July-September 1997 Issue


Options for Using China's Coal in Cars: A Life-Cycle Assessment

hinese policymakers face a bewildering array of options as they consider using their vast coal deposits to fuel vehicles. Coal can be used to produce gasoline, methanol, or electricity. But each of those fuels has its advantages and disadvantages. MIT researchers and their Chinese and American collaborators have now estimated how shifting from petroleum to each coal-based fuel would change consumer costs, environmental impacts, and energy efficiency. Their life-cycle analyses included the complete history of each type of fuel and vehicle, from extraction of raw materials through production, use, and disposal. As expected, changing to any of the coal-based fuels would reduce efficiency, increase carbon dioxide emissions, and cost more. However, the added cost is a relatively small fraction of the total cost of owning a vehicle--except in the case of electricity, which is prohibitively expensive unless batteries improve dramatically. Effects on emissions other than carbon dioxide vary from fuel to fuel. This type of comprehensive assessment represents a methodology that can also be used to compare technology options in other fields.

China is the world's largest coal producer. Yet in recent years this coal-rich nation has had to import petroleum, largely to fuel its rapidly growing transportation sector. Concerned about energy security, Chinese policymakers are now considering ways to use coal in place of petroleum, particularly as an automotive fuel. Processes exist for converting coal into gasoline, methanol, and electricity, any of which can run a car. The challenge is to identify the option that best meets the policymakers' three main goals: to minimize costs to the consumer, to minimize environmental damage, and to use the nation's domestic resources most efficiently. Unfortunately, no single option is the obvious winner on all fronts.

In 1996, the Chinese government initiated a study to clarify and evaluate the costs and benefits associated with the various options. Collaborating in the study were more than forty technical experts from MIT, Ford Motor Company, Tsinghua University, and several agencies of the government of the Peoples Republic of China. MIT's part of the work was coordinated by the Energy Laboratory and led by Malcolm A. Weiss, senior research staff member at the Laboratory. The Ford team was led by Walter M. Kreucher, manager, Advanced Environmental and Fuels Engineering, and the Chinese team by Qiming Zhu, professor at Tsinghua University.

The goal of the study was not to make specific recommendations but to provide Chinese leaders with sound scientific information on the potential consequences of different approaches to using coal as an automotive fuel. The approach ultimately adopted by Chinese policymakers will, of course, take into account other societal concerns not examined in the study.

Switching from petroleum-derived gasoline to a coal-based fuel is not a simple matter. It may affect the entire history of the fuel, the car, and their use together. Therefore, the researchers performed their assessment using a technique known as life-cycle analysis. The schematic below shows the elements considered in the analysis. The top boxes follow the fate of the fuel-extraction from the ground, processing into a form useful in an automobile (including electricity), and delivery to the vehicle. The bottom boxes trace the vehicle's history, including extraction of the raw materials and manufacture of a vehicle suited to the fuel being assessed. The last two boxes represent the use of the fuel in the car and the disposal of the car at the end of its useful life.

To perform a life-cycle analysis of a given fuel "scenario," the researchers had to determine the economic cost, energy efficiency, and environmental impacts of each of those steps (each box in the diagram). But it is not possible to produce absolute quantitative values for all scenarios that everyone will agree on. Consider just one variable: the cost of manufacturing cars that run on methanol rather than on gasoline. Two automotive experts are unlikely to agree on the exact cost of making a gasoline car, much less the more unfamiliar methanol car. Debate will arise about even broad assumptions. (Will the cars be made in China? Will the engines be imported from Japan?) But experts can agree on roughly how much it costs to manufacture a gasoline car. Moreover, they can agree on generally what changes are needed to make a methanol car instead. And they can calculate how much those changes will cost or save.

The researchers therefore performed their assessment as a comparative analysis. They assumed as their base case a Ford Escort fueled by petroleum-derived gasoline, equipped with an emission-control catalyst, and operated for 12 years. (Catalysts are not used in China now because unleaded gasoline is not generally available. Both are expected to come into widespread use by the year 2000.) They then estimated the performance measures of interest: the cost to the consumer, the environmental impacts, and the energy efficiency associated with each step in the diagram. For each coal-based scenario they determined how each of those three measures would change from the base case, first within each step and then cumulatively. They thus ranked the options relative to the base case and relative to one another on each of the performance measures.

The analysis focused on fuel options of particular interest to the Chinese participants. It considered two petroleum-based fuels--gasoline (the base case) and diesel--and three coal-based fuels--gasoline, electricity, and methanol. For each coal-based option, they made certain assumptions.

• Gasoline: Coal can be converted into gasoline by indirect liquefaction (the solid coal is converted to a gas and the gas to a liquid) or by direct liquefaction (the solid coal is converted directly to a liquid). The analysis assumed indirect liquefaction using the Fischer-Tropsch technique. Direct liquefaction is also of interest and will be analyzed in a future supplement to the study.
  
  
• Electricity: The impacts of choosing electricity depend critically on the future cost and efficiency of batteries for electric vehicles. Those factors are highly uncertain. Therefore, the evaluation included battery technologies ranging from today's conventional lead-acid battery to a highly efficient battery representing the long-term goal of the US Advanced Battery Consortium. The main focus was on a medium-term technology, the nickel metal hydride battery (NiMH).
  
  
• Methanol: The vehicle assumed for this scenario was a "flexible-fuel" vehicle that operates on 85% methanol and 15% gasoline. Dedicated methanol vehicles would be feasible only when methanol is widely available throughout China. The main fuel-processing technique assumed was advanced technology involving coal gasification, then methanol synthesis. Several other minor potential sources of methanol were also considered, the most important being methanol made from methane gas that can be drawn from cracks in coal beds. While "coal-bed gas" could produce relatively little methanol nationwide, it could be an important supplementary source of methanol in certain regions.

All the data used in the study are intended to reflect conditions in China, which often differ from corresponding conditions in the United States. Chinese participants in the study provided data on costs, air emissions, and energy efficiency for each fuel scenario. Ford developed data for vehicle manufacturing and operation. The analyses were performed using a variety of existing spreadsheet-based models and computer simulations, specially tailored for this life-cycle analysis.

Chinese policymakers are concerned about energy efficiency not only because they do not want to waste resources but also--perhaps more importantly--because energy efficiency can affect both economic cost and environmental impacts. The most direct effect of energy efficiency is on emissions of carbon dioxide (CO₂): the more efficient a fuel scenario, the lower its emissions of CO₂.

To determine the energy efficiency of each fuel scenario, the analysis tracked energy input versus energy output in each step of the life-cycle diagram. Converting coal to automotive fuels is inefficient, so the lifetime energy efficiencies of all the coal-based scenarios are lower than those of the petroleum scenarios (both gasoline and diesel). Electricity does best of the coal-based scenarios because of the higher efficiency of operating electric vehicles. But that advantage is offset by the inefficiency of generating electricity from coal.

In contrast to energy efficiency, cost plays a direct and important role in deciding the attractiveness of the options. The analysis focused on cost to the consumer. What would switching to coal mean for the owner's out-of-pocket expenses during the lifetime of the vehicle? The expenses included no government-imposed taxes or fees, and costs incurred in different years were not discounted.

A consumer's total cost consists of three components: buying the car, buying fuel, and operating the car. (Operating costs include insurance, maintenance, license and registration fees, parking, and finance charges on car loans.) The table below shows how each of those components differs from the base case for each scenario, yielding a life-cycle cost differential to the consumer.

As shown in the table, the consumer's expenditure on fuel varies dramatically from scenario to scenario. However, the cost of buying fuel is a small part of the total lifetime cost associated with having a car. As shown in the pie chart below, buying fuel is only about 10% of the total cost, with the remainder split about equally between buying the car and operating it.

The total incremental costs for the non-electricity coal-based scenarios do not differ greatly--from one another or from the gasoline base case. The distributions of costs among the three components differ, but the bottom line is about the same: having a (non-electric) coal-fueled car costs $2000-3000 more than having a conventional petroleum-fueled car. That added cost, spread over the 12-year lifetime of the car, is relatively small: the total cost of a conventional petroleum vehicle is estimated at $40,000, so the variation is only about 5%.

The outcome is quite different with the electric option. Assuming an electric car with an NiMH battery, the consumer pays $700 less for fuel than in the base case. But that saving is offset by dramatically higher costs to buy and operate the vehicle. Overall, the consumer will pay close to $15,000 more to own and drive an electric car--a substantial increase, even spread over 12 years. And despite the high cost, the consumer will still get a car with the power and range limitations that make electric vehicles unmarketable today. Analyses of the other battery options show that the cost of switching to electricity is comparable to that of the other coal-based options only if batteries meet the Battery Consortium's long-term goals for cost and storage capacity.

Comparing the environmental effects of the different scenarios is more complex. Again, the analysis looked at emissions during the processing of the fuel and the manufacturing of the vehicle as well as "on-road" vehicle emissions, based on a standard urban driving cycle defined by the US Environmental Protection Agency. The table below presents detailed results for key pollutants. (The values are absolute rather than comparative to show which emission levels and changes are large enough to be of concern.)

A few generalizations can be made. Coal-bed gas is the winner in most areas--not a surprise, as starting with methane gas is always cleaner than starting with either coal or oil. As already stated, all of the coal-based options generate more CO₂ than the petroleum-based options do. Thus, the switch from petroleum to any of the coal-based fuels would further increase China's already substantial contribution to greenhouse gas emissions.

As for the other emissions (most of whose impacts are regional or local), the outcome varies from pollutant to pollutant, with no one scenario being the across-the-board winner. For example, methanol from coal generates lower sulfur dioxide and nitrogen oxides emissions than do the petroleum-to-gasoline base case and the coal-based options, but it produces lots of carbon monoxide. Electric power beats all the other options (coal and petroleum alike) on carbon monoxide and hydrocarbons, but it generates the most sulfur dioxide. Petroleum-based diesel does well in most categories but has high emissions of nitrogen oxides and particulate matter. Only gasoline from coal is relatively straightforward: it is among the worst on almost all emissions. With its high cost and high CO₂ emissions, it ranks low in the overall assessment.

The researchers did not expect to find a best option. They intended only to quantify the trade-offs for the Chinese policymakers. They do, however, emphasize that the findings depend on many assumptions. For example, if the cost of oil was higher than they assumed or if the cost of manufacturing vehicles for the coal-based fuels was lower, the economic viability of all the coal-based options would improve. They also caution that other factors must be examined. For example, for each scenario they estimated water requirements but not whether those requirements could be met. Land use has yet to be addressed, and a detailed examination of the technical feasibility of each option is needed. Nevertheless, the results thus far should provide a useful foundation for Chinese policymaking.

Another important contribution is the analytical methodology itself. The computer models developed are simple, flexible, and transparent so that users can easily make clear and consistent comparisons of the impacts of different technology options and can readily see the results of changing underlying assumptions such as the cost of oil. This type of life-cycle analysis can also be used in other areas in which the technology options are many, their impacts are uncertain, and their costs and benefits are variable.

This research at MIT was supported by the Ford Motor Company. Further information can be found in references.