The dream of clean, usable energy needs to reflect practical reality.
Published in Nature of July 10, 2003
By Paul M. Grant
In a speech directed mainly at the woes of the US economy, the looming conflict with Iraq, and the AIDS crisis in Africa, President Bush's State of the Union address earlier this year also contained a figurative hydrogen bomb, a metaphor descriptive of the surprise experienced by many of us in the energy community. The President said "Tonight I'm proposing $1.2 billion in research funding so that America can lead the world in developing clean, hydrogen-powered automobiles. A single chemical reaction between hydrogen and oxygen generates energy, which can be used to power a car -- producing only water, not exhaust fumes. With a new national commitment, our scientists and engineers will overcome obstacles to taking these cars from laboratory to showroom, so that the first car driven by a child born today could be powered by hydrogen, and pollution-free."
The White House later revealed the core program to be a five-year, $720 million initiative designated "FreedomFUEL," aimed at developing technology to support a national hydrogen production and delivery infrastructure, directed principally toward the replacement of petroleum as individual vehicular fuel. The Bush Administration deserves much praise for championing US determination to aggressively pursue realization of what is surely the ultimate 'clean' fuel. We'll soon see if Congress agrees -- as the old adage goes, "the President proposes, but the Congress disposes".
The key enabling technology for hydrogen-driven cars and trucks will probably be fuel cells, already under development for many years, and beginning to show some signs of reaching maturity in cost and performance. So let's concede our wildest dreams will come true the day finally comes when your newborn drives an economical fuel-cell powered SUV with enough on-board hydrogen storage to rove 500 km without having to frequently refill at any one of always available and nearby hydro-stations. Where will the hydrogen come from? After all, this is fuel you can't mine or drill for.
The "FreedomFUEL Fact Sheet" issued by the White House and within which one might hope to find answers to this rather important question, gave only one clue: "Hydrogen can be produced from abundant domestic resources including natural gas, coal, biomass, and even water" (my italics). Well, these resources will have to be staggeringly abundant indeed.
For hydrogen to completely supplant petroleum for transportation, an enormous outlay in capital plant and significant land area set-asides will be necessary for its production. Even with a Herculean effort on the model of the Apollo Program, the hydrogen economy will arrive slowly, and will require vast infrastructure investment over a long period of time that may not be driven by the market economy, but more likely may evolve from a series of government energy policy directives and legislative action that will mandate a blend of petroleum, methane, ethanol and hydrogen on the pathway to an eventual all-hydrogen transportation supply. This interim will certainly also see a migration across several generations of internal-combustion engine technology, such as petroleum-electric hybrid vehicles.
People in the United States consume nearly 20 million barrels of petroleum daily, around 12 million being used for surface transport. A major portion of this is wasted, as an internal combustion engine has an efficiency (the energy that actually goes to turning the shaft) of 20 - 30%. 'Only' 3 million barrels, therefore, are 'usefully' consumed. Because a fuel cell is not a heat engine, and, theoretically at least, escapes the bottleneck of the Second Law of Thermodynamics governing internal combustion motors, only enough homemade hydrogen is needed to offset the energy content of these three million barrels of oil, imported or otherwise, with the remaining, wasted nine million now saved.
Future fuel cells may be able to convert about 80% of the Gibbs free energy released by combining hydrogen with oxygen to make water into electrical energy (at present, this factor is around 50%). Folded in must also be the losses in both electricity conversion and electric motor efficiency, around 20%, to 'shaft energy' to move the car. Thus the overall efficiency is 64%, much better than can be obtained from gasoline or diesel engines. Hence, we need to generate daily some 230,000 tonnes of hydrogen, enough in liquid form to fill 2,200 space shuttle booster rockets, or, as a gas, to lift a total of 13,000 Hindenburg dirigibles.
Hydrogen is not a 'primal' energy source, that is, unlike fossil fuels or uranium, we have to expend more energy to extract the hydrogen from whatever it is chemically combined with than we can recover in its end use. There are a number of ways we can extract hydrogen from methane and even steam passed over coal. However, for simplicity, and to bypass issues of carbon and carbon dioxide sequestration, let's assume we obtain the hydrogen by 'splitting' water with electricity - electrolysis. Although this isn't the cheapest industrial approach to 'make' hydrogen at present, it will nonetheless illustrate the enormous production scale that must be addressed: we would have to add about 400 gigawatts continuously available new electric power generation, just about doubling the present US national average power capacity of a mere 440 gigawatts! The number of new power plants that would have to be built, based on presently available technologies, to produce 400 gigawatts is, say, 800 500-megawatt natural gas-fired combined cycle units, or 500 800-megawatt coal-fired units, 200 Hoover Dams (2 gigawatts each), or 100 French-type nuclear clusters (approximately four reactors, one gigawatt each)
The average capital cost of electric power plant construction is $1,000 per kilowatt (with considerable variance), resulting in a new investment of at least 400 billion dollars (one-twentieth of current US GDP), not including storage and delivery costs, to enable a complete transformation to current surface transport running on hydrogen instead of petroleum. A daunting prospect, but not impossible. To get the daily hydrogen ration of 230,000 tonnes, slightly over 2 million tonnes of water will be required. Nevertheless, when this amount of water is expelled as 'exhaust' it will be recycled to the environment in several days, unlike carbon dioxide.
A popular justification for moving from petroleum to hydrogen, often invoked by US political pundits, is "freedom from dependence on Middle-East oil." Yet, according to the latest figures published by the US Energy Information Agency, only 14% of US oil imports comes from the Gulf states: Europe (30%) and Japan (75%) are much more dependent on this region's resources (source US EIA).
According to EPRI studies, about 97% of the hydrogen produced in the United States derives from the thermocatalytic 'splitting' of natural gas or refinery gases, or 'coal gasification', the reaction of water (steam) with carbon to yield hydrogen and carbon monoxide. However, the heat to effect both processes must be produced somehow, usually through fossil fuel combustion or, presently under consideration, heat from a nuclear reaction, neither of which can escape the eventual carbon sequestration challenge either in the primary fuel consumed, and/or the residue from extracting hydrogen from a carbonaceous host. Moreover, using fossil combustion to derive hydrogen electrolytically for transport would yield carbon emissions likely to offset most of the benefits of saving 9 million barrels of petroleum per day. We are thus left with only water as a source of hydrogen and carbon-free generation of electricity or heat to get it out.
What's it going to take?
So, what about using renewable energy resources to do this job? Sweden's Norsk Hydro currently build the largest electrolyzers targeting the European hydroelectric industry, and BC Hydro and Hydro Quebec have done studies on hydrogen production using off-peak hydroelectricity. BC Hydro recently completed a pilot project, which is now making hydrogen for use in fuel-cell powered buses in Vancouver. But I doubt whether you'll see even one major new North American hydroplant built in the near (or far) future expressly to generate hydrogen, let alone 200 Boulder Dam equivalents!
Advocates of hydrogen have often pointed to the promise held out by wind and photovoltaic solar technologies. It's expected that these technologies will approach power densities of about 10 watts per square meter (1 watt per square foot) and 100 watts per square meter (10 watts per square foot), respectively, when the wind is blowing hardest and the Sun shining the brightest. Imposing an annual duty factor of 20% on solar (the Sun doesn't shine all the time) and 30% on wind (the wind doesn't blow all the time), the Earth surface area required to produce 400 gigawatts to manufacture and store sufficient transportation hydrogen over one year turns out to be 130,000 square km for wind and 20,000 square km for solar technologies, the former is about the size of New York State, the latter about half the size of Denmark.
Capital investment costs for these two renewable alternatives are by no means well quantified, but currently can run twice 'conventional'. Equally impressive would be the ecological and environmental havoc wrought, even if the generation facilities were dispersed. However, if residential and industrial roof areas, land already in use, were vigorously exploited for photovoltaic solar generation, much less ecological damage would ensue.
Another option, extracting energy from biomass through combustion, is attractive in principle because the carbon emitted is recycled through future plant growth. Presumably, one could create electrical energy or thermal energy to split water from firing biomass. Right now, approximately 15% of the land area of the United States is under cultivation, mostly for food. Some estimates put the annual biomass energy equivalent of these crops at roughly 23 million gigawatt-hours, or about 2,600 gigawatts continuous power production. (R. A. Ristinen and J. J. Kraushaar, "Energy and the Environment," (John Wiley & Sons, 1999)) To add an additional 400 gigawatts of electrical power to produce hydrogen for transport would require another 3% to undergo cultivation, an area about the size of Nevada.
Some photochemical reactions produce hydrogen directly from biomass, usually algae. One approach under study uses green algae bioengineered to release H2, in conjunction with fermented effluent feedback to sustain the algae population. The sole feedstock is water, with no net carbon emission. The predicted electrical equivalent power production of hydrogen is 6.4 megawatts per square mile, requiring an area of about 70 square miles, or "three-quarters Nevada equivalent" for 400-gigawatt continuous production.
No room at the top - or bottom
Whenever you do something 'massive' to ecology -- like pumping gigatons of CO2 into the atmosphere, or try to find somewhere else to put it, or propose million-acre wind or solar farms -- the environmental consequences are uncertain at best and disastrous at worst. With respect to carbon recovery or sequestration, an example of such uncertainty (or pending disaster) is brought out the experiment on iron fertilization of the oceans near Antarctica to capture carbon, which resulted in huge algae blooms followed by emission of methyl bromide, an ozone poison.
The lesson is that if you have a way to make both electricity and hydrogen without making carbon dioxide, it's probably wise to use it. And we do. Its science is sound and solved. Its technology is mature and safe. We know how to bound its problems, and prevent repetition of past accidents as well as forestall future ones. *** OK, I've added a few sentences at the end of this paragraph to make that clear. I disagree that there is worldwide mistrust, not that there are many pockets of radical, and in my opinion, scientifically irrational fear. If radiation release is a concern, then there should be concern about the recently published EC finding that oil and gas drilling in the North Sea releases "naturally" into the marine environment 30 times that of Europe's entire nuclear fleet We do know how to prevent future accidents. By virtue of the physics involved in its design, a meltdown simply cannot happen in a high-temperature gas-cooled reactor of the design under consideration in the US, Russia, Japan and South Africa. In addition, the ALWRs coming on line have so many years of safety experience behind them that the risk of a partial meltdown of the kind that occurred at Three Mile Island is vanishingly small, and it did occur, it would be contained, just as it was at Three Mile Island.
Chernobyl was the worst accident that could possibly happen so we know what the worst is already and it was the only power plant accident that took lives (there's a grisly Republican joke in the US to the effect that "more people lost their lives in Ted Kennedy's car than at Three Mile Island"). There will never be another CBRK design built again, known to be flawed from the very first. As far as I know, the populations of the US, Canada, France, Japan, and many others are still allowed to vote, and they vote to keep nuclear power, if only implicitly. In the last week or so, the Swiss explicitly voted to keep their nuclear power plants. As I write these comments, the United States Senate is in the middle of a debate on a portion of the FY04 energy omnibus bill. The bill contains a provision for the Department of Energy to lend 50%, lend, not give, of the cost of construction of new nuclear power plants in the US. Not one Senator has attacked this new nuclear initiative on the basis of safety not one. The Senators from the two most environmentally active states in the Union, New Hampshire and Oregon, have spoken against the upfront loan guarantee and suggest a rate subsidy like wind and solar get. But nuclear doesn't need it. In California, the cheapest electricity comes from our two nuke plants and throughout the rest of the country, nuclear power is as cheap as coal about half the time. In the fall of 2000, after the spike in electricity prices, a Field Poll revealed 72% of Californian would approve a third nuclear power plant in the state. Tomorrow the debate continues on the Senate floor. I could go on and on about the nuclear issue, but I believe I've made my point for issues surrounding hydrogen sufficiently in this piece and I hope you guys agree! It's called nuclear fission power. Its fuel costs are fixed and supply assured for at least 500 to 1,000 years by which time, maybe, we'll have neutron-free 'clean' fusion...maybe. My opinion is likely to shock many readers accustomed to the belief that nuclear power is inherently dangerous and forever unsafe. Be that as it may. I only invite them to re-examine the facts and come to their own conclusions. Toward that end, may I recommend they start by reading "The Need for Nuclear Power," by Richard Rhodes and Dennis Beller, published in the January-February 2000 issue of the journal Foreign Affairs.
On the west coast of Japan, at Kashiwazaki Kariwa, Tokyo Electric Power Corporation, has spent 20 years building perhaps the most modern nuclear power complex in the world. Its eight reactor units generate of total of 8 gigawatts of electric power, available 90% of the time, on a site that occupies slightly less than 4 square km). This area includes a visitor center, machine shops and on-site 'spent' fuel storage. More than one quarter of the plant area is green space, sand dunes and native growth are left undisturbed. The power density is still an amazing 1,800 watts per square meter. Thus, the 400 gigawatts worth of electricity needed to extract hydrogen from water to power US surface transport could be cumulatively generated on land only occupying 90 square miles, an area less than 10 miles on edge, containing more than enough room for electrolyzers and storage. There is much hand-wringing over the issue of nuclear waste disposal Sorry, again I have to strenuously disagree. I think this view reflects more the British attitude than elsewhere. In the US, the establishment of Yucca Mountain was a long overdrawn process, never strongly opposed by many Nevadans, especially those near the sight employed by its construction. Now it's a done deal, even though their will be the inevitable demonstrations when "spent" fuel begins to be transported to the site remember, in the US we move nuclear weapons around all the time, including processing waste from their manufacturing, but for some reason the power plant waste has become the target of extremists. Moreover, our nuclear Navy is welcomed in every port in the world except in New Zealand, of course. Nobody wants it? Well, Russia does and the Japanese believe their $26 B facility under construction at Rokkasho will prove quite profitable, not only for their own waste disposal and reprocessing, but that from Europe as well but what is easier to keep under control for any length of time? Massive geological and oceanic deposits of carbon dioxide, or highly compacted vitrified material after reprocessing that mainly consists of short-lived isotopes?
Any form of highly concentrated potential energy, from dynamite to deuterium, provides many opportunities for humans to misbehave and make mischief. Nonetheless, I believe an eventual resurgence of nuclear power is absolutely necessary for the continuing industrialization of world society with minimal environmental impact and eco-invasion, one in which hydrogen will supplant fossil sources. FreedomCar and FreedomFuel cannot be had without FreedomNukes - and FreedomNukes cannot be had in a world order that continues to permit the unrestricted proliferation of nuclear weapons. International laws and institutions must be established that control and vigorously enforce use of actinide materials for peaceful purposes from minehead, through recovery and breeding, to eventual disposal, and prevent diversion to rogue weapons programs. Only then can be realized the vision the fathers of the atomic age foresaw and desired, a world where 'atoms for peace' would prevail, creating a clean energy source independent of any geographically accidental richness of fossil reserves.
Paul Grant is a Science Fellow at the Electric Power Research Institute and an IBM Research Staff Member Emeritus.