Hydrogen from Sunlight

John Read wrote in this ABlog post (I hate group blogs):

The Aussies have done it. And not a moment too soon! Using special titanium oxide ceramics, they can now use sunlight to split water, and produce hydrogen fuel for unlimited energy. The device rivals still-unrealized “controlled fusion”

This is an interesting development in the research area of Photoelectrochemical water splitting, but it's not about whether you can get hydrogen from sunlight. You could already do that, if you were willing to go from photovoltaics to electrolysis, or use solar energy for thermal water splitting. So, I think John is a bit over-enthusiastic.

The real question is whether you can produce energy at economical rates, and I haven't seen any information on that. Given a competitive efficiency, it's likely that you could create such a system that's quite a bit cheaper than photovoltaics, but you still have to deal with the fact that solar energy is a fairly diffuse energy source. Time for a little math - how much space would it take to support an average household. Feel free to check my numbers...

Picking a spot in California as a test case, we find that it averages about 5 KWh / sq meter / day.

The average US household uses around 12000 KWh / year = 32.8 KWh / day

So, if our system were 100% efficiency, we'd need about 6.5 meters of collector area. That sets a lower bound. Picking some more reasonable numbers:

Light to hydrogen efficiency = 10%

This seems like a fairly reasonable guess. Photovoltaics are in the same realm, and photosynthesis clocks in at around 8% for efficient plants.

Hydrogen to electricity efficiency = 75%

This is a harder number to come up with. There are claims of up to 80% here, so I'm going to choose 75% as a reasonable number.

Power inverters (to go from DC to AC) are approximately 90% efficient.

So, that makes our sunlight to AC power efficiency =

0.1 * 0.75 * 0.9 = 0.0675

Dividing our 6.5 meters by this factor gives us around 96 square meters of collection area required to take a California house off-grid. That's a collector that's properly aligned with the sun - being at the wrong angle reduces the efficiency.

That's a pretty large amount of space.

If, however, you can do this in large desert areas, you can get a megawatt of daytime hydrogen out of a collector that's about 150 meters square.

There are also some interesting space applications. Presuming you can get water from asteroid sources, you could perhaps generate hydrogen and oxygen fairly simply.

Comments (21)

  1. Jerry Pisk says:

    This all nice but the problem with electricity is that it’s not easy to store for later use. So you would need to plug your house back into the grid at time when you use a lot of electricity – after dark, when you get home from work.

  2. Kevin says:

    You can always use a flywheel (http://www.powerware.com/UPS/PF2_Flywheel.asp)

    I read a lot about Kinetic storage about 5 years ago, but not much recently.

  3. Per Soderlind says:

    An (better?) alternative is to get hydrogen from the wind, take a look at the Utsira Project:


  4. Jerry: one reason this may be better than simple photovoltaics is that hydrogen could be easier to store than electricity.

    Disconnecting from the grid sounds good but this is not going to happen for a long while. However, adding an auxiliary power source to a significant number of houses would be quite an improvement.

    For example, it should be possible to power air conditionning units from solar energy without drawing from the grid at all. In general, the more sun you get the more energy you get, but also the more air conditionning you need. Add a relatively efficient way to store some of the energy (like hydrogen-absorbing compounds) to tide you over through the night and you’re all set.

    But this is more of a policy problem than a technology problem. It’s already possible to save a lot of energy heating water big stringing a few tubes between two glass plates on housetops, yet nobody does it because the price of electricity doesn’t reflect its true cost (including environmental and political costs).

  5. Chris says:

    Who says the process needs to occur on Earth?

  6. jaybaz [MS] says:

    "That’s a pretty large amount of space. "

    96 square meters is just over 1000 square feet. (Thanks google.) Thinking about my house, that’s less than the area my house occupies. So if I cover the south-facing parts of my roof with collectors, I can probably cover 1/2 of my energy consumption right there.

    "The real question is whether you can produce energy at economical rates"

    There’s a lot more complexity, though:

    – How much energy does it take to manufacture and install "special titanium oxide ceramics"? Is it anywhere near how much energy they will harness?

    – How much waste is produced to manufacture the collectors, and dispose of them when they die? How does that waste compare to the waste from other energy sources?

    – Given the possibility of getting 80% of your energy directly from personal solar collectors, would you be willing to change how you live to reduce energy consuption by 20%? Perhaps living "off the grid" is worth that.

    – As resources dwindle, and global warming increases, we’ll find both financial and environmental motivations to build more energy efficient homes.

    – If research raises the light->hyrdogen efficiency from 10% to 15%, the whole picture changes.

    – Much of our energy is used to create heat. Going solar->H2->electricity->heat is always going to be less efficient than just collecting solar heat. By splitting my sun-facing roof surface between heat collection & hydrogen collection I should get an incremental improvement.

  7. Jim Argeropoulos says:

    Auxiliary power sources bring a host of issues all their own. One of the main concerns is how to turn them off when you need to do maintenance. Now this may not be an issue for you and your home, but if not installed correctly, they can be a serious hazard for utility workers.

    That said, I would love to be able to afford solar and wind power for my home.

    My priorities would be

    1. better insulation

    2. passive solar (I live in a Northern/cooler climate)

    3. wind generator fair distant third

    4. active solar a long distant fourth

  8. Darren Oakey says:

    as mentioned above, 96 squares doesn’t sound infeasible when you think about roof space. Throw on a windmill in the middle, get some hamsters in treadmills, and you’re pretty well there 🙂

    Trouble is, at the moment, it’s cheaper just to get the electricity from the grid 🙁

  9. DM McGowan II says:

    Stirling Engines.

  10. Kal says:

    Your math may be OK, you do lose something at every transition step. But no one would design a alternative energy powered system that continues to use/waste all that energy.

    With good design most of the heat/cool costs can be eliminated.

    And think of the power grid as one large battery. Your individual solar may have times when it is producing excess and will have times when it is not producing enough. Wind can fill some of those gaps. Every source will play a part.

    Current cost is important if you are making a purchase, but for societal planning purposes it makes sense to apply the semiconductor cost curve model. Currently the quantity of solar being produced is trivial compared to the need and the cost reflects that.

    If a large percentage of the population has solar collectors on their roof, and if we tie it all together with a fairly efficient grid, we can produce a large percentage of our rationally used electricity.

    All that is lacking is vision and the will to replace oil/gas/coal with something more sustainable.

  11. Eric Brown says:


    The power grid is *NOT* one large battery. At any given moment, supply MUST equal demand. If supply exceeds demand to any significant degree, or if demand exceeds supply, circuit breakers trip to prevent large transformers from exploding. Sometimes the circuit breakers don’t trip soon enough.

    What’s lacking from alternative energy is a sufficient energy density so that we don’t need to pave the Mojave Desert.

    You might try reading _The Solar Fraud_, by Howard Hayden (available at Amazon.com), or some of Steven Den Beste’s essays on alternative energy, to see that vision and will are no match for cold hard equations.

  12. Ron the Swan says:

    I tried reading _The Solar Fraud_ but found the integrity of the author sorely lacking. Sorry. Howard is clue-less.

    Anyone who thinks the sun’s energy isn’t dense hasn’t been to the tropics lately. Energy from sunlight hitting the earth in one 24 hour day is greater than all oil _ever_. All the electricity for the USA could come from photovoltaics using an area 100 miles square. It doesn’t all have to be in one place. There is plenty of desert space available. It can be found over our heads (where less grows than in the Mojave by far) in deserts that are also known as "roofs".

    Humanity is about to experience an unparalleled transformation away from fossil fuels. We have no time to spare. Hydrogen is not a source, and its storage is very ineffecient, so it will play only a limited role (except in presidential speeches).

  13. Kal says:


    From the point of view of your individual power needs the grid is a battery equivalent.

    Of course total supply has to equal total demand. The same is true of a battery. But at any point in time and even in the aggregate, your supply to the grid does not have to equal your demand from the grid.

    Cold hard equations often yield answers unrelated to the real questions.

Skip to main content