio. Presently, such systems are in their earli- est demonstration phases, and much work remains to determine whether they can be produced for a wide range of operating requirements and environments. Figure 4-11 presents the European, Japanese, and U.S. emissions standards, along with the demonstrated capabilities of typical 

PRESENT AND FUTURE AUTOMOTIVE TECHNOLOGIES 89 diesel and spark ignition systems. NOx emission levels are depicted in grams per kilogram of fuel burned, a unit that serves to represent the level of technology required for the removal of nitrogen oxides. Because emissions standards are set on a grams-per-mile or grams-per-kilometer basis, as fuel mileage improves, more stringent emissions standards can be achieved. Thus, as shown in Figure 4-11, a car achieving 20 miles per gallon (mpg) might have difficulty meeting the Euro III diesel standard, but if the mileage for that diesel engine were improved to 40 mpg, it could clearly meet the proposed Euro III diesel standard. Figure 4-11 indicates that some level of exhaust aftertreatment would be required to remove nitrogen oxides for diesel engines as the emissions limits are lowered. With three-way catalyst technology, the gasoline engine can achieve or further reduce emissions as required by the U.S. 2007 light-duty standard. The diesel engine will benefit from many of the possible improve- ments described earlier for the spark ignition engine—for example, ve- hicle weight reduction, improved component efficiencies, recovery of ki- netic energy during braking, and engine turn-off when stopped. Vehicle hybridization is one way in which to realize the latter two improvements (see the section Hybrid Vehicle Technologies). The advantages of the diesel engine for hauling heavy loads will con- tinue to make it the leading option for trucks. Now that the pollution produced by trucking is gaining more attention, new emissions standards are being applied to heavy-duty vehicles. The technologies developed for improving the emissions performance of heavy-duty vehicles also may become available for lighter vehicles in the future. Transmissions Matching the torque and speed requirements of a vehicle’s drive wheels with that of the engine’s capability requires a transmission that permits a variety of operational gear ratios. Such transmissions can be shifted manually, electromechanically, or automatically from one gear ratio to the next. Mechanical transmissions usually have the highest over- all transmission efficiencies, typically between 90 and 97 percent. In re- cent years, hydrodynamic transmission drives have been augmented with torque converter lockup mechanisms so that the hydrodynamic losses can be minimized when the overall transmission gear ratio is 1:1. Most ve- hicles in the United States use hydrodynamic automatic transmissions. Substantial improvements in transmission performance are under way (DeCicco et al., 2001). The main developments are the addition of extra gear ratios to conventional transmissions (five- and six-speed auto- matics), motor-driven gear ratio shifting (which allows smart electronic control and eliminates the torque converter—a source of friction losses, especially in urban driving), and continuously variable transmissions. 

90 PERSONAL CARS AND CHINA 25 Euro II diesel standard 20 NOx emissions (grams per kilogram fuel) Diesel engine out 15 capability with Euro III diesel standard cooled EGR 2002 Japan diesel standard 10 1997 U.S. SI standard Euro II SI standard Euro IV diesel Euro III SI standard 5 2000 Japan SI standard Euro IV SI standard 2007 U.S. diesel standard 2007 U.S. SI standard 0 SI tailpipe out capability with hot three-way catalyst -5 10 20 30 40 50 60 70 80 90 100 110 Vehicle mileage (miles per gallon) FIGURE 4-11 Fuel-specific nitrogen oxide (NOx) emissions standards versus capability. NOTE: SI = spark ignition; EGR = exhaust gas recirculation. Fuel-spe- cific gravity gasoline = 0.75; diesel fuel = 0.85. SOURCES: Standards: Table 4-1; capabilities: Flynn (2001). Added gear ratios allow the engine to turn at modest speeds (near the maximum efficiency zone shown in Figure 4-7) over a range of vehicle speed and acceleration conditions. The continuously variable transmis- sion uses more complex technology to approach an infinite number of gear ratios in order to optimize engine speed over variable driving condi- tions and to permit the engine to always operate in the maximum effi- ciency zones. The cost implications of these advanced transmission sys- tems are uncertain, but there appear to be enough potential improvement possibilities so that the costs of evolved transmission systems will not have a major impact on vehicle costs. 

PRESENT AND FUTURE AUTOMOTIVE TECHNOLOGIES 91 Some European and Asian vehicle manufacturers have offered pro- duction versions of continuously variable transmissions. Such devices typically use tapered belts and pulleys to change effective gear ratios so that the engine is always operating at maximum efficiency. Torque is transmitted through frictional forces at the belt pulley interface, and speed is varied by changing the relative diameter of the input and output pul- leys. Today such devices are being offered only on very small vehicles because of the limited torque-carrying capability of such frictional drives. Electrical Systems In the early days of automobiles, batteries were used first for lights and then, after the invention of the electric starter in 1912, to start the engine automatically. The generator was developed concurrently. The 6- volt (V) dry cell or lead acid battery was the standard in the first half of the twentieth century. As cars evolved, the 6 V systems proved inad- equate, and in the 1950s the industry changed to a 12 V lead acid battery. Since then, the addition of control systems and more electrical amenities in vehicles has steadily increased the electrical power requirement for cars, but large current draws cause voltage drops and significant parasitic power losses in today’s electrical systems. It is estimated that a 12 V sys- tem can support only up to 3.5 kW electrical loads. The present average power demand in an automobile is about 1.2 kW. Emerging technologies that will require even greater electrification are discussed later in this chapter. The industry is considering a transitional dual voltage system, though the car of the future is likely to use a single 36 V (42 V output generator rating) power distribution bus with provision for hybrid operations (di- rect current and alternating current) and intelligent controls with multivoltage power distribution and load management. Some current pro- duction vehicles already are using 42 V machines for the starter-genera- tor. Although these new systems are likely to be introduced first in luxury vehicles because of cost, they also have significant performance advan- tages for smaller cars. Transition to the higher voltage could reduce both power losses and the weight of wiring harnesses. However, advanced power control and distribution systems will be needed to operate the more complex vehicles of the future. For vehicles that use electric motors for propulsion, such motors could, in addition to powering accessories, serve as the starter-generator and support regenerative braking, producing higher efficiencies over a wider speed range. The traction motors, however, are still in the develop- mental stage. Permanent magnet motors require costly materials and are limited in operational speed at 42 V by back emf (back voltage) problems. 

92 PERSONAL CARS AND CHINA Induction motors tend to overheat and thus lose efficiency at 42 V (it is hard to cool a moving rotor). Both induction and permanent magnet mo- tors could benefit from an even higher voltage system, but that creates other problems, including ones of safety. Switched reluctance motors are under development to avoid the problems of back emf and overheating, but the present generation of motors has noise and vibration problems. Further research and development, however, will likely uncover solutions to all these challenges, opening the possibility that over the next decade the rate of introduction of 42 V automotive technologies into passenger cars will increase substantially (Ehsani et al., 2001). Electronic Controls Because they provide the increased sophistication needed to meet higher emissions standards while providing good fuel economy and good vehicle drivability, electronic controls have become ubiquitous in modern automobile technology. In fact, in recent years these controls have evolved to digital systems that are fully programmable. At the current level of technology, a programmable digital computer is coupled to advanced sen- sors that provide real-time data to allow the inference of various engine operating parameters such as specific engine pollutants, fuel consump- tion, engine horsepower, and engine torque. Using these inputs, the com- puter instantaneously controls spark, fuel delivery, quantity of exhaust gas recirculated to the engine, and, in some cases, the transmission. Am- bient parameters such as temperature and atmospheric pressure also may be measured. The computer industry can easily provide the hardware to accom- plish these tasks, but the bigger challenge is developing the control strat- egy that will optimize vehicle performance under a wide variety of ambi- ent and driving conditions. Doing so will involve utilizing complex analytical models that relate emissions, fuel economy, and vehicle drivability to detailed operating parameters of the vehicle power train— for example, its engine, transmission, and driveline. These parameters depend on the specific driving conditions and vary over wide ranges be- tween engine idle and full power. Usually, a limited amount of hardware-specific information is col- lected in the laboratory on various operating parameters, such as fuel con- sumption and emissions under specific operating conditions. These data are then fed into analytical models to provide the final control algorithms for each specific engine/vehicle combination. Because of the extraordi- nary effort devoted to creating the individual algorithms and because the control algorithms are of critical importance to the satisfactory operation of the vehicle, manufacturers consider both the details of the procedure 

PRESENT AND FUTURE AUTOMOTIVE TECHNOLOGIES 93 and the algorithms to be highly proprietary. Algorithms for hybrid de- signs are significantly more complex and require additional levels of de- sign skills and sophistication. Emerging Capabilities The growth in instrumentation and control capabilities in vehicles is leading to new possibilities for diagnosing vehicle problems and expedit- ing repair—perhaps even while the vehicle is in use. Global positioning system (GPS) capabilities already are available to help drivers identify their location and navigate. In the future, these capabilities may extend to automated driving on specially equipped highways and other advanced technologies. Research is under way on automated highways, magneti- cally levitated vehicles, “drive-by-wire” systems, and other long-term technologies. However, most of these systems are at least a decade or more from realization and probably will require rethinking the traditional ways of delivering mobility to a diversity of consumers. Fuels and Onboard Fuel Storage Present Fuel Technologies China’s present road transportation sector is almost completely de- pendent on petroleum fuels, with some very limited use of electricity and compressed natural gas. Gasoline and diesel fuels are used widely in motorized vehicles because of their high energy storage density, compact and lightweight storage systems, and low cost relative to other fuels. Table 4-3 shows the energy storage volume and density values for various auto- motive fuels. Energy equivalent to fifteen gallons of gasoline (1800 MJ) is used as the basis for comparing alternative liquid fuels. For hydrogen, it is assumed that 6 kg of fuel is used (735 MJ) in a smaller, more efficient hydrogen fuel cell vehicle (Westbrook and Chase, 1988; Stodolsky et al., 1999; Wang, 1999; also see Chapter 5). The last four columns indicate the volume and weight of the fuel stored onboard. The last two rows in Table 4.3 show some battery data for comparison. A liquid fuel storage tank in an automobile weighs less than the fuel it contains, whereas metal hydride storage systems, the heavy-walled pres- sure tanks required to store compressed gas, and the insulation systems needed for liquid hydrogen storage at very low temperatures can add significant weight penalties. Researchers are aiming for hydrogen storage systems that store about 5–10 percent hydrogen by weight and volume. Much research is under way on the storage possibilities of carbon fibers and other novel media, but significant progress is still needed to make a 

94 PERSONAL CARS AND CHINA hydrogen storage system that does not impose too high a cost, volume, and weight penalty on a car. The present petroleum fuels (domestic and imported) also will con- tinue to change. To meet the Euro II emissions standards and the even more stringent ones to come, the gasoline (from petroleum) of the future will move toward very low sulfur content, with possible changes in volatility, aromatics, or other specifications driven by the increasingly stringent en- vironmental emissions standards. Diesel fuel (from petroleum) will evolve from its current properties toward very low sulfur content, with possible changes in volatility, polynuclear aromatics, cetane, and other specifica- tions Some of the near-term alternative fuels might be feasible in China. Natural gas is being used now as compressed natural gas in limited quantities for bus and taxi fleets in specific locations such as Beijing. CNG (primarily methane) is typically stored in vehicles as a pressurized gas (at about 200 atm pressure). Methanol from coal or coal bed gas could be used as an additive to gasoline in an M15 (85 percent gasoline–15 percent methanol) mixture or as a pure fuel. Because methanol has a lower energy density than gaso- line, it requires a larger fuel tank. Moreover, it is corrosive and so requires special metallurgy, and it is a contact poison and so must be handled with care. Finally, methanol from coal or other sources (such as natural gas) is generally more expensive than gasoline. Methanol fuel can be reformed on board to make hydrogen for a fuel cell. Overall, methanol from coal bed gas does not save net energy or reduce GHG emissions, but it could serve as a domestic fuel alternative for China. Dimethyl ether (DME) from coal or coal bed gas offers a good alterna- tive to gasoline and diesel. Wang (1999) estimates that DME, even if manu- factured from remote gas, would be quite expensive, with a delivered (to the vehicle tank) cost (in 1995 U.S. dollars) ex tax of $1.89 per gallon (RMB4.14 per liter) of gasoline equivalent. For comparison, diesel (50 per- cent Fischer-Tropsch) would cost $0.65 per gallon (RMB1.42 per liter) and methanol $1.20 per gallon (RMB2.63 per liter) when made from remote natural gas valued at about $0.50 per gigajoule (GJ). Making DME from coal would cost considerably more than making it from remote natural gas. Moreover, manufacturing energy losses would be higher, and carbon dioxide (CO2) emissions would be much higher. DME also is a pressur- ized gas at normal temperatures (boiling point of –24°C) and thus would require an entirely new and expensive fuel distribution system and new and costly changes in all vehicle fuel tanks and fuel systems. Vehicle range would be lower because of fuel storage limitations, especially for heavy- duty trucks. Biofuels such as ethanol and methanol can be made from agricultural 

PRESENT AND FUTURE AUTOMOTIVE TECHNOLOGIES 95 or other combustible wastes or from farmed energy crops such as corn, grain, or fast-growing cellulosic materials. Basically, biomass energy con- tent is the result of a solar energy conversion process that operates at about 1–2 percent efficiency, so the energy content is relatively low per unit of planted area. Fossil fuels today are the result of eons of biomass conver- sion. Moreover, account must be taken of the costs and benefits of using land to produce fuel instead of food and of the pollution problems associ- ated with producing biofuels. Liquefied petroleum gas (LPG), mostly a mixture of propane and bu- tane, has been used for years as a convenient “bottled gas” for remote locations, for camping, and for vehicle fuel. At 5–10 atmospheres (atm) pressure, saturated liquid can be stored at typical ambient temperatures. LPG is usually produced along with oil or natural gas, but it is not as plentiful. If over the next two decades hydrogen achieves widespread use as a transportation fuel, it will likely be manufactured most economically by reforming natural gas at “service stations” located on a gas pipeline net- work. It probably will be stored and dispensed as a gas at about 350–400 atm pressure. Other, more expensive options include generating hydro- gen from coal (Williams, 1998), electrolysis of water, or reforming natural gas in large, centralized facilities and piping compressed hydrogen or trucking liquid hydrogen to service stations. Any of these methods, how- ever, will require large investments in infrastructure. Although cars that utilize electric power are considered “zero emis- sions vehicles,” emissions may still be associated with the original pro- duction of the electricity. Although other technologies are being developed, nickel metal hy- dride (NiMH) batteries are the technology of choice for automotive appli- cations today, both for hybrids and electric vehicles. Advanced lead acid batteries are less expensive than the NiMH batteries, but have a much shorter operational life. Electric vehicle batteries currently have specific energy of about 70 watt-hours per kilogram (Wh/kg) and specific power of about 150 W/kg (U.S. Department of Energy, 1999; GMC, 2001). It is assumed that by the year 2020 electric vehicle battery performance will improve, especially the specific energy, and battery performance will be close to meeting the Advanced Battery Consortium’s commercial goals of 150 Wh/kg and 300 W/kg (U.S. Council for Automotive Research, 2000). These commercial goals are judged to be the battery performance required to produce ac- ceptable electric vehicle performance. Although the NiMH battery prob- ably cannot reach this potential, another technology, such as the lithium- ion battery, may. Its specific energy is significantly higher than that of the NiMH battery technology. Batteries are not intended to be fully dis- 

96 PERSONAL CARS AND CHINA charged, because such a step shortens their lifetime and decreases their capacity. Also, topping off a battery at a high state of charge is inefficient because of its internal resistance. Thus cycled battery applications tend to operate within a state of charge range of 20–80 percent. For the entirely electric vehicle, both battery performance and charge density constraints (specific power and specific energy) are important. In addition to providing the power needed for peak motor power, the bat- tery energy storage capacity must be sufficient to give adequate vehicle range. When a battery’s specific energy is too low, the extra battery weight needed adds to the vehicle mass and thus requires additional structural support and increased motor power, generating an undesirable com- pounding effect. Given this constraint, the battery pack is selected based on its power capacity, and no effort is made to augment vehicle range beyond what the available electric vehicle battery technology can pro- vide. The physical size of the battery also must be considered because of its possible intrusion into the vehicle’s interior space.2 Although some of the fuel alternatives just described may have niche uses in China, a practical view suggests that petroleum fuels will be the main choice for the automobile fleet as it develops over the next few de- cades. Even for this choice, China will have to make significant invest- ments in new fuel transportation and distribution infrastructure, includ- ing cleaning or replacing old facilities that are incompatible with the new, cleaner fuels. HYBRID VEHICLE TECHNOLOGIES Hybrid Power Trains A typical combustion engine delivers peak efficiency only at a par- ticular power level. Today’s cars are designed with oversized engines that are able to provide peak power for acceleration in passing or climbing hills at highway speeds. The engines, therefore, operate inefficiently at low urban speeds with low part-load efficiency (Figure 4-8) and burn fuel while idling in traffic. Table 4-4 lists the time and fuel use shares during 2 For hybrid systems, which are discussed next in this chapter, only the battery’s specific power is critical, because discharged batteries can be recharged during operation of the internal combustion engine. High-power hybrid electric vehicle NiMH batteries currently have a specific power of about 400 W/kg and a specific energy of about 40 Wh/kg (at a 3- hour rate.) It is assumed that battery performance will improve over the next decade or so, especially in specific power, and that the goals of 800 W/kg and 50 Wh/kg are well within reach (F. R. Kalhammer, Electric Power Research Institute, personal communication, 2000). Again, lithium-ion battery technology may well surpass this goal. 

PRESENT AND FUTURE AUTOMOTIVE TECHNOLOGIES 97 vehicle stops and braking decelerations for a typical U.S. car for urban driving cycles in the United States, Japan, and Europe. The table reveals that a vehicle spends a significant amount of time and fuel during both stops and braking decelerations and that the fuel saving potential for en- gine idle-off is very large. If engine start-stop is designed to recover en- ergy losses during vehicle stops only, about 12–19 percent of fuel could be saved for a vehicle operating in these cycles. The fuel savings would be even higher if engine start-stop were designed to recover all engine idling losses. This savings can be facilitated through hybridization. Many different types of hybrid vehicles have been developed (Rovera and Mesaiti, 1999). In general, a hybrid is designed with an engine that is smaller than that needed for a similar nonhybrid car. The hybrid’s smaller combustion engine (spark ignition or diesel) operates closer to its peak efficiency, which occurs near its maximum power output (a larger engine would be operating at lower efficiency at the same power output). The engine can be shut off during vehicle stops, braking decelerations, and even low-power driving, depending on the specific design of the hybrid system. A rechargeable battery system is used to provide extra power on demand. At low speeds or while the vehicle is stopped, unless the engine is shut off, the combustion engine uses excess power to recharge the bat- tery system. The most efficient hybrid vehicle configuration also captures the regenerative energy from braking the car and uses it to recharge the battery. This type of hybrid offers major improvements in efficiency in urban driving cycles and results in lower total emissions because of the smaller engine—which is switched off at idle if the battery system is fully charged. The disadvantages of hybrids are the extra materials and weight asso- TABLE 4-4 Percentage of Time Spent and Fuel Consumed by a Typical U.S. Car during Vehicle Stops and Braking in Different Urban Driving Cycles Vehicle Stops Vehicle Braking Total Engine Idle Driving Cycle Time Fuel Time Fuel Time Fuel FTP 19.2 11.6 23.6 14.2 42.8 25.8 Japan 29.2 19.1 23.1 15.2 52.3 34.3 Europe 24.9 14.5 15.7 9.2 40.6 23.7 NOTE: FTP = Federal Test Procedures (U.S.). SOURCE: Authors’ estimate; also see An et al. (2002). 

98 PERSONAL CARS AND CHINA ciated with a parallel electric drive system and with the battery system. Costs and some vehicle efficiency penalties are associated with the added components, and the larger the battery system, the higher the added costs and the greater the weight. For urban driving, a relatively small battery system matches the stop-and-go driving that allows frequent draw-down and recharging of the battery. In highway driving, the battery may not recharge at all, and there is the danger that it may be discharged, at which point the car loses power because it is operating on an underpowered combustion engine. For this reason, the Toyota Prius sold in the United States has both a larger engine and a larger battery system than the Japa- nese model. Today’s hybrid electric vehicles (HEVs) have three basic configura- tions: 1. The series HEV configuration, in which the engine drives a genera- tor that produces electricity, which in turn powers a motor to drive the wheels and, during periods of low power demand, charges the battery. Braking energy also can be used to charge the battery. This configuration is called a series hybrid because the power flows along a single path. 2. The parallel HEV configuration, in which both the engine and mo- tor drive the vehicle wheels. Because the power flows along two paths, this configuration is called a parallel system. This system also allows the engine to charge the battery on board and recover braking energy. 3. The power-split HEV configuration, which is closer to the parallel configuration. It differs in that a planetary gear system combined with a starter-generator can transfer power between the internal combustion engine and electric motor, both of which are coupled to the drive shaft. In this configuration, the internal combustion engine provides the pri- mary power, with a power-split device (planetary gear with starter-gen- erator) sending power to both the drive shaft and the electric motor. This system is sometimes called an electrically variable transmission system (Toyota, 1997). A series HEV is technically the simplest; however, it usually requires large electrical components, and thus it is heavier and more expensive. Most hybrid city buses and heavy-duty urban trucks use series configura- tions. The most popular choice for today’s commercial and prototype light-duty HEVs is the parallel configuration. It requires more compli- cated system integration, but it is lighter, more efficient, and less costly than the series system (yet more expensive than the nonhybrid technolo- gies). The power-split system is technically more complicated, but poten- tially can achieve the highest efficiency. 

PRESENT AND FUTURE AUTOMOTIVE TECHNOLOGIES 99 Categories of Hybrid Vehicles In principle, hybrid propulsion can take many forms, from slight de- grees of hybridization (e.g., using an integrated starter-generator with engine start-stop capability) to designs that drive the wheels only electri- cally. HEVs can be classified according to the portion of their maximum propulsion power provided by an electric drive: • Minimal (or mini) hybrid—fraction of onboard electric power less than about 10 percent. It can provide engine idle-off capability, but no regenerative braking and electric-only driving. The minimal hybrid usu- ally uses a low-voltage integrated starter-generator, and its fuel economy benefit under U.S. driving conditions is about 10 percent.3 • Mild and medium hybrid—fraction of electric power ranges from 10 to 25 percent; idle-off and some regenerative braking, but no significant electric-only driving (example: Honda Insight). The fuel economy benefit of such HEVs under U.S. driving conditions is about 10–30 percent. • Full hybrid—fraction of electric power ranges from 25 to 50 percent; some electric-only driving but no real trip range, and battery not designed for plug-in recharging (example: Toyota Prius). Full HEVs are sometimes called “power” hybrids. The fuel economy benefit of such HEVs under U.S. driving conditions is about 30–50 percent. Energy HEVs (also called “charge depletion hybrids”) have a useful all-electric driving range (50 miles or more) and plug-in recharging abil- ity. But they require a large battery pack and are considerably more ex- pensive than the other hybrids. No energy hybrids have been announced for mass production. Another reason may be that battery technologies remain too limited to provide adequate combinations of efficiency and performance even when supplemented by an engine-powered generator. Batteries are certainly a major cost factor for all hybrids, and so like pure electric vehicles energy hybrids will carry a very substantial cost premium for the foreseeable future. A full hybrid is considered a more radical change than the other hy- brids from the conventional internal combustion engine vehicle, whereas a mini or mild hybrid is considered a more natural evolution from a con- ventional vehicle, resulting from a historical trend of increasing vehicle onboard electric power. The vehicle electric power growth rate was about 3 Hybrids can have different types of propulsion systems (spark ignition, diesel, or fuel cell) with varying efficiencies; the fuel economy benefit noted results only from the hybrid- ization. 

100 PERSONAL CARS AND CHINA 6 percent from 1920 to 1940, 2 percent from 1940 to 1970, and again 6 percent from 1970 to 1990 (An et al., 1999; Moore, 1999). Industry projec- tions indicate this electrification trend will continue and probably grow. Typical U.S. cars have an onboard electric power requirement of about 1.2 kW. This electric power requirement will increase to 3–5 kW over the next few years because of the addition of features such as heated seats and windows, multimedia, water/oil pumps, power steering, HVAC (heat- ing, ventilating, air-conditioning) fans, electromagnetic valves, and heated catalysts. Currently, most major auto manufacturers and suppliers are working on the so-called “integrated starter-generator” system, which will increase onboard electric power to about 10–15 kW and support features such as fast crank, torque smoothing, engine idle-off, and launch assist, and a cer- tain degree of regenerative braking. The Honda Insight hybrid vehicle falls into this category. When electric power increases to 20 kW, the vehicle’s internal combustion engine can be further downsized, with such added features as electric HVAC, power assist, fast heating, and limp- home capability (if out of fuel for the engine). DaimlerChrysler’s ECX2 is in this category. Vehicles with onboard electric power capability of 10–20 kW are often called mild hybrid vehicles. All major U.S. and European manufacturers are pushing for this concept of hybrid vehicle, along with standardization of a 42 V electrical system. Toyota recently released a mild hybrid option for its luxury Crown model. When onboard electric power increases beyond 20 kW, as in the Toyota Prius and in proposed fuel cell hybrid vehicles, it finally reaches the so-called full hybrid vehicle territory. A full hybrid vehicle with a significantly downsized engine and large electric motor, combined with electrically variable transmission technologies like those developed by Toyota and Nissan (Toyota, 1997), will achieve the maximum benefit from vehicle hybridization. But consumers may find full hybrid technology too costly, placing the technology at high risk of weak customer acceptance without government or manufacturer subsidies. Comparing Commercial and Concept HEVs In recent years vehicle manufacturers have made great progress in developing and demonstrating commercially available and concept hy- brid electric vehicles. These vehicles include commercially available gaso- line hybrid cars (Toyota Prius and Honda Insight) and the diesel hybrid concept cars (Ford Prodigy, GM Precept, and DaimlerChrysler ESX3) that emerged from the Partnership for a New Generation of Vehicles (see Chapter 8). Table 4-5 summarizes some basic characteristics of selected commercial and concept HEVs in the United States (An et al., 2001b). Data 

101 sources are specified in the table because some reported figures are not always consistent among different sources, and some figures may not rep- resent official or certified figures. For example, only Prius and Insight have fuel economy ratings certified by the U.S. Environmental Protection Agency. Other fuel economy figures are based on manufacturers’ claims. In Figure 4-12, the fraction of electric power of selected HEVs is esti- mated by dividing peak motor power by total combined peak motor and internal combustion engine power. Note that this figure gives only an ap- proximate measure because the power split during operation is variable. Assessing the Benefits of Hybrid Vehicle Fuel Economy Because PNGV and other hybrid vehicles have such expensive drive trains, they have employed unprecedented levels of conventional ve- hicle fuel economy technologies such as aggressive load reduction mea- sures. These measures include reducing weight and improving air/tire resistance as well as using lightweight materials and diesel engines to reduce weight. The choice of diesel engines for PNGV vehicles is a re- sult of setting high efficiency goals within a five-year time horizon. The Fraction of electric power (percent) 50 45 40 35 30 25 20 15 10 5 0 Honda DC Ford U.S. GM Insight ESX3 Prodigy Prius Precept Mild Full hybrid hybrid Increasing electrical power fraction FIGURE 4-12 Fraction of electric power of selected hybrid electric vehicles. SOURCE: An et al. (2001b). 

102 TABLE 4-5 Characteristics of Commercial and Prototype Hybrid Electric Vehicles Curb Power Engine Engine Motor Trans- Weight Plant Size Power Battery Peak mission CAFE 0–60 HEV Type Status (lbs) Type (liters) (hp) Type (kW) Type (mpga) Time (s) U.S. Prius Gasoline hybrid Commercial 2,765 SI I-4 1.5 70 NiMH 33 CVT 58 12.1 Honda Insight Gasoline hybrid Commercial 1,856 SI I-3 1.0 67 NiMH 10 M5 76 10.6 Ford Prodigyb Diesel hybrid Concept car 2,387 CIDI I-4 1.2 74 NiMH 16 A5 70 12.0 DC ESX3 Diesel hybrid Concept car 2,250 CIDI I-3 1.5 74 Li-ion 15 EMAT-6 72 11.0 GM Preceptc Diesel hybrid Concept car 2,590 CIDI I-3 1.3 59 NiMH 35 A4 80 11.5 a CAFE fuel economy rating represents combined 45/55 miles per gallon (mpg) highway/city fuel economy and is based on an unadjusted figure. b On the basis of the starter-generator rated 3l kilowatts (kW) for continuous, 8 kW for three minutes, and 35 kW for three seconds (s); 16 kW is assumed for a 12-second 0–60 acceleration. c The front motor is 25 kW and rear motor 10 kW. Therefore, the total motor peak power is 35 kW. NOTE: HEV = hybrid electric vehicle; CAFE = Corporate Average Fuel Economy; SI = spark ignition; CIDI = compressed ignition direct injection; NiMH = nickel metal hydride; Li-ion = lithium-ion; CVT = continuously variable transmission; M5 = manual five gear; A5 = automatic five gear; EMAT = electro-mechanical automatic transmission; hp = horsepower. SOURCES: U.S. Prius: EV News (2000); all HEVs except U.S. Prius: NRC (2000); Honda Insight: Automotive Engineering (1999); ESX3: Automotive Engineering (2000); GM Precept: Precept press release. 

PRESENT AND FUTURE AUTOMOTIVE TECHNOLOGIES 103 associated emissions and cost impacts were considered secondary in this context. It is important to understand to what extent the gains in fuel economy are actually achieved through the conventional technologies, on the one hand, and how much are obtained through “pure” hybrid technologies and systems, on the other. Four common elements that contribute to gains in fuel economy for these commercial and concept HEVs are: 1. choice of high-efficiency diesel engines for the three PNGV hybrid vehicles 2. aggressive load reduction measures that lower vehicle air and tire drag losses, as well as overall vehicle weight 3. engine downsizing to utilize a smaller, more advanced onboard combustion engine, as well as implementation of a more advanced trans- mission system 4. system electrification and hybridization to utilize electric power to optimize system efficiency, turning off the engine during idling, and pro- vision of regenerative braking. It is not easy to estimate the fuel economy benefits of “pure” hybrid technologies. First, hybrid benefits depend strongly on the driving cycle. The fuel economy benefits of hybrid electric vehicles are much higher under stop-and-go traffic conditions than under free-flow highway driv- ing conditions. Thus the fuel economy benefits of HEVs in China’s large cities should be higher than those in the United States. Second, there is no doubt that conventional technologies include all the measures in items 1 and 2 and that hybrid technology includes all the measures in item 4. It is less clear, however, how item 3 should be categorized, because hybrid technologies usually require synergies from conventional technologies to deliver the best benefits. It would then seem that the hybrid benefits are more than item 4 alone, but somewhat less than the combination of items 3 and 4 (An et al., 2001b; DeCicco et al., 2001). Figure 4-13 presents the improvement in U.S. fuel economy by each technology element for the selected HEVs. Issues Surrounding Fuel Cell Hybridization Uncertainties are associated with whether hybridization would ben- efit fuel cell vehicles (Table 4-6). In terms of system efficiency, unlike in internal combustion engines where the engine efficiency increases with the power demand, the efficiency of fuel cell stacks peaks at a low power point and decreases as power demand approaches either the maximum or 

104 PERSONAL CARS AND CHINA Baseline conventional vehicle (gasoline) Dieselization Load reduction Engine downsizing 90 Hybrid optimization 80 Miles per gallon 70 60 50 40 30 20 10 0 U.S. Prius . Honda Insight Ford Prodigy DC ESX3 GM Precept FIGURE 4-13 U.S. fuel economy (gasoline equivalent) through elements of die- selization, load reduction, engine downsizing, and hybrid optimization. SOURCE: An et al. (2001b). minimum level. For vehicles with onboard reformers, the efficiency of the reformer’s transient operation can differ dramatically from that of its steady-state operation. As for performance, the transient response time of fuel reformers is a critical issue and has a profound implication for the benefits of hybridization (the addition of batteries or other energy storage devices). In terms of system costs, the trade-off between the reduced cost associated with downsizing a fuel cell system and the added cost of bat- teries and other hybrid components should be assessed. The benefits and TABLE 4-6 Trade-offs for Hybridizing Fuel Cell Vehicles Trade-off Fuel Cell Vehicles Hybrid Fuel Cell/Battery Fuel cell stack High Reduced because of downsizing efficiency Regenerative braking None 10–20+ percent in urban driving Cold start Slow Rapid Specific weight High High to medium (kilograms/kilowatt) Price Fuel cell price high, Battery price relatively low, but not but dropping rapidly dropping as quickly as fuel cell Overall Long term, favorable Near to medium term, favorable SOURCE: Society of Automotive Engineers (2001, 2002). 

PRESENT AND FUTURE AUTOMOTIVE TECHNOLOGIES 105 trade-offs of hybridizing fuel cell vehicles not only differ from those from hybridizing conventional internal combustion engine vehicles, but also differ among different kinds of fuel cell systems. Hybridization Summary In spite of the many advantages of the hybrid vehicle, hybrid tech- nologies are still quite expensive and add up to 20 percent to the cost of a vehicle. Hybrid electric vehicles are many years from taking a significant market share from the conventional internal combustion engine vehicles. In turn, the economies of scale needed to reduce their cost disadvantage probably will not be realized for many years. In the meantime, some of the first steps toward hybrid technologies—the integrated starter-genera- tor, for example—will begin to appear in commercial products. In the United States, expensive features such as hybridization are more likely to appear in higher-priced (or subsidized) vehicles before they make their way into the wider commercial markets. SYSTEM INTEGRATION AND MANUFACTURING The preceding sections have described the status of the major compo- nents of a modern automobile. In China, many car components are al- ready being manufactured for assembly in local plants and for export. Moreover, local vendors are involved in technology transfer and improve- ment on the component level, although entry into the World Trade Orga- nization (WTO) will create pressure on component manufacturers to be more cost-competitive. Yet car design requires sophisticated techniques for integrating all these components—and ultimately for satisfying cus- tomer requirements for performance, safety, comfort, and convenience— to produce a cost-competitive, reliable vehicle. The leading global automakers have had years of experience in producing and marketing cars, and they continue to refine their system design and integration skills in order to maintain or improve their competitive position in the market- place. Consequently, these skills are considered highly proprietary and are carefully guarded. At present, China’s automotive industry is well behind global auto- motive industry leaders in design and system integration capabilities. China has some world-class auto production plants, but these rely on the expertise of foreign partners who are reluctant to transfer this expertise to their Chinese partners. If China is to compete in the future as a world- class auto manufacturer, it must build expertise in these critical areas. In future joint ventures and similar partnerships with foreign automakers, it 

106 PERSONAL CARS AND CHINA will be important for the Chinese partner to demand training and full inclusion in design and integration activities—at least for cars that will be developed for the Chinese market. Because Chinese automakers are likely to better understand Chinese consumers and their buying patterns, for- eign partners may be willing to relax proprietary concerns in the code- velopment of models that meet the needs of Chinese consumers. Yet un- der WTO such partners also can compete directly in Chinese markets. Therefore, the Chinese should be prepared to bring some special capabili- ties into partnerships in exchange for training and knowledge transfer. Manufacturing Implications World-class producers of vehicles must not only design high-quality vehicles, but also produce them in efficient facilities. In any discussion of manufacturing, it is useful to separate the vehicle assembly operations from the manufacture of major components such as engines and trans- missions. Major components are normally manufactured in high-capac- ity, highly automated plants. Because a large fraction of the tools found in these facilities are made by a few worldwide machine tool manufacturers, any facilities built in China can possess the latest technologies. Further- more, because of the high cost of the facilities, several vehicle manufactur- ers may use the same components purchased from external suppliers, al- lowing an economy of scale in production that benefits all participants. As noted in Chapter 3, component suppliers Robert Bosch, Engelhard, and Corning already have subsidiaries in China that are following this pattern. Whether the Chinese industry chooses to compete with the large component suppliers or to allow foreign manufacturers to dominate this part of the market is largely an economic issue and not one of technical capability. China has experience with vehicle assembly in modern plants, but these plants have been mostly ones transplanted from overseas through joint venture partners. Because the volumes for the Chinese car market will not at first be large for any single manufacturer, these facilities must be flexible enough to allow the assembly of more than one vehicle type. This flexible technology is just appearing in foreign facilities, but it is ex- pected to find greater uses as the market continues to fracture into smaller niches. Besides, as just noted, the latest tools for an assembly plant can be purchased from the worldwide machine tool industry. Unfortunately, even having the latest tools in the most modern fa- cilities does not guarantee efficient operation. Japanese manufacturers, through their highly developed “lean” manufacturing processes and through the colocation of major component and assembly plants, are able to assemble a vehicle in about half the number of worker-hours 

PRESENT AND FUTURE AUTOMOTIVE TECHNOLOGIES 107 characteristic of American manufacturers (Dassbach, 1994). Through years of sustained effort, Japanese manufacturers also have developed a reputation for the highest quality products. To compete worldwide, the Chinese automotive industry will need to develop the management ca- pability to operate its factories in the most efficient manner while pay- ing close attention to quality. China has a major asset in its intelligent, hard-working people, who provide skilled labor at wages that are low relative to the average in the member countries of the Organisation for Economic Co-operation and Development (OECD), but it will need to examine carefully its capability to manage that workforce in the most efficient manner. Choosing Competitive Technologies As discussed in Chapter 3, a variety of cars are already being pro- duced successfully in China. Most of these employ foreign technology, adapted to some extent to meet special Chinese needs, through a variety of partnerships and joint ventures with Chinese companies. If the Chinese automotive industry wishes to compete in open international markets or to compete with products coming to China from these markets, it must find technologies or market niches in which it is able to compete success- fully. In Figure 4-14, the estimates by Weiss et al. (2000) are replotted to show life cycle energy as a function of life cycle cost for the technologies considered. If China is to compete successfully in the manufacture of lower-cost cars, it would seem prudent to select technologies that do not start out with a potentially significant cost penalty. As shown in Figure 4-14, the overlap in the energy use and cost pro- jections for technology alternatives is large, especially for the advanced technologies for which the uncertainty ranges are largest. These uncer- tainty ranges will be reduced over the next decade or so as advanced tech- nologies mature from the present stage of development. Therefore, much more risk is associated with future competitiveness if major investments are made now by China to develop hydrogen fuel cell or electric cars, or even full hybrids, rather than in the future. A near-term focus on more conventional technologies still allows many opportunities for innovations that will enhance the competitiveness of a China car. APPENDIX: OTHER TYPES OF FUEL CELLS Of the many types of fuel cell, some are being developed for station- ary power generation. These systems are more appropriate for steady- state operation because they require the maintenance of a special environ- 

108 PERSONAL CARS AND CHINA 1996 Toyota Camry Electric vehicle Baseline 2020 4 Hydrogen fuel cell Advanced body gasoline Energy (MJ/km) 2 Advanced body diesel Gas, diesel, or CNG hybrids 0.3 0.4 0.5 Cost (dollars per kilometer) FIGURE 4-14 Estimated energy versus cost ranges for selected technologies per kilometer. NOTE: Shaded areas represented uncertainties in prediction; CNG = compressed natural gas. SOURCE: Weiss et al. (2002). ment. Thus such systems may not yet meet the rapid start-up requirement of an automobile and are likely to perform inefficiently in a lightly loaded automotive duty cycle. Descriptions of some of the fuel cells likely to find commercial uses over the next several decades follow: • Phosphoric acid. This type of fuel cell, which is commercially avail- able, is used mostly for stationary power sources. Phosphoric acid fuel cells generate electricity at more than 40 percent efficiency—and nearly 85 percent of the steam this fuel cell produces is used for cogeneration. Oper- ating temperatures are in the vicinity of 400°F or 200ºC. • Molten carbonate. Molten carbonate fuel cells, which promise high fuel-to-electricity efficiencies, operate at about 1,200°F or 650ºC. To date, molten carbonate fuel cells have been operated using hydrogen, carbon monoxide, natural gas, propane, landfill gas, marine diesel, and simu- lated coal gasification products. Molten carbonate fuel cells of 10 kW to 2 megawatts (MW) have been tested on a variety of fuels. Carbonate fuel cells for stationary applications have been successfully demonstrated in Japan and Italy. • Solid oxide. Another highly promising fuel cell, the solid oxide fuel cell (SOFC), could be used in large, high-power applications, including 

PRESENT AND FUTURE AUTOMOTIVE TECHNOLOGIES 109 industrial and large-scale central electricity generating stations. Some re- searchers also foresee the use of solid oxide in motor vehicles and are developing fuel cell auxiliary power units with SOFCs. A solid oxide sys- tem usually uses a hard ceramic material instead of a liquid electrolyte, allowing operating temperatures to reach 1,800°F. Power-generating effi- ciencies could reach 60 percent. One type of SOFC uses an array of meter- long tubes, and other variations include a compressed disc that resembles the lid of a soup can. Tubular SOFC designs are closer to commercializa- tion and are being produced by several companies. Demonstrations of tubular SOFC technology have produced as much as 220 kW. Still other fuel cells are not likely to be commercialized in the next few decades: • Alkaline. Long used by the U.S. National Aeronautics and Space Administration (NASA) on space missions, these fuel cells can achieve power-generating efficiencies of up to 70 percent. They use alkaline po- tassium hydroxide as the electrolyte. Until recently, these fuel cells were too costly for commercial applications, but several companies are examin- ing ways to reduce costs and improve operating flexibility. • Direct methanol fuel cells. These cells are similar to the proton ex- change membrane fuel cell (see the section Emerging Propulsion Tech- nologies) in that they use a polymer membrane as the electrolyte. How- ever, in the direct methanol fuel cell, the anode catalyst itself draws the hydrogen from the liquid methanol, eliminating the need for a fuel re- former. Efficiencies of about 40 percent are expected with this type of fuel cell, which would typically operate at a temperature of between 120° and 190°F or 50º and 90ºC. Higher efficiencies are achieved at higher tempera- tures. • Regenerative fuel cells. Still a very young member of the fuel cell family, regenerative fuel cells would be attractive as a closed-loop form of power generation, in conjunction with a solar power source. Solar power can be converted directly to electricity, but there may be applications where the energy losses associated with a regenerative fuel cell are justi- fied. During solar collection, water is separated into hydrogen and oxy- gen by a solar-powered electrolyzer. The hydrogen and oxygen are then available to be fed into the fuel cell, which is able to generate electricity, heat, and water when direct solar power is not available. The water is then recirculated back to the solar-powered electrolyzer and the process begins again. NASA and others are currently conducting research on these types of fuel cells. 

110 PERSONAL CARS AND CHINA REFERENCES An, F., F. Stodolsky, and D. Santini. 1999. Hybrid Options for Light-Duty Vehicles. SAE Technical Paper No. 1999-01-2929. Society of Automotive Engineers. An, F., J. DeCicco, and M. Ross. 2001a. Assessing the Fuel Economy Potential of Light-Duty Vehicles. SAE Technical Paper No. 2001- 01FTT-31, SAE Special Publication (SP-1637) on New Energy Systems and Environmental Impacts. Society of Automotive Engineers. An, F., A. Vyas, J. Anderson, and D. Santini. 2001b. Evaluating Commercial and Prototype HEVs. SAE Technical Paper No. 2001-01-0951. Society of Automotive Engineers. An, F., D. Freidman, and M. Ross. 2002. Near-Term Fuel Economy Potential for Light-Duty Trucks. SAE Technical Paper No. 2002-01-1900. Society of Automotive Engineers. Ashley, S. 2001. Fuel cells start to look real. Automotive Engineering (March). Automotive Engineering. 1996. Life cycle analysis: Getting the total picture on vehicle engi- neering alternatives. (March):49–52. ———. 1999. ESX3. (October):55. ———. 2000. Dodge’s mild hybrid. (May):32. Chinese State Environmental Protection Agency. 1999. Technical Policy on Prevention and Control of Motor Vehicle Pollution. Beijing. Dassbach, C. 1994. Where is North American automobile production headed? Low-wage lean production. Electronic Journal of Sociology 1(1). Davis, B. H. 1999. Transportation Energy Data Book: Ed. 19. ORNL-6958. Oak Ridge Na- tional Laboratory, September. Dec, J. E. 1997. A Conceptual Model of DI Diesel Combustion Based on Laser-Sheet Imaging. SAE Technical Paper No. 970873. Society of Automotive Engineers. DeCicco, J., F. An, and M. Ross. 2001. Technology Options for Improving the Fuel Economy of U.S. Cars and Light Trucks by 2010–2015. American Council for an Energy Efficient Economy, Washington, D.C., April. De Witt, M. J., and C. Z. Wan. 2000. Establishing capabilities to evaluate reductants for NOx adsorber technologies. Paper presented at the 6th Diesel Engine Emissions Reduction Workshop, August. Ehsani, M., A. Emade, and H. Gao. 2001. 42V Automotive Power Systems. SAE Technical Paper No. 2001-01-2465. Presented at the Future Transportation Technology Confer- ence, Costa Mesa, Calif., August. EV News. 2000. June. Flynn, P. F. 2000. Diesels—Promise and issues. Paper presented at the 6th Diesel Engine Emissions Reduction Workshop, August. ———. 2001. Diesels 2007—Promise and problems. Paper presented at the 7th Diesel Engine Emissions Reduction Workshop, August. Flynn, P. F., R. P. Durrett, G. L. Hunter, L. A. Farrell, and O. C. Akinyemi. 2000. Minimum Engine Flame Temperature Impacts on Diesel and Spark-Ignition NOx Production. SAE Technical Paper No. 2000-01-1177. Society of Automotive Engineers. Flynn, P. F., R. P. Durrett, G. L. Hunter, A. O. zur Loye, O. C. Akinyemi, J. E. Dec, and C. K. Westbrook. 1999. Diesel Combustion: An Integrated View Combining Laser Diagnos- tics, Chemical Kinetics, and Empirical Validation. SAE Technical Paper No. 1999-01- 0509. Society of Automotive Engineers. General Motors Corporation (GMC). 2001. Well-to-Wheels Energy Use and Greenhouse Gas Emissions of Advanced Fuel/Vehicle Systems: North American Analysis. Executive Summary Report, Detroit, April. Höhlein, B., J. Nitsch, and U. Wagner. 1998. Brennstoffzellen-Studie (Fuel Cell Studies): Vorhaben Nr. 686, Ganzheitliche Systemuntersuchung zur Energiewandlung durch Brennstoffzellen. Forschungsvereinigung Verbrennungskraftmaschinen e.V., Frankfurt am Main. 

PRESENT AND FUTURE AUTOMOTIVE TECHNOLOGIES 111 Horton, E. J., and W. D. Compton. 1984. Technological trends in automobiles. Science 225:588. Jost, K. 2002. Spark-ignition engine trends. Automotive Engineering International (January). Kummer, J. T. 1981. Catalysts for automobile engine control. Progress in Energy and Com- bustion Science 6:177–199. Moore, B. 1999. Advanced energy systems. Paper presented at the SAE Topical Technology Workshop on Hybrid Electric Vehicles: Here and Now (TOPTEC). SAE (Society of Automotive Engineers) International, Warrendale, Pa. Naber, J. D., and D. L. Siebers. 1996. Effects of Gas Density and Vaporization on Penetration and Dispersion of Diesel Sprays. SAE Technical Paper No. 960034. Society of Automo- tive Engineers. National Research Council (NRC). 2000. Review of the Research Program of the Partnership for a New Generation of Vehicles. Sixth Report. Washington, D.C.: National Academy Press. ———. 2002. Effectiveness and Impact of Corporate Average Fuel Economy (CAFE) Stan- dards. Washington, D.C.: National Academies Press. Office of Technology Assessment. 1995. Advanced Automotive Technologies: Visions of a Super-Efficient Family Car. OTA-ETI-638. Washington, D.C.: Government Printing Office. Ogden, J. M., M. M. Steinbugler, and T. G. Kreutz. 1999. A comparison of hydrogen, metha- nol, and gasoline as fuels for fuel cell vehicles: Implications for vehicle design and infrastructure development. Journal of Power Sources 79:143–168. Pembina Institute. 2000. Climate-Friendly Hydrogen Fuel: A Comparison of the Life-Cycle Greenhouse Gas Emissions for Selected Fuel Cell Vehicle Hydrogen Production Sys- tems. Pembina Institute, Drayton Valley, Alberta, March. Rovera, G., and D. Mesaiti. 1999. The general purpose hybrid vehicles. Paper presented at the SAE Topical Technology Workshop on Hybrid Electric Vehicles: Here and Now (TOPTEC). SAE (Society of Automotive Engineers) International, Warrendale, Pa. Sher, E., ed. 1998. Handbook of Air Pollution from Internal Combustion Engines: Pollutant Formation and Control. Boston: Academic Press. Siebers, D. L. 1999. Scaling liquid-phase fuel penetration in diesel sprays based on mixing limited vaporization. Paper presented at 1999 Congress of the Society of Automotive Engineers, Detroit, February. Sierra Research. 1997. Automotive fuel economy potential using cost-effective design changes. Sierra Research, Buffalo, N.Y. Singh, M., R. Cuenca, J. Formento, L. Gaines, B. Marr, D. Santini, M. Wang, S. Adelman, D. Kline, J. Mark, J. Ohi, N. Rau, S. Freeman, K. Humphreys, and M. Placet. 1998. Total Energy Cycle Assessment of Electric and Conventional Vehicles: An Energy and Envi- ronmental Analysis. ANL/ES/RP-96387, Argonne National Laboratory. Society of Automotive Engineers. 2001. Fuel Cell Power for Transportation 2001. SAE SP- 1589. ———. 2002. Fuel Cell Power for Transportation 2002. SAE SP-1691. Stodolsky, F., L. Gaines, C. Marshall, and F. An. 1999. Total Fuel Cycle Impacts of Advanced Vehicles. SAE Technical Paper No. 1999-01-0322. Society of Automotive Engineers. Toyota. 1997. Prius Product Information 2000. Toyota Hybrid System, Toyota Press Infor- mation 1997. U.S. Council for Automotive Research. 2000. U.S. Advanced Battery Consortium Battery Criteria: Commercialization. Online. USCAR. Available at www.USCAR.org. Accessed May 2000. U.S. Department of Energy. 1999. Hybrid electric vehicle component information. Online. Available at www.hev.doe.gov/components/batteries.html. Accessed June 13, 2002. ———. 2001. Fuel Cells for Transportation. 2001 Annual Progress Report, December. 

112 PERSONAL CARS AND CHINA Wang, M. Q. 1999. GREET 1.5—Transportation Fuel-Cycle Model, Volume I: Methodology, Development, Use, and Results. ANL/ESD-39. Argonne National Laboratory, August. Wang, M. Q., and H.-S. Huang. 1999. A Full Fuel-Cycle Analysis of Energy and Emissions Impacts of Transportation Fuels Produced from Natural Gas. ANL/ESD-40. Argonne National Laboratory, December. Weiss, M. A., J. B. Heywood, E. M. Drake, A. Schafer, and F. F. AuYeung. 2000. On the Road in 2020: A Life-Cycle Analysis of New Automobile Technologies. MIT Energy Labora- tory Report #MIT-EL-00-003. Massachusetts Institute of Technology, Cambridge. Wengel, J., and E. Schirrmeister, eds. 2000. The Innovation Process from the Internal Com- bustion Engine to Fuel Cells. Fraunhofer-Institute, Karlsruhe, Germany, February. Westbrook, C. K., and L. K. Chase. 1988. Chemical Kinetics and Thermochemical Data for Combustion Applications. Lawrence Livermore National Laboratories Report UCID- 17833. Westbrook, C. K., W. J. Pitz, and W. R. Leppard. 1991.The Autoignition Chemistry of Paraf- finic Fuels and Pro-Knock and Anti-Knock Additives: A Detailed Chemical Kinetic Study. SAE Technical Paper No. 912314. Society of Automotive Engineers. Williams, R. 1998. Fuel cells, coal, and China. Paper presented at the Ninth Annual U.S. Hydrogen Meeting: Implementing a Global Solution, Washington, D.C. Zhan R., R. Hurley, W. Han, J. A. Lymburner, D. Schuetzle, J. Li, G. Wu, L. Xu, H. S. Gandhi, R. Huang, and D. Deng. 2001. The Development of Low Precious-Metal, Rare-Earth Oxide (Reo) Catalysts for Vehicle Emission Control in Emerging Markets. SAE Techni- cal Paper No. 2001-01-0225. Presented at the SAE 2001 World Congress, March, Detroit. Society of Automotive Engineers.