Ethanol plant in Tucumcari, New Mexico produces 3 million gallons of biomass fuel annually from corn or milo; the by-product of which is a valuable cattle feed. The author argues that biomass fuels are a more practical, economic pathway to a cleaner energy future than hydrogen

Practical, Clean Energy For Future Transportation

White paper on feasibility of hydrogen economy vs other energy options

By F. David Doty, PhD

Introduction: There can be no doubt the status quo cannot continue for more than a few decades in automotive transportation, but neither can we ignore the economic and social lessons of the past forty years from other dreams, such as fission, fusion, and orbiting microwave power stations. It is becoming increasingly more clear with each passing year that hydrogen fuel cells will not be practical, and our only viable, long-term option is renewables.

It was encouraging to see Nobel Laureate George Olah (Chemistry) recently point out that a "Methanol Economy" seems much more practical than a "Hydrogen Economy" [1]. In recent years, very efficient methods of dehydrating bio-methanol into bio-ethylene (C2H4) and water have been developed. Ethylene, in turn, can be used to efficiently produce all hydrocarbon fuels and products currently obtained from fossil sources. Thus, bio-gasoline, bio-diesel, bio-kerosene, and bio-methane can be efficiently and safely produced, stored, and transported within our current infrastructure. Direct Methanol Fuel Cells (DMFC), which convert methanol directly into water and CO2 while producing electricity, have advanced to the point that they are now beginning to appear in commercial products such as cellular phones, laptop computers, and some military equipment. Here, the power source only has to be competitive with lithium batteries (which is much easier than competing with diesel engines), but that is still an impressive achievement and a promising beginning.

While there will undoubtedly be some economically viable applications for fuel cells (especially DMFC, but also hydrogen), economically viable solutions for the following fuel-cell challenges in automobiles seem highly unlikely within the next four or five decades.

1. Fuel costs. Current U.S. H2 production is enormous – about 2x1010 kg/yr. Yet, the current pre-tax cost of liquid nitrogen (LH2) in the U.S, delivered in 15,000 gallon (4300 kg) tankers to high-volume customers is about $5/kg. The cost of pressurized H2 for consumers has been in the range of $100/kg (current dollars) for over forty years. Some studies have concluded mini-reformers (if mass produced) at corner filling stations could provide hydrogen at $3.4-4.3/kg, assuming natural gas priced at $5/GJ [2], but other data suggest the price would be four times higher at likely commercial prices for natural gas ($14/GJ) 15 years from now. Moreover, current proven domestic natural gas reserves will last only 10 years [3], and there is increasing pessimism about future domestic discoveries [4]. The price of natural gas has increased by a factor of 10 in the past 30 years and a factor of three in the last six years.

On the other hand, the current U.S. pre-tax cost of diesel for the individual consumer at the local station is about $0.5/kg. Of course, we need 2.8 kg to equal the energy of one kg of H2, but that still leaves an order of magnitude cost advantage for diesel per unit energy. Realistic estimates suggest the pre-tax price of bio-diesel from bio-methanol could ultimately be below $0.6/kg, though it will be quite a while before bio-methanol competes.

2. Fuel-cell engine costs. The only possible type of hydrogen fuel cell for automotive applications is the proton exchange membrane fuel cell (PEMFC), (also called PEFC, polymer electrolyte fuel cell), as all other types are either far too massive or have unacceptably short lifetimes. The cost of PEMFC engines (fuel cells, power conditioning, electric motors, etc.) is often reported to be in the range of 3,000-8,000 U.S. dollars per kilowatt – 100 times that of the common diesel engine, or forty times that of the advanced diesel, which will soon exceed 58% LHV efficiency [5]. Inspection of the sales and financial data from the largest current producer of PEMFCs for non-mobile use (Plug Power) suggests current costs of PEMFCs (not including real R&D) are actually in the range of $15,000-$30,000/kW [6]. It is worth noting that polymer FCs have been in use and development for over forty years, and costs have not yet begun to drop significantly – notwithstanding many assertions to the contrary that use artificial costs from heavily subsidized projects or cite costs of massive, stationary fuel cells that are unsuitable for vehicles. Over the past ten years, Ballard Power has furnished 85% of all vehicle fuel-cell engines world wide. By some methods, one could conclude that the manufacturing cost, not including true R&D, of their latest fuel-cell engines has been well over $1M each.

3. Fuel storage mass, volume, and safety. Safety-approved low-cost compressed gas cylinders currently achieve 1.5% H2 storage by mass at 34 MPa (5000 psi). A $15,000 carbon-fiber-wrapped fuel tank achieving 11% H2 storage seems impractical for the small private car, and liquid hydrogen doesn't keep long. Moderate-priced tanks of aircraft-grade aluminum alloy are likely to be competitively priced and achieve 6% storage, and titanium-alloy tanks may eventually permit 9% storage at practical prices. Hence, the practical energy density of hydrogen, after including some extra structure needed for protection in the event of a collision, may be 15% that of diesel. At 5000 psi, the volumetric energy density is only 10% that of diesel. Just the mechanical energy (forget the chemical energy) stored in the hydrogen tank may be 5 times that of a 50-calliber artillery shell, and the impact strength of light-weight tanks is not high. The risks associated with carrying this mechanical bomb around are probably two orders of magnitude greater than we are accustomed to accepting in our gasoline-powered cars today. The huge mass and volume penalties associated with practical H2 storage seem likely to keep the mileage of hydrogen-powered automobiles (of acceptable range, acceleration, cost, and cargo capacity) relatively low, and the risks associated with either pressurized gas or cryogenic on-board storage are completely unacceptable for personal vehicles [7].

4. Fossil CO2 release. The only economically viable sources of H2 in the U.S. (and most other countries) are natural gas and coal. The nearly adiabatic partial-oxidation/reformation/shift reactions use 3 kg of natural gas (90% CH4) to produce 1 kg of H2 plus 9.5 kg of CO2 [7]. Then, over 3 kg of coal must be burned (releasing another 10 kg of CO2) to generate the 10 kWhr (36 MJ) needed to purify and liquefy 1 kg of H2, which will usually be required for efficient distribution for at least the next two decades. The energy efficiency in producing LH2 is under 50%. (This number has not budged in 15 years and will not in the next 50. We're near Carnot limits.) The energy content of 1 kg of H2 is equivalent to 2.8 kg (1 gal.) of diesel, which contains only 2.3 kg of carbon.

At 70 miles per gallon, the advanced fossil-diesel hybrid achieves 7 miles per kilogram of total CO2, while the bio-diesel vehicle could achieve infinite miles/kg of fossil CO2. The production-grade hydrogen fuel-cell automobile won't get very good mileage because of the order-of-magnitude penalty in fuel energy density. At 40 miles per kilogram of hydrogen, it achieves about 2 miles/kg of total CO2. Hence, when miles/kg of fossil CO2 release ("fossil mileage") is more fairly calculated, the total CO2 generated per mile by a hydrogen vehicle is likely to be 3.5 times that of a comparable fossil-diesel-powered hybrid vehicle. Interestingly, the hydrogen path recently advocated by the DOE (distributed, local reformation) is simply not compatible with CO2 sequestration, and it would extend our intense dependence on fossil fuels as long as possible.

5. Infrastructure Development. Some have estimated that the development of an efficient hydrogen distribution infrastructure would cost only $300B, though other estimates are four times that amount. However, the fuel distribution infrastructure is very small compared to the required manufacturing infrastructure and vehicle replacement costs.

Conclusions and Discussion. It is interesting to note that just four years ago, the DOE expected 10,000 fuel-cell-powered vehicles to be in use by now. But today, there are about 100, and they now project there will be "fleets" six years from now. At least four companies have each invested over $350M into trying to develop PEMFC manufacturing, but with little real success. Undoubtedly, if the DOE invests $2.1B (as planned) over the next 7 years, more demonstration vehicles (at perhaps $400K each) will be on the road, but that really doesn't get us much closer to a practical solution for private transportation for the future.

How could such a well-intentioned scientific endeavor as clean energy stray so far from reality? Perhaps like this. For three decades, it has been pretty clear to many concerned scientists that our world's oil supplies would be largely depleted within their lifetime and major changes would be forced upon us. Moreover, it was common knowledge in the mid-1970's that hydrogen fuel cells for more than a decade had achieved up to three times the efficiency of gasoline engines, and our very cheap natural gas resources (hence, hydrogen) seemed inexhaustible. It was reasonable to expect that major progress could be made in reducing the manufacturing cost of fuel cells, so the notion of a hydrogen economy, ultimately based on nuclear power plants, seemed to have economic merit. Undoubtedly, the fact that the basic reaction was conceptually based on fifth-grade chemistry also contributed to its popularity.

It would then take more than two decades (until mid-2001) for five major realities to begin to be appreciated by a few scientists. First of all, order-of-magnitude cost reductions in manufacturing processes are almost never realized after the fourth decade of development (recall, hydrogen fuel cells were used in the space program in the early 1960's). Secondly, the efficiency of the gasoline and diesel engines would steadily improve. (Diesel engines now exceed 50% and will soon exceed 58% efficiency. Gasoline engines have exceeded 30% for two decades and will soon exceed 38% [5].) Thirdly, global warming would have to be seriously addressed much sooner than most had expected. Fourthly, the price of natural gas (hence, hydrogen) in the U.S. would skyrocket early in the 21st century as demand began to exceed domestic supply. Finally, nuclear power would not likely be accepted again.

If we can believe the reports that large, low-efficiency (31%) 75 kW PEMFCs for non-mobile use can finally be produced (by UTC) for under $3000/kW [7], then it seems reasonable to project that the break-even cost for vehicle-qualified PEMFC systems might be ~$7000/kW ($500K for a 100 hp engine) at a similar (small) production scale. History suggests a factor-of-six reduction in cost is about the limit that can be expected in going from small-scale production to large-scale production. The catch is that high-volume production is unlikely ever to materialize, even if a fully adequate fueling infrastructure were completed (at a cost of over $500B) by 2030, given the (A) short time (perhaps 10 years) that hydrogen could be available at a tolerable price, (B) limited driving range of hydrogen vehicles, (C) serious global warming issue with hydrogen, (D) serious safety issues with hydrogen fuel, (E) short FC life, (F) poor tolerance of PEMFCs to normal environmental conditions, and (G) competition from liquid biofuels.

While the DOE/EIA are still predicting no increases in the price of fossil fuels over the next 15 years, it is worth noting that they have been forced to revise their price estimates upward (for gas, oil, and coal) every year for the past six years, and there is currently no question that another upward revision will be required in their next annual projection [8]. (Most experts outside the U.S. have done a much better job.) We must take seriously the fact that the world will soon be running out of cheap, fossil oil; and if we don't prepare by developing viable alternatives, the economic consequences will be severe. However, focusing all our efforts on a single dream that seems less and less likely to be of any practical benefit is worse than doing nothing at all because of the false hope it engenders.

We simply cannot ignore global warming, and we must accept the fact that hydrogen will do nothing to reduce CO2 emissions. Rather, it will increase CO2 emissions – not just for a few decades, but until we're ready to build hundreds of new nuclear power plants, and then only for several more decades until we run out of viable uranium ore [8, 9]. It's time we start putting some serious money into real options for renewable energy to address global warming and our future transportation needs. It's surprising how many people don't understand that bio-fuels needn't contribute to global warming over the long run. Cellulosic bio-ethanol (from poplars and switch grass) could be economically competitive within six years [10, 11]. Bio-diesel and high-alcohols (butanol-pentanol-hexanol) from waste also have considerable promise [12, 13].

Bio-methanol is coming from behind and has a steep uphill fight to compete with methanol from natural gas, but ultimately it may be our best long-range option. Efficient, moderate-sized methanol plants have been built to produce about 83,000 tonnes (28 million gallons) of bio-methanol per year, which require about 12,000 to 40,000 acres for sustainable harvesting, depending on the climate [10, 14]. A large part of the operating cost with biomass can be the transportation of the feedstock to the processing plant, even for dedicated biomass land use. For dual-use lands, the transportation costs can go way up, so smaller bio-mass plants could be more cost effective. Hence, we may see a future with thousands of bio-methanol, ethanol, and high-alcohol plants scattered throughout rural and suburban areas. (Obviously, there are huge economic and social implications involved here – especially, enormous rural job creation and the long-awaited recovery in the chemical industry.) Some of the product would be used locally, but most of the bio-methanol would be trucked or piped to biofuel refineries to be converted to diesel, gasoline, and jet fuel [13].

An even more promising source for biodiesel may be high-oil algae. Estimates suggest 100 billion gallons of biodiesel could be produced from 8000 square miles of salt-water ponds in relatively small farms in the desert, and the economics appear quite promising. Perhaps enough biodiesel to replace all petroleum transportation fuels in the U.S. could be grown in less than 10% of the area of the Sonora desert in the Southwest [15]. The primary feedstock would be hundreds of millions of tons of CO2 from coal power plants.

And what about the longer range outlook, seventy years from now, will bio-mass be adequate? Undoubtedly, wind, solar, and clean coal with carbon sequestration will need to play a larger role, and we'll see electric vehicles with more advanced batteries finally competing. But the primary intermediary in transportation will still not be hydrogen. Hydrogen simply isn't the best way to distribute energy or to power vehicles. It will never compete in the transportation arena. For now, our first priority should be practical solutions for the next thirty years.

Finally, I suspect majority-opinion recommendations from an independent, blue-ribbon review panel of our current energy programs would include: rapidly ramp up renewables funding, begin winding down hydrogen funding, and begin major investments into diesel hybrids, bio-methanol, high-oil algae, and cellulosic ethanol.


1.George Olah, "The Methanol Economy", C&E News, September 22, 2003. See also, http://www.usc.edu/assets/college/faculty/profiles/60.html.


Julian Darley, Post Carbon Institute, http://www.postcarbon.org/ , 2004.

C. J. Campbell and Anders Sivertsson, "Updating The Depletion Model", ASPO Meeting, Paris, 2003, http://www.peakoil.net/.

James J. Eberhardt, "Fuels of the Future for Cars and Trucks", presented at DEER, 2002. See http://www.osti.gov/fcvt/deer2002/deer2002wkshp.html.


The Hydrogen Economy: Opportunities, Costs, Barriers, and R&D Needs, Committee on Alternatives and Strategies for Future Hydrogen Production and Use, National Academy of Engineering, National Academies Press, http://www.nap.edu/books/0309091632/html/ 2004.

F. David Doty, http://www.dotynmr.com/PDF/Doty_H2Price.pdf , 2004.

J. W. Storm van Leeuwen and P. Smith, http://beheer.oprit.rug.nl/deenen/ , 2003.

L. B. Lave, W. M. Griffin, and H. MacLean, "The Ethanol Answer to Carbon Emissions", Issues in Sci. and Tech., Winter, 2001. http://www.nap.edu/issues/18.2/lave.html.

http://www.novozymes.com/cgi-bin/bvisapi.dll/biotimes/one_article.jsp?id=27092&lang=en .

http://www.generalbiomass.com/renew2.htm , http://www.biodiesel.org/ .

F. David Doty, http://www.dotynmr.com/PDF/Doty_FutureFuels.pdf , 2004.

Mikael Ohlstrom, T. Makinen, J. Laurikko, and R. Pipatti, "New concepts for biofuels in transportation: Biomass-based methanol ...", VTT Energy, Finland, 2001. http://www.inf.vtt.fi/pdf/tiedotteet/2001/T2074.pdf .

Michael Briggs, http://www.unh.edu/p2/biodiesel/article_alge.html , 2004.

Times Article Viewed: 7988
Published: 13-Mar-2004


blog comments powered by Disqus