ã 2002 Darryl McMahon (all rights reserved)
Fuel Cell Electric Vehicle (FCEV) Efficiency
(or Should We Be Betting the Future of EVs on Hydrogen?)
I would really like to believe in the hydrogen energy economy. Part of me would like to see hydrogen fuel cells in every vehicle, because that would make them electric vehicles - EVs (FCEVs for the taxonomists among us). I have been a believer in EVs for over twenty years. I converted and drove my first in 1978, and have owned at least one ever since. Yet I can't quite work up that enthusiasm for hydrogen fuel cells.
I see many reasons to believe in hydrogen. I have followed it for years, from the early days of Dr. David Scott in Canada and Dr. Roger Billings in the U.S. I would like to see companies like Ballard and Stuart be smashing successes. There is no question that hydrogen is a clean fuel when used in a fuel cell, producing just water, electricity and heat as outputs. That appeals to the environmentalist in me.
I see the value in hydrogen being both the storage medium and the fuel. That should permit intermittent, non-dispatchable energy sources like wind, solar and tidal become more attractive. And goodness knows I would like to see a return on all the tax dollars my federal government has spent on fuel cells. The government of Canada has already invested millions of dollars into fuel cell research, and recently committed to spending at least Cdn$100,000,000 more.
I hoped that interest in, and the success of, FCEVs would lead to a renewed interest in more conventional electric traction technologies. Sadly, that has not been the case to date. As both Canadian and U.S. (and other) governments spend more tax dollars supporting more research on hydrogen fuel cells, the automotive industry has simply used the FCEV as a justification for further stalling any efforts at putting zero-emissions EVs on the road in the short term. So instead of bolstering the on-road EV industry, the FCEV has become another hammer to smash it out of existence. The FCEV has also become a convenient excuse for severely reducing research into batteries and other energy storage mechanisms (e.g. flywheels) that could be viable as on-board storage for EVs.
Suppose that half the Cdn$100,000,000 promised by the Canadian government for short-term future fuel cell R&D were spent instead on EV production using existing technology. That's enough to put about 10,000 NEVs on the road, or perhaps 3,000 high-speed urban EVs into operation, or twice that number if the prospective owners paid half the cost. Even more if this amount were applied as a tax credit for zero-emission vehicles. Now that would put a Canadian EV industry on its feet with a real solid start, and make a measurable contribution to environmental quality in our country. It would also create a real market for viable hydrogen fuel cells as retrofits, presuming they become an economic reality.
Still, the conservationist in me just can't get past the efficiency issues. Hydrogen fuel cell EVs may be more attractive than internal combustion engines from an efficiency and environmental perspective, but they just don't stack up against conventional battery EVs on efficiency. Eventually, efficiency has to lead to pricing.
I am going to start from what I consider the optimal case for hydrogen as an environmental transport fuel. Wind turbines in areas remote from major urban centres will produce electricity intermittently, but relatively steadily over periods of weeks. This electricity will be used exclusively to produce hydrogen from water via electrolysis, avoiding the issues of dispatchable power, backup generation facilities and increasing the cost of the facilities. I am assuming exclusively so as to avoid any costs associated with building new transmissions lines. Besides, if the facility was grid-connected, there would be no surplus power to be used for generating hydrogen - the amount of wind-generated electricity in North America will remain so small for decades to come that it could not meet even baseload demand on the continent.
Commercial electrolysis operations appear to top out at about 75% efficiency, based on a quick review of literature on the subject. 60% seems to be more realistic in real-world operations. Assuming a generous 75% efficiency, the cost is a 25% energy loss as a result of the conversion from electricity to hydrogen. Now we are going to liquify the hydrogen to make it compact to transport (because liquid hydrogen is the most compact storage form). Liquifying hydrogen from room temperature uses approximately the equivalent of 40% of the energy stored in the hydrogen (including some leakage and evaporation losses). Let's simplify our calculations by simply assuming that only 60% of the hydrogen we put into the liquifying process appears as liquid hydrogen. A special transport truck, a FCEV vehicle, is used to transport the liquid hydrogen from the production site to the distribution depot. It consumes 4% of the energy it is carrying making the round trip each day. (This is likely generous; pipelines and electrical transmission lines are typically considered to have losses in the range of 5 to 8% depending on the distance, and industry prefers pipelines to trucks when they have the option.) The liquid hydrogen is transferred to the storage facility with no losses. However, keeping the storage facility cooled to 20 degrees Kelvin is estimated to cost 1% of the volume per day, and we will assume that 3 days supply is retained on average in storage, which costs an additional 3%. The hydrogen is pumped from storage into on-board fuel tanks in each vehicle, which may be cryogenic, but probably are not. We will assume zero losses here as well. Once in the vehicle, which is fueled on average once a week, there is another 3% losses in storage due to continued refrigeration or warming/evaporation losses. Then the hydrogen is fed to a fuel cell. We'll assume the fuel cell has a 40% efficiency. Robust, commercial fuel cells capable of use in road vehicles have not reached this figure yet, and some look upon it as an unreachable, theoretical figure. (Larger, non-mobile fuel cells are now achieving 35% efficiencies under ideal operating conditions, but not continuously.) This seems to be the deep, dirty secret of the FCEV industry. They tout the 90-plus per cent efficiencies of the electric drive train, but are curiously silent about the efficiency of the fuel cell itself. Fuel cell statistics are almost always presented in terms of kW of capacity as peak power output, not efficiency. The nascent FCEV industry points out that the fuel cell plus electric drive train is more efficient than an internal combustion engine, but conveniently ignore efficiency comparisons with battery electrics. However, we'll give the hydrogen fuel cell every break we can in this analysis and use the 40% figure. However, the fuel cell is not sized to provide peak power for the vehicle, but something just over typical operating power demand. Supercapacitors or high-power batteries are used as a buffer to even out demand on the fuel cell, and this charging and discharging consumes some power, which I am assuming to be about 2%. Only now do we have the equivalent of the power that comes from batteries in a battery-only EV. So let's recapitulate all those losses, in terms of efficiencies, starting from the power from the wind turbine.
Process Efficiency (loss)
Electrolysis 75% 25%
Liquification 60% 40%
Transport 96% 4%
Bulk storage 97% 3%
Vehicle storage 97% 3%
Fuel cell 40% 60%
Supercapacitor 98% 2%
How does that stack up? Well, doing the math (75% x 60% x 96% x 97% x 97% x 40% x 98%=15.93%) comes out to a total efficiency of under 16%. If that same kWh were transmitted via high tension lines and the local distribution grid (8% loss) and used to charge lead-acid batteries (10% loss), the equivalent efficiency (from turbine to motor controller) would be 83% efficient; more than 5 times better than the hydrogen storage and transport cycle. And I have been making pretty generous assumptions all along to the benefit of the hydrogen fuel cell energy cycle, so the 16% figure is undoubtedly generous, and still ignores the cost of building the electrolysis facility, the transport vehicle, the bulk storage facility and the fuel cell itself. (I am assuming that the batteries and the on-board hydrogen storage will have similar prices for the foreseeable future.) This analysis has also ignored the R&D remaining related to the safe and economical handling, storage, transport and dispensing of hydrogen.
Is it worth having to produce at least 5 times as much electricity to travel a kilometer in a FCEV than the same distance in a BEV? Before answering that question, remember that more electricity is produced in North America from burning coal or burning natural gas than from all wind, solar and tidal power combined. Each of us will have to reach our own conclusion; for me the answer is "No".
And there are those who will argue that the efficiency of fuel cells will improve, and exceed even the 40% figure I have granted in the analysis above. That could be, but I suspect that fuel cells will never reach the over 300% efficiency they would have to achieve to match the energy efficiency of the battery EV already on the road today. To my undoubtedly limited imagination, that still remains a physical impossibility.
In December, 2002, Honda and Nissan introduced FCEVs in California with great mass media fanfare. Lost in the small print was the fact that these companies see mass marketing of such vehicles as being at least another decade away. Honda will lease a limited number of 2003 Honda FCXs in the U.S and Japan by the end of this year. During the first two-to-three year period, the company will lease about 30 fuel cell vehicles in California and Tokyo--areas in which hydrogen fueling is available. The company currently has no plans, however, for mass-market sales of these vehicles. Nissan only intends to start public road testing of the zero-emissions X-TRAIL FCV in Japan in 2003.
Is it really a good idea to let EVs be put on hold for a decade or more while we wait to see if the FCEV, the hydrogen FCEV in particular, pans out? Or, is there a reasonable chance that the FCEV is yet another panacea that will never become a market reality, and end up as another automotive industry stall tactic like the USABC, the PNGV, etc? I don't think it is time to rest and believe that this time, the automakers really mean to deliver a clean-air vehicle out of the goodness of their hearts. Note the lack of economic information that is currently being provided by the promoters of FCEVs, which is how they eventually strangled the recent reincarnation of battery EVs (e.g. the GM EV-1, the Ford Ranger EV, the Chrysler EPIC, the Honda EV+, the Nissan Altra, the Toyota E-com forced out of the automakers by CARB in the late 1990's) and are now stifling their very limited ventures with hybrids (Toyota Prius, Honda Civic Hybrid and the never-quite ready Daimler-Chrysler Durango / Contractor Pickup, the GM Triax, etc). In fact, at the January 2003 Detroit Auto Show, the big 3 North American automakers are already backing away from fuel cells and going back to "hybrids" as their PR strategy (in reality, electric-assist gasoline vehicles, not serious hybrids with robust electric drive-trains).
Realistically, do we really want to deal with raw hydrogen? Nature doesn't. Hydrogen is highly reactive; it's as if it wants to be anything else but hydrogen. It seems to me that if someone really wants to make a mass-market fuel cell for in-vehicle use, it will not be based on using pure hydrogen. Instead, it would be based on using something easier to handle, store and transport. I don't mean gasoline, as Daimler-Chrysler was wasting money on a couple of years ago. I am thinking of a fuel cell that will use methanol or ethanol directly. A liquid fuel, easy to retrofit into the existing fuel structure. In fact, many fuel depots already dispense ethanol/gasoline blended fuels (aka gasohol). High hydrogen content, low carbon content, possibly not as environmentally pure as a hydrogen fuel cell, but were folks really going to collect the water vapour exhaust to drink it anyway? I am aware of two companies currently working on very small fuel cells that use methanol as their primary input.
Where does that leave those of us that want to see zero-emissions, clean-air, electric drive displacing a significant number of pollution-mobiles? Back at looking for the better battery, continuing to make incremental improvements in existing technologies, and showing that current, off-the-shelf technologies can already do the job in many on-road and off-road applications.