Harvard Grid Computing Project Discovers 20k Organic Photovoltaic Molecules

Lucas123 writes "In June, Harvard's Clean Energy Project plans to release to solar power developers a list of the top 20,000 organic compounds, any one of which could be used to make cheap, printable photovoltaic cells (PVCs). The CEP uses the computing resources of IBM's World Community Grid for the computational chemistry to find the best molecules for organic photovoltaics culled the list from about 7 million. About 6,000 computers are part of the project at any one time. If successful, the crowdsourcing-style project, which has been crunching data for the past two-plus years, could lead to PVCs that cost about as much as paint to cover a one-meter square wall." The big thing here is that they've discovered a lot of organic molecules that have the potential for 10% or better conversion; roughly equivalent to the current best PV material, and twice as efficient as other available organic PV materials.

Comments

Organic compounds

[girlintraining]

Yes, it could lead to an organic compound that could do that. It could also lead to an organic compound like the one recently installed into BMWs that, when exposed to fire, converts in an aerosol of the deadliest acid known to man. It was marketed as a "green" alternative to existing refridgerants... and it was approved by the EPA. Twenty thousand molecules sounds impressive -- but the odds of finding one that meets safety requirements and is still effective isn't good. Pharmaceutical companies test thousands of compounds every year... and very, very few of those find a medical application. It's the same story here.

So yes, good first step. Good exploratory research. Don't get your hopes up.

Re:Organic compounds

[kromozone]

Are you referring to 2,3,3,3-Tetrafluoropropene? That HF production scenario involved Daimler spraying HFO-1234yf over a burning hot engine block. The conditions were tuned to disqualify it. There's a bit more to that story than the surface. German industry vs. US industry pushing different alternatives and each trying to warp the science their way. PVs aren't going to be aerosolized and sprayed over 500C engine blocks while mixed with compressor fluid. Considerably easier to predict the behavior of an organic molecule in this case.

Re:Organic compounds

[girlintraining]

That HF production scenario involved Daimler spraying HFO-1234yf over a burning hot engine block.

Okay, am I the only one that thinks that putting a chemical that, when exposed to high heat or fire, converts to one that can cause death if it comes in contact with a patch of skin smaller than the palm of your hand for a few seconds in a car's engine compartment is a really dumb idea? In the event of a front-end collision, you've got shit spraying and leaking everywhere, smoke, flames, people dead, dying, or injured... and you're suggesting that we should introduce into an already inherently dangerous situation for first responders to walk into... the risk of exposure to an airborn acid that can kill them if they come in contact with it and likely wouldn't know at the time they did?

I'm sorry, but I'm with Congress on this: The woman that approved this was a flaming retard that, on no account, should be put in a position of authority over approving other compounds that could potentially save a company a few bucks at the expense of people's lives and health.

Re:Organic compounds

[girlintraining]

Can you provide a citation that doesn't come from a fear-mongering rag of an excuse for journalism?

Sure can!

(trollface)

Re:Organic compounds

[Anonymous Coward]

Seeing as the current refrigerant we use will have equally BAD results if tested in this manner, I wouldn't go nailing anyone to any crosses just yet.

Don't believe me? Run out, buy a can of air duster, make sure you are in a room without much air circulation, turn the can upside down and light the resulting liquid spray on fire. Breath in REAL deep.*

They both make into Hydrogen Fluoride when burned, just add water for your dreaded hydrofluoric acid. Hey, wait a minute, aren't your lung tissues made up of lots of water?

To further make a point: The Dymler engineers mixed the HFO1234yf with compressor oil to increase its burning potential, then sprayed it over a large area on a hot engine block. In our experiment up top we ignited pure R-152a with nothing but a Bic lighter. Food for thought.

*Don't do this.

Re:Organic compounds

[gman003]

Now, I'm not particularly good with metric, nor am I particularly experienced with engine repair (having done nothing more complex than replacing a water pump), but I would think 500C is a bit unusual for an engine to operate at. That's roughly 900F, well above the melting point of, say, lead, and getting close to that of aluminum or magnesium.

According to some brief googling, the typical operating temperature for an engine is under 250F (120C), and gasoline auto-ignites at 280C (540F). So by the time your engine block has reached 500C, you should already have run a good ways away.

Not to mention that, just by the name, tetraflouropropene sounds like a hard chemical to aerosolize, which is also a condition needed for it to release HF.

So to recap:
You first need to get your engine block to a temperature far beyond what it's designed to handle. Then you need to be in a crash violent enough to aerosolize a decent-sized organic compound, *and* that aerosol has to land on that engine. Finally, all the above has to happen in sufficient quantities to produce a dangerous amount of HF gas, which I will note is not quite as holyfuckballswereallgonnadie lethal as you seem to think (it is very dangerous, and rightly feared, but you aren't going to die from a milliliter of it).

Yeah, I'm fine with that. Can't be much more dangerous than gasoline, which can kill you under far less unusual circumstances.

Re:Organic compounds

[bill_mcgonigle]

You first need to get your engine block to a temperature far beyond what it's designed to handle.

As I understood the point, they mixed this stuff with oil and then sprayed that mixture over the engine block. The hot engine ignited the oil and the burning oil reached the required temperature, presumably.

Re:Organic compounds

[girlintraining]

but I would think 500C is a bit unusual for an engine to operate at. That's roughly 900F, well above the melting point of, say, lead, and getting close to that of aluminum or magnesium.

If you pop open the hood and look along the sides of the engines, you'll notice that even though your block is aluminum, your exhaust manifold is not. The operating temperature of that will vary from 500 to 1000F for a V6 or V8. It will be higher if it is a rotary engine, or turbo-charged. A turbo-charger works by taking the pressure of the exhaust and using that to drive a turbine that compresses air and feeds it into the intake -- as a result, the exhaust will be at a much higher pressure, typically 9-12 PSI, and that results in the excess heat not dissipating as quickly. 1000F is easily attainable in a turbo-charged engine, like those typically found on the higher-end vehicles this refrigerant was/is installed in.

So by the time your engine block has reached 500C, you should already have run a good ways away.

As indicated earlier, the engine block is not the only source of heat under the hood, nor is it the hottest location. Also, the ignition temperature of gasoline can be much lower than 280C -- it can be as low as 232C (495F).

tetraflouropropene sounds like a hard chemical to aerosolize, which is also a condition needed for it to release HF.

It is in a closed loop refrigeration system. The typical pressures for the "high" side of a typical system is 200-350 PSI. Needless to say, a leak in the system would result in already-heated liquid that is designed to vaporize at 15-25 PSI being released into the atmosphere (at zero PSI)... which makes converting it to a gaseous state a simple matter of poking a hole somewhere in either loop; Though it would be somewhat more disasterous on the "high" side of the compressor.

So to recap:
Your understanding of physics is based on incorrect assumptions, and is incomplete as well.

Can't be much more dangerous than gasoline, which can kill you under far less unusual circumstances.

Yes, if you drink it I suppose. But many people have been doused in gasoline and unless they are lit on fire, find that it simply stinks and itches. And in many cases, people have survived being burned by gasoline spills that have caught fire. The same can not be said of anyone exposed to hydrofluoric acid. The CDC has a few things to say about it... namely that it can be used as a chemical weapon and is exceptionally toxic and fatal even in small amounts. Gasoline on the other hand...

Misleading statement in TFA

[Taco Cowboy]

... they've discovered a lot of organic molecules that have the potential for 10% or better conversion; roughly equivalent to the current best PV material

The current best PV materials have 20% or better conversion rate
 
Even the garden variety stuffs from China gets you about or above 15% conversion rate
 
I reckon the organic compounds are better, in the sense that they do not pollute the environment as much, but to that they are "equivalent" to the "best PV material" in terms of conversion rate, tastes a little bit funny to me

Re:Misleading statement in TFA

[mspohr]

Nitpicking point taken.
However, the other part of the equation, cost, has the potential to make these very attractive compounds. If you could turn the side of your house into a solar panel for just the cost of paining it, this would be a very attractive value proposition. Even if the efficiency was only half that of a conventional PV panel, the cost per watt would be much lower.
Good to see this research.

Re:Misleading statement in TFA

[Immerman]

Indeed. They're probably including labor costs, but still. One gallon of paint covers ~32m^2, * 2-6kWh/m^2/day insolation for a vertical equator-facing surface outside the tropics (relatively constant throughout the year, neglecting cloud cover and other shade sources) * 10% = 6-20kWh/day year round for a one-gallon paint job. Wouldn't have to last long at all to pay for itself.

Re:Misleading statement in TFA

[djh101010]

Nitpicking, my ass. The difference between 10% and 17% is huge. Wake me when someone finds a mass-producable solar cell that has better than that efficiency, and decades of life at that output. Current (heh...) silicon PV cells have reached the point of a 6 year payback on investment, and where the rails to mount them on cost more than the panels they hold. Until someone finds a hypothetical breakthrough, anything less than this efficiency is a waste of time and money other than for pure research.

Re:Misleading statement in TFA

[TheLink]

Maybe you can paint your wall cheaply with these compounds, but how would you get the electricity out?

I daresay doing it in a way where you actually get electricity would make the costs go up by a lot more. And depending on how its done it could make the efficiency go down too.

Merely painting your wall with crushed/blended solar panel material won't produce much usable power.

Re:Misleading statement in TFA

[mspohr]

I'll take a wild guess and say they use wires.

Re:Misleading statement in TFA

[wagnerrp]

Real world efficiency is on the order of 30-35%, and that's for multi-junction GaAs cells that only see use in concentrated solar and space-based power systems. The best crystalline silicon units do in the 20-25% range, and amorphous silicon units do around 15%. Most plants run around 2-3% of conversion of sunlight into biomass, however sugar cane tops the list at close to 10%. Note that's not just chlorophyll activation, but the whole process of using that to drive ion pumps and produce storage molecules.

Re:Misleading statement in TFA

[bluefoxlucid]

"As cheap as paint" doesn't mean it is paint. Why are there so many flaming dumbasses here assuming this must be paintable solar cells because it compares the cost to the cost of paint? I can compare the cost of a loaf of bread to a blowjob from a $2 hooker, maybe the bread will give me a blowjob?

Re:Roughly equivalent my ass.

[Colonel Korn]

Try half. High efficiency silicon cells are up to 20%.

The best are now sitting at 44% (http://en.wikipedia.org/wiki/File:PVeff%28rev130307%29.jpg). That doesn't mean cheaper solar cells don't have lots of potential, but it does mean the editors here screwed up again. There are a few other errors in TFS as well, but this one really got me:

could lead to PVCs that cost about as much as paint to cover a one-meter square wall."

Huh? So does this mean a PV coating will will have the same cost per area as paint. Personal expertise tells me no. Does it mean a postage stamp of PV coating will coast as much as a square meter of paint? That's actually more realistic for the midterm future, but the language in TFS shows such a basic lack of understanding of both numbers and units that it's impossible to tell what the editor or submitter really meant to say.

Excuse me

[viperidaenz]

While I go and file 20,000 patents...

Re:Roughly equivalent my ass.

[viperidaenz]

Those 44% cells use an optical concentrator, aka magnifying glass, and require a substantial cooling system (concentrating the suns energy 418x creates a lot of heat)
They're also not commercially available, although neither are any cells using one of these 20,000 different molecules.

PV with 40% conversion rates

[Taco Cowboy]

Actually, depending on how you define best (do laboratory samples count?) I think they're past 40% now

If you read closely on how they achieve the more than 40% efficiency you would see that they are not "1 sun" PV structure

Case in point, Solar Junction ( http://www.sj-solar.com/ ) came out with the "lattice matched 942X" which has 44% conversion rate

That "942X" denotes "942 suns", which means, the PV from Solar Junction is not flat panel type of PV, but rather, achieve its high efficiency with the use of focusing optics

Allow me to quote from http://www.renewableenergyworld.com/rea/news/article/2012/10/solar-junction-does-it-again-sets-new-cpv-efficiency-record

" ... Think of a magnifying glass. Basically, you have a very high performance solar cell that sits at the focal point of these focusing optics. The solar cell converts the photon flux into electrons, and power

Re:Roughly equivalent my ass.

[Khyber]

We haven't used optical concentrators for 40% cells in quite a while. Now we use focused-bandgap absorption and silicon nano-structures to act as waveguides, no concentrator needed.