23 July 2006

Hydrogen's Death Knell?

Update: Ben has now posted an interview with Ulf Bossel. If you're masochistic you can also listen to me try to make the same points here.

A major announcement (hat tip to theWatt.com) came out of the Lucrene Fuel Cell Forum two weeks ago. The president of the conference, Ulf Bossel, presumably with the support of the organizing commity, announced that the pre-eminant European fuel cell conference would no longer be providing a forum for the discussion of hydrogen fuel cells. This is due to the unsuitability of hydrogen as a fuel to power our economy.
Fuel cells are energy converters, not energy sources. They will be part of a sustainable energy solution only if they can compete with other conversion technologies. This includes system parameters, fuels and applications. Time has come for a critical assessment.


The European Fuel Cell Forum is committed to the establishment of a safe energy future. Therefore, it will continue to promote fuel cells for sustainable fuels, but discontinue supporting the development of fuel cells for hypothetical fuel supplies. Time has come for decisions. Keeping all options open is not an adequate response to mounting energy problems.

Therefore, the schedule of the European SOFC Forum will be continued in 2008 with an extended conference every second year. Beginning 2007 (July 2 to 6) sustainable energy topics will be emphasized in odd years. Despite earlier announcements the European PEFC Forum series will not be continued.

A series of technical reports on the subject by Ulf Bossel and others is available here. I discussed this on this week's theWatt.com podcast. I would like to reiterate some of the arguments here.

Just to provide a bit of background there are basically two general categories of fuel cells: ones that operate at low temperature − typically below the boiling point of water − and ones that operate at a high temperature of at least several hundred degrees. The proton exchange membrane or PEM fuel cell is the low temperature fuel cell, along with the direct methanol fuel cell. There are many more high temperature fuel cells − examples would be the solid oxide or molten carbonate types. The PEM is unique in that it's the only type that can only burn pure hydrogen; all of the other types can directly burn hydrocarbons of one type or another.

The proponents of the proton exchange membrane fuel cell have promulgated this concept of the 'hydrogen economy' that I'm sure all my readers have seen references too. Basically the hydrogen economy is the idea that we can shift from using fossil fuels to hydrogen as a chemical fuel to power our economy. The transformation would begin with the transportation sector and eventually propagate to providing residential fuel cells that could provide combined heat and power.

There are three major shortcomings of the hydrogen economy concept:
  1. Production.
  2. Storage.
  3. Distribution.
Distribution is basically a chicken and egg problem. No one wants to buy a hydrogen car until there are fueling stations available and no corporation wants to invest in hydrogen fueling stations until there are customers on the road. Building the hydrogen economy would require an absolutely massive capital investment. For example, none of our current natural gas pipelines are capable of handling hydrogen because it’s a highly corrosive substance.

The storage problem is partly technological and partly the laws of physics. The basic difficulty comes from the fact that hydrogen has an extremely low density: liquid hydrogen a density of only 80 k/m3. In order to get hydrogen from a lighter than air gas to some usable stored form it needs to be compressed, liquefied, or chemically bonded. All of these means need to consume a large fraction of the energy of the hydrogen to get it to that state. Hydrogen is not like gasoline. You cannot pull up to a station and pump your tank full in a couple of minutes. A lot of people don't realize that pumping up a high-pressure compressed hydrogen tank can easily take 30 - 60 minutes.

The last problem is that of production; it is the most fundamental problem and it’s the basis of the schism that's occurred at the Lucerne Fuel Cell Forum. Unlike fossil fuels hydrogen doesn't exist in nature on Earth − we can't poke a hole in the ground and pump out hydrogen formed from long-dead plants. Hydrogen isn't an energy source; it's an energy currency, like electricity. Elemental hydrogen has to be produced from other compounds such as water or hydrocarbons.

When it comes to producing hydrogen from fossil fuels the high temperature fuel cell guys rightly get out their signing voice and break into their best rendition of, "Anything you can do I can do better..." from Mary Poppins. The solid oxide fuel cell and other types can all burn natural gas, gasified coal and biomass at a higher efficiency than converting those feedstocks to hydrogen and using a PEM cell.

That leaves producing hydrogen from the electrolysis of water, which is the supposedly ‘green’ option. The reality is that the electrolysis to fuel cell path is a terribly inefficient method to convert solar, wind, and nuclear energy to useful work. Let us consider the production of hydrogen from wind power. First you have to rectify the alternating current to direct current to power the electrolyzer, which is about 90 % efficient. An electrolyzer is optimistically 75 % efficient so you lose another quarter of your energy there. Then you need to store the hydrogen, by say compressing it under high pressure. This would consume about 20 % of the energy content of the hydrogen, and distributing it perhaps another 10 %. Now we finally have the hydrogen at the fuel cell but then we have to remember that the fuel cell is maybe 50 % efficient. The product of the fuel cell is direct current electricity, so in the end we’ve gone through a whole bunch of steps in a big circle. When you multiply all these factors together you find that the well-to-wheel (or source-to-sink) efficiency is only about 25 %.

The obvious question that Ulf Bossel and people such as myself ask is why go to all that trouble? Why not just transmit and use the electricity directly? High-voltage direct current electricity transmission is just as efficient as pipelining hydrogen. If we allow for 90 % efficiency for rectifying and 90 % for transmission we end up with 3.3 times more energy for the electricity economy than the hydrogen economy. If you want to include batteries the math doesn’t change much because the round-trip efficiency of batteries is really very high – 90 % for lithium-ion batteries. As Bossel states, hydrogen cannot compete with its own fuel source − in this case, electrons. This poor efficiency of the hydrogen economy that I’ve talked about is not something that has a solution through improved technology. The laws of thermodynamics maintain the limiting factor here. All the extra steps in the hydrogen case produce entropy, and there’s no way to get past certain theoretical limits to the efficiency of each stage.

The inefficiency of hydrogen isn’t something that we can afford environmentally. Would anyone consider it better to have three wind turbines rather than one, or three nuclear power plants rather than one? If you try to figure out how many power plants we would need to implement the hydrogen economy it becomes readily apparent what a fantasyland it is. The USA uses approximately 20 million barrels of oil per day. If we were to replace every gallon of gasoline with a kilogram of hydrogen we would require 1.4 TW of continuous power. However, there’s only 0.9 TW currently installed, of which about 2/3rds is used on average, so the idea of trying to use nighttime power to produce hydrogen won’t work. The existing infrastructure is incapable of powering a hydrogen economy – we’re talking about 1500 large nuclear power plants. So not only would we have to change our entire distribution network but we would also have to massively ramp up electricity production. The expense of the whole idea is terrifying.

This is sharply contrasted with trying to develop plug-in vehicles, electrified rail and such. Because the electricity path is so much more efficient we can actually dump almost all of our transportation energy needs on the existing electricity grid. If we throw in some efficiency improvements to the residential and commercial sectors then everything is peachy. The existing electrical grid that we have may not seem as sexy as hydrogen but it’s definitely the better option.

What's happened at Lucrene is that the rest of the fuel cell community have gotten tired of the empty promises of the hydrogen economy and are fighting back. PEM fuel cells have been receiving a disproportionate share of funding into all alternate energy technologies. What Ulf Bossel is saying is that is we have to refocus our efforts and monies onto technologies that we know will actually work rather than some idea put forth by a special interest group. Of course, the PEM researchers don’t want to hear this. Careers are at stake so I wouldn’t expect abandonment of the hydrogen economy concept quite yet.


ben said...

I agree.

Companies working on hydrogen fuel cells argue that we as a people will never accept a reduction in our standard of living...something that BEVs will bring us because of their limited range. I've also heard Ballard (both the G. and the company) argue first hand that the world needs hydrogen fueld vehicles because the developing world is going to surpass North America in terms of number of vehicle sales and so we better be providing them with "clean" transportation methods. I think this is BS.

I've calculated that Honda's hydrogen fuel cell car requires 0.48kWh/km (http://www.thewatt.com/article-1162-nested-1-0.html), but battery vehciles existing today (not using Li-ion batteries), require around 0.23kWh/km (the new Tesla using Li-ion uses just 0.125 kWh/km). Basically what this means is that if we drive 35km/day, then the amount of additional electricity required to make hydrogen would be the same as the per capita electricity consumption of Germany. BUT, if we had BEV cars that use 0.23kWh/km, then the amount of extra electricity consumption would be equivalent to the per capita consumption of Poland. Is this still an option for developing countries? My first guess would be no.

I know some places that have actually given up on researching SOFCs in favor for PEMFC...Norway put a lot of money into their SOFC program in the 90s but then got very discouraged, dumped it and started up a PEM program. They'll get discouraged again.

In the end, the only long term application for fuel cells that I can envision is a small CHP unit for a small community running off of bio-gas from waste products.

ben said...

Here's me replying to my own comment:

"Is this still an option for developing countries? My first guess would be no."

I'm leaning towards yes now. Using Li-ion batteries, the amount of additional electricity consumption is equivalent to a country with a human development index of only 0.6. I think this is viable for the world.

Auros said...

Hello there,

A friend of mine was telling me about the Li-Ion fire hazard issue, and I remembered having heard that the battery technologies have improved significantly since '99, so I went looking for info, and found this page, where you remarked that this problem had been solved by changing the cathode material...

Any chance you could expand on the details of how this problem was solved or ameliorated?

Auros said...

Oh, also, "Anything You Can Do, I Can Do Better" is from Annie Get Your Gun, not Mary Poppins. *g*

Engineer-Poet said...

Heh heh heh.  Vindicated.

David said...

Isn't it correct that hydrogen fuel cells (more specifically: their associated storage tanks) would provide higher energy density than batteries?

If they don't, then it's indeed hard to see much future for this technology.

I did see something recently about railroad interest in hydrogen fuel-cell-powered locomotives, which doesn't appear to be a PR-driven project.

Robert McLeod said...

Certainly hydrogen has a superior mass energy density when compared to electric vehicles − neg on the volume energy density. When comparing the energy density of batteries to hydrogen fuel cells one needs to add the weight of the tank and fuel cell itself to that of the hydrogen. Batteries stand alone in the energy density accounting. Both still need an electric motor.

In terms of power density I think that vehicles like the Tesla or Eliica have conclusively shown that they have the edge.

Robin Hood H20 said...

Pls review my web page
Note all 50 / 60 Hz diesel generators run at 1500 rpm - makes it easy to convert the injectors to diesel fuel (dun deal). No H2 storage problems just sell the renewable electrical energy and use oil as a lubricant in very small ammounts.

Thomas said...

... and yet you hear about the Hydrogen Economy as if it were a sure thing... :-(

The only compelling thing about the hydrogen economy is that it is portrayed as if you are just switching from one liquid fuel to another, thus you can go about your life as always.

That and the supposition that it is the only way to utilize wind energy efficiently. This latter argument is a direct extrapolation of how we use electricity today way into the future. But the truth is that using electricity in a more flexible and intelligent manner, would save a tremendous amount of money (even in a business-as-usual scenario) while enabling a much greater share of renewable energy at the same time.

Robert, you touched upon thermal storage in another post. Thermal storage is great, because 1) we use a lot of thermal energy, either as heating or cooling, and 2) you can store a lot of it very cheap!

drdrapp said...

Overall, the hydrogen economy looks pretty bad. Too many inefficiencies. Furthermore, hydrogen leakage (which will be substantial) will cause significant atmospheric impacts. The one really good aspect is that in principle (though maybe not in practice) it offers a way to store energy from intermittent renewable energy resources. This could enable a great expansion of the application of such energy sources.
The electric car has been widely touted as environmentally beneficial, but when you look at the whole end-to-end system: mine the coal, transport the coal, burn the coal, have power lines all over the landscape, deliver the power, reprocess acids and metals in batteries, etc. it may not look so good.
The problem with most of these discussions is that they center on how we are going to USE ENERGY - whereas our really big problem is providing the ENERGY SOURCE. In the longer run, this may depend on technologies that appear very futuristic at this juncture (fusion, solar power from space, ...) but there seems to be a rather extended intermediate run ahead of us where none of these will be available. And in order to extend that intermediate run period to give us enough time to develop long term solutions, we must drastically cut usage and improve efficiency wherever we can. I note that General Motors demands the right to make behemoths with V8 engines and claims that we will lose the auto business to Japan if we don't. I have news for GM: they have already lost.

Celia said...
This comment has been removed by a blog administrator.
Omar Cruz said...
This comment has been removed by a blog administrator.