Fuel Cells

A Developing Transportation Technology

By Alan F. Montemayor      image of PDF button


Alan F. Montemayor is a principal engineer working in fuel cell systems within the advanced vehicle technology section of the Engine and Vehicle Research Division. His current projects include introducing fuel cells into a tractor-trailer that will run in Palm Springs, Calif., and procuring, installing and running three fuel cells at Brooks Air Force Base in San Antonio.


Southwest Research Institute's modeling and simulation program has put SwRI in the vanguard of cleaner, more efficient fuel cell technology for transportation applications.

The application of fuel cell technology to heavy-duty trucks is a timely issue in vehicle research and development. Total U.S. truck fuel consumption continues to increase and has surpassed fuel usage by automobiles in the past decade. Applying fuel efficient technology to trucks offers a higher rate of return than applying similar technology to automobiles. Thus, the transportation industry must give careful consideration to improvements in overall fuel efficiency that can result in lower operating costs -- even if it means a small increase in capital costs.

SunLine Services Group, under contract with the U.S. Army Tank-Automotive and Armaments Command (TACOM), has subcontracted with SwRI to develop a heavy-duty truck powered by a fuel cell using hydrogen derived from reformed diesel fuel. SwRI's ability to predict performance efficiency and emissions is one of the key reasons SwRI has been awarded its largest vehicle integration contract to date.

Modeling and simulation work at SwRI grew from development sponsored by internal research. Subsequent sponsorship has been provided by the U.S. Defense Advanced Research Projects Agency (DARPA); U.S. Council for Automotive Research (a consortium comprised of Ford, GM and Daimler-Chrysler) under the title of Partnership for a New Generation of Vehicles Systems Analysis Toolkit; the U.S. Army National Automotive Center in support of the Army's transition to fuel cell power; and, most recently, SwRI has developed a new generation of simulation software, which has been funded in part by Daimler-Chrysler.

The vehicle powertrain simulation program will be available commercially and will model and simulate various vehicle powertrain configurations, including hybrid series, parallel, combination series and parallel, fuel cells and conventional automotive powertrains.

Background

Why haven't fuel cells in stationary or transportation applications -- with the promise of thermodynamic efficiencies of 50 percent -- already obtained widespread market penetration? The answer, according to engineers at SwRI, is largely due to system complexity and fuels storage efficiency, which affect cost, long-term reliability and size. However, the belief that fuel cells can have a positive impact in many power-producing applications is driving ever-increasing research and development in this area.

With the help of a catalyst, fuel cells combine hydrogen and oxygen to produce direct current electricity without combustion. Cells can be combined into modular units that are extremely efficient, theoretically capable of converting up to 70 percent of the fuel's energy into useable electricity under ideal circumstances. The cells require little maintenance and can produce electricity without harmful byproducts, making them suitable for use in residential and environmentally sensitive areas.

Although British scientist Sir William Grove discovered the physical principles behind power generation from fuel cells in 1839, it was not until the space race of the 1960s that investment into research and development paid off with the selection of a fuel cell to power U.S. spacecraft. Since that time, research and development has continued, with the result that the transportation industry is beginning to accept fuel cells as a power production alternative to the internal combustion engine.

The most popular fuel cell in development is the proton exchange membrane or polymer electrolyte membrane (PEM) fuel cell. Another candidate fuel cell receiving considerable development is the solid oxide fuel cell (SOFC). The PEM fuel cell's advantage is that it operates at a lower temperature than the SOFC, which results in a faster response to braking and acceleration. However, the PEM fuel cell is intolerant of carbon monoxide and sulfur in its gaseous hydrogen fuel supply. The choice of fuel cell -- and there are many other types being developed -- is constrained by the system into which it is to be integrated.

PEM fuel cells feed in hydrogen and air and produce electricity directly. Water vapor and heat are the only other byproducts. The hydrogen must be pure because impurities such as sulfur and carbon monoxide can damage the catalyst in the cell. In this diagram, protons (yellow arrows) are produced by the interaction of hydrogen and the catalyst and pass through the permeable membrane (blue). The protons combine with oxygen in the presence of a catalyst and the reaction produces water. Electrons are conducted through the catalyst support structure and flow to an external load.

Systems integration -- modeling and simulation-based activity

Primary objectives in developing fuel cells for the transportation industry are to improve efficiency and to reduce emissions. Fuel cells operating on hydrogen produce electricity, water vapor and heat as end products. If the hydrogen has been "reformed" from other, more complex fuels, the resulting emissions theoretically contain only carbon dioxide and fuel-borne substances such as sulfur compounds.

Because pure oxygen is used only in spacecraft as the oxidation agent and air is used in all other applications, some oxides of nitrogen and other pollutants are produced. This level of oxides of nitrogen is considerably lower than a diesel engine might produce because the maximum temperature produced by the reaction is considerably lower for a fuel cell than for a diesel engine.

A significant problem with using fuel cells in the transportation industry is providing a continuous stream of hydrogen to the fuel cell. Hydrogen must either be stored on the vehicle in a tank or produced through reformation of petroleum-based fuels.

SwRI is working with a variety of companies to develop a reformation process that is commercially viable for use in heavy trucks and other vehicles to preclude the need for a large hydrogen storage tank on vehicles powered by fuel cells.


In the United States, trucks consume more energy than automobiles.This trend will continue according to data from the U.S. Department of Energy. Increasing the efficiency of trucks can significantly reduce overall energy usage and help reduce air pollution.


Fuel reformation

Providing hydrogen to fuel cells through reformation is not an insignificant task.

Natural gas is usually the fuel of choice for stationary reforming applications because it consists of at least 90 percent methane, which produces four hydrogen atoms per carbon atom and because no initial vaporization of the fuel is required before reformation.

Transportation applications, however, require liquid hydrocarbon fuels to be processed before they are provided to the fuel cell. Fuel processing typically begins with vaporizing the fuel and removing sulfur that might otherwise poison the reformation process and fuel cell catalysts. The next step is to produce hydrogen through one of two fuel reformation techniques.

Steam reformation reacts the hydrocarbon fuel with water. If the steam reformer is run with excess water, a water-gas shift produces some additional hydrogen in a secondary reaction. Alternatively, partial oxidation (POX) reformation can be accomplished either with a burner or a catalyst. Partial oxidation reacts fuel with a limited amount of oxygen. Diesel fuels under high temperature and pressure yield hydrogen and carbon monoxide. Steam reformation of diesel fuel can produce approximately twice as many molecules of hydrogen per molecule of fuel as POX (about half of the hydrogen comes from the steam), with the same amount of carbon monoxide produced in either case.

Steam reformation, however, is a heat-absorbing (endothermic) process, while POX is a heat-producing (exothermic) process. Therefore, steam reformation requires a constant supply of additional heat, which must come from the fuel supply, thus reducing the overall efficiency of the steam reformation reaction toward or below that of the POX reaction.

A third promising technique for producing hydrogen, often referred to as autothermal reforming (ATR), combines POX and steam reformation, with the partial oxidation reaction providing heat for steam. SwRI and other industry researchers are evaluating the efficiency of these systems for the heavy-duty fuel cell truck application.


Liquid fuels must be "reformed" into a hydrogen-rich gas used in fuel cells. There are three processes to accomplish the reforming: steam reforming, partial oxidation reforming and autothermal reforming (a combination of steam and partial oxidation). This idealized diagram shows some of the steps that may occur in a steam reforming process (steam injection is not shown on the diagram).


Fuel cell optimization

Development of an optimal design for a fuel cell to power trucks requires an analysis of multiple subsystem configurations with the aim of optimizing the performance of the truck over its expected life cycle.

The powertrain of an optimized system will include fuel cells, one or more electric motors, power converters and possibly batteries or other electrical energy storage devices. Power converters will be needed to convert the fuel cell direct current to the alternating current used by many traction drive motors. Additionally, batteries likely will be used to buffer the output of the fuel cell to optimize its response to braking and acceleration.

Engineers in SwRI's Vehicle Systems Research Department use genetic algorithms to perform an optimization search of all of the powertrain components and component sizes that could be used in the fuel cell-powered truck.

From a purely statistical viewpoint, the problem involves many variables that must be adjusted until an optimum configuration is found. Whereas traditional optimization techniques require closed form equations describing the system with continuous functions for derivation analyses, the use of genetic algorithms can be likened to a structured approach of trial and error. In this approach, the parameters used to describe the system (such as fuel cell maximum power, battery pack maximum current, electric motor maximum speed and torque, etc.) are defined and their values are converted to a binary string of 1s and 0s. These parameters are joined to form a so-called "chromosome." Each chromosome offers a possible solution to the overall optimization problem.

Multiple chromosomes are generated either at random or by expert insight. Each possible solution is used in modeling the fuel cell truck performance and evaluated against the performance criteria that must be established for the system, with each chromosome ranked according to its performance. Next, combinations of chromosomes are "mated" to define the next generation of possible solution sets and their performance is modeled and ranked.

Additionally, each generation of chromosomes receives some specific number of random mutations. It is the injection of random mutations that ensures that the optimum solution of the entire function is found and not merely a local optimum as may result with traditional optimization techniques. SwRI already has applied this technique to optimization involving conventional vehicle powertrain optimization, control system optimization and engine design.


Genetic algorithms mimic the process of evolution in that each new generation (iterative computer simulation) contains mutations that may or may not be beneficial. This powerful technique can find true optimal solutions and avoid local minima or maxima.


Conclusion

Fuel cells are likely to be applied to the transportation industry in the near to midterm future. They likely will be applied as an auxiliary power source for heavy trucks, which idle to provide power for drivers using their trucks as a temporary dwelling. Fuel cells likely will be a viable alternative power source on automobiles, trucks and buses early in this century. The nation's major automakers already have developed prototype vehicles powered by fuel cells. Analysis work performed at SwRI -- which has yielded three patents on fuel cell designs and applications for four more patents on fuel cell designs and sensors -- likely will lead to more efficient fuel cell designs and better control of current systems. No matter how the fuel cell is adapted for future use, SwRI will contribute to its development.

Comments about this article? Contact Montemayor at (210) 522-6940 or amontemayor@swri.org.

Published in the Fall 2001 issue of Technology Today®, published by Southwest Research Institute. For more information, contact Maria Stothoff.

Technics Fall 2001 Technology Today
SwRI Publications SwRI Home