Page 20

2 Development of Vehicle Subsystems

CANDIDATE SYSTEMS

The success of the PNGV program depends on integrating R&D programs that can collectively improve the fuel efficiency of automobiles within the very stringent boundary conditions of size, reliability, durability, safety, and affordability of today's vehicles. At the same time the vehicles must meet emission regulations, be largely recyclable, and use components that can be mass produced and maintained similar to current automotive products.

To achieve the Goal 3 fuel economy target of 80 mpg (1.25 gallons per 100 miles), the energy conversion efficiency of the chemical conversion system (e.g., a power plant, such as a compression-ignition direct-injection [CIDI] engine or a fuel cell) averaged over a driving cycle will have to be at least 40 percent. This challenging goal requires developing and integrating many vehicle system concepts. For example, the primary power plant will have to be integrated with energy-storage devices and the vehicle structure will have to be built of lightweight materials. Every aspect and function of the vehicle will have to be optimized, both individually and as part of the vehicle system. No aspect of the vehicle function can be left untouched, from minimizing the energy expenditure for maintaining comfort in the passenger cabin to significantly improving the conversion efficiency of the exhaust-gas after-treatment systems.

The USCAR partners have chosen the hybrid-electric vehicle (HEV) as the power train in their respective concept cars. The HEV uses stored energy in the battery to drive an electric motor that provides power boost to the engine, thereby permitting a smaller engine that can be operated closer to optimum conditions. This results in increased energy conversion efficiency, reduced emissions, and



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 20
Page 20 2 Development of Vehicle Subsystems CANDIDATE SYSTEMS The success of the PNGV program depends on integrating R&D programs that can collectively improve the fuel efficiency of automobiles within the very stringent boundary conditions of size, reliability, durability, safety, and affordability of today's vehicles. At the same time the vehicles must meet emission regulations, be largely recyclable, and use components that can be mass produced and maintained similar to current automotive products. To achieve the Goal 3 fuel economy target of 80 mpg (1.25 gallons per 100 miles), the energy conversion efficiency of the chemical conversion system (e.g., a power plant, such as a compression-ignition direct-injection [CIDI] engine or a fuel cell) averaged over a driving cycle will have to be at least 40 percent. This challenging goal requires developing and integrating many vehicle system concepts. For example, the primary power plant will have to be integrated with energy-storage devices and the vehicle structure will have to be built of lightweight materials. Every aspect and function of the vehicle will have to be optimized, both individually and as part of the vehicle system. No aspect of the vehicle function can be left untouched, from minimizing the energy expenditure for maintaining comfort in the passenger cabin to significantly improving the conversion efficiency of the exhaust-gas after-treatment systems. The USCAR partners have chosen the hybrid-electric vehicle (HEV) as the power train in their respective concept cars. The HEV uses stored energy in the battery to drive an electric motor that provides power boost to the engine, thereby permitting a smaller engine that can be operated closer to optimum conditions. This results in increased energy conversion efficiency, reduced emissions, and

OCR for page 20
Page 21the potential to recover a fraction of the vehicle's kinetic energy during braking. Not only were the concept cars demonstrated by the PNGV partners in 2000 a great technological achievement, they also helped clarify the remaining hurdles to achieving success for the PNGV program. It is apparent that the fuel cell will not be feasible in a production-prototype vehicle by 2004. This leaves the internal combustion engine as the primary energy converter, and even using the most efficient one, the CIDI diesel engine, the three-times fuel economy target remains a stretch goal. If maximization of fuel economy is the design target, the diesel engine is the first choice; however, meeting the mandated emission standards is a major challenge for the diesel engine. Therefore, a critical consideration to maximize fuel economy is reduction of nitrogen oxides (NOx) and particulates, the two emission standards that are most difficult for the diesel engine to meet. Reducing the cost for manufacturing and moving the concept technology into current and future vehicles has now become a central factor, so there has been a notable shift in emphasis during 2000–2001 to cost reduction and manufacturability. Resources are being focused on continued development of enabling technologies, such as exhaust-gas after-treatment systems, fuel composition effects on system performance, advanced battery energy storage systems, and power electronics and component cost reduction. The investigations of promising longer-term prospects, such as advanced combustion systems and fuel cell technologies, are also continuing. The PNGV presented to the committee an overview of the status and critical development issues of the candidate energy-conversion and energy-storage technologies that survived the 1997 technology selection process. Overviews of candidate electrical and electronic systems and advanced structural materials for the vehicle body were also presented. This chapter addresses the following technology areas and related issues: Four-stroke internal-combustion reciprocating engines; Fuel cells; Electrochemical storage systems (rechargeable batteries); Power electronics and electrical systems; Structural materials; Vehicle safety; and Fuels. The committee reviewed R&D programs for each of these technologies, along with related vehicle safety and fuels issues, to assess progress so far and the developments required for the future. In the committee's opinion the PNGV continues to make significant progress in developing the candidate systems and identifying critical technologies that must be addressed to make each system viable. The committee is pleased that, since the introduction of the concept vehicles in 2000, there has been a shift in program focus to include affordability.

OCR for page 20
Page 22 INTERNAL COMBUSTION RECIPROCATING ENGINES The internal combustion engine continues to be the primary candidate power plant for meeting near-term PNGV program goals. To meet the fuel economy target (80 mpg) of Goal 3 the internal combustion engine will have to be integrated into an HEV configuration. The CIDI engine, using diesel fuel, is the most efficient of the internal combustion engines. Consequently, in the near term, taking advantage of the high efficiency of the diesel engine and integrating it into an HEV is the most promising way to attain maximum vehicle fuel economy. However, the challenges of meeting the new California Air Resources Board (CARB) and the U.S. Environmental Protection Agency Tier 2 emission standards are a major hurdle for the CIDI engine (NRC, 1999, 2000), even when used in an HEV power train. In maintaining its quest for a vehicle with fuel consumption of 1.25 gallons per 100 miles (80 mpg) the PNGV has continued its focus on the diesel engine as the primary energy converter for the vehicle. As a result, PNGV has continued its aggressive investigation of different approaches to emission reduction for CIDI engines. To achieve the emission targets will require an integrated approach to further refinements in engine design and operation, aggressive development of exhaust-gas after-treatment and its integration into the power-train system, and modification of the fuel to allow optimum engine performance while facilitating the exhaust-gas emission-reduction technologies. The most notable fuel modification is the need to reduce the sulfur content in the fuel. These three aspects of reducing NOx and particulate matter were discussed in the committee's sixth report (NRC, 2000). The partnership continued its emphasis on the system approach during 2000–2001. The primary emphasis continues to be exhaust-gas after-treatment systems; however, work continues on fundamental combustion systems, such as homogeneous charge compression ignition (HCCI), basic injection, and combustion analysis, that if successful could offer some emissions reduction and fuel economy improvement. There is a general sentiment that emission standards can be met with incremental development of known technology applied to homogeneous-charge spark-ignition engines and probably with direct-injection spark-ignited gasoline engines. Using the spark-ignited engine, however, would result in a reduction in the vehicle's fuel economy compared with that of a diesel engine. Each of the partners has proprietary in-house programs on both homogeneous and direct-injection spark-ignition engines. The fuel economy of gasoline direct-injection engines and the challenge of reducing their emissions to meet CARB low-emission vehicle (LEV 2) and federal Tier 2 standards falls between those of the homogeneous spark-ignited gasoline engine and the diesel engine. Because the in-house programs are proprietary, those efforts are not reported here. This review focuses on the status of the partners' joint R&D efforts for diesel engines and the identification of the critical barriers that need to be surmounted for success in the PNGV program.

OCR for page 20
Page 23 Program Status and Plans The unveiling of the concept cars by each of the three USCAR partners in 2000 represented a transition in the PNGV program. The concept cars are a successful technology demonstration that also serves to focus more sharply the critical technical hurdles remaining for successful completion of the program. Clearly, manufacturability and affordability are two critical issues on which PNGV has now placed increased emphasis. The post-concept-car aspect of the program also signifies a shift to a higher degree of in-house proprietary effort by each of the partners as plans are made to transfer the concept car technologies into their respective products. The committee was given individual proprietary briefings by each of the partners to help it understand the extent of these efforts. In addition to the proprietary internal work PNGV continued its programs of collaborative work on precompetitive fundamentals. This involves work on the interaction between fuel composition and engine performance, investigations of combustion fundamentals and diagnostics, and emission control systems, primarily exhaust-gas after-treatment. The partnership's activity in the four-stroke direct- injection (4SDI) engine technical area is a major collaborative effort of the individual partners, the national laboratories, and a few universities. The Pacific Northwest National Laboratory, Lawrence Livermore National Laboratory, Lawrence Berkeley National Laboratory, Sandia National Laboratories, Los Alamos National Laboratory, National Renewable Energy Laboratory, Argonne National Laboratory, Oak Ridge National Laboratory, and the Department of Energy Headquarters are all active collaborators with the partnership. University participation includes Wayne State University, the University of Wisconsin, and the University of Michigan. It is important to remember that development of a vehicle power train that maximizes fuel economy while minimizing emissions requires a power-train systems approach. There will be interactions among the fuel, the engine, and the exhaust-gas after-treatment subsystems. PNGV is addressing this systems issue; however, for the purposes of the following discussion, it is convenient to divide it into three components: engine combustion, emission control systems, and engine-fuel interactions. Engine-Combustion System Developments The challenge of meeting the CARB LEV 2 and Tier 2 emission standards with a diesel engine was highlighted in the 2000 committee report (NRC, 2000). A critical requirement is that tailpipe NO x and particulate matter (PM) emissions will need to be drastically reduced for the diesel engine to become a viable PNGV power plant. It is unlikely that requisite emission reductions will be achieved through in-cylinder combustion modification alone; exhaust-gas after-treatment will almost certainly be necessary. However, a reduction in engine-out emissions

OCR for page 20
Page 24reduces the after-treatment conversion efficiency required. Furthermore, it may be necessary to tailor the engine exhaust composition to the after-treatment device, which means being able to control the combustion process. Therefore, understanding combustion fundamentals continues to be important. The engine combustion investigations reported to the committee this year had as objectives: further understanding of in-cylinder combustion phenomena, development of diagnostics that would facilitate optimization of in-cylinder combustion, and demonstration of a fuel system for dimethyl ether, which is a non-sooting fuel. The partnership has established a draft technical R&D plan and engine technology performance targets for the 80 mpg (1.25 gallons per 100 miles) PNGV vehicle. The Technical R&D Plan Draft Objectives for CIDI engines operating on diesel fuel are (Howden, 2000): By 2002 develop and validate NO x (0.2 g/mile) and PM (0.02 g/mile) emission control technologies; By 2004 develop and validate NO x (0.07 g/mile) and PM (0.01 g/mile) emission control technologies; and By 2007 demonstrate the capability to meet Tier 2 Bin 3 emissions (NO x : 0.03 g/mile; PM: 0.01 g/mile). 1 The engine technical targets are given in Table 2–1. The emission targets in the table are for the tailpipe output, considering the combined system of the fuel, engine, and after-treatment devices. The 4SDI technical team also quantified emission targets for the engine alone: 0.36 g/mile for NO x and 0.04 g/mile for PM. The difference between these levels and the 2004 targets of 0.07 and 0.01 g/mile for NO x and PM will need to be achieved through the exhaust-gas after-treatment system. It is convenient to categorize the projects in the 4SDI program as those that address short-term issues and those that address longer-term issues. In the engine combustion subprogram, the near-term projects include (1) developing advanced fuel injection systems; (2) studying the cylinder-to-cylinder distribution and transient response of exhaust-gas recirculation (EGR); (3) developing in-cylinder combustion and PM measurements; and (4) performing detailed comparisons between combustion data obtained in an optical research engine, a similar metal engine, and the predictions of sophisticated three-dimensional computer simulations. Projects in the longer time frame focus on continued investigation of HCCI 1The California LEV 2 targets for NO x and participates are LEV NO x =0.05 g/mile at 50,000 miles and LEV PM=0.01 g/mile at 100,000 miles. This is simplified for illustrative purposes; the California emission regulations are a complex set of rules.

OCR for page 20
Page 25 TABLE 2–1 CIDI Engine Technology R&D Plan Technical Targets for an 80-mpg PNGV Vehicle Year Characteristics 2002 2004 2007 Best brake thermal efficiency (%) a 44 45 46 Best full load thermal efficiency (%) b 42 43 44 Displacement power density (kW/L) c 42 45 47 Specific power (W/kg) 590 625 650 Durability (hours) >3,500 >5,000 >5,000 Emission control cost ($/kW) d 5 4 3 Exhaust emission control device volume (L/L) e 3 2 1.5 Engine cost ($/kW) d , f 30 30 30 NO x emissions g (g/mile) h 0.2 0.07 0.03 PM emissions g (g/mile) i 0.02 0.01 0.01 Fuel economy penalty due to emission control system (%) <5 <5 <5 a Ratio of mechanical power out to fuel energy rate (lower heating value) in. b Ratio of mechanical power out to fuel energy rate (lower heating value) in at peak power. c Ratio of the peak power output divided by the volumetric displacement of the engine. d Assumed high-volume production of 500,000 units per year. e Volume of emission control system (in liters) per liter of engine displacement. f Constant out-year cost targets reflect the objective of maintaining engine system cost while increasing engine complexity. g Full-useful-life emissions using advanced petroleum-based fuels as measured over the federal test procedure as used for certification in those years. h NOx values given are tailpipe emissions; the engine-out target for NOx is 0.36 g/mile. The difference between this number and the tailpipe emission target will need to be achieved through after-treatment systems. i PM values given are tailpipe emissions; the engine-out target for PM is 0.04 g/mile. The difference between this number and the tailpipe emission target will need to be achieved through after-treatment systems. combustion, with assessments of how advanced variable compression ratio engine concepts and electromagnetically or hydraulically actuated engine valves—often referred to as camless engines—could enhance combustion control. It is generally agreed that EGR will have to be employed to meet emission standards. Typical engines tend to exhibit a quadratic increase of PM emissions with EGR increases. Furthermore, the engine's tolerance for EGR varies inversely with load. The exact reasons for these behaviors are not known. Through its fundamental research programs PNGV is developing a diagnostic for use in production engines for measuring the cylinder-to-cylinder EGR distribution. Being able to minimize or eliminate the maldistribution of EGR between different

OCR for page 20
Page 26engine cylinders should enable maximum exploitation of this emission reduction technology. More insight into the fuel-injection and air-entrainment processes was gained through optical diagnostics in a high-pressure, high-temperature constant-volume vessel. For a further understanding of in-cylinder particulate phenomena a collaborative effort was started between PNGV and Sandia National Laboratories to address real-time particulate diagnostics for size, number density, and volume fraction. Real-time measurements of the particulate emissions during a cold start have been made for both a turbocharged diesel-powered vehicle and one with a port-injected spark-ignition engine. Even though the measurements are qualitative, the results of the particulate measurements for relative size distributions and mass concentrations for the two engines are consistent with those obtained using conventional measurement techniques during steady-state operation. The particulate levels from the spark-ignition engine are approximately one order of magnitude below those of the diesel engine. The size data are important, as there is growing concern about a correlation between particulate size and adverse health effects. If consensus is reached on the importance of the particulate size and adverse health effects, the partnership will be in a position to assess the status of the engine it is developing. The collaborative effort between Sandia National Laboratories, Wayne State University, and the University of Wisconsin, Madison, continues. In this program an optical engine, metal engine, and advanced computer simulation are being used to enhance PNGV's fundamental understanding of in-cylinder processes. The combination of optical diagnostics, metal engine operation, and computational comparison are being used to improve the capabilities of the simulation, which in turn is being implemented for design optimization. A more basic approach to in-cylinder emission reduction is being pursued through PNGV's program on HCCI combustion. PNGV's research and an exhaustive literature review have confirmed that HCCI combustion is limited to light-load operation. Therefore, before HCCI combustion can be employed in an automotive engine, methods of control must be developed. Commonly used methods to control HCCI combustion include variable valve timing, variable compression ratio, variable residual gas retention, active ignition via hot surfaces or auxiliary injection, variable mixture preparation schemes, and variable fuel chemistry. The PNGV is active in evaluating the potential of these techniques through combustion diagnostics (being done at Sandia National Laboratories) and fuel composition-engine interactions, and investigations of variable valve actuation and variable-compression-ratio systems. The challenges are many and great; however, if successfully developed, these combustion approaches could be integrated into a more conventional engine-operating scheme to improve emissions and fuel economy, but probably not for the 2004 time frame. In 2000, it was reported that the 4SDI technical team, working with the Department of Energy, awarded Cummins Engine Company and Detroit Diesel

OCR for page 20
Page 27Corporation cooperative agreements to develop PNGV-size CIDI engines. Since the announcement the respective companies have formed alliances with catalyst manufacturers and are developing complete CIDI engine systems, including exhaust-gas after-treatment. After-Treatment and Controls and Sensors The most difficult challenge facing the 4SDI technical team is to develop power trains that meet the legally mandated emission standards. At this point in the program the only hope for meeting these standards is through the development of very effective exhaust-gas after-treatment systems. As was the case during the committee's sixth review, the 4SDI technical team devoted a majority of its effort to this activity in 2001. The primary focus of its effort is reducing the NO x and PM emissions from the diesel engine. NO x After-Treatment Systems. The 4SDI technical team presented its strategy for NO x control technology development to the committee. The plan is to pursue extensive in-cylinder EGR, which will require optimizing the EGR systems in conjunction with NO x -reduction after-treatment systems. After-treatment techniques being studied include selective catalytic reduction (SCR) systems using urea (NH2CONH2); active lean-NO x reduction systems using fuel as the reductant; NO x absorbers or traps with catalyst systems; and nonthermal-plasma catalytic systems. Passive lean-NO x catalytic systems have been dropped from consideration. The 4SDI technical team believes that the major breakthrough in conversion efficiency needed to make them viable is unlikely in program's time frame. The highest NO x conversion efficiency and the least fuel economy penalty have been achieved using an SCR system with urea as the reductant. Conversion efficiencies of approximately 80 percent over the Federal Test Procedure (FTP) driving cycle have been obtained with simulated feed gas. This resulted in a projected fuel economy penalty of less than 0.5 percent. Engine tests using a 1.2-liter DIATA (direct-injection, aluminum, through-bolt assembly) engine indicated 70 percent NO x conversion efficiency. The remaining technical issues for SCR with urea are avoiding production of ammonium nitrate and ammonium sulfate under low-temperature operation; eliminating ammonia slip at high temperatures; size and cost of the system; developing a viable way to store, deliver, and replace the urea onboard the vehicle; and infrastructure issues for urea distribution and delivery to the vehicle. A urea SCR system will also require tailpipe NO x and NH3 sensors for closed-loop control of urea injection. Active lean-NO x catalyst systems use hydrocarbon injection, most likely fuel, upstream of the catalyst to promote the NO x reduction. The best performance of such systems to date yields conversion efficiencies of less that 50 percent, with fuel economy penalties of 5 to 8 percent. This efficiency is too low to meet emission standards. In addition, the temperature range for peak conver-

OCR for page 20
Page 28sion is too narrow and occurs at too low a temperature to match the maximum efficiency operating range of the engine. In addition, the catalysts are susceptible to moisture degradation. For active lean-NO x catalyst technology to be viable, the conversion efficiencies, moisture resistance, and light-off temperature range will all have to be improved (i.e., new catalysts must be developed). Nitrogen oxide traps have been demonstrated to work very well. However, they are sensitive to sulfur contamination. This sulfur sensitivity does not appear to have a minimum threshold. Any sulfur in the exhaust will contaminate the trap. Reducing the sulfur level in the fuel but not eliminating it will only reduce the rate at which the trap becomes contaminated. Test data indicated that even with a fuel sulfur level of 3 ppm, the absorber's NO x -trapping efficiency was reduced from 95 percent to 80 percent in just 250 hours. Regeneration of the trap to get rid of the sulfur requires temperatures on the order of 650 to 700°C for up to 10 minutes and adversely affects the trap's durability. Even if the sulfur contamination problem can be eliminated, the NO x traps, like the active catalyst systems, require regeneration by injection of hydrocarbons, again most likely fuel. Consequently there is a fuel economy penalty associated with NO x traps. The 4SDI technical team estimates that this fuel economy penalty would be approximately 5 percent. Adding a sulfur trap would extend the lifetime of current NO x traps. This means that either a sulfur trap regeneration procedure must be developed or a routine maintenance schedule for sulfur trap replacement must be mandated. Successful development of a viable sulfur trap would then pave the way for use of other sulfur-sensitive technologies (e.g., continuously regenerating particulate traps). To address this critical need DOE's Office of Advanced Automotive Technologies (OAAT), working with the PNGV, has included sulfur-trap development and sulfur-tolerant catalyst material development in a recent solicitation. The request for proposals details specifications for the sulfur trap and the sulfur-tolerant catalysts that are directly aimed at PNGV needs. The funding levels are $1.5 million, with a 20 percent industry cost share, for each topic. The program duration is two years. Although it is still in the research stage, there is growing optimism about the potential of nonthermal plasma catalysis for NO x , and possibly particulate, control (much of the work is being conducted at the Pacific Northwest National Laboratory). Creating a plasma within a dielectric catalytic material enhances NO x reduction. It is believed that the plasma converts some of the NO in the exhaust gas to NO2 and partially oxidizes the hydrocarbon reductant to intermediate hydrocarbon species, aldehydes, and carbon monoxide (CO). These intermediate hydrocarbon species and the NO2 are then involved in the reduction reactions at the NO x catalyst. An oxidation catalyst is then needed downstream of the plasma catalyst to remove the residual hydrocarbons (HC) and CO. To date the best performance achieved in the laboratory has been a NO x conversion of approximately 60 percent with a fuel economy penalty of approximately 5 per-

OCR for page 20
Page 29cent. (A conversion efficiency of greater than 95 percent will be required to meet Bin 2 of the federal Tier 2 emission regulations.) Development work is continuing to investigate different catalyst combinations and is pursuing techniques for reducing the energy consumption. One interesting experiment was performed that applied the plasma system to a urea SCR system while simulating engine cold start. The temperature at the end of the test was only 140°C, and excellent low-temperature performance was achieved. One of the reasons for the high level of interest in plasma catalysis is that the NO2 generated by the plasma as part of the NO x reduction process is also known to be an oxidant for PM. It may be possible to incorporate a plasma catalysis system that will reduce both NO x and particulates in the same device. An experimental plasma-catalyst diesel particulate filter was tested. The catalyst bed served as a trap for the larger particles and seemed to oxidize the smaller ones. It successfully reduced the particulate mass emissions over a particle size range of 10 to 350 nm. When the device was tested without the plasma operating, the passage of particles smaller than 200 nm became excessive, peaking at 700 million particles per cubic centimeter at a particle size of approximately 50 nm. When the device was operated with the plasma, the particle flux was on the order of 1 million particles per cubic centimeter for all size ranges. This is approximately the particle emission rate for a homogeneous spark-ignited engine. Chemi-cal analysis of the particulates indicated that the plasma was also effective in reducing the polycyclic aromatic content of the particulate matter. Such characteristics of the particulate matter are not currently subject to regulation; however, these data are important as they establish a database for evaluation relative to the findings of current investigations into the possible link between particulate composition and size and adverse health effects. The above results are encouraging. However, the catalyst bed to which the plasma is applied eventually became clogged with particulate matter and needed to be regenerated, a common problem for all diesel particulate filters. In total, the plasma-catalyst after-treatment technology is an exciting program. The plasma catalyst system offers the advantages of NO x removal over a wide temperature range, low NO2 production, and very-low-temperature NO x removal with urea SCR and possibly even PM removal. Because it is a new technology, there is reason for optimism that it will improve with further maturity. The significant challenges at this time are its electrical power consumption and the question of whether future improvements in conversion efficiency will be sufficient to meet future emission standards, which will undoubtedly be lower than the current CARB LEV 2 and Tier 2 standards. Particulate Control Systems. PM reduction strategies include fuel injection optimization, regenerative particulate traps, and fuel and lubricant modifications, which will be discussed below in the “ Engine-Fuel Interactions ” section. The degradation of fuel economy, as well as the expense, is a concern for all these

OCR for page 20
Page 30approaches. Most vehicles will require participate traps to meet the new PM emission standards. The two major issues for the particulate traps are how well they filter out very small particles (d < 250 nm) and how easily they can be regenerated. The 4SDI technical team is pursuing particulate trap regeneration technology and is closely following the activity in Europe, where in 2001 Peugeot is planning to introduce a regenerative particulate trap that uses the Rhodia Eolys™ cerium-based fuel additive. In addition to the cerium-based fuel additive, techniques for particulate trap regeneration include microwave ignition, electrical ignition, and catalytic regeneration with a catalyst in the trap (continually regenerating trap [CRT]). The CRT oxidizes the PM using NO2, which is produced catalytically at the entrance of the trap. The catalyst to produce NO2 is sulfur sensitive, and consequently the CRT requires low-sulfur fuel or a sulfur trap. The PNGV is actively investigating or evaluating all these approaches to trap regeneration. Although particle size is not a component of the emission standards, increased attention is being given to the size distribution of the PM emitted from combustion systems. Of concern is the complex relationship between particle size, particle number, and total particulate matter mass and adverse health effects with a suggestion that particle number may be especially important for ultrafine particles less than 500 nm in diameter. The 4SDI technical team has started to build such a database. The particle sizes measured range from approximately 50 to 500 nm. Tests were conducted for steady-state operation with a filter but without regeneration. The results indicated that the particle size distributions before and after the filter were similar in terms of the number of particles as a function of size. However, the number density and volume fraction of particulates were reduced by over 99 percent. The assessment of the filters as a PM control technology remains optimistic. The primary issues that must be addressed are the durability, regeneration efficiency, and cost of these devices. Vehicle Testing. In an effort to move closer to practical demonstration of the different after-treatment systems and to address the issues of integrating them into the entire power-train system, the 4SDI technical team is performing integrated power-train vehicle testing. Two different research vehicles fitted with experimental after-treatment systems, particulate traps, and a NO x absorber or urea SCR system are being tested on a special low-sulfur fuel (4 ppm sulfur). Results from these tests indicate that under ideal conditions Tier 2 emission standards are achievable with small diesel vehicles and advanced controls. Of course, these results are from highly controlled research tests. The life of the NO x absorber and the extent to which it is affected by sulfur in the exhaust remain problematic. Reducing sulfur sensitivity, enhancing absorber performance, and developing improved regeneration will be the focus of further efforts. Much work is still required.

OCR for page 20
Page 60 ~ enlarge ~ FIGURE 2–2 Polymer composite pickup box. SOURCE: Mehta, 2000. ~ enlarge ~ FIGURE 2–3 Cost of polymer composite pickup box relative to the cost of steel pickup box. SOURCE: Mehta, 2000.

OCR for page 20
Page 61the box is far superior to that of steel truck beds. Also, Ford has used sheet molding technology to produce an all-composite, one-piece SUV truck bed for the Ford Explorer Sport Trac. The steel counterpart was 30 percent heavier and composed of 40 separate parts. Another approach to composite body technology is being pursued by DaimlerChrysler in its proprietary PNGV program (NRC, 1999, 2000). Its process produces large, one-piece parts by injection molding of glass-fiber-reinforced polymer (GFRP) resins. The integrated moldings, which are 25 to 35 percent lighter than mild steel, are attached to an aluminum space frame, resulting in a BIW 50 percent lighter than steel (NRC, 2000). As indicated in Table 2–9, this technology was used in the DaimlerChrysler ESX3 concept vehicle, which achieved the lowest curb weight of the three concept vehicles developed by the USCAR partners (see Table 2–9). Projects Under Way Focal Project III (FP III) supported by the Advanced Composites Consortium (ACC) involves the development of BIW structure in GFRP composites. This entails constructing the body from approximately 11 large liquid molded parts, as shown in Figure 2–4. Thus far, computer-aided design (CAD) and coarse finite element analysis (FEA) models have been generated for initial concepts ~ enlarge ~ FIGURE 2–4 The ACC Focal Project III body-in-white structure. SOURCE: Mehta, 2000.

OCR for page 20
Page 62and have been evaluated for performance, processing, and assembly. Bending and torsional stiffness of the BIW structure have been optimized. Detailed finite element models of chosen concepts have been generated for analysis of all load cases, including stiffness, strength, durability, and modal performance. Key processing and material options that require further development in order to manufacture the FP III structure have been identified. In the past, work by the ACC showed that it was difficult to achieve good frontal crash response from polymer composite front-end structures. These early problems have been overcome. The ACC's Focal Project 1 demonstrated that a composite structure can be designed to meet Federal Motor Vehicle Safety Standard (FMVSS) requirements. In this case the front end of an Escort was built and successfully tested in a frontal crash experiment. Although the ACC projects are aimed at glass FRP, where weight savings are in the range of 25 to 35 percent, the technology of liquid molding of composites can be adapted to carbon FRP, where weight savings approach 65 percent. Projects are in place with suppliers and ORNL to develop low-cost carbon fibers through lower-cost precursors and by microwave processing. Potential exists to reduce fiber cost by 20 percent. Aluminum metal matrix composite (Al-MMC) applications in chassis and power-train subsystems could account for 30- to 50-lb weight savings. The major hurdles to developing applications of this material are feedstock costs ( Table 2–10) and the development of a reliable process for making the composites. A low-cost powder metal process and a casting process are under development. In the casting process a new and improved batch technique for mixing silicon carbide in the aluminum melt has been developed, and the performance of a lower-cost (silicon carbide) reinforcement has been demonstrated. Equipment that can prepare a 600-kg sample batch of composite material was expected to be operating in January 2001. In the area of recycling, a project on Aluminum Alloy Scrap Sorting has been initiated to demonstrate the commercial practicality of sorting shredded scrap, using color and laser spectroscopy. This process is capable of separating cast from wrought materials and distinguishing among wrought alloys. Program Assessment As the PNGV program moves toward 2004 and the requirement of producing affordable production-prototype vehicles, the PNGV team should attempt to balance the opposing requirements of weight reduction and affordability. The arguments presented in the “ Materials Selection, Design, and Manufacturing ” section above lend credence to this position. The PNGV team should carefully follow the progress on the new ULSAB-AVC project. This modeling study appears to be on track to demonstrating that a 2,275-lb steel-intensive vehicle, operating on a gasoline-powered ICE engine, can be attained. It may be possible

OCR for page 20
Page 63to add aluminum and/or magnesium components (hybridize) and/or polymer composites and achieve the vehicle weight target at a much lower cost than with the aluminum-intensive vehicle approach. These design and manufacturing techniques could be used on aluminum-intensive vehicles as well as hybrid steel-aluminum structures. The challenge in being able to adopt low-density materials on a broad scale is the development of new, low-cost processing methods, including feedstock materials. The committee is satisfied that the PNGV materials technical team has developed a portfolio of R&D programs that is directed toward producing lower-cost, low-density materials. Having said this, the committee finds it difficult to track progress on a given project on a year-to-year basis. Each year the PNGV materials technical team selects a different group of projects on which to report. For example, last year magnesium and titanium projects were discussed, but this year they are barely mentioned. An alternate reporting approach is needed. Recommendations Recommendation. Because affordability is a key requirement of the 2004 production-prototype vehicles, the committee believes that more attention should be paid to the design and manufacturing techniques being worked on by the American Iron and Steel Institute in the Ultralight Steel Auto Body Advanced Vehicle Concept project. These techniques should be applied to aluminum-intensive vehicles, as well as hybrid-material body construction. More broadly, the committee urges a systematic, critical examination of the prospects to achieve cost goals for all key vehicle subsystems and components. Recommendation. Given the difficulty of tracking progress on more than 30 materials R&D projects, PNGV should devise a new reporting approach that measures progress against objectives. VEHICLE SAFETY The economic feasibility of PNGV-class vehicles will depend on meeting or exceeding consumer expectations in vehicle function, comfort, and safety. The Goal 3 safety requirement in the 1995 PNGV Program Plan was to meet all Federal Motor Vehicle Safety Standards (PNGV, 1995). Since 1995 consumer choice has shifted to heavier vehicles that also are perceived to provide increased safety. The PNGV vehicle is designed to replace a heavier passenger car of the same size in a fleet that is dominated by even heavier cars and light trucks. This environment may be an impediment to selling these vehicles. Also, since 1995 family vehicles have introduced a profusion of safety features beyond those required by the government, some of which may have to be incorporated into

OCR for page 20
Page 64these new vehicles for them to be competitive. In addition to these considerations, the new technologies, introduced by PNGV-type cars, will create new failure modes and safety concerns that will have to be considered. Status and Progress The PNGV concept cars were designed to meet all existing standards with the exception of FMVSS 111, which requires outside rear-view mirrors (CFR, 2000). To reduce aerodynamic drag on these vehicles, outside mirrors were replaced with an indirect viewing system that uses digital cameras and a flat screen display inside the vehicle. These designs offer opportunities to increase the amount of information available to the driver. However, it is not known how well a flat screen display will replace the external mirror presentation to the driver under all conditions. Research by the PNGV is planned during 2001 to assist the National Highway Traffic Safety Administration (NHTSA) in evaluating human factors associated with alternatives to exterior mirrors. The results will provide NHTSA research information to assist in a possible upgrade to FMVSS 111. The 1995 PNGV Program Plan calls for NHTSA involvement to “help assure that any vehicles offered for sale possess structural integrity, include occupant protection systems, and do not compromise safety levels” (PNGV, 1995). In December 1999 the Safety Working Group was established to involve the PNGV industry partners and NHTSA more actively in safety issues. The working group has identified and prioritized five safety research needs by PNGV-class vehicles. The working group has recognized the need to address safety concerns that extend beyond the goal of meeting the minimum FMVSS. These issues involve the safety of PNGV-related designs that may not be addressed by present standards. For example, in general, customers associate lighter vehicles with decreased safety. This may impede the sales of PNGV-type vehicles. It has been well documented that smaller, lighter vehicles provide less occupant crash protection (DOT, 1997). The PNGV-type vehicle will use lighter materials and new construction techniques to reduce weight but maintain the vehicle size. Consequently, the effect on occupant safety may be mitigated due to increased crush space available. This important issue is being addressed by analyzing field accident data and is funded by the industry partners. The study will examine the relationship between safety and vehicle size, independent of vehicle weight. The absence of PNGV-type vehicles in the database makes it more difficult to separate the effects of vehicle size from vehicle weight. A second study of vehicle size and weight that uses computer modeling of vehicle structures in crashes has been defined but not funded. In its sixth report the committee recognized this research need and recommended that it be accelerated (NRC, 2000).

OCR for page 20
Page 65 Other as yet unfunded safety research projects identified by the Safety Working Group include (1) mathematical models for joints in integrated materials structures; (2) HEV power trains (power electronics, energy storage); and (3) alternative fuels. Changes in Environment The PNGV safety goals developed in 1995 apparently did not anticipate the changes that have subsequently occurred in consumer attitudes toward vehicle weight and safety. Most consumers (and some physicists) equate increases in vehicle weight with increases in safety. In today's market a lightweight vehicle like the PNGV-type vehicle may be unacceptable to safety-conscious consumers. However, the overall safety of a vehicle is not governed totally by the vehicle's weight. Improved crash-avoidance features, higher levels of crash protection, and increased vehicle size can offset the weight disadvantage. The marketability of the PNGV-type vehicle will depend on demonstrating to consumers that they are not trading safety for fuel economy. To date, the PNGV project has not adequately addressed this issue. In addition, since 1995 a large number of new safety features beyond those required by federal standards have been introduced. These features were supported by a competitive market rather than government mandates. Examples of market-driven safety features on today's family vehicles include anti-lock brakes; automatic traction control; automatic stability control; low-tire-pressure warning; and many new occupant restraint devices. By the time PNGV-type vehicles enter the marketplace, the level of vehicle safety technology will be quite different from the 1994 family sedans that were the baseline for PNGV function and safety, and this will have to be reflected in the ultimate designs. As noted above, consumer choice has shifted during recent years to heavier vehicles that are also perceived to provide increased safety, and marketing for PNGV-type vehicles will have to deal with this perception. In addition, the introduction of new technology has occasionally created new safety concerns, such as unintended injuries and fatalities from air bag deployments in unusual circumstances. These concerns include not only the crashworthiness of new designs and materials but also new safety issues in flammability, fire and explosion, toxicity, sources of heat and chemical burns, and electrical shock. Defining the safety issues associated with the new technologies for PNGV-type vehicles will be a priority of the recently established Safety Working Group. The committee believes that funding for this work will have to be increased to adequately address these issues. Also, research support from the government appears to be nonexistent in this area, which should be corrected.

OCR for page 20
Page 66 Recommendations Recommendation. Because of the unresolved issues related to safety, the safety research in support of PNGV should receive a higher priority. Critical projects identified by the Safety Working Group should be accelerated and funded appropriately. More effective research support should be provided by the U.S. Department of Transportation. Recommendation. Any initial low-volume production fleet vehicles should be monitored for early signs of any safety issues. FUEL ISSUES All the power system options under evaluation in the PNGV program appear to require fuel modifications that would have significant impacts on the petroleum industry, as pointed out previously (NRC, 2000). Mindful that fuels compatible with efficient engine technologies must be widely available to produce meaningful reductions in fuel use, the PNGV has continued to investigate the interactions of power systems and fuels. There have been important additional testing programs involving auto and petroleum companies, including the Ad Hoc Diesel Fuel Research Program, the Coordinating Research Council (CRC) Advanced Vehicle, Fuel, Lubricant (AVFL) Committee, and the Advanced Petroleum-Based Fuels (APBF) Program, as well as efforts through the CARB Fuel Cell Program, and EUCAR/USCAR Cooperative Fuels Research (see Chapter 2 section, “ Internal Combustion Reciprocating Engines ”). The primary power plant options under consideration in the PNGV are the CIDI engine in the HEV configuration and fuel cells; the fuel implications of each of these are discussed below. Fuels for CIDI Engines The CIDI engine continues to be potentially attractive because of its high efficiency, but emissions of PM and NO x are of concern, particularly in view of the EPA Tier 2 standards. PNGV is pursuing options for reducing engine-out emissions, as well as after-treatment methods. Fuel composition influences both of these approaches. Sulfur content is probably the most important fuel parameter, affecting both engine-out emissions (primarily particulates) and the performance of after-treatment systems to reduce either NO x or particulates. In the past year priority in the program has been on identifying systems with the potential to meet the Tier 2 standards and, while available data indicate that any sulfur is detrimental, the PNGV has not determined just how low the sulfur content must be to ensure long-

OCR for page 20
Page 67term operation. Results show that reducing sulfur improves the effectiveness of SCR systems for NO x control. Work on sulfur traps has continued and indicates that such traps might make it feasible to use sulfur-sensitive after-treatment with sulfur-containing fuels. Upstream trapping of sulfur made it possible to achieve 90 percent reduction of NO x , using a NO x adsorber, which is very sensitive to sulfur. However, available data from Toyota indicate that the trap regeneration temperature will be on the order of 650°C for periods of 10 minutes or more, and even then satisfactory regeneration was obtained only with a feed sulfur content of 8 ppm (Toyota, 2000). Much more work on this concept is required to provide relief from the feed sulfur restriction. Sulfur traps would require periodic maintenance with a frequency depending on the fuel sulfur level, and this would raise the issue of ensuring consumer participation in such maintenance. Lower-sulfur commercial fuels will be available by June 1, 2006, because of the EPA (Federal Register, 2001) final rule requiring refiners to produce diesel fuel with a maximum sulfur content of 15 ppm. Based on answers to committee questions, the PNGV expects that, to ensure 15 ppm maximum sulfur, the average sulfur content will be 10 ppm (PNGV, 2001). A recent study (DOE, 2000a) indicates that less than 10 ppm sulfur may be required to achieve “minimum acceptable effectiveness” of CIDI engine emission control devices. Nevertheless, in discussions with the committee, PNGV members expressed the view that an average sulfur content of 10 ppm and a maximum level of 15 ppm would be marginally acceptable. If current efforts fail to develop systems that can tolerate the level of sulfur mandated by EPA, a major program re-assessment will be required. As addressed in the section “ Internal Combustion Reciprocating Engines ” in Chapter 2, fuel composition can impact the production of regulated emissions, most notably PM and NO x . It has been concluded that, even under the most favorable conditions, the reduction of in-cylinder emissions is not sufficient to preclude extensive exhaust-gas after-treatment. In addition to the assessment of fuel composition impacts on regulated emissions, PNGV also investigated the impact of fuel composition on potential health-impacting toxic chemicals. One program completed in 2000–2001 was directed toward chemical characterization of engine-out diesel emissions, including compounds that might have toxicological effects, 4 using advanced fuels. This study showed that use of a Fischer Tropsch “diesel fuel” low in aromatics, or a mixture of an oxygenate and a low-sulfur petroleum fuel, significantly reduced the emissions of hydrocarbons, particulate matter, polyaromatic hydrocarbons (PAHs), and aldehydes, when compared with emissions using conventional diesel fuel. In addition, these fuels produced the lowest overall “toxic” gas and PAH exhaust emissions. Thus, avail- 4Emissions of 4 potentially toxic gaseous emissions, 11 gaseous PAH compounds, and 17 particulate-soluble organic-phase PAH compounds were measured.

OCR for page 20
Page 68able data indicate that gross changes in fuel composition (i.e., reducing aromatics content or adding oxygenates) can affect engine-out emissions in significant ways. The PNGV told the committee that a fuel cetane number of at least 45 will probably be required, and this will restrict the aromatics content to some extent. The committee notes that reducing sulfur to 15 ppm could lead to a reduction in aromatics content and an increase in cetane number since the hydro-treating process used to reduce sulfur will also directionally saturate aromatics. No definitive data on the extent of this effect are available to the committee. Another program determined the contribution of lubricating oil to the emissions of PM from an advanced diesel engine. Lubricants studied included synthetic and mineral-oil-based materials. CARB diesel and a mixture of an oxygenate and low-sulfur petroleum were used as fuels. These lubricants had a smaller effect on particulate emission than the fuels studied. However the lubricant contribution could become important at very low exhaust emission levels. The supply of aqueous urea would quickly become an important issue if urea SCR were to become the preferred NO x -control strategy. While this is not a fuels issue, it is mentioned here since fueling stations would be one potential means of providing this urea. Pursuing this possibility would require the cooperation of the petroleum industry. The committee believes that the most important fuel issue is the sulfur content required to meet long-term emission certification. If the preferred power systems were to require sulfur levels lower than those mandated by EPA or other compositional changes, additional government action would be needed to coordinate the efforts of the auto and petroleum industries to ensure timely availability of commercial fuel. The committee believes that a very important issue is the maximum sulfur level required to meet the long-term emission certification. Fuels for Fuel Cells While some work has been initiated on the direct methanol fuel cell, the PNGV has focused primarily on the hydrogen fuel cell with a gasoline fuel processor to provide the hydrogen. The program also includes some work on a flexible-fuel processor as well as distributed hydrogen generation at service stations, combined with onboard storage of hydrogen. This program is developing technologies that could be applied to different fuels, but in applications the fuel processor would be optimized for one fuel only. The reforming catalyst that converts gasoline to hydrogen is sensitive to sulfur. While progress has been made to increase the sulfur tolerance of this catalyst, it is not clear what sulfur level will be acceptable. In addition, reforming is more difficult with more aromatic gasolines. The systems under development might require a reduction of sulfur and aromatics content; in fact, the required aromatics content might be low enough to require a different petroleum fuel, such

OCR for page 20
Page 69as naphtha, as the committee pointed out last year in its sixth report (NRC, 2000). A presentation on the California Fuel Cell Partnership indicated to the committee a lack of optimism for the ultimate success of the gasoline (or modified gasoline) reformer (Wallace, 2001). Because of the issues with gasoline and the fact that methanol is much easier to reform to hydrogen, reforming of methanol or other oxygenates continues to be in the program. A multiyear program (DOE, 2000b) to address these issues and to explore the use of reformulated diesel, methanol, ethanol, and Fischer-Tropsch liquids in fuel-flexible fuel processors is under way. While the committee feels that additional efforts are needed to identify fuel and reformer combinations that are feasible, it is not optimistic about the future for fuel-flexible processors, in view of the added complexity introduced by the requirement for multiple fuels. The program is comprehensive in scope. As results are obtained, it will be important to narrow the scope quickly, using decision points that are included in the program plan. The study should include an update of the cost and availability outlook for methanol and other oxygenates based on an updated analysis of the long-term supply and cost of natural gas and from Fisher-Tropsch fuels. As pointed out by the committee last year (NRC, 2000), the capital cost for hydrogen generation at large, central facilities followed by widescale distribution is roughly $3,500 to $6,700 per vehicle. 5 Last year the committee recommended consideration of distributed hydrogen generation at service stations, and DOE has included this in its program mentioned above. The committee notes that reforming of gasoline or methanol is not a clear winner to reduce carbon dioxide emissions and increase overall efficiency of fuel use. As options for off-vehicle hydrogen generation are evaluated, the effects of each approach on life-cycle efficiency and carbon dioxide emissions should be kept in focus. The DOE program also includes work on chemical storage systems, such as metal hydrides, which have the potential to reduce onboard storage volume or storage pressure, or, with additional storage and associated weight, increase the vehicle's driving range. In view of the difficulties with supplying hydrogen for the hydrogen fuel cell, it is significant that DaimlerChrysler announced in 2000 its first test vehicle, a go-cart equipped with a 3-kW direct methanol fuel cell (with no reformer) (Muller, 2000). The vehicle is reported to have a top speed of 22 mph, and the company has already built a 6-kW, 60-V version. This represents an important step along the way toward development of fuel-cell-powered vehicles that do not require hydrogen. The company estimates that direct-methanol fuel cell vehicles could be available in about 10 years. 5In comparison, the onboard gasoline processor would cost $4,350 per vehicle based on PNGV targets (a 50-kW system at a cost of $300/kW), 29 percent of which is for the fuel processor, and would eliminate the need for a hydrogen distribution system.

OCR for page 20
Page 70 As pointed out last year, methanol is not widely available at the present time, but it could be distributed through existing service stations with modifications to the supply and distribution system, for example, to maintain product segregation, manage water, and replace any materials not compatible with methanol. Issues related to its corrosive properties and potential public health effects would require further investigation, and time would be required for fuel suppliers to make the facility modifications necessary to supply an additional fuel. In addition, new plants to manufacture methanol could be required depending on the extent of commercial use of this option, and methanol cost would be affected by natural gas cost. On the other hand, because methanol is used to make methyl tertiarybutyl ether (MTBE), if the use of MTBE in gasoline decreases, more methanol could become available without new plants. It is clear from the foregoing discussion that the hydrogen required by fuel cells could be generated from gasoline, another petroleum liquid fuel, methanol, or natural gas. Some options call for onboard generation, and others, for generation in centralized facilities. Given the program goals of increasing efficiency and reducing fuel consumption, it will be important to make all-inclusive, “well to wheels” analyses (Weiss et al., 2000) to account for all factors associated with providing hydrogen. The present DOE Hydrogen Program has a total of 77 projects that deal with hydrogen production, storage, and use for both stationary source applications and vehicular use (DOE, 2001). Recommendations Recommendation. High priority should be given to determining what fuel sulfur level will permit the preferred compression-ignition direct-injection (CIDI) engine and its after-treatment system to meet all regulatory and warranty requirements. An enhanced cooperative effort between the auto and petroleum industries should be undertaken to ensure that the fuels needed commercially will be available on a timely basis. Recommendation. Given the breadth of the multiyear Fuels for Fuel Cells R&D Program, the go/no-go decision points should be closely followed to facilitate identification and timely development of the preferred options.