All Night Long

August 1, 2008 4:51 AM

MIT researchers have overcome a major barrier to large-scale solar power: storing energy for later use.

"If you can only have energy when the sun is shining, you're in deep trouble. And that's why, in my opinion, photovoltaics haven't penetrated the market," Daniel Nocera, an MIT professor of energy, said in an interview at his Cambridge, Massachusetts, office. "If I could provide a storage mechanism, then I make energy 24/7 and then we can start talking about solar."

Solar has been growing as a power source in the United States -- last year the nation's solar capacity rose 45 percent to 750 megawatts. But it is still a tiny power source, producing enough energy to meet the needs of about 600,000 typical homes, and only while the sun is shining, according to data from the Solar Energy Industries Association.

Most U.S. homes with solar panels feed electricity into the power grid during the day, but have to draw back from the grid at night. Nocera said his development would allow homeowners to bank solar energy as hydrogen and oxygen, which a fuel cell could use to produce electricity when the sun was not shining.

"I can turn sunlight into a chemical fuel, now I can use photovoltaics at night," said Nocera, who explained the discovery in a paper written with Matthew Kanan published on Thursday in the journal Science.

posted by chuckdarwin (52 comments total)

If you can only have energy when the sun is shining, you're in deep trouble.

Thus explaining why there is so little plant life on earth.

Pardon my rant, but I see this ridiculous argument everywhere. "If it can't solve 110% of the problem, it's crap!" Listen. Diversity of energy sources is good. We can take baby steps. Get 10% of our energy from solar, even if only during the day. Get another 10% from geothermal, but only in volcanic areas. Get another 10% from biofuels, but only in areas where we can grow/harvest them. Etc. It adds up. It gets us away from an energy monoculture where increases in gas prices hit the consumer multiple different ways (i.e. it not only costs more to drive to the store, but the food that's there is also through the roof).
posted by DU at 5:06 AM on August 1, 2008 [12 favorites]

In the absence of other material, one link would probably have done it; the rest is largely repetition of the same press release.

The CNET commentary is the most revealing. John Turner from the NREL injects a sensible dose of caution between doses of hype.

Also: "It's not totally practical, but because it's easy, people are already all over it," Nocera said. "Being a scientist, I can play outside the rule box." No Daniel, you can't. It needs to be practical above all.

Still, progress is progress, all that aside.
posted by le morte de bea arthur at 5:07 AM on August 1, 2008

If I could provide a storage mechanism, then I make energy 24/7 and then we can start talking about solar

Doesn't this mechanism already exist in the form of batteries? As far as I can tell none of the articles mention using batteries to store captured solar power, and how this would be better than that. It seems like new process would need to be more efficient and/or cheaper than existing battery storage for it to be useful.
posted by burnmp3s at 5:58 AM on August 1, 2008 [1 favorite]

This is great news.

Let's hope this breakthrough makes it to the little man.
posted by AzzaMcKazza at 6:02 AM on August 1, 2008

I was kind of reminded of the "Hydrogen Hoax" article which was posted in the comments of this thread. The gist of the article is that hydrogen power will not become anything like affordable while it is so expensive to make hydrogen. If this provides a dramatically cheaper method then great - however I wonder if the requirement for expensive metal catalysts at both the splitting and re-combining stages could be a significant problem.
posted by rongorongo at 6:03 AM on August 1, 2008

In the late 90s UK energy firm National Power (now part of RWE npower) developed a 100MWh energy storage system called Regenesys.

I understand it was shelved because of cost. However increasing energy prices globally will make energy storage more economically viable.

It's clearly also massively useful for wind or wave power, which are both internittent in their load.

Strange though, I always thought the main reason photovoltaics have not succeeded is that they themselves are too expensive to produce.
posted by mossiman at 6:09 AM on August 1, 2008

How on earth is either storing energy from solar or using solar to create hydrogen a new concept? Large-scale solar is probably better served by thermal storage IMO (cracking water is too expensive) but in any case this is kind of, you know, obvious.
posted by Skorgu at 6:10 AM on August 1, 2008

le morte de bea arthur writes "It needs to be practical above all."

Most inventions aren't practical at first, especially those requiring an infrastructure: automobiles required macadam roads and petrol stations, the telegraph and phone and broadband internet required stringing copper or cable, steamships required coaling stations, the electric light required power stations, transformers, electric linesand wiring homes -- and each of thos without the other was merely a useless curiosity.

But as in all those cases, you start in a few places, and you build out. There were electric carbon-arc lamps before Edison, but they were dim and ugly and burnt out quickly. That's existing photovoltaic solar. With this breakthrough (and the orange glass also invented at MIT), we're getting to the Edison point for solar, where it's now worth investing in.

It'll be a novelty for a few more years, then something only the enthusiasts have, then something the rich have to show off with, and then -- like the light bulb -- it'll be a commodity item that everyone purchases at Wal*Mart.
posted by orthogonality at 6:11 AM on August 1, 2008 [5 favorites]

Isn't this essentially the same
idea.
posted by evilelf at 6:13 AM on August 1, 2008

Metafilter post 2052: "Peak Platinum."
posted by netbros at 6:19 AM on August 1, 2008

Most inventions aren't practical at first...

I would go even farther. Most inventions shouldn't be practical at first. With the constraint of practicality at every step, we can never escape a local tech maxima. But with the ability to take backwards steps in practicality, we can rich a more global maximum.
posted by DU at 6:31 AM on August 1, 2008 [4 favorites]

I wonder if the requirement for expensive metal catalysts at both the splitting and re-combining stages could be a significant problem.

There are small quantities of platinum in automotive catalytic converters. Not so small as to be unattractive to thieves, apparently.



Large-scale solar is probably better served by thermal storage IMO (cracking water is too expensive)...

If this works out, cracking water will be free.
posted by Kirth Gerson at 6:36 AM on August 1, 2008

Thus explaining why there is so little plant life on earth.

Ahem. The *whole* point of photosynthesis? Storing energy. It takes sunlight as the primary energy source, various nutrients from the ground, and CO₂, and produces carbohydrates and oxygen. Those carbohydrates are stored, and allow the plant to live when there isn't adequate sunlight. Indeed, since single source energy is easier, the plant doesn't use sunlight directly at all -- it converts to carbohydrates, and uses those. If sunlight's plentiful, it'll store the excess.

If you relied on Sunlight alone for the energy needed to sustain life, you'd die at dark. If you rely on sunlight to *produce* energy sources that are storable, and use those stores at night, you live through the night. Indeed, a healthy plant can easily live for a week without sunlight -- by living off the stored energy in the form of carbohydrates in the plant.

Why do you think fruits are sweet? Stored Carbs.
posted by eriko at 6:41 AM on August 1, 2008 [7 favorites]

Why do you think fruits are sweet? Stored Carbs.

But not for the plant. This is like saying the whole point of the manufacturing industry is storing energy, since you can extract energy by burning children's toys and clothing.

However, you do have a point--the plant does store energy for sunless times. However, a plant has kind of limited options. Humans have more options. It certainly makes no sense at all to eschew a major energy source simply because it solves less than 100% of the problem. That's the classic definition of the perfect being the enemy of the good.
posted by DU at 6:48 AM on August 1, 2008

yeah, better example would be maple syrup.

mmm, pancakes.
posted by yort at 6:53 AM on August 1, 2008

I wonder how long it will be before big oil and big energy make claims of this being too expensive and unreliable.
posted by Fizz at 6:56 AM on August 1, 2008

Here is an extract from the Science article that describes what they have done in a little more detail.

"The hard part is finding catalysts that can orchestrate this dance of electrons and protons. The anode, which links oxygens together, has been a particularly difficult challenge. Platinum works but is too expensive and rare to be viable on an industrial scale. "If we are going to use solar energy in a direct conversion process, we need to cover large areas," Turner says. "That makes a low-cost catalyst a must." Other metals and metal oxides can do the job but not at a neutral pH--another key to keeping costs down. In 2004, Nocera's team reported in the Journal of the American Chemical Society a cobalt-based catalyst that did the reverse reaction, catalyzing the production of water from O2, protons, and electrons. "That told us cobalt could manage multielectron and proton-coupled reactions," Nocera says.

Unfortunately, cobalt is useless as a standalone water-splitting anode because it dissolves in water. Nocera and his Ph.D. student Matthew Kanan knew they couldn't get over this hurdle. So they went around it instead. For their anode, they started with a stable electrode material known as indium tin oxide (ITO). They then placed their anode in a beaker of water, which they spiked with cobalt (Co2+) and potassium phosphate. When they flipped on the current, this created a positive charge in the ITO. Kanan and Nocera believe this initially pulls electrons from the Co2+, turning it first to Co3+, which pairs up with negatively charged phosphate ions and precipitates out of solution, forming a film of rocklike cobalt phosphate atop the ITO. Another electron is yanked from the Co3+ in the film to make Co4+, although the mechanism has not yet been nailed down. The film forms the critical water-splitting catalyst. As it does so, it swipes electrons from hydrogen atoms in water and then grabs hold of lone oxygen atoms and welds them together. In the process, the Co4+ returns to Co2+ and again dissolves into the water, and the cycle is repeated.

The catalyst isn't perfect. It still requires excess electricity to start the water-splitting reaction, energy that isn't recovered and stored in the fuel. And for now, the catalyst can accept only low levels of electrical current. Nocera says he's hopeful that both problems can be solved, and because the catalysts are so easy to make, he expects progress will be swift."
posted by zeoslap at 6:58 AM on August 1, 2008 [1 favorite]

DU, I don't think it's possible to just get rid of the "idea of" an energy monoculture that fuels general havoc and wars. We can't take baby steps, because that monoculture is a big ugly giant that will stomp any baby flat. There needs to be a replacement monoculture that fixes some of the problems associated with the present one but which can't be easily stomped. And honestly, this sounds pretty damn awesome to me.
posted by gorgor_balabala at 7:07 AM on August 1, 2008

When they start taking maple syrup to make fuel, I'm joining the armed insurrection.
posted by Kirth Gerson at 7:19 AM on August 1, 2008

There are only four(!) primary sources of energy available:

1. Solar, which includes:
a. Heat (thermal) energy
b. Evaporation energy (ie. rain forms rivers which turn water wheels)
c. Wind and water currents
d. Plant and animal decay (biogas, etc..)
e. Fossil fuels (decayed plants fossilized)

2. Nuclear fuels

3. Geothermal heat

4. Motion of Earth and Moon (tidal energy)
posted by stbalbach at 7:31 AM on August 1, 2008

Huh, there seems to be something missing: the MIT article doesn't say that the cobalt catalyst is a replacement for platinum. It says it's the oxygen catalyst. So what is the platinum replacement?
posted by gorgor_balabala at 7:45 AM on August 1, 2008

stbalbach: If we want to be technical, the Motion of Earth and Moon are primarily solar. In a gravitational sense.
posted by Ian.I.Am at 7:47 AM on August 1, 2008

Thanks zeoslap, that makes more sense. I really don't get why this story is being presented as having anything to do with solar energy at all. If it works and is easily manufactured it'll be great for any kind of energy storage. The mainstream articles don't even /mention/ the catalyst.
posted by Skorgu at 7:50 AM on August 1, 2008 [1 favorite]

If we want to be technical, the Motion of Earth and Moon are primarily solar.

No, pure momentum -- all the sun does is (constantly) change the vector. We could replace the sun with an non-radiating lump of the same mass, and the Earth/Moon System would still have the same KE.

Technically, Nuclear fission is stellar power -- not solar power, but the materials we use for fission reactors were created in stars. Geothermal is partially solar -- the materials were heated as they orbited the sun before the planet aggregated, but that's probably a minor component.
posted by eriko at 7:55 AM on August 1, 2008

Besides what DU said solar is a good match for A/C load. Many places you need A/C when the sun is shining and then the load drops dramatically with the setting sun.
posted by Mitheral at 7:59 AM on August 1, 2008

I'm really really really hoping for affordable solar, which has been said by many to be just in the offing. I'll even go without storage capacity and draw from the grid at night. I'd still save a hell of a lot of money that way. I live in southern Arizona, so if there's one thing we have in abundance, it's sunlight. We get 350 days a year of sun here. The sun, of course, is the biggest driver of our intense summer heat, and so daytime is when daily power demand spikes. I run a swamp cooler, so my power demands are lower than most people running AC. But I do have to keep it running on high in the summer. To be able to run that - and everything else in the house - off of solar during the day would be awesome. Affordable batteries would be icing on the cake. (Yes, battery systems do exist already, but they are big, clunky, expensive, and they don't store enough energy for the typical homeowner, at least not without learning to manage power demands.)
posted by azpenguin at 8:00 AM on August 1, 2008

This kind of process (water photolysis using a catalyser) is rather well known. What all this seems to be about is merely the development of a somewhat better catalyser. It may be interesting, but unfortunately it seems to be buried under a ton of hype. Why do I smell an IPO and a skilled press officer behind all of this?
posted by Skeptic at 8:07 AM on August 1, 2008

Doesn't this mechanism already exist in the form of batteries?

Hydroelectic dams routinely pump water back up into the lakes behind them at night when the power grid is underutilised. TVA has been doing this for decades.

Storing energy like this is cheap, clean, completely low-tech, and highly efficient. Unused solar energy could easily be used to pump water above some turbines to be released later when the sun goes down.

Leave it to the people at MIT to come up with a complicated solution to a simple problem.
posted by three blind mice at 8:08 AM on August 1, 2008

then something the rich have to show off with

I remember seeing on TOH or some other renovation show that, in the first houses that were electrified, the outlets were placed at head height so they would be obvious and could be shown off to guests.
posted by Turtles all the way down at 8:10 AM on August 1, 2008

Skorgu I really don't get why this story is being presented as having anything to do with solar energy at all.

The main energy source is light (I say "main" because apparently they also need some electricity to make it work). It is about turning light into hydrogen, hence the solar connection.
posted by Skeptic at 8:12 AM on August 1, 2008

Metafilter post 2052: "Peak Platinum."

You joke, but we're already mining copper, tin, precious metals, etc. to the point where there really will be a supply problem in the next few decades. Once all the metals are stripped from the Earth's surface, there's not much to do but recycle what can be reclaimed, with diminishing returns.
posted by Blazecock Pileon at 8:14 AM on August 1, 2008

three blind mice Leave it to the people at MIT to come up with a complicated solution to a simple problem.

Actually, the problem they are adressing is a complex one, and the solution quite simple. Unfortunately both points are lost in the breathless coverage.

First to the problem:

The problem is not simply energy storage, but about portable energy storage. Unfortunately, while dams are quite efficient in producing and storing energy, they are not very portable. Batteries also have their own drawbacks (weight, efficiency, logistics, etc.) These are the reasons why fluid fuels are popular, in particular for vehicles. Unfortunately, most fluid fuels used today are non-renewable fossil fuels, and contain carbon. Hydrogen has long been seen as an alternative, even if it also has a few drawbacks, the biggest one being that the Second Law of Thermodynamics dictates that it requires more energy to be produced than it will return in use. How much more? Currently, a lot more. Producing hydrogen from the most abundant renewable source of energy available (sunlight) is massively inefficient. That's because it requires turning sunlight into electricity first, usually through very inefficient and expensive photovoltaic panels, and then electricity into hydrogen through also very inefficient water hydrolysis.

Now to the solution:

The PV+hydrolysis combo means that you first turn sunlight into chemical energy (in the semiconductor layers of the solar cells) and this chemical energy into electricity, which then you turn back into chemical energy in the hydrolysis. And you lose energy in each step, thanks to that cursed Second Law. Now, what if it was possible to cut out the electricity step altogether, just as it is done in photosynthesis, so that the sunlight directly splits the water? Bingo, it turns out that there are substances that act as catalysers in just such a "photolysis" reaction.
posted by Skeptic at 8:30 AM on August 1, 2008 [6 favorites]

The breakthrough here seems to be in finding a functionally stable catalyst for one of the two terminals... which is certainly important - but electrolysis of water is nothing new by itself.

Is this really a major breakthrough in power storage, or just a small step towards a hydrogen economy?
posted by TravellingDen at 8:32 AM on August 1, 2008

This is interesting, some of the storage types (strictly thermal) I've read about recently were molten salts as they hold their heat load very well when the sun is on the other side of the blue ball.
posted by iamabot at 8:37 AM on August 1, 2008

Skeptic: That makes more sense, thank you. Will someone go kick the media in the teeth and teach them to do science journalism?
posted by Skorgu at 8:38 AM on August 1, 2008

Very cool, but I'll believe it when the researchers start committing suicide or die in single car accidents.
posted by Mr_Zero at 8:58 AM on August 1, 2008 [2 favorites]

Solar Hydrogen Energy Company has been trying to commercialize photolysis for years, with no luck. Maybe they've got the wrong catalyst.
posted by anthill at 9:45 AM on August 1, 2008

Yeah the crux of the innovation is not only that this catalyst is cheap, abundant and resistant to impurities (all nice things that would make a practical device simpler and cheaper) but that it purportedly uses ~90% less energy to split water. Having a more efficient catalyst will enable cheaper hydrogen because it will take far less power to produce. That's the breakthrough. Hopefully the publication of the results will show rigorously how much more efficient it is.
posted by no_moniker at 9:46 AM on August 1, 2008

Skorgu : Actually, I've had a second look at the articles, and they appear to be about electrolysis, not photolysis as I thought. So you were right, this actually has nothing to do with solar energy. Moreover, the only apparent advantage is having a better efficiency in neutral water.
Revolutionary? I don't think so.
posted by Skeptic at 9:48 AM on August 1, 2008

Unfortunately, while dams are quite efficient in producing and storing energy, they are not very portable.

*Scratches plan for dam-powered car.*
posted by kirkaracha at 9:50 AM on August 1, 2008 [2 favorites]

This is pretty slick, now we just need to come up with some clever ways of selling it when it finally comes to market. I'm going to have to trademark 'fauxtosynthesis', 'roboplant', and 'cobaltalyst'.
posted by quin at 9:54 AM on August 1, 2008

Skeptic the articles are all remarkably unclear about what, exactly, the "breakthrough" is. They all make it sound like nobody has ever thought of using solar power to generate hydrogen. The EETimes makes it sound like it's just really really good electrolysis. Which is nice, but phrases like almost 100-percent efficiency make my bullshit detector sit up, bark and fetch the morning paper.

So, in conclusion, I'll believe it when I can buy one at Home Depot and as usual science reporting fucking sucks.
posted by Skorgu at 9:58 AM on August 1, 2008

Moreover, the only apparent advantage is having a better efficiency in neutral water.
Revolutionary? I don't think so.

Please correct me where I'm wrong skeptic and others, but isn't the whole point that this new technique of electrolysis to produce hydrogen effectively makes storing the power generated as electricity by conventional photovoltaic cells massively more productive and efficient? Since efficient long-term storage of solar-generated energy has been the major short-fall of the technology all along (and currently, the only way to store solar energy is by converting it and storing it in hydrogen fuel cells through much less efficient methods), this seems to go a significant way toward solving that problem and making solar energy more practical, doesn't it? How is that not exactly the break-through the technology's been waiting for? Production of photovoltaic cells has seen dramatic gains in efficiency and affordability in recent years, and most accounts I've seen agree the only remaining challenge is the efficient long-term storage problem. This potentially solves that. Or if it doesn't, please explain why. And don't tell me it's because solar power can't be used to generate the electrical charge used in the electrolysis process that produces the hydrogen. This guy who already produces hydrogen to run his car using a less efficient form of solar-powered electrolysis would disagree, as would anyone who's ever wired their stomper 4X4 truck to run off a solar cell instead of a battery for an elementary school science fair when they were a kid. And don't tell me it's just because it "sets your bullshit detector off" either. Moon landings set some people's bullshit detectors off, too. BS detectors, while useful, are piss-poor substitutes for a good argument.
posted by saulgoodman at 10:11 AM on August 1, 2008

the only way to store solar energy is by converting it. Solar energy can be stored the same way normal electricity can, batteries, (super/ultra)capacitors, SMES, pumping water uphill, whatever.

This sounds like a very efficient way to generate hydrogen via a catalyst. Normal electrolysis is either horribly inefficient or insanely expensive and twitchy. If this does turn out to 1) exist, 2) be readily mass produced and 3) be economically viable it will solve one aspect of the hydrogen-as-battery problem.

There are, of course, others. Hydrogen is a pain in the ass to store. By weight H₂ is a very efficient energy storage, but by volume it's ridiculous. You need tens of thousands of psi if you want to store it in gaseous form, stupidly expensive cryo setups if you want to liquify it, all sorts of complex, fiddly, expensive bits to transfer the stuff and all that storage stuff is heavy so you're eating away at your joules-per-kg.

There's some really fascinating stuff being done with metal hydrides among others but none of them are even fully sorted experimentally much less commercially viable.

The Hydrogen Economy is Nuclear energy wearing a hat and I'm all for that. This is one part to a very complicated chain of events that has to be complete from cradle to grave before it can really take off. My bet is on biologically generated synthetic hydrocarbons personally, and I'm sure a cheap, easy source of H₂ could be useful in that process.

But again, the paucity of details and the breathless prose does set off my finely-honed bullshit detector and as usual it's the burden of those making extraordinary claims (*cough* near 100% efficiency *cough*) to provide extraordinary evidence.
posted by Skorgu at 11:04 AM on August 1, 2008

good clarifications, skorgu. thanks!
posted by saulgoodman at 11:57 AM on August 1, 2008

Very sweet!
posted by jeffburdges at 2:35 PM on August 1, 2008

The NPR blurb I first heard this via had exactly Zero interesting information in it. Breathless huff huff MIT huff wheeze solar energy, the most friendly green technology available. For a second there I was afraid I'd actually have to go look up the article.

And, Indium isn't exactly cheap (not to mention, it's a toxic heavy metal).
posted by gensubuser at 5:30 PM on August 1, 2008

the really important innovation here is the catalytic action of the base metal cobalt, which is a huge deal in terms of the potential for widespread use.

the platinum group metals used in e.g. automotive catalysts are undergoing an enormous price spike (don't you wish you'd bought some rhodium back in 2003?), and there is a big push to develop efficient zero-precious-metal catalysts.

now, if only it weren't toxic!
posted by sergeant sandwich at 6:39 PM on August 1, 2008

What's with all the articles claiming that photosynthesis is cracking water? Photosynthesis is splitting CO₂ in to O₂ and C for use in cellulose.

So it's efficient water-cracking. Nice. But it still costs way too much to store and transport H₂, which is the primary drawback to a hydrogen economy. I'd be much more impressed if they had an efficient way to safely dissolve hydrogen (like we do with acetylene for welding) for transport, or if they'd made a more-efficient PV cell or directly cracked water with solar radiation.

Cracking water from a PV source is still not so great when the PV cell is only about 20% efficient and then what does one do with all the hydrogen so produced?
posted by polyglot at 8:48 PM on August 1, 2008

then what does one do with all the hydrogen so produced?

Zeppelins!
posted by TwelveTwo at 1:03 AM on August 2, 2008

This is all well and good, but what we really need is a single zero maintenance system that efficiently stores solar energy in an accessible, safe and stable form.

Ideally, these terrestrial regenerative energy efficient systems would have negative carbon emissions and when fully deployed almost universally accepted by the populace as beautiful. If we are really going to push the design brief, then we should demand tasty food and versatile building materials as a by-product.

Back to the drawing board chaps.
posted by verisimilitude at 2:08 PM on August 2, 2008

Some more analysis of the claims. Assuming JoulesBurn (*ehem*) has his math right (which I haven't checked) I don't even see that this warrants more than a one-liner in some obscure journal.

#include <std_media_science_rant.h>, etc.
posted by Skorgu at 11:09 AM on August 6, 2008