May 12, 2005
Study Sees Hydrogen Problems Requiring Decades To Solve

Use of hydrogen to transport and store enerfgy is still a distant prospect.

WEST LAFAYETTE, Ind. – Researchers conclude in an article to be published in June that it could take "several decades" to overcome daunting technical challenges standing in the way of the mass production and use of hydrogen fuel cell cars.

The article notes that "success is not certain" in efforts to develop inexpensive, hydrogen-powered fuel cells and to create the vast storage and transportation infrastructure needed for the vehicles, stressing that hydrogen's "wide-scale use is laden with potential technical, economic and societal impasses." In case fuel cells never do become practical for cars, the researchers conclude, it would be wise for the nation to "maintain a robust portfolio of energy research and development" in other areas.

"In my mind, developing practical hydrogen fuel cells for cars is definitely doable, but we must solve very daunting technical challenges," said Rakesh Agrawal, Purdue University's Winthrop E. Stone Distinguished Professor of Chemical Engineering.

The article will appear as the cover story in the June issue of the AIChE Journal, a publication of the American Institute of Chemical Engineers. The article was written by Agrawal, Martin Offutt, from the National Research Council, and Michael P. Ramage, a retired executive from ExxonMobile Corp.

Fuel cells cost too much to build and have short operating lifetimes.

"Today's fuel cells generate power at a cost of greater than $2,000 per kilowatt, compared with $35 per kilowatt for the internal combustion engine, so they are more than 10 times more expensive than conventional automotive technology," Agrawal said. "At the same time, fuel cells have an operating lifetime for cars of less than 1,000 hours of driving time, compared with at least 5,000 hours of driving time for an internal combustion engine.

"That means fuel cells wear out at least five times faster than internal combustion engines. If I buy a new car, I expect it to last, say, 10 years, which equates to about 3,000 hours of driving time. If my fuel cell only lasts 1,000 hours, you can see that's not very practical."

Cheaper and longer lasting catalysts are needed. Plus, in order to use fuel cells to burn hydrogen the hydrogen transportation and storage problems need to be solved.

To bring down the cost of fuel cells, less expensive catalysts and membrane materials are needed, Agrawal said.

Developing an infrastructure of hydrogen storage and transportation represents other significant challenges.

"A fuel-cell car built with today's technology would cost about $250,000, but you would have no place to fill up the tank," Agrawal said.

Hydrogen is a light gas, which makes it more expensive to transport and store. Because its molecular weight is only 2 – compared with heavier gases, such as methane, which has a molecular weight of 16 – less hydrogen is contained in the same space as heavier gases, making its transport more expensive.

Agrawal sees hydrogen vehicles starting to show up on the road in the year 2020.

"I believe we can probably solve the technological problems related to making hydrogen fuel cells practical as a replacement for the internal combustion engine, but it won't be easy and it likely won't happen very soon," Agrawal said. "An optimistic prediction would be that a significant number hydrogen fuel cell cars will be entering the marketplace around 2020, and by 2050 everybody will be driving them."

But that is an optimistic prediction. A lot of problems must be solved to even start hydrogen deployment in 2020. In the meantime the market for gas-electric hybrid vehicles is going to become quite large. Many of those hybrids will be pluggable and some people will be charging them from their home outlets. Photovoltaics might drop to the point that a portion of that car battery recharging will be done using electric generated right at home.

Suppose nuclear power experiences a resurgence. Hydrogen could be generated at nuclear plants. But if superconductor technology continues to improve and battery technology does as well then superconducting power lines which suffer no resistance might deliver nuclear power to electric vehicle batteries more conveniently at home at a lower cost than a hugely expensive infrastructure for delivering hydrogen to fuel stations.

In my view hydrogen's eventual role as primary vehicle fuel is by no means assured. Future solutions to hydrogen's technological problems will not compete with today's other energy technologies. Hydrogen's supporting technologies will compete with tomorrow's batteries, superconductors, and other energy technologies. Those competing technologies will be delivering benefits decades before hydrogen begins to do so and therefore industry, academic, and government labs will continue to refine those other technologies. By the time hydrogen is ready the competiton might be too firmly entrenched and cheap to be dislodged.

Fuel cells have a future independent of hydrogen. If the cost and durability problems with fuel cells could be solved for burning hydrocarbon liquid fuels then fuel cels could be adopted much more rapidly as a more efficient way to burn fossil fuels. Liquid fuel burning fuel cells could even work in hybrid vehicles with batteries providing increased efficiency through regenerative braking.

Share |      Randall Parker, 2005 May 12 11:18 AM  Energy Tech

Philip Sargent said at May 12, 2005 12:40 PM:

Right on. I'm with you all the way.
Confusing technical feasibility with commercial viability is usually not a serious error over a 40-year timeframe, but in this case, from what we know now, I think it is.

Joseph said at May 13, 2005 1:31 AM:

I personally do not find the attempt wastefull in itself. Hydrogen is a clumsy way to power transportation. I do think that even if a major breakthrough occured in say batteries the basic research into fuel cells will still pay for itself in the future if nothing else than by way of serendepity. I should have known things would go south as soon as Al Bore opened his mouth in 95(?) with his glowing endorsement of FC's and onboard fuel reformers. I'd love to have that guy beside me at a racetrack, I'd make a fortune betting contrary to his picks.

Assuming FC vehicles using hydrogen were the only option I would hazard that most of the system could be made practical. Say SOFC's for the vehicle, nuclear as the primary producers for hydrogen. There has been some progress that shows potential for end-user hydrogen storage but the vexing problem of the wholesale transport has no really good answer yet. It could be accomplished by liquification yet then you've just added another energy/dollar sink into the system. Ammonia might be a good method or even ethanol as far a bulk shipping but then you have the problem of cracking it at the retail sales point. I'm really not happy with ammonia, it's done all the time yet if used for wholesale transport as a vehicle fuel carrier the danger goes up appreciably due to the huge volume increase. To be honest I'd rather have a whole tanker load of liquid hydrogen go up versus a split open ammonia tanker. Pipelines might work in some cases but I can't see it as a really widespread method. Perhaps I'm missing something on the wholesale distribution scenarios.

odograph said at May 14, 2005 7:46 AM:

I'm a semi-retired engineer. In my career I've seen a lot of technologies tried. Some work right away, and some hit dead ends. Even those that hit dead ends may (years later) find some kind of end-run, or alternate path.

The plan that doesn't work, has never worked in my experience, is just throwing more money at a dead end. Spending money to promote, or to build out infrastructure for, a technology which is at a dead end is just laughable.

I think the real, sensible, engineering strategy would be to spend some moderate amounts on research, directed at hydrogen's technical challenges. I would spend NO money on promotion or deployment or "hydrogen highways" until those problems are solved, or a new path is found.

I hate to bring up the arch-typical examples of technology we wanted, but never got - the flying car and personal jet pack - but they provide a cautionary example. It wouldn't have mattered if we'd spent billions on those or not, they just were not going to work with our technology. Just because you want something, that doesn't mean you can engineer it.

Sometimes you hit a dead end, and when you do, you try other paths.

martin g said at May 15, 2005 6:32 AM:

I’m coining the phrase H2ype.

The ‘Hydrogen Economy’ has nothing to do with ‘green tech‘ and everything to do with ‘Big Oil’ finding a product that consumers will think is PC. ( Oh, and it’ll be easily taxable too ! )

See :

KenS said at May 15, 2005 12:41 PM:

martin g:

I like H2ype. It is H2ypermania. I think big oil does promote H2 as a PC diversion from the real problems.
But they like it even more because it is, at best, years away.

Thus, everyone except oil is blamed for not delivering the miracle fuel: scientists, engineers, fuel cell and car makers, and government.

Dave said at May 16, 2005 11:03 AM:

The hydrogen distribution system already exists and extends to every house and business in the developed world. Water.

Within 3 years solar powered cleaving of water will become competitive with the grid.

instead of propane or oil we can then store this hydrogen in tanks and run stoves, HVAC, fireplaces etc on the hydrogen. Within 5 years fuel cells will be able to generate electricity from this h2.

within 5 years, home hydrogen fuel pumps will fill the tanks of cars with zero pollution hybrids that run the internal combusion engine on hydrogen.


Randall Parker said at May 16, 2005 11:24 AM:

Dave Gobel,

You say:

we can then store this hydrogen in tanks

Ah, but that is the hard part.

Gaseous hydrogen has to be stored under very high pressure because hydrogen has very little energy per volume at low pressures. But it takes energy just to compress the hydrogen to put it into a high pressure bottle. The bottle has to be made of thick steel or something else that can handle the pressure. Hydrogen molecules, owing to their small size, can get out through very small leaks. So seal and valve quality has to be very high.

Hydrogen can be stored in a small area if liquified. But liquification requires energy. Plus, the container that holds cold hydrogen liquid has to be very well insulated and continually air conditioned to keep it cold enough. That requires additional on-going energy expenditure.

The third possibility is to come up with nanomaterials that hydrogen will stick to. This is an area of active research by many labs. Some promising discoveries are being made along these lines. But this is still an unsolved problem.

I don't know when photovoltaics will become cheap to make. I've read a number of reports in the last few years that sounded very promising. But some of the earlier announcements forecasted production of cheap solar cells by now. Well, that hasn't happened.

odograph said at May 16, 2005 11:24 AM:

With predicutions like that Dave, I think you should show us some math. How much hydrogen will you need, and how many solar panels (in square feet and dollars) will it take to make it?

As a help, I'll mention that a kilogram of hydrogen is often stated as the equivalent of 1 gallon of gasoline, and about 33 kWh is required to produce one kilogram. So a middle range home solar system (3kW, and a $27,000 investment) would take ten hours of good sunlight and dedicated use to produce the equivalent of one gallon of gasoline.

data from:


Dave said at May 16, 2005 2:02 PM:

<against a background of armwaving and grandiose marching music< Well - all I can do is say this is my intuition. Then look at my track record of midwifing disruptive concepts and technologies.

The water is there. The solar energy is there. Darpa and VC's are spending huge sums on nano-tech to reduce logistics tails for both energy and water delivery. so, it's intuition. Feel free to ignore me. There's no downside :-)

Oh - by the way, there's no way someone without a degree in biology could have a real influence on aging...unless you could make a prize to reverse aging and have plain folks grow it exponentially...maybe that would work?
<music and armwaving ends>

whimsy intentional!

odograph said at May 16, 2005 4:46 PM:

Solar "being here" is actually a good parallel to the hydrogen thing. I remember the 70's man ;-), and the really cool projections we had for solar power cheaper than fossil fuel energy "in ten years" or "in twenty years" ... we've just passed the long bar, and guess what ... it's wind energy that is (in some places with a lotta wind, like Texas) that is cheaper than fossil fuel.

That illustrates two things. First of all, saying you are going to do it fast, doesn't mean you will. And second, "innovation" often comes from a direction we didn't expect.

Maybe we could say that wind and solar were in a race (still are) and rather than the "breakthroughs" early optimists had hoped for, we are instead making steady incremental progress ... with wind power ahead on price.

(and given all the work in 30+ years of solar, I don't think I'd expect the 100-fold price/performance increase you'd need in a year or two.)

hemmer said at May 18, 2005 11:14 AM:

Please give some feedback, what you thing about this engine, if its running on hydrogen.

TM Lutas said at May 21, 2005 5:50 PM:

The DOE has made some milestone goals for 2010 regarding hydrogen. Ballard Power Systems has a roadmap out that shows how it expects to meet or exceed those goals. One interesting fact is that Ballard has already demonstrated a fuel cell stack that runs 2000 hours, twice the 1000 hours that this pessimistic report says is the current state of the art. The 2010 goal is 5000 hours, about where ICE engines are today. The DOE power density goal for 2010 is 2000 Wattsnet/liter. Ballard is currently at 1200 and predicts it will be at 2500 by 2010. The DOE FC stack cost goal is $30/KW as is Ballard's. Ballard's 2002 cost figure was $150 and it's 2004 figure was $103. This contrasts with the claim in this paper that the figure is $2000.

These goals all point to 2010 being a realistic point where vehicle makers can start turning out fuel cell cars. The roadmap has a few years of historical data and, I'm sure, it'll be updated as time goes on and we'll be able to see whether Ballard's views are realistic. This is very different from any other alternative energy scheme that I've seen in the past. Hard numbers and graphs showing commercial viability in the near future are the sort of hard data that we need to see whether hydrogen fuel cells are realistic or pie in the sky.

The fact is that just a cursory reading gave me two figures that are simply not factually accurate. If this is published with those bad figures in place, it really is a failure of the peer review process. Ballard has real cells that are being sent out to car manufacturers all over the world. They're probably the leading company in the field. If they say that they've demonstrated fuel cells of these characteristics, you can bank on it that they exist and that Ford, GM, Toyota, BMW, et al have all seen them in their testing labs.

Philip Sargent said at June 26, 2005 2:00 PM:

The Lithium Economy just looks a whole lot easier to me. Every problem with the Hydrogen economy has a parallel problem in the Lithium economy, and if you compare problem with problem, the Lithium option is less risky, cheaper and earlier in almost every case.


Randall Parker said at June 26, 2005 3:06 PM:


Here you say:

The problems with the Lithium economy are (1) insufficient electrical distribution infrastructure to cope with a sensible proportion of our land transport coming from rechargeable cells, (2) low volumetric energy density of today's cells, (3) cost of the cells.

I don't see how we lack the distribution infrastructure. Recharging batteries for most vehicles could be done at night while we sleep. There is a huge difference between late night and peak day time electric demand, something like a factor of 2 or 3 isn't there? So the infrastructure that is built for peak demand can be used more hours of the day.

The key here is to move to electric price ratings in kilowatt-hours that are different at different hours of the day. Charge more for peak usage and less for off-peak usage.

So I think the main obstacle is the batteries. Everything else would fall into place if we had great battery technology.

You also say:

There is a an ecominic path-dependency transition issue: the early uses all use electricity for rechargeable batteries. This is more expensive (per km of car travel) than diesel. Only later, when we get into the mature Lithium anode economy, do we probably move from making Lithium electrochemically to (probably cheaper?) pyrometallurgical routes.

I think one thing we have going for us is the laptop rechargeable battery market. It is a big source of demand for better batteries. But that market has been there for years and still hasn't yielded a big step forward. I think we need much more funding for basic photochemistry research.

Philip Sargent said at June 27, 2005 8:58 AM:

Yes, if we can schedule all recharging to be done at night, then the tranmission capacity would seem to be there (I'm not a transmission expert; maybe some distribution transformers have to cool off at night). However, in the UK, 25% of total energy demand is land transport (33% in USA?) - anyway, quite a bit more than the total energy delivered by the electricity supply industry.

So in terms of GWh/year, we need to double the electricity generation capacity. This means increasing baseload plant (nuclear, coal+CO2seq., methane) because we really don't want to use expensive peaking plant for this. The reason electricity is cheaper at night is largely (a) to encourage people not to use electricity at peak times if they have the choice, rather than (b) because baseload is cheaper than daytime plant, let alone peaking plant. Evidence in favour of this in the UK is that Dinorwic (Welsh large pumped hydro scheme) is not used for load shifting these days since real costs of electricity in day and night are so similar (it is used to provide short-term power support, but that's another matter).

Re rechargeable battery developments - in fact there has been huge development over recent years, and there is a lot more to come. See the report: (1165kB) which is admittedly a long read. Maybe someone could write a summary?


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 ©