UCLA scientists double efficiency of novel solar cell | Hacker News Reader

UCLA scientists double efficiency of novel solar cell

phys.org

61 pointsm_class23 comments

I was thinking about this the other day, in that since the photovoltaic effect is a probabilistic thing (you need your photon to be absorbed so that you can spit off an exciton), if you had transparent cells you might be able to boost your efficiency by creating a light trap with mirrors to bounce them through the cells 10 - 20 times and thus increase the chance that they generate an exciton by that much.

A transparent cell works because it basically eliminates the chance for higher energy photons to be absorbed. So I suspect this won't work.

Right – but reflectors at the base of your cell still help. Any photons that pass through the first time, not being absorbed, are reflected back, effectively increasing the path length of your photon in the material.

Indeed; I believe they currently do that in some solar cells. I personal favourite are "distributed Bragg reflectors" – they're mirrors, but don't work in the traditional sense:

http://en.wikipedia.org/wiki/Distributed_Bragg_reflector

Ok, that is just too cool. Thanks for that link.

it's important to mention that they doubled the efficiency of their own previous solar cell, clocking in at 7.3%. a mass produced amorphous silicon cell achieves around 8%. however the advantage of polymer solar cells might be much lower production costs.

So put a regular cell immediately behind the transparent cell? It'd be equivalent to a multi-junction standard cell, just one that goes out into the infrared instead of stopping at red.

Can anyone here comment on whether there is enough energy in the infrared spectrum to make this useful?

Just using the articles numbers 40% of IR energy -> 4% of sun's energy, which would mean 10% of the energy is in IR, which matches fairly well with eyeballing this: http://commons.wikimedia.org/wiki/File:Solar_Spectrum.png

This shows the solar spectrum:

https://en.wikipedia.org/wiki/File:Solar_Spectrum.png

Visible light ends at about 700nm, so it looks to me (doing an approximate area comparison) as if there is about twice the total power in the spectrum available in the infrared as compared to in the visible.

The article (sorry, it's behind a journal subscription) doesn't say outright in words what frequencies are most absorbed, but from looking at their absorption spectra I would assume that the band gap is between about 650nm and 850nm. I am a materials scientist, but I don't know enough of the acronyms they're using to say more than that. The only other major part I understand and knew to look for is that those efficiencies are measured under one sun of illumination (a lot of the really good looking numbers are done under tens or hundreds of suns).

Convoluting those absorbance spectra with the solar spectrum to get an approximate maximum power input looks to me like something along the lines of 1/2-2/3rds of the total power as is in the visible spectrum. Definitely not negligible.

Is anyone looking into manufacturing this? I think it would be awesome to have this as a film that could be put on windows. Of course the main issue there is getting that energy back into the home and power grid.

This is totally uninteresting for anything other than PR unless it is cheap. Land for solar-cells is not the primary cost of solar energy and won't be for quite some time.

But isn't the idea to decentralize power generation?

I don't see big advantages to a slightly decentralized power generated system, and a fully decentralized system won't be feasible at a minimum until we have much better energy storage (and even then, there is insufficient insolation for most tall buildings to be self-sufficient with solar).

I see a huge benefit in that if this is in my house then my power consumption off the grid, in aggregate, is less.

You say land isn't the issue, but access to it for the average person still is. Yet if I can coat the windows of my own domicile I consider that a net positive.

Seems pretty silly to not absorb visible light, given that that's where solar energy is peaked. I don't really see the point of this.

Did you see this part? placed on windows, sunroofs, smartphone displays and other surfaces to harvest energy from the sun. It can turn surfaces that need to be transparent into solar cells. And for windows, the visible energy is already being put to good use, reducing the need for interior lighting.

Yes, I understand. I'm just questioning whether it's worth the effort trying to collect solar energy in these scenarios.

It will provide power that never existed, additionally it makes things like windows be a less of a heat source for things like air conditioners to pump back out.

I don't know if its economical yet (nothing product wise available) so it is somewhat premature to even speculate if it is or isn't worth the effort. The idea is sound however to take advantage of unproductive real estate.

Fair enough. But there's so much unproductive real estate that could get full-spectrum solar cells put on it, and generate a lot more power per unit area. I just think there's a lot of lower-hanging fruit.

Agreed, just pointing out that we don't know what we don't know with this technology. Say its cheap, might as well take advantage of the opportunity. I doubt any one methodology will be the best approach, and even if it isn't cheap now, maybe it will be in the future.

Eh, whatever, beer time!

"Economical" is the sum total of the problem with solar.

"Lack of space to put the panels" is not a problem at all - even in densely populated city centers there are rooftops galore; everywhere else, the cost of panels far exceeds the cost of land.

So it's a solution to a problem that doesn't exist.

I'd say that the cost of efficient PV array systems is a real issue which currently exists.

Unfortunately the article doesn't say much about the costs of these new cells.

In my area (Latitude 20 deg. N.) solar begins paying for itself after only a couple, or more, years depending on the type of system.

Obviously, one wouldn't choose to power heavy inductive load appliances with solar power, but for small items such as electronics and lighting, even refrigeration, solar makes more sense than paying $0.47 per KWh.

Then there are reliability issues inherent to municipal power grids...

When my neighbors experience a power outage my home is not affected, and to me this alone is worth the price I paid for my stand-alone solar power system.

Reply on news.ycombinator.com