E-Lab October - December 2000

Today's diesel vehicle is the subject of much debate. It offers high fuel efficiency and therefore low emissions of carbon dioxide (CO2). But it produces large quantities of other pollutants, notably NOx and particulate matter (PM). Can diesel vehicles be made to meet regulatory limits that are now being set? And are those limits sufficiently strict-or are they in some cases too strict?

In July 2000, the Energy Laboratory hosted a two-day symposium to consider the diesel debate. The focus was on scientific aspects of diesel engines and fuel, including what is known, what uncertainties remain, and what research is needed. The symposium was part of a series of annual conferences begun in 1993. Their purpose is to inform decisionmakers about scientific aspects of important air pollution issues and to improve communication between the scientific and regulatory communities. Invited participants at last summer's symposium included almost 90 scientists, regulators, and industry and public interest representatives.

A recurring theme throughout the symposium was the widespread and varied use of diesels throughout our economy. Diesel engines power freight trucks, farm equipment, logging trucks, school buses, liquid carriers, transit buses, and fire trucks. "Off-road" uses include construction and mining equipment, heavy rail, and ships. Speakers noted that more than 80% of US freight is transported by diesel trucks. Participants agreed that diesels are critical to our economy and lifestyle-and they will not disappear soon. Diesels offer a combination of characteristics not available from gasoline, compressed natural gas, liquid petroleum gas, or methanol engines. They can deliver power at a heavier load and over a wider range of loads than other engines can, and they are more fuel efficient than other options are (see the chart below). They seldom break down; they last for hundreds of thousands of miles; they have the lowest lifetime costs of any powertrain; and the infrastructure for distributing diesel fuel is well established. Finally, their CO2 emissions are low. Indeed, some experts view the diesel as the best option for increasing the fuel efficiency and reducing the CO2 emissions of our motor vehicle fleet in the near term.

energy effiecency of various vehicle/fuel combinations

A comparison of the energy efficiency of various vehicle and fuel combinations, presented at MIT's symposium on diesels by David J. Rickeard of ExxonMobil Petroleum & Chemical (Belgium). The data indicate the percentage of the energy in the source crude oil or natural gas that produces useful work at the wheels of the vehicle. The "missing" energy includes energy used to produce and refine the crude and to distribute the fuel to service stations, plus energy lost in the vehicle. The diesel is a clear winner, but maintaining its high fuel efficiency while reducing its emissions of particulate matter and nitrogen oxides is a challenge.

However, today's diesels have a potentially fatal flaw. They are major contributors to ambient concentrations of both PM and NOx, pollutants that have adverse health and environmental impacts, including the formation of ozone. Diesels are dirty in part because-until recently-relatively little attention was given to cleaning them up. According to one symposium speaker, in the 1950s diesels were deemed small contributors to air pollution; so for decades regulatory and engineering attention focused on improving gasoline engines. As a result, today's diesel fleet is substantially dirtier than today's gasoline fleet. And total diesel fuel consumption is increasing by about 5% per year-about twice as fast as gasoline consumption is. Today, about a quarter of the total fuel used in vehicles and off-road equipment is diesel.

Diesels are now in the emissions-control spotlight. The "Phase 2" rules issued by the US Environmental Protection Agency (EPA) in June 2000 aim for a 90% reduction in PM, NOx, and hydrocarbon emissions in new vehicles. The PM standard takes effect in 2007 and includes finer particles than previously limited. The NOx standard will be phased in from 2007 to 2010, and a sulfur cap of 15 ppm for highway diesel fuel will be phased in starting in 2006. Regulations address both on-road and off-road uses, and emissions limits apply to the vehicle and fuel in combination. Speakers from Europe and Japan noted that the new US regulations are somewhat stricter than those in other countries, though the gap is narrowing. Many European countries typically place more emphasis on limiting emissions of CO2 and other greenhouse gases, and tax incentives are used to encourage diesel use and fuel cleanup.

Symposium participants disagreed considerably about the timing and appropriateness of the evolving US diesel regulations. Some speakers cited studies indicating the mortality and morbidity rates associated with diesel exhaust. But others questioned the science used in assessing diesel's health impacts. EPA's rating of diesel exhaust as a "probable" carcinogen in 1989 was based on animal studies. However, many experts now believe that animal studies are not always appropriate for assessing human risk. And epidemiological studies performed to date are far from definitive. Most have involved problems identifying actual exposures and excluding the impacts of other "confounding" factors such as smoking and nutrition patterns.

Nevertheless, most participants agreed that considerable evidence does suggest that diesel exhaust causes adverse health effects. Recognizing that our understanding of such health issues will always be imperfect, they recommended that we act on what we know so far and do what we can to reduce exposures, especially as the use of diesels is increasing.

Unfortunately, there is no easy way to reduce diesel emissions. According to engine experts, modifying the diesel engine's design or operation so that critical pollutants do not form is difficult. In the typical diesel engine, there are trade-offs between fuel economy and the formation of PM and NOx. Adjusting for maximum fuel economy increases NOx, and vice versa. Increasing temperatures to reduce particulate levels also increases the formation of NOx. And certain engine modifications that eliminate large sooty particles increase the formation of small particles-the ones believed to be most harmful to human health. One speaker noted that proposed regulations should be carefully assessed to foresee and prevent such unintended effects.

Given those difficulties, symposium participants agreed that meeting the regulations for 2007 will require the use of aftertreatment devices that remove pollutants after they leave the engine and before they arrive at the exhaust pipe. Filters are available that can capture PM, but preventing small particles from condensing in the exhaust after release remains a challenge. Capturing NOx is even more difficult. An NOx-absorbing catalyst and urea-injection devices are now being developed but are not yet available.

A key problem with existing and potential control technologies is that they are poisoned by sulfur present in diesel fuel. (Crude oil naturally contains sulfur compounds, some of which remain in diesel fuel if special steps are not taken during the refining process.) Sulfur is also a culprit in forming small particles in diesel exhaust streams. Discussions of the new regulations limiting fuel sulfur occurred throughout the symposium, and the viewpoints expressed diverged considerably. Vehicle engine manufacturers argued that EPA's sulfur cap of 15 ppm is not low enough. The Engine Manufacturers Association (EMA) deemed a cap of 5 ppm necessary to "enable" NOx aftertreatment technologies and to ensure their durability. (US on-road fuel now generally contains 350 ppm sulfur.) While sulfur traps might be developed for individual vehicles, removing the sulfur at the refinery to whatever low levels are set seems preferable. Unless better fuel becomes available, the EMA contends that new diesel engines will effectively be choked out of the market.

Fuel industry representatives protested that achieving that level of desulfurization would require enormous investments in refinery modification, and companies would get no payback. If the limit is too stringent, refineries may switch to other products such as home heating oil, where sulfur limits do not apply; or they may simply go out of the refining business. A diesel fuel shortage could occur. Eventually prices would go up and new supply would come on line, but in the meantime there may not be adequate supplies for our growing diesel fleet. Fuel industry representatives also observed that some refineries will wait until the last minute to invest in upgrades in hopes that EPA may issue waivers or delay the regulation.

A more optimistic outlook was offered by a representative from Corning, one of the first companies to make catalytic converters. He observed that setting extremely stringent standards has frequently led to "technology forcing." In the past, automotive companies have deemed new emissions standards technically and economically infeasible. Yet they subsequently came up with a better gasoline engine, making development of a new aftertreatment device unnecessary. "Raising the bar" by setting "unrealistic" standards can thus be a highly effective regulatory approach.

Another recurring concern was how EPA should determine whether a vehicle manufacturer or fuel supplier is complying with relevant regulations. For example, does a single truck plucked from an assembly line have to meet the emissions limit? Does the limit apply to an average of ten trucks tested once a month at each facility? Or does it apply to a vehicle after the consumer has been driving it for six months or six years? And how does one design an inspection and maintenance program that is effective and enforceable? Even with a given engine design, tuning it to reduce one pollutant may increase another. And after inspection, some drivers may be tempted to undo certain pollution-control devices or adjustments in order to increase fuel efficiency and save money.

Finally, increased regulatory attention must focus on controlling emissions from diesel vehicles already in use. Several speakers noted that retrofit programs can significantly reduce PM, hydrocarbons, carbon monoxide, and toxic emissions and potentially NOx. According to one speaker, PM filters have already been retrofitted on 10,000 vehicles in Europe. The outcome has been efficient reduction of PM and decreases in other toxins, with minimal operating problems.

The list of research needs identified at the symposium is considerable. We need better epidemiological studies focusing not only on cancer but also on non-cancer health effects such as asthma and premature death. We need to monitor people to see if health problems decrease as current regulations begin to reduce diesel emissions. Ultimately, we need a better understanding of the biological mechanism whereby diesel exhaust or its constituents affect people's health. Such knowledge would clarify what aspects of particulates make them harmful-their composition, number, total mass, size, or coatings of other substances on their surfaces. The result could be more narrowly focused regulations that provide the same protection at a lower cost.

We also need a better understanding of diesel emissions (especially particulates), including how they form, what happens to them in the environment, and what people are actually exposed to. Computer models now being developed promise better predictions of emissions from on-road diesel sources, and dynamometer tests are providing critical data on emission rates under various operating conditions. Better information is also needed about off-road emissions-a challenge because off-road diesel uses are diverse, few emissions measurements are available, and data on equipment populations, use, and activity are limited. Ambient measurements of pollutants, especially ultrafine particulates, are also inadequate.

Finally, we need to continue the type of open discussion and debate that occurred at the summer symposium. Diesel regulation is an extremely complicated subject that spans technology and policy. Affected stakeholders must have an opportunity to hear the concerns, capabilities, and research needs of one another. Regulations may have to be issued before the science gives a clear answer. And even the best possible scientific understanding may not define a "right" regulatory strategy. There will always be a trade-off between cost and the level of control. Agreeing on the appropriate level of regulation may be a greater challenge than actually engineering the vehicles to meet it.