July 20, 2003
Hydrogen Not Good Short, Medium Term Form Of Fuel

UC Berkeley academics throw cold water on the prospects for hydrogen.

In a paper appearing in the July 18 issue of Science magazine, Alex Farrell, assistant professor of energy and resources at UC Berkeley, and David Keith, associate professor of engineering and public policy at Carnegie Mellon University, present various short- and long-term strategies that they say would achieve the same results as switching from gasoline-powered vehicles to hydrogen cars.

"Hydrogen cars are a poor short-term strategy, and it's not even clear that they are a good idea in the long term," said Farrell. "Because the prospects for hydrogen cars are so uncertain, we need to think carefully before we invest all this money and all this public effort in one area."

Farrell and Keith compared the costs of developing fuel cell vehicles to the costs of other strategies for achieving the same environmental and economic goals.

"There are three reasons you might think hydrogen would be a good thing to use as a transportation fuel - it can reduce air pollution, slow global climate change and reduce dependence on oil imports - but for each one there is something else you could do that would probably work better, work faster and be cheaper," Farrell said.

The biggest problem with hydrogen as a means to reduce pollution is that it has to be produced from another energy source. But the most cost competitive energy sources are all forms of fossil fuels. The production of the hydrogen is not 100% efficient and producing it from fossil fuels produces pollution. The transportation and storage of the hydrogen also use substantial amounts of energy.

Hydrogen is also more difficult to store and transport and takes up much more space than liquid hydrocarbon fuels. It is not the only conceivable approach to pursue for reducing net pollution from vehicles for the purpose of reducing green house gasses. Another approach to reduce the net production of green house gasses would be to develop a light-driven chemical process that would fix carbon out of atmospheric carbon dioxide to make hydrocarbon fuels. Or if cheap photovoltaic solar cells could be developed then another approach would be to use electricity from solar cells to drive the chemical process to fix carbon from carbon dioxide. Effectively gasoline would be generated from solar power. Then the gasoline could be burned in cars. This artificial carbon cycle would eliminate the net addition of carbon dioxide gas to the atmosphere.

Back in 2000 the MIT Sloan Automotive Laboratory report On The Road: A life-cycle analysis of new automobile technologies by Malcolm A. Weiss, John B. Heywood, Elisabeth M. Drake, Andreas Schafer, and Felix F. AuYeung registered reservations about the future of hydrogen fuel. (PDF Format)

Continued evolution of the traditional gasoline car technology could result in 2020 vehicles that reduce energy consumption and GHG emissions by about one third from comparable current vehicles and at a roughly 5% increase in car cost. This evolved “baseline” vehicle system is the one against which new 2020 technologies should be compared.

More advanced technologies for propulsion systems and other vehicle components could yield additional reductions in life cycle GHG emissions (up to about 50% lower than the evolved baseline vehicle) at increased vehicle purchase and use costs (up to about 20% greater than the evolved baseline vehicle).


If automobile systems with drastically lower GHG emissions are required in the very long run future (perhaps in 30 to 50 years or more), hydrogen and electrical energy are the only identified options for “fuels”, but only if both are produced from non-fossil sources of primary energy (such as nuclear or solar) or from fossil primary energy with carbon sequestration.

A more recent MIT study released in March 2003 voices even greater doubts about the viability and desireability of hydrogen as a fuel source in the next couple of decades.

Published in MIT Tech Talk, March 5, 2003.

Even with aggressive research, the hydrogen fuel-cell vehicle will not be better than the diesel hybrid (a vehicle powered by a conventional engine supplemented by an electric motor) in terms of total energy use and greenhouse gas emissions by 2020, says a study recently released by the Laboratory for Energy and the Environment (LFEE).

And while hybrid vehicles are already appearing on the roads, adoption of the hydrogen-based vehicle will require major infrastructure changes to make compressed hydrogen available. If we need to curb greenhouse gases within the next 20 years, improving mainstream gasoline and diesel engines and transmissions and expanding the use of hybrids is the way to go.

These results come from a systematic and comprehensive assessment of a variety of engine and fuel technologies as they are likely to be in 2020 with intense research but no real "breakthroughs." The assessment was led by Malcolm A. Weiss, LFEE senior research staff member, and John B. Heywood, the Sun Jae Professor of Mechanical Engineering and director of MIT's Laboratory for 21st-Century Energy.


However, the researchers do not recommend stopping work on the hydrogen fuel cell. "If auto systems with significantly lower greenhouse gas emissions are required in, say, 30 to 50 years, hydrogen is the only major fuel option identified to date," said Heywood. The hydrogen must, of course, be produced without making greenhouse gas emissions, hence from a non-carbon source such as solar energy or from conventional fuels while sequestering the carbon emissions.

The full text of the March 2003 MIT study Comparative Assessment Of Fuel Cells is available as a PDF document.

Curiously, in spite of the drawbacks of hydrogen as a way to store and transport energy hydrogen produced in cars for immediate burning may be a way to increase the efficiency of internal combustion engines.

But the researchers want to take the concept a big step further, using plasma technology to turn cars into small-scale hydrogen- producing plants - and sharply boosting the spark-ignition engine's efficiency along the way.

"Spark-ignition engines are roughly 30 percent efficient and diesels are about 40 percent efficient," notes Cohn. "We want to approach a diesel level of efficiency while avoiding diesel's pollution problems."

The plasmatron - about the size of a half-gallon milk carton - would convert about a third of a vehicle's gasoline stream into hydrogen. In doing so, it would boost efficiency in varied ways.

I think the hydrogen fuel hype is vastly overblown. The US government spending on hydrogen development is money that would be better spent developing photovoltaic materials that can be made much more cheaply than current photovoltaics. The goal of US government-funded energy research ought to be to obsolesce fossil fuels by developing cheaper competitors.

Update: A big step forward in battery tech would lower the cost and weight of batteries far enough to make hybrid vehicles competitive would allow reductions in emissions and in fossil fuel use in a way that would use all the existing infrastructure. Donald Sadoway of MIT says that a big step forward in battery tech is achieveable. On the subject of whether much better batteries could be developed for use in hybrid vehicles see my Energy Tech archives and in particular see my post Is Hydrogen The Energy Of The Future? for the bottom part of the post where I link to Sadoway's views.

On the question of whether photovoltaics would have to take up too much space, first of all, it will eventually be possible to achieve fairly high solar photovoltaic cell efficiency. See my post Material Discovered For Full Spectrum Photovoltaic Cell about some LBNL researchers who found a material that is 50% efficient. Surely nanotubes will be able to achieve a still higher effiency.

Also, I've done rough calculations on surface area needed for photovoltaics and the energy needed looks like it is achieveable with a fairly small portion of the Earth's surface. On my Parapundit.com blog in the Grand Strategy archive see the comment section of my post Energy Policy, Islamic Terrorism, And Grand Strategy where I introduce some rough calculations on the area needed for photovoltaics. I'd appreciate it if anyone could point to more accurate calculations of how much energy the United States currently uses and how much space in southern parts of the US would be needed to be used to collect enough energy for current consumption rates.

You can also follow a debate about this post on Arnold Kling's EconLog.

Share |      Randall Parker, 2003 July 20 10:14 PM  Energy Tech

Nick Schulz said at July 29, 2003 5:28 AM:

Can you email me more bio info about yourself? Thx.


TM Lutas said at September 4, 2003 7:26 AM:

Hydrogen fuel cells are starting to come on market as replacements for batteries and generators. It isn't fair to the technology to just look at it as an ICE replacement and not as a battery replacement.

Dean Lehrke said at October 8, 2004 7:02 PM:

Lets face it, an H2 economy would be an environmental disaster! It is truly the "emperor's new clothes" since there are way too many problems with storage and a costly pipeline infrastructure that would cause more global warming than present C02 emissions. H2 is hyped mainly by the fuel cell companies who are sucking up huge amounts of federal money doled out by government beaurocrats ignorantly lobbied into the scheme. Ethanol made from biomass through Fischer-Tropsch gasification or pyrolosis conversion (ie GTL or Gas to Liquids Technology and not by "frankenbacteria" fermentation)is the only truly environmentally benign transportation fuel,period. There is nothing more to argue. If you are still committed to fuel cells then use simple reformers powered by ethanol, but don't use on-board H2. Ethanol is the only fuel that occurs naturally in nature and is totally biodegradable, and does not pollute groundwater. It also has a low vapor pressure so it will not pollute the atmosphere during storage and handling like H2 or gasoline does.

Concerned Scientist said at May 4, 2005 10:24 PM:

In response to Dean Lehrke:

1. Ethanol as a transportation fuel requires an unsustainable use of U.S. agricultural land (www.news.cornell.edu/releases/Aug01/corn-basedethanol.hrs.html).
2. Leaked H2 cannot pollute because it achieves escape velocity in the atmosphere (www.astronomynotes.com/solarsys/s3.htm).

Benefits of H2 looked over by the MIT guys:

1. A land area 10,000 mi square in Arizona can support all U.S. electricity needs using solar panels of 12-15% efficiency (www.epa.gov/cleanenergy/renew.htm), and the amount of energy consumed in the U.S. as electricity is about the same as that for oil (add coal & nuclear together and compare to oil on energy.cr.usgs.gov/energy/stats_ctry/Stat1.html#CONSUMPTION, see also www.eia.doe.gov/emeu/cabs/usa.html). If technology is developed to use solar energy to make hydrogen from water, then the U.S. has ample domestic solar supply to fuel the transportation needs of the entire country. Though the Berkeley fellows say "for each one there is something else you could do that would probably work better," I challenge them to envision an energy-independent U.S. based on a gasoline or diesel economy.

2. H2 from solar power not only eliminates CO2 emissions, but also has no NOx or SOx emissions, which usually come from cars and coal plants, respectively (www.earthsci.org/energy/fuelcell/fuelcell.html, search for "nox"). These two pollutants are the leading causes of acid rain (www.epa.gov/boston/eco/acidrain/causes.html).

3. H2 can likely be stored effectively in the future thanks to the development of advanced storage systems that are based on neither metal hydrides nor liquid hydrogen (www.nrel.gov/hydrogen/proj_storage.html).

4. Efforts are underway nationwide to efficiently produce H2 using solar power (www.nrel.gov/hydrogen/proj_production_delivery.html, search "solar hydrogen production" at www.google.com).

It is of note that the MIT Sloan Automotive Laboratory receives its funding from GM, DaimlerChrysler, Ford, and a Saudi corporation (http://web.mit.edu/sloan-auto-lab/sponsors/). Why would they recommend anything besides keeping the U.S. hooked on oil-burning gasoline engines?

Joey Russell said at June 16, 2005 4:01 PM:

You tell him Concerned Scientist. I like how you cited your sources. I checked them out and they fit the bill. I think Dean Lehrke and the team at Berkeley who defend fossil fuel use have to think about one thing. Fossil fuels are made by plants and animals that died MILLIONS of years ago. I say this because fossil fuels aren't going to be around forever. Some generous estimates put the rate at 50-70 years before supplies are finally depleted. Even if this isn't correct, people need to think about the very real fact that eventually fossil fuels will be gone. It doesn't matter about how efficient an engine is then if there is no fuel to buy. When the rate gets to around 7 to 10 dollars a gallon, people will most likely cut their driving way down due to enormous prices. This will affect their lives so much that they would look for alternate sources of fuel. Electric cars, solar powered vehicles and hydrogen cars are the obvious solution.

There is an argument about the cost of the infrastructure for hydrogen power. Well I'm quite sure that when fossil fuels are gone, the government isn't going to go back to horse and buggy. So they will make the necessary changes beforehand by creating a pipeline or whatever they need to do to make hydrogen filling stations a reality

Sione Vatu said at September 7, 2005 5:32 PM:

Hydrogen requires a lot of difficult to master technologies, most of which provide inferior utility to the end user.

Hydrogen is difficult to store (more so than a liquid hydrocarbon fuel) and, nano-tubes or not, it is much more demanding to deal with than liquid hydrocarbon fuels. It has wide explosion limits, it leaks where other substances wouldn't, it embrittles metals (this is very dangerous) and so on.

Hydrogen is difficult to distribute and transport. Remember, it embrittles. It is a pernicious discoverer and exploiter of imperfections, cracks and porosity in the components it comes into contact with. Small imperfections are deadly with this stuff. Massive new and perfect quality infrastruture is necessary. Think NASA. Think Shuttle. Oh oh. Another MEGAPROJECT.

H2 is difficult to manufacture economically. The electrolysis of water is not a trivial task to achieve reliably on an industrial scale. And makinge H2 in the quantities needed is going to require titanic investment in production faciltites (along with all the supporting industries required- all built from scratch). Oh oh. Another MEGAPROJECT (several actually).

And then there are the photovoltaic cells. Billions of them. Thousands of square miles of them. OK. Let's see. Making ethanol is "unsustainable" but making hydrogen is not! Carbon fuel bad, hydrogen OK! Academic BS designed to keep all the grant money coming! As an aside, scientists are better at absorbing tax dollars than just about anyone or anything. Best part is they do not have to achieve anything. It just isn't considered necessary!

But let's assume we build a huge photovoltaic cell array in Arizona. Now we have a wee problem. The dirty little photovoltaic secret is the question, how much energy is constructing and maintaining and cleaning all this photcell going to consume. Check it out. Then calculate how long the cells will last (this is known as a TBR or time between replacements). Now calculate just how much energy will be actually generated and then do some maths and....it ain't good brothers. It's another misinspired statist MEGAPROJECT. Kill it. Kill it quick! We should avoid all these reich-building activities at any cost! Private and local efforts are much more appropriate.

In the end what is required is something to provide a little energy to transport Ma and Pa Kettle. Giga-dollars of hi-tech is not the solution to their requirement. The very idea of big govt "science" providing what people want is naive and ill inspired. That approach is really about keeping political classes (including tame "scientists") fed, well groomed and in power. Beware of concerned scientists on the govt. payrolls!

Go look at the hydrocarbon fuels again. The carbon cycle is a cycle after all...

Dean Lehrke said at September 19, 2005 6:35 PM:


How on earth did you come up with the notion that I was in favor of fossil fuels?! I am in the camp of bio-fuel promoters, not of fossil fuels! Or did you think that ethanol is a fossil fuel?

The H2 economy can only exist using cheap coal or natural gas for feedstocks, and this is of course not sustainable. And PV-electrolysis is very expensive and unsustainable.

On the contrary, bio-fuel like ethanol can be produced using no fossil fuel. It is now being done, and is sustainable. See: http://www.ethanolmarketplace.com/082405_news2.asp

John Bailo said at August 4, 2008 7:57 AM:

Care to update this post now that Nocera's work has been published?

Randall Parker said at August 4, 2008 7:55 PM:

John Bailo,

Nocera's work promises to lower the cost of hydrogen production 10 years from now. But I've yet to come across a report of an advance that will make hydrogen storage in cars practical for longer ranges. Plus, the fuel cells remain far too expensive. I am more optimistic about the fuel cells than about the hydrogen storage. Maybe hydrogen will become usable for shorter ranges. But Honda does not expect hydrogen cars to become affordable in the next 10 years.

Stationary hydrogen storage seems easier to achieve given storage material weight is not a problem. Though materials cost is still a problem.

Post a comment
Name (not anon or anonymous):
Email Address:
Remember info?

Go Read More Posts On FuturePundit
Site Traffic Info
The contents of this site are copyright ©