Slashdot Mirror
Quantum Dots Could Double Solar Energy Efficiency
dptalia notes the recent publication in Science of research demonstrating a way to use hot electrons in solar cells, resulting in an overall energy conversion efficiency of 66%. Here is the abstract in Science; access to the full article requires a subscription. "A team of University of Minnesota-led researchers has cleared a major hurdle in the drive to build solar cells with potential efficiencies up to twice as high as current levels, which rarely exceed 30 percent. ... Tisdale and his colleagues demonstrated that quantum dots — made not of silicon but of another semiconductor called lead selenide — could indeed be made to surrender their 'hot' electrons before they cooled. The electrons were pulled away by titanium dioxide, another common inexpensive and abundant semiconductor material that behaves like a wire."
Seems like I keep hearing of breakthroughs, but nothing ever seems to fucking change!
"The next step is to construct solar cells with quantum dots and study them. But one big problem still remains: “Hot” electrons also lose their energy in titanium dioxide. New solar cell designs will be needed to eliminate this loss, the researchers said."
If I had a tribble for every time one of these solar energy articles came out with their pages full of nothing.... I could make a lot of fur coats.
Could someone in the research field please hold on to their excitement until they can post a report that has words like "WILL, SHALL, DEFINITELY, HAS, IS" instead of the wimpy "could, may, might, has potential to, in 5 years if all goes well....."
I got sucked in by one years ago and pestered the company for information about their new "product" which was due out "soon".... and that was nearly 10 years ago.
So many advances in tiny little cells on a research bench, and so many promising advances. Yet none of them seem to show up at the local hardware store.
I understand that advances in quantum materials science is cool, and can change everything just like the invention of the transistor did once. But seriosly folks - the number of speculative postings based on
these barely germinated lab experiments seem a little bit like the kid who cried "Solar revolution", or was that wolf?
RTFA, about 66%
Not a problem; lead's not nearly as bad as the arsenic in some panels!
(T>t && O(n)--) == sqrt(666)
Not a problem; lead's not nearly as bad as the arsenic in some panels!
We need to find a way to get usable arsenic out of contaminated soil. There are literally thousands of tons of it around the world, much of it slated for "cleanup" (secure burial). It's a lot less dangerous when you make it into a solar panel than when it's free to get into groundwater.
"You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
You're confusing energy conversion efficiency with energy production. The main connection there is that less efficiency means more raw resources for the same result. They're certainly not the same thing.
I think what the GP was getting at is something like, "This sounds way better than past solar conversion efficiency. Can we know build viable solar power stations? What about orbital solar power satellites? Where does this leave coal and nuclear power stations? What will the overall energy production strategy be, once this comes to market, given projected energy needs WHEN it will come to market?"
That's not a set of questions you want to answer too hastily.
Same as fusion: Ten years from now, for all possible values of "now."
I've fallen off your lawn, and I can't get up.
I'm no expert, but probably a lot more. Some facilitating factors:
1. PbSe is pretty easy to synthesize as nanorods. TiO2 is even easier. Lower production cost.
2. Higher efficiency (theoretically) than the 40% record achieved using triple junction cells (which have extreme costs and are likely never going to be practical) and the ~25% achievable using single-junction silicon cells (maximum theoretically about 31%).
This should lead to a great increase in the achievable power. The only thing that I'm unsure of is whether you can concentrate the light in nano-confined cells as much as you can in bulk material cells. The (I believe) issue becomes current density saturation either within the material or at the connector interface. Not altogether familiar with the R&D in this area. Since high-efficiency cells can be concentrated efficiently by a factor of ~1000x, this could be a significant effect if nano-confined cells can't be concentrated very much.
I have left slashdot and am now on Soylent News. FUCK YOU DICE.
I wrote a paper on this idea last semester, and as interesting a find as it is, I don't think it's ever going to lead to enhanced power conversion efficiencies (PCE). The "Double Solar Energy Efficiency" is actually a theoretical doubling of the thermodynamic limit on PCE, and it doesn't take non-radiative losses into account. These losses have been minimized in the record breaking silicon and multi-junction solar cells, but quantum dots bring lots of problems with them.
It's definitely worth further investigation, but currently I'm not convinced that it will ever bring improvements.
In my experience, rules set by regulatory bodies are never that useful, and generally only become so when their proposed regulations threaten the bottom line of someone with a lot of money. Unfortunately in that case, they often get diluted to worthlessness...
It's messy.
If you're lucky, you have a clear goal, and that goal is reachable though no one knew it beforehand of course. Also far from assured is that no one else has already done it. Not at all easy to find it out either, as others may have approached some problem from a completely different direction, and you will not be looking in the right places when you try to find out what others have done. And you could discover something entirely unrelated while in pursuit of your objectives, something worth taking a massive detour to pursue. Reporting on science skips over the difficulties to the point that it's like the joke about how to put an elephant in a refrigerator. You open the door, put the elephant in, and close the door, duh! Even scientific journals omit "extraneous" details about wrong turns taken, because "thinking is vulgar". Researchers naturally don't want to look dumb, and neither do journals. Plus, journals are always trying to save space and time. Streamlining articles to talk of only the successful experiments is an obvious cut to make. There's a STNG episode that touches on the problem. Lore mocks Dr. Noonien Soong, calling him "Often Wrong", but there it is clear that Lore's criticism is unfair. Science is all about doing things that will probably turn out to be mistakes, and learning from them.
Most people think science is about finding answers. That's only half of it. More important is asking good questions. Finding out which questions to ask and hopefully answer. Yes there are bad questions, don't believe that dogma about there being no such thing as a dumb question. They're still worth asking if you don't know or can't know they're bad. What they aren't is worth answering, and discovering that is the problem. There are questions that turn out to make no sense, or are too simple, or too complicated, or pointless, or misleading, or are based upon or arise from inferior modelings or from misunderstandings, or are predicated on assumptions that actually aren't valid, or are unanswerable. Math is full of shifts in representation to avoid such problems.
For instance, how do you deal with infinity in, say, the slope of a line? Vertical lines have infinite slope. You could pull out calculus and limits to handle these infinities, but if you do, you've just been sucked into doing things in a much harder way, been pulled down a rabbit hole to figure out all kinds of problems with working with infinities and division by 0. So much easier to use vector representations from linear algebra and avoid those infinities. All those questions about handling infinities turn out to be pointless for this problem. Mind you, infinites crop up in plenty of other places, so having ways to handle them is still useful, it just isn't necessary for this problem of representing a line. That's an old, well known problem for which we've known good answers for a long time now. There are also things like the famous Fourier Transform, and Lagrangian basis, not to mention Calculus itself to handle problems that are very nasty in classical geometry. But in new, unexplored territory it's not so easy to tell which questions will turn out to be bad.
Most things that scientists try turn out to be wrong. History has a few famous ones, such as "Squaring the Circle". The public hears little more than the 1% that turned out right.
Intellectual Property is a monopolistic, selfish, and defective concept. It is "tyranny over the mind of man"
Mushrooms such as the matsutake and shaggy mane accumulate arsenic. These could be grown on arsenic contaminated lands and then processed to extract the arsenic: http://books.google.com/books?id=NPI8_-omzvsC&pg=PA103&lpg=PA103&dq=mycelium+running+arsenic&source=bl&ots=38qB71dh2N&sig=jWBgriT6LmoI7ELXTWuy6Vgfm3I&hl=en&ei=uI0bTJOlAZuKNc6gwbcM&sa=X&oi=book_result&ct=result&resnum=1&ved=0CBIQ6AEwAA#v=onepage&q&f=false
Joe Lencioni | Shifting Pixel
The only thing that I'm unsure of is whether you can concentrate the light in nano-confined cells as much as you can in bulk material cells.
I would think that quantum dots might be ideal for use in a grid array of something like the dye-sensitized collectors that have recently been developed. I don't think that current saturation would be an issue, as the leads will be distributed evenly at each quantum dot. The problem I see is that increasing the area used for contact wiring will mean increased non-radiative losses. The article states that the wire contacts will be made of semi-conductor material as well, and semi-conductors typically have lower conductivity/higher loss than conductors. Supposedly they are trying to overcome those losses by using a partially generative material as an intermediate between the conducting leads and PbSe cells?
True. But installation and area and transport and many other components of the price, are independent of efficiency. So if you could make and install a 50% efficient solar-cell for less than twice the price of a 25% one, you'd have a win.
But sure, twice-as-effective ten-times-as-expensive isn't interesting.
who cares about efficiency?
If you make a solar cell that is 100% efficient but costs $10billion / watt to make, who cares?
On the other hand if you make one that is only 1% efficient but only cost .01c / watt to make, that would change the world.
I seriously wonder why people keep promoting micro-generation. Electricity distribution is very efficient ( you lose maybe 10% of the electricity on average ) , and you lose a lot of economies of scale when going micro. Even assuming you'd manage to make some super cost-efficient solar cells, why would you go destroy it by putting them on poorly aligned roofs, as opposed to building a designated plant where they can be made to track the sun throughout the year?
Seriously , for EVERY energy source centralized generation will come out on-top. It's a consequence of how easy electric power is to transport, as well as the fact that most energy generation schemes scale very well. The only real exception is where you're burning something for combined heat and power, thereby allowing you to recover the spill heat from the power-plant. However, even in that case district-heating will probably work out better, and does in many regions. We use it extensively in Sweden.
Essentially the only way micro-generation is going to be competitive with centralized generation is if government fucks up big time to make centralized generation inefficient. Granted that is of course plausible, and I'm sure there's many people who are willing to say that it is happening many places, but this is a political problem that could just as well ( and probably will ) hit micro-generation. It does nothing to alter the fact that on technical merits, micro-generation is inferior for all places that are connected to the electric grid. It just doesn't make any sense to take technologies that scale very well and deploy them in as small units as possible.
This is good, solid condensed matter physics research, and it is important. But, the 66% efficiency they're quoting comes from an old calculation of what is effectively a perfect solar thermal converter where the electrons are "hot" and the temperature of the crystal doesn't really matter because there's no interaction between the electrons (or photons) and the crystal. This theory has been the motivation for the hot electron transfer work on quantum dots, but at some point it should come up that they can't get hot electron transfer to work at much above around 100 K. That's kind-of really important (not to mention impractical). The device they're *trying* to make shouldn't need to be cooled at all. That's a giant clue that the theory they're basing this amazing 66% on is more than a little oversimplified for what they actually have. It's annoying that people are using an out of date 28 year old paper to sell the direction of their research; it makes life hard on their competition using more honest efficiencies.
I skipped the yearly trip to Vegas and invested some time and money in small scale solar for the house. Best 1K I ever spent. Running the entire outdoors media system and lights on solar.
DIY Solar project.
http://www.flickr.com/photos/nsaspook/sets/72157622934371746/show/
In GOD we trust, all others we monitor.