October-December 1999 Issue


Fuels and Vehicles for 2020:
How the New Technologies Measure Up

n Energy Laboratory assessment of likely new technologies for passenger cars in 2020 has come up with no overall winners in the race for cars with lower greenhouse-gas and other emissions. Initial results from the assessment show that the gains from continued work on conventional fuels and vehicles are so great that emerging technologies like the fuel cell will have trouble competing. By 2020, conventional vehicles will be twice as efficient, half as polluting, and cost little more. New technologies will provide somewhat greater efficiency and emissions gains but at a much higher cost. With little or no private benefit to purchasers, the new technologies are unlikely to succeed in the marketplace unless government action or public pressure calls for major reductions in greenhouse-gas emissions. The MIT researchers compared the technologies using data from various sources, adjusted so that key assumptions were consistent. Calculations included energy use and emissions not only from driving the vehicle but also from making and delivering the fuel--a change that dramatically reduced the attractiveness of some technologies. The assessment examined how vehicle purchasers, fuel manufacturers, vehicle distributors, and all other major "stakeholders" would trade off dozens of characteristics of different technologies, from cost and safety to convenience and familiarity. The analysis confirmed that what is unimportant to one stakeholder group can be a real "show-stopper" for another. Identifying critical showstoppers and developing strategies to overcome them is the researchers' ultimate goal. They are now completing their initial assessments and are beginning to examine other technologies, including biomass fuels and battery vehicles, and other types of vehicles, including light and heavy trucks.

Demand for road transportation is constantly growing, especially in developing countries with expanding economies. Yet the environmental consequences of even today's level of demand are significant. In the United States, on-road vehicles emit huge amounts of greenhouse gases, including nearly a third of all domestic carbon dioxide (CO₂) emissions and about a fifth of global CO₂ emissions. Vehicles also emit large quantities of hydrocarbons, nitrogen oxides, carbon monoxide, and particulates, all of which pose local and regional environmental and health problems. The best way to reduce those emissions--and to meet new, stricter regulatory limits--may be by using new fuel and vehicle technologies. Many promising technologies are now being developed, and studies that examine and compare their costs and benefits abound. However, in many cases the data and assumptions on which those studies are based are inconsistent, incomplete, or unclear. Knowing which of the technologies are most likely to succeed is therefore difficult.

For the past year, Dr. Malcolm A. Weiss, Professor John B. Heywood, Dr. Elisabeth M. Drake, Dr. Andreas Schafer, Felix AuYeung, and Darian Unger have been assessing promising new road technologies using a methodology designed to be comprehensive, systematic, and easy to understand. The assessment covers various combinations of fuel and vehicle technologies over their entire life cycles; and it considers all characteristics of each combination from the perspectives of the organizations and people whose decisions influence its success or failure.

The figure below shows the system considered. The shaded boxes indicate the major stakeholders--groups of people with similar interests and values. The left side of the figure shows the fuel cycle. It starts at the energy source (the oil in the ground, the trees in the woods), moves through the fuel manufacturer (the refinery, the manufacturing plant), and ends with the fuel distributor, who delivers the fuel to the vehicle user. The right side of the figure shows the vehicle cycle. It starts with raw materials and parts (iron ore, scrap metal), moves through the vehicle manufacturer, and ends with the vehicle distributor, who sells cars to the vehicle purchaser and maintains and repairs cars for the vehicle user. The vehicle user operates the car throughout its lifetime and then disposes of it, recycling or scrapping parts as appropriate. Overseeing the entire system is the government, which has concerns about environmental, legal, infrastructure, and other issues.

The focus of the assessment is the year 2020--far enough ahead for new technologies to be developed and introduced but not so far that identifying the potential competing technologies is impossible. For the initial analysis, the researchers selected fuel and engine technologies that can be developed by 2020 with "reasonable diligence and effort" and no need for technical "breakthroughs." Each new technology was assumed to make up a few percent of all the cars being sold in 2020--enough to benefit from most economies of scale. The initial assessment thus focuses on where we want to be in 2020 and ignores (for now) the possible difficulties encountered in getting there.

The technologies considered fall into three categories: fuels, power trains, and vehicle bodies. (Power trains include combinations of fuel systems, propulsion systems, and drive trains, which are used to transfer work from the engine to the wheels.) The specific technologies are listed in the table below. Bio-mass fuels and battery vehicles were not included in the initial analysis because they are less likely to be developed and widely commercialized by 2020. Seven new technology options were defined, each employing a selected fuel and power train in a lightweight, "advanced" vehicle body that uses a new design and new materials, including abundant aluminum. In all cases, the vehicle's capacity, range, and performance were comparable to those of today's typical family car.

As a basis for comparison, the researchers defined a car similar to a Toyota Camry with a gasoline-fueled internal combustion engine (ICE), not as it exists today but as it is likely to have evolved by 2020. The cumulative effect of "evolutionary" changes in the fuel, power train, and vehicle body yields major advances similar to the advances of the past twenty years or so. The 2020 "Camry" has the same capacity, range, and performance as the 1996 Camry; but it weighs 22% less, uses half as much fuel, and emits half as much CO₂, all for a cost increase of just 5%, or $1000 (in constant dollars).

For each technology option, the researchers estimated energy efficiency, CO₂ emissions, costs, safety and health effects, reliability, convenience, and other characteristics that can make or break a road technology. The full fuel and vehicle cycles were included (though energy use and emissions during the vehicle's manufacture were not accounted for in this initial assessment). Data came from recently published reports, follow-up interviews with authors, and selected unpublished studies. Rather than averaging data from different studies, the researchers used their experience and judgment to select data that appeared both reasonable and based on careful analysis. They used models developed at MIT and at ETH-Zürich to define the characteristics of vehicles.

The researchers then compared each of the seven new technology options with the 2020 Camry, one characteristic at a time. Evaluations for a given characteristic could range from much more favorable to not materially different to much less favorable. The outcomes of such comparisons can vary substantially among the stakeholder groups, and a negative evaluation by just one group can kill a new technology. Therefore, the researchers made up a large grid, or "template," for each stakeholder group showing how that group evaluates more than thirty characteristics of each new technology option, as compared to the base case 2020 Camry. A supporting template explains every evaluation in a few words.

Given the mass of information and opinions, the researchers do not attempt to summarize the collected data or to aggregate it in order to merit-rate or rank the technologies. However, they do make several generalizations that they believe will hold up as they continue to gather additional data. For example, the assessment shows clearly that comparisons based on published studies must be treated cautiously. Studies are often based on assumptions and constraints that are inconsistent or not explicitly defined. As a result, outcomes can vary widely. For example, estimates of the amount of energy consumed in the fuel cycle varied from study to study by more than a factor of two.

Significantly, some studies leave out the fuel cycle entirely. Yet reliable data show that the energy consumed in making and distributing fuel can be a large fraction of the energy consumed when using the fuel to run the vehicle. For example, several studies conclude that a provider of methanol or hydrogen made from natural gas uses well over half a megajoule of energy to make each megajoule of the fuel ultimately delivered to the customer. Leaving out the fuel cycle can therefore alter the efficiency rankings of various technologies. One study compares the fuel efficiency of five technologies. When the fuel cycle is excluded, the hydrogen-based fuel cell is most efficient. When the fuel cycle is included, a diesel ICE is best. The methanol-based fuel cell is second best when the fuel cycle is excluded but last when fuel cycle is included. Thus, comparisons of technologies can be misleading if the technology assessments do not use a "well-to-wheels" approach that includes both the fuel and vehicle cycles.

The MIT assessment also demonstrates the importance of evaluating technologies from the perspectives of different stakeholders. Where one stakeholder group sees advantages, another may see enormous barriers--and another may be totally indifferent. As an example, the researchers considered how several stakeholder groups might view methanol-based fuel cell vehicles compared to the 2020 Camry. The government would strongly favor the fuel cell vehicle because it would cut CO₂ emissions in half (per kilometer driven) and reduce other air emissions to negligible values. Fuel suppliers, on the other hand, would need to invest $12-15 billion to supply enough methanol to equal 5% of today's US gasoline energy; and most of the investment would have to be in remote locations abroad, where natural gas is relatively inexpensive. The risk would be enormous. Finally, from the vehicle owner's point of view, there is no incentive to buy such a car. Fuel savings would be trivial. In 2020, as today, the cost of buying fuel (not including taxes) will be a negligible part of the cost of owning a car; so any potential fuel savings will not influence the consumer's technology choice. But buying a fuel cell vehicle would cost over $26,000--far more than the $20,000 Camry.

Indeed, there appears to be no economic reason for US consumers to demand any of the new technologies and therefore no reason for manufacturers to make them or for distributors to offer them. New fuel and vehicle technologies may reduce environmental insults, but the consumer must pay the extra cost while receiving little direct private benefit. The figure below compares three advanced-body 2020 vehicles with the evolved 2020 Camry on the basis of three characteristics: vehicle cost, energy use, and CO₂ emissions. (The tops of the bars are corrugated to imply that these numbers are estimates.) The 2020 advanced-body gasoline ICE vehicle costs a little more than the 2020 Camry but will cut energy use and CO₂ emissions. The gasoline hybrid (a gasoline ICE supplemented by a battery) costs still more but reduces energy use and CO₂ emissions still more. The methanol fuel cell vehicle costs still more; and energy use and CO₂ emissions are not quite as good, although the differences are within the uncertainty of the results. The evolved 2020 Camry is hard to beat, and the incentive to invest in the new technologies simply is not there. Thus, it seems unlikely that these new technologies will enter the market unless the government takes action to provide the needed incentive--or unless purchasers come to believe that the extra cost is worthwhile for the public benefit.

Of all the engine-fuel combinations considered, the researchers found no single technology that proved an across-the-board winner in 2020. The diesel hybrid looks best for high efficiency and low CO₂ emissions, but it is not clear that particulate and nitrogen oxides emissions from diesels can be reduced enough to meet likely future regulatory standards. The hydrogen-based fuel cell--a favorite of many people--actually appears less attractive than the diesel hybrid. Its efficiency and emissions characteristics are comparable, but it would cost more and its adoption would require major infrastructure changes to make compressed hydrogen widely available. However, special opportunities may arise for certain technologies. For example, hydrogen fuel cells may be adopted in urban buses--a situation where a single station for manufacturing and compressing hydrogen could serve a whole fleet of vehicles.

In the longer run--say, 50 years--hydrogen-powered fuel cell or battery vehicles would be attractive goals, but developing those technologies will involve overcoming formidable technical and economic obstacles. The researchers stress that research activities in the near and intermediate term should be consistent with long-term goals. For example, devoting considerable resources to developing electrical drive trains is appropriate. Electrical drive trains are used in hybrid systems, which promise substantial gains in energy use and emissions well before 2020. Likewise, electrical drive trains are used in both fuel cell and battery vehicles. Developing and introducing electrical drive trains would therefore serve both intermediate-term and long-term needs.

The researchers are now revising and completing their data tables and templates through further analyses of the literature and through discussions with various experts. They will consider other emissions and other technologies, including biomass fuels and battery vehicles; and they will examine not just passenger cars but also sport utility vehicles and light and heavy trucks. Based on their systematic analysis of stakeholder reactions and attitudes, the researchers hope they can design strategies for overcoming critical barriers that may otherwise keep promising new fuel and energy technologies from succeeding in the marketplace.

Malcolm A. Weiss is a senior research staff member in the Energy Laboratory. John B. Heywood is the Sun Jae Professor of Mechanical Engineering and director of MIT's Sloan Automotive Laboratory. Elisabeth M. Drake is associate director of the Energy Laboratory. Andreas Schafer is a research associate in the Center for Technology, Policy, and Industrial Development. Felix AuYeung is a master's degree candidate in the Department of Mechanical Engineering. Darian Unger is a PhD candidate in the Technology, Management, and Policy Program. This research was supported by the V. Kann Rasmussen Foundation and by Chevron, Exxon, Mobil, and Norsk Hydro. Publications are forthcoming.