Energy Collective blog power policy climate - the conversation happens here

Tuesday, April 25, 2006

Why I'm Not a Big Fan of Hydrogen

As you know, assuming you're a regular reader around here, I'm not a huge fan of hydrogen and the reason why, is because I believe there are better alternatives to nearly all of the potential applications of hydrogen. This paper, does a good job of thinking through most of the reasons why I believe that:

Carring the Energy Future: Comparing Hydrogen and Electricity for Transmission, Storage and Transportation.

I would highly recommend taking a look at the paper. I'd love to hear other's thoughts on what I think is a pretty convincing job of illustrating the viable alternatives to potential hydrogen applications. Its time to cut through the hype and get down to exploring alternative before we start (or continue) investing in the what may be a costly mistake.

What follows is a discussion on this thread between myself, and University of Oregon Professor of Physics, Greg Bothun. I would welcome and encourage the continuation of this discussion in the comments section below.


It seems that the only time that hydrogen is better than the alternatives is perhaps when you are trying to tap remote renewables that cannot be linked to the grid in any other manner (e.g. high voltage DC lines). In that case, tapping what would otherwise be an unusable renewable resource by making electorlytic hydrogen on site and piping it to demand centers would be a worthwhile model.

However, it still makes little sense to use that hydrogen as a transport fuel in fuel cell vehicles (FCVs). A better alternative would be to utilize the hydrogen near the demand center in more centralized hydrogen fuel cell power plants that can use more efficient high temperature fuel cells (like solid oxide FCs) rather than the low temp ones (PEM FCs) found in FCVs. The heat from the fuel cells can also be utilized in this model (and is wasted in PEM FVCs), making the process significantly more efficient.

Here's the back of the envelope math, starting with the H2 arriving at the demand center in pipelines and on through the same electric drive train to power the vehicle. This should allow a direct comparison between the two options:

Option A: Hydrogen FCVs utilize the gaseous H2 at fueling stations:

  • Gaseous H2 at filling station...

  • H2 compression for on-board storage: 92.5% (Liquification is much worse at 70% and has a host of problems regarding dealing with boil-off)

  • PEM Fuel Cell: 45% eff (I think that's about average for PEM cells)

  • Useful electric energy for electric drive train...

  • Total efficiency: 41.6%
    Assuming 1 mmBtu of H2 input, we get 416,000 btu to the electric drive train

    Option B: Electric vehicles (or plug-ins) utilizing electricity from central Fuel Cell Power Plants running on gaseous H2:

  • Gaseous H2 at fuel cell power station...

  • Solid Oxide Fuel Cell: 55% efficiency (commercially available SOFCs usually operate in the 50-60% range)
    Electric transmission (very short distance as SOFC power plants can be situated within the city, near the point of end use): 99%

  • AC-DC inverter and charge controller on EV: 95%

  • Li-ion battery storage: 99% (99%+ culoumbic efficiency for Li-ion. This is only 66% for NiMH batteries)

  • Battery self-discharge: 92% (assumes 8% per month self-discharge rate)

  • useful electric energy for electric drive train...

  • Total efficiency: 47.1%
    Assuming 1 mmBtu of H2 input, we get 471,000 btu to the electric drive train.

    So, it looks like nearly a wash here. The SOFC/EV pathway is better but not significantly so. Why make a fuss? Well, I of course didn't figure in the recapture of most of the waste heat from the SOFC plant that goes to waste in the PEMFC pathway.

    We can probably recapture 2/3rds of the waste heat from the SOFC power plant for co-gen (supply it as process heat or steam to a nearby industrial facility or maybe situate each plant within a neighborhood to provide electricity and heat for the whole neighborhood or some such ... SOFC power plants are not shabby neighbors to have, considering they produce zero emissions and produce useful heat or steam for export).

    That means that if we assume the same 1 mmBtu of H2 input as above, we will get 300,000 btu of useful heat on top of the 471,000 btu to the electric drive train (66.7% waste heat capture * 45% SOFC waste heat = 30% useful heat; 30% * 10^6 btu = 300,000 btu). Now we end up with a total efficiency for the SOFC/cogen/EV pathway of 77.1% (i.e. (300,000 btu + 471,000 btu)/10^6 btu). That IS something to make a fuss about as it is a significantly more efficient use of the H2 than the PEM FCV pathway above which only manages to eke 416,000 btu of useful energy out of the H2.

    Finally, as a briefly mentioned above, if HVDC lines can instead be used to reach the remote renewables, this option would probably be even better as the transmission losses from HVDC and gaseous H2 pipelines are just about a wash (with GH2 pipelines having a slight edge). The fuel cell is the bottle neck in both of the above pathways and the HVDC option avoids it completely. The question here, as with most proposed hydrogen uses, is why go through the phase change from electricity to H2 if you are just going to go back to electricity?

    To which Prof Bothun replied:
    There is a lot to respond to here, but I will stick to the first paragraph for now.

    I am already familiar with this paper and I fully agree with it conclusions within the constraints that it imposes upon itself. Any argument involving efficiencies and traditional economies of scale will render Hydrogen as a poor investment strategy and not a wise choice. This is transparent.

    What is left out of the discussion is the following critical point:

    --> What happens when the extant infrastructure for delivery electricity is fully saturated and expansion of that infrastructure is neither economically or technially feasiable. At that piont we will require another mechanism to move energy. This critical aspect is left out of the discussion, continually.

    If such saturation never occurs, then we will be fine, [but] the grid is not scalable forever and we already seeing that. In the same way that when the electiricty goes out on a cold winter night all night, sometime in the future, I will need to use wood as my energy source, so too would I need an alternative like Hydrogen. As long as we can continue to generate and distribute electricity on a grid basis then we should simply invest in mechanism that pump electricity into the grid. This could work.

    Its my personal belief, (and others) that grid saturation and ineffectiviness and unrealiability is going to be a very large problem in the long term (30 years) and will start to be come more of a problem in the near term to the point that will be unable to effectively distribute all the electricity that we are generating. AT that point, everyone will want a hydrogen generator in their house, to kick in when the grid is down. If no one ever believes that the "grid will go down nationally" then there is no need for this proactive solution and investments are better made with just direct electric.

    So yes, if direct electric is our future then hydrogen is not part of the generation equation, for pretty obvious reasons. But, if we are more limited by electricity delivery, then we will have problems and I don't believe that advanced battery storage packs powered by PV's on roofs is as scalable of a solution to this problem as using hydrogen as the means of energy transport. We can generate more "energy units" of hydrogen from large scale renewable projects in remote locatoins, than we can access sufficient material for building enough batteries.

    This obviously plays in the transportation. If you can't recharge your electric vehicle because the grid is down for the day, then your stuck.

    To sum up, if you believe that reliable and continuous electricity delivery will continue far into the future, then the direct electric economy is the right way to go. That could turn out to be the case. The advantage of hydrogen, however, is it can be used by the individuall locally for both electricity and for transportation and so it has flexibility. I liken this very much to the Strategic Petroleum Reserve (which is now down to 35 days capacity). We very much need a strategic Hydrogen reserve.

    Please, weigh in with your comments and we can continue this discussion here. What do you think?...


    Heiko said...

    Very sensible post, as usual. I agree with all you say, basically.

    I am baffled that a Professor of Physics would think that the grid represents a hurdle to EV's. There's more than enough off-peak idle capacity there.

    I am still sceptical about hydrogen + fuel cells for powering cars. But, there are many sensible uses for hydrogen, many sensible uses for fuel cells, and many reasons to look at storage of gases (not least the use of nat gas to power transportation).

    One addition to your math:

    If we are using compressed hydrogen, why shouldn't we be able to recover the energy used for compression?
    There are "compressed air cars", presumably we could power a car from both the energy available from depressurising the hydrogen and the chemical energy available from oxidising the hydrogen?

    Jesse Jenkins said...

    Thanks for the comment Ben. I have also read a couple of Ulf Bossel's papers and agree that he presents several good arguments against a hydrogen economy. In fact Bossel is cited several times in the Carrying the Energy Future paper I linked above. However, I included that paper rather than Bossel's as Bossel writes in a bit more technical and less accessible language than the authors of Carrying the Energy Future.

    I guess I disagree as to the promise of battery electric vehicles. I believe that given the funding currently provided for fuel cell research, the remaining hurdles for advanced batteries could be solved much sooner than the remaining technical challanges for fuel cells. And on the time-frame that fuel cells will be available at, we could have carbon-based ultracapacitors with the energy density of Li-ion batteries ready to go, which represents more or less the end-game in my opinion.

    In the near-term, plug-in hybrids will have the energy density, performance and range US customers are used to and can realize a tremendous savings in petroleum and fossil energy and greenhouse gas emissions in a the near-term future, long before a production-model fuel cell vehicle ever hits the road.

    I also agree that fuel cells have merit in specific situations, like providing electricity (and heat!) for grid-independent communities from renewable sources, but, as you said, the focus for research should be shifted away from transportation.

    Anonymous said...

    Great debate here. I'm sure all of you would like to know about and be quite fascinated to see the new documentary "Who Killed the Electrc Car?" being released by Sony Pictures Classics in June 2006.
    David Lane

    Anonymous said...

    Sorry to interrupt your speculation on the future of the car, but I've a word or two -- sorry if I tread on any toes here.

    I agree basically, myself, but presently I'm disappointed with the work being done on electric cars.

    The problem I find with "plug-in" electrics, primarily, is that if they're focussed on performance, they're terribly impractical anyway -- and either way they eliminate many of the key driving points enjoyed by someone who drives for entertainment.

    Me, for instance.

    When it comes to getting to work, I catch the bus. It's cheap, I don't have to park, and there's exercise to be had.

    But on the weekend I often go, get into my car and blast up a twisty bit of road, delighting in the supple feedback, listening to the revs rise, and just how much fun it is to snick the selector into the next gear -- It puts a smile on my face, and wipes away the dreary work week, the paper and the rigmarole.

    Bear with me, I know this is kind of odd to find in a thread about future cars.

    But the day I give up my irresponsible, planet-killing 2-litre sports hatchback, built in the era of big hair and Devo, is the day there's something harmless that I can get that same rush of sensation from. To that end, I'd love to see BMW continue its hydrogen internal-combustion work.

    Because sure, it's not that efficient as an EV, and it isn't as quiet, and it's not as high tech as a fuel cell powered vehicle, nor does it quietly refuel overnight in my garage.

    But it also isn't a planet-killing petrol-drinking iron horse, that came with a tapedeck installed, and so much vinyl that you think it was made from RCA's head office.

    And it's a car. I'm in love with a car -- and I hope, one day, in love with an H2 powered car.

    I wouldn't mind paying more for the privilege -- and I hope I get that option. Petrol even smells horrible. I hope I can live without it soon.

    Jesse Jenkins said...

    Neal, what it looks like your looking for is one of these babies!

    As for the H2-ICE BMW, while its not a 'earth killing' gas guzzler, you've got to be very careful where the hydrogen comes from. If it comes from electrolysis of water using electricity from the US average generating mix, its actually quite a bit worse than gasoline in terms of total fossil energy inputs as well as greenhouse gas emissions and several criteria pollutant emissions.

    If it comes from natural gas, its about as good as petroleum, except now we're reliant on natural gas and North American supplies are already tight while we use basically none of it for transport. If we start using natural gas for transport, we're gonna have to import it and guess who's got natural gas: all those same very stable countries that have oil left - Russia, Iran, Qatar, Saudi Arabia, United Arab Emirates, etc...

    Basically, unless that hydrogen comes from renewables, it isn't the wonder drug we're looking for.

    (In contrast, because electric vehicles and plug-ins are so efficient, they can overcome the predominatly fossil fueled and GHG-intesive nature of the U.S. electricity mix and offer reductions in not only oil use but also fossil energy use, GHG emissions and emissions of most criteria pollutants - this is all based on my research for my thesis which I will post online soon....)

    Anonymous said...

    I hope the Venturi has a gearbox. That's just essential for me by now! Engine braking that gives me back power sounds nice, too --

    But anyway, I live in Australia. We've got a whole lot of LNG and LPG so hydrogen here's not so bad an idea, and is a little less dirty in the end. Turbo LPG is an option, but again, planet-killing and noxious. .. if fun. And we may be going nuclear, soon, which is a bit controversial, but most would agree fairly clean power, for electrically-derived H2.

    I'd love a car that ran on ethanol, which used say, geothermal heat in its production... Even solar heat, as it's not specifically electricity that's needed.

    Heck, my car could run on ethanol! And I could keep my tape deck, too.

    Anonymous said...

    What would the efficiency of H2 ICE's be like? ICE's tend to be wasteful contraptions, but perhaps designing it specifically, from the ground up, for H2 combustion could yield higher efficiency? If these things are just toys (for sunday drives) it won't matter much of course... but 20% ICE average combined with 70% electrolysis plus trans losses plus storage... that's hideous!

    Jesse Jenkins said...

    Anonymous, the efficiency of an H2 ICE engine would presumably be roughly equivalent to the efficiency of a traditional gasoline-powered ICE. I believe that H2 ICE engines would be spark-ignition engines (like your gasoline engine - Otto cycle engines) rather than the more efficient compression-ignition (Diesel cycle) ICEs.

    That means you're probably looking at a maximum efficiency of about 30-35% for an H2 ICE engine and perhaps as low as 15-20%.

    And as you said, combining 70% electrolysis + transmission losses + 15-35% efficienct H2 ICE engine = hideously wasteful!