Science Interviews


Fri, 4th May 2012

Harvesting Energy from all forms of Sunlight

Per-Anders Hansen, Department of Chemistry, University of Oslo

Listen Now    Download as mp3 from the show Naked in Norway

Per -  Sunlight consists of small energy particles called photons and these can have a lot of energy like ultraviolet light or blue light or they can have medium energy like red light or visible light, or they can have very little energy like infrared or heat radiation.  When these photons hit the solar cell, what they do is that they knock loose one electron and this electron can then travel around in our circuit and give us power.  And to do this, the photon needs a minimum amount of energy.  But if they have more than this, they still only knock loose one electron and this is why solar cells are not more efficient.  They could be like two or three times more efficient.

Chris -  So if we had a way of taking some of those photons that have verySolar panel high energy, and converting them into more photons at a lower energy, we could get more electrons out and therefore, the overall efficiency of the cells would be much higher.

Per -  Exactly.  If you take these high energy UV photons and simply cut them in two, cut the energy in two so we have two particles of half the energy, then you would get twice the amount of power out of those photons.

Chris -  Can we do that?  Can you convert light of one wavelength like UV into a longer wavelength like that, like you're saying?  Is that possible?

Per -  Yeah, itís possible and itís actually quite easy.  Itís easy to turn one photon, one ultraviolet photon, into one visible or one red photon, that's actually quite easy.  That's what's going on in fluorescent tube lightings actually.  So that we sort of know how to do.  The trickier part comes when you want to cut it in two so that you get actually twice the amount of light or photons out from the ultraviolet light.  That's where the trouble is.  We can do that for very high energy photons.  That has been done very efficiently.  The trick is to do it for photons that actually exist in the sunlight.  That, no one has done yet.

Chris -  So, we know how to do this for high energy photons like ultraviolet but because there's not much ultraviolet reaching us here at the ground, you're saying that although the materials exist to do it, we havenít got really the light that can make use of that.  So we need a new material that will work with the photons we do have and cut them in this way.

Per -  Yeah.  We know how to convert very high energy ultraviolet light that exists like in space outside our atmosphere, but the ultraviolet light that comes down to Earth that we get sunburnt off, it doesnít have enough energy to work with those kind of materials.

Chris -  How are you grappling with it?  What are you doing to try and overcome that?

Per -  Well, I use very peculiar synthesis method called atomic layer deposition.  It gives me a very high control of which atoms sit next to each other.  You usually have two components in this kind of material, something that absorbs the light you want, and something that emits the light you want.  And you need these to be sort of in every other position, so you have something that absorbs and around it has to sit these emitters and around the emitters, it has to sit absorbers.  So, if they are mixed up randomly so that you have one bunch of emitters over there and one bunch of absorbers over there, itís not going to work.  But if you kind of really control what's next to each other, you have a lot more control of what's going on and the physics that's going on in the material.

Chris -  So would that array work because the absorber soaks up the energy and passes it to the emitter next door which then spits out the photons of the wavelength you do want and they're then carried off and absorbed into the semi-conducting material to actually make the current.

Per -  Yeah, that's exactly how it will work.

Chris -  So what materials can we use?  Are you basically trying different combinations of anything to see if you can get the right sorts of absorption and emission or do you already have some candidates that you know are the right sorts of chemicals to use?

Per -  Yes, to both.  We know what we need.  We just need something that absorbs ultraviolet light and blue light, and many things do that so for example, sun cream absorbs ultraviolet light.  Sun cream consists of titanium oxide for example.  OK, titanium oxide, that could maybe be a good candidate.  Then you just need to find something that emits light very well, so then I thought of, ok, what emits red, a TV emits red.  In a TV, what is it that makes TV go red is europium.  So, combine those two and that works.  So of course, that is not my big secret.  My big secret is more fancier than that, but that is kind of the quest youíre going through Ė just what can absorb it and what can emit what I want and then just combine it.  And then you need to combine it in these very particular ways that everything sits at the right place.

Chris -  So those atoms are deposited in just the right configuration.

Per -  Yes.

Chris -  If you can get this to work, what sort of step change in efficiency could we achieve with a photovoltaic?  What will it go from to?

Per -  Well, a normal silicon solar cell that you can buy on the market today is between 16 and 20% efficient usually, maybe a bit more sometimes. For red light or light between red and 900 nanometres or something like that, itís actually much more efficient. Itís more like 60, 70, 80% efficient.  So, if you actually can take all that other energy, convert it to red light then itís actually no problem to just take the normal solar cell that you have at your cabin and you suddenly get 50% efficiency out of it.

Chris -  Which is a dramatic improvement.  I think the figure is something like the amount of sunlight that falls on the Earth in one day is equivalent to the amount of energy that the entire world consumes in a year.  So if we can get cells up to those sorts of efficiencies, we actually could do quite a lot to offset what weíre concerned about, sort out climate change and all that kind of thing from the energy industry.

Per -  Yeah, we have plenty of sunlight.  If we can just convert it efficiently and of course, cheaply, then actually we don't really need anything else.

Chris -  So, how far away do you think you are from realizing the technology youíre working on?  How long before we see these novel coatings going on to photovoltaics so that you can harness the extra energy weíre currently just chucking away?

Per -  There is some testing going on now around the world doing this Ė one ultraviolet into one red or something like that. But this one to two photon cutting, that hasn't been done yet.  I hope that we can maybe see that around the world in maybe 5 years, 10 years, I hope.

Chris -  So I need to come back in 5 or 10 years and see if you were right.

Per -  Yeah, come back then and see how I'm doing.

Chris -  Iíll do that.


Subscribe Free

Related Content


Make a comment

Interesting. I understand also that the solar cells used on some space probes are much more efficient than the kind usually found on Earth, and I believe the reason is that they have three layers, each responding to a different wavelength band. I understand they also are much more expensive. Also, not long ago there was a report of an enhanced solar collector that is a hybrid design: part photovoltaic and part thermal. It uses a curved mirror to concentrate sunlight on a tube that carries a fluid, but the tube is also covered by photovoltaic cells. And then there was another report about an improved system that relies upon a curved mirror collecting sunlight upon an advanced photovoltaic collector of unknown design, which is actually rather expensive, but because it is small compared to the total collecting area, the system actually ends up being cheaper per watt than a standard solar panel. This particular device, as I recall, was invented by Roger Angel the optics expert of the University of Arizona. Atomic-S, Fri, 22nd Jun 2012

Interesting approach.

Can one use infrared to generate the energy, as it is easy to make IR emitters.  Perhaps carbon nanotubes which would absorb broad spectrum light, and re-emit IR, although it may not be in a single wavelength.

One problem is that I assume emissions are omnidirectional.  So, if one thinks like a cube, then at least 1/6 of the light would be re-emitted back towards the source, and thus would be lost.  On a plane, the loss could be much higher.

Adding more places for loss of energy also seems dangerous, so while the absorption of red might be twice as efficient for energy input as violet, that added efficiency could be quickly lost.  Nonetheless, it is worth trying.

As far as the triple junction cells, they are most efficient on Earth with concentrated light which means that they must have a solar tracking system.  It is likely that one of the differences between terrestrial solar panels and space solar panels is the broad frequencies of light that is absorbed by our atmosphere.

It is quite possible that the Per-Anders Hansen system would also benefit from concentrated light. CliffordK, Fri, 22nd Jun 2012

See the whole discussion | Make a comment

Not working please enable javascript
Powered by UKfast
Genetics Society