Listen to Podcast

Scott Bilby: We’re keen to listen to our first interviewee for today. That person is Dr. Mark Wanlass. He is a researcher at the National Center for Photovoltaics, a division of the U.S. National Renewable Energy Laboratory, ( aka NREL).

Dr Wanlass has been working on a type of photovoltaic cell called multi-junction solar cells since the early 1980’s and he is the principal scientist in a new type of multi-junction cell that improves performance, and reduces weight and reduces cost. This year it won an award from the prestigious (R&D) Research & Development Magazine. The R&D 100 Awards recognize the most promising new products, processes, materials, or software developed throughout the world and introduced to the market.

Good morning Mark. Thank for joining us from Colorado.

Mark Wanlass: Hello Scott, how are you?

Scott Bilby: I’m very good, thank you. Matt and I are eager to ask you about this amazing new technology. Just this morning we actually saw a media release about the new cell. Can you just tell us about the award you won for the new cell and just a little bit of the background to it? …briefly introduce it to the audience.

Mark Wanlass: Sure. Actually this is an invention that I had dating back to the early 1990’s but we only starting working on it in about 2003. We got some good results in 2004 and had our first publication at the Photovoltaics Specialist Conference in January 2005.

I’ll tell you about the award first. Basically, it had to do with the results that we achieved prior to the recent advancements, working with Emcore Photovoltaics in Albuquerque and they were supported by the Airforce Research Lab, also in Albuquerque. So it was kind of a joint nomination. But really it had to do with the fact that we had achieved very high performance and also had transferred the technology to industry. So we’re very proud of that; very happy to see that happen.

In the meantime we’ve made more advancements on this in the lab. This is a technology that really has many variants and can be expanded to more complex structures and higher performance in the future. But very recently a colleague of mine named John Geisz led a group here that took the same basic device concept and demonstrated a 40.8% result, which is a new world record at about 326 suns concentration. More or less that is what the award was about. This approach really differs from conventional multi-junction cells in a number of ways and we can get into that as we go here.

Matthew Wright: Just for listeners, you mentioned the 326 suns. Now, I know what that means, but I'm wondering if you could just explain the difference between one sun and 326 suns?

Mark Wanlass: Sure. These cells are expensive to make and the expense is justified because of the performance. Originally these were adopted for space power because space power applications have a very high cost tolerance. What they’re really after is a very high specific power which is watts per unit mass or watts per kilogram. So they were first deployed there, but for terrestrial applications we’re really looking at what are called Terrestrial Concentrator Systems. And the gain there is that we want to mitigate the cost of the cell by using relatively low cost optics and support structure and tracking devices to gather light from a large area and concentrate it on a small area. And that means we can use a very tiny solar cell that produces a lot of power. What this means is that we can mitigate the cost by making the cells very small and using optics. As we make cells more and more efficient what it means is that we’re transferring basically more of the available money to spend, which is typically something like two and three (U.S.) dollars per installed watt, towards the balance of system which is the support structure and optics, etc. So, that’s really why we are pushing for more and more efficiency. It really just adds more value to the balance of the system. So 326 suns relates to the fact that we’re taking the power that’s incident per unit area from the Sun and concentrating it 326 times onto a spot.

Scott Bilby: Now previously on this show we’ve made mention of an institution called Spetrolab, is it a government body or something?, and they’ve come up with really high efficiencies in the past which are not that far off the efficiencies you guys have got, but I assume that the big advantage is the reduction in cost and the lighter weight you’re able to get from this new cell.

Mark Wanlass: The excitement around this technology is the fact that this is really a major departure from conventional technology. Conventional multi-junction cells are typically grown on germanium substrates and they’re grown in what’s called an upright configuration so you’re growing from low-band gap to high-band gap. You’re adding in typically three band gaps, or subcells, in these structures and when the sunlight comes through that sort of structure, (when it’s grown upright) it’s getting selectively absorbed by these different band gaps as it cascades through this crystalline structure.

But there are some limitations in this conventional technology and they’re pretty serious. One is that the substrate is basically a part of the finished device product and therefore you can’t get rid of its weight or its cost because the germanium is actually an active subcell in the cell structure. There are some other things that also constrain the conventional technology to obtaining higher performance and lower cost, but I really won’t go into those.

This new technology that we developed is really based on two fundamentally different things from the convention. The first one is that the structures are grown inverted. What this means is that we started out with a crystalline substrate, such as gallium arsenide or germanium, and we grow the high-band gap materials first and then cascade to lower band gaps.

The second part of it is that we actually incorporate materials that are both lattice-matched and lattice-mismatched with respect to the substrate. What this does is for us is that it allows us to expand the available band gaps that we can get into these multi-junctions, and this allows higher performance.

Now, beyond that there are a number of other advantages and one of then is that because we grow this thing upside down we have to affix it to what we call a handle or secondary carrier. That means we necessarily have to remove the original growth substrate. Even though that may sound more complex, it actually opens up the possibility of reclaiming that substrate and reclaiming its cost somewhat.

The second thing that happens by mounting these cells up to a secondary carrier is that what you’re left with are the ultra-thin layers that you grew. So, these structures are in the order of ten microns thick which means that they are inherently flexible. When we made that to a carrier of choice that can be engineered, we can engineer into that carrier a wide range of properties that are very advantageous. For example, it can be robust, it can be low cost, it can be flexible, it can be optically transparent if you needed that, it can have a high thermal conductivity, etc.

So, one of the reasons why the space power people are so interested is because we can make this very thin, ultra-lightweight cell on a flexible substrate which means that we’ve got a very high specific power and it opens up new ways to deploy these things because the whole cell itself can be flexible, if it’s put on a flexible carrier.

For the terrestrial problem, first of all we get higher performance, but there are other problems inherent to concentration, namely thermal management for example is a big one and if we can manage these things thermally better we can actually get higher performance out of the cells. If you’ve got the conventional approach with the thick substrate in place, those kinds of substrates are inherently thermally poor in terms of their conductivity, so it’s a nice thing to be able to get rid of that parent substrate.

Matthew Wright: OK, so basically in summary, you get the flexibility of being able to install this stuff on surfaces that may not conform to being perfectly flat and square, and also the potential is that the efficiency can go up a lot higher and the cost can go right down.

Mark Wanlass: Yes, I’d say the cost can come down somewhat. It’s going to have some sort of impact. There are costs outside of just the growth and substrate cost itself; you’ve got other processing that you have to put into these things to put on things like metals for contacts and reflection coatings, and things of that nature which are kind of a fixed cost, but it will help, it should help.

Scott Bilby: So, you’re comparing the cell that your team has developed compared to other types of other multi-junction cell. Does that mean that there’s the potential for your cell to have even higher efficiencies where the other more traditional multi-junction cells simply can’t go a lot further with their efficiency rate?

Mark Wanlass: Yeah, basically what happened with the conventional cells there was a hope ten years or so ago that they could come up with what I would call the missing band gap. There was, for the germanium-based cells, there was a real need for a subcell that had a band gap of one electronvolt.

A number of materials were researched to try to serve that need, and these materials contained sort of exotic elements such as nitrogen and boron and thallium and so forth. It turns out that none of them worked out, and so that next subcell that would have taken this thing to the next level with the conventional approach just didn’t pan out. And so this older idea of using mismatched materials kind of caught on because the options were drying up. It turns out that I’ve worked on lattice mismatched epitaxial growth now for 28 years, so I had a lot of experience in that area and I always believed that it would have some kind of an application. Then when this idea came along for the inverted cell, it kind of came together and it’s been a big success story.

I think there’s still some issues around this technology. Reliability in general is an issue we’re exploring for concentrators because they just haven’t had the 20 year lifetime experiment in the field; silicon’s had that kind of thing. There have been modules on site at various test locations for a few years, but not in the range that people are interested in so we’re doing things like accelerated lifetime testing.

Scott Bilby: Is that going to give you the possibility of getting efficiency rates somewhere up near around 60% or so?

Mark Wanlass: Well, I don’t think in my lifetime, but I would say that I think that with this approach realistically I think we could see some efficiencies that are maybe 45 to 50% for a 4 band gap version that is optimised in terms of the band gaps. I’m actually working on a version of that right now that I’m hoping will eventually get over 45%. It’s going to take a lot of research.

Matthew Wright: So, that means for listeners that 50% of the light coming and hitting that material becomes electricity and pretty damn impressive.

Mark Wanlass: I was going to say about concentrators, you have to deploy these things in arid regions such as the desert regions of the southwest of the U.S. or in Spain or Italy or the Mediterranean or say northern Africa, and also in Australia as you know, because the optics can essentially only accept direct rays of light so you can’t have any scattering. So, these kinds of concentrator systems don’t work well where you’ve got cloudiness and humidity and things like that because they tend to scatter light. But still there’s a huge opportunity, and the nice thing about it is that in a lot of these regions people don’t want to live there anyway so you could deploy these things in areas where it’s very hot and dry and generate a lot of power.

I should mention too that the thermal management issue is a big issue because even if you’re performing at 50% efficient, you’ve got 50% of the power that’s going on the cell is not being converted to electricity, it’s being converted to heat. So, you have to be able to pull that out because the hotter a solar cell gets the poorer it performs. So, there’s a big advantage there with these ultra-thin cells, and I think that’s something we’re going to try to take advantage of.

Matthew Wright: Does that mean as you actually have the direct electricity efficiency go up, you’re actually reducing some of the heat problem as well?

Mark Wanlass: You certainly are. There’s no doubt about it because energy’s got to be conserved.

Matthew Wright: That’s a fantastic outcome. You get double the value when you increase the efficiency because you help control the management issues around the heat.

Mark Wanlass: That’s right.

Scott Bilby: Now, Dr. Wanlass, you’ve spoken about the technology and you mentioned very briefly the opportunities, can you talk a little more about the opportunities? Because that’s what Beyond Zero is very interested about, solutions to climate change and we want to see stuff rolled out on a very large scale. Can you speak with us a little bit about, because we’ve already spoken to David from Ausra whose got plants that are already in the making in the United States, where is your technology going to fit in on a big scale?

Mark Wanlass: Well, I think eventually it’s going to fit in in the same way that conventional cells are being used now. I mean, they’re being used because they have been developed to a very high and reliable level for space power, and so they’re adapting them to be concentrator cells instead of in space they’re operating as one Sun cells.

That technology exists, the production for it exists. I can tell you right now that for example Emcore is slated to produce this kind of IMM (Inverted Metamorphic Multi-junction) cell by the end of next year, it’s what they’ve been saying. And we’ve got a whole host of other companies that have actually approached NREL to adopt this technology and do things such as co-operative research and development agreements such that we can transfer the technology and they can put it into production.

Scott Bilby: OK. So, it’s really moving along. Can you just mention that name again; was that ‘Emcore’?

Mark Wanlass: Emcore, yes. Emcore Photovoltaics. Emcore and Spectrolab are the two multi-junction cell producers in the U.S right now.

Matthew Wright: And they’re doing the same type? Or ones doing the new type of multi-junction, and the other one the old type?

Mark Wanlass: Well, they’re essentially doing the same kind of conventional cell right now, but when I first published on this back in 2005 Emcore immediately jumped on it and were very interested, as was the Airforce Defence Lab, I think principally for space application, but they recognised the efficiency potential and how this thing is expandable to even more band gaps and higher efficiency and some of the other advantages. So, they kind of got a jump I think on the rest of the world on this, but I can tell you right now that based on what I’ve seen at other conferences Spectrolab’s already working on it and there are some other companies. There’s a company called Microlink Devices in Chicago that’s interested in it. I think the interest is coming up pretty strong right now.

Matthew Wright: Great. It looks like we’re going to see that commercial deployment. In Australia we actually have a company called Solar Systems who uses, I don’t think they’re up to these kind of efficiencies yet, but I assume they can just swap the modules out from their existing concentrating photovoltaic units which look like dishes, giant satellite dishes, they could swap these in place I expect, could they?

Mark Wanlass: Yes, I’m certain of that. I think they’re getting their cells from Spectrolab; I believe that’s true and so eventually with a future generation system these kind of cells would look just like what’s in there now basically except they would be ultra-thin and on a thermally conducted carrier. Essentially they’d look the same and they would swap right into place.

Matthew Wright: And in doing that they’d then have thermal management set up for a less efficient system, which means that the whole thing would run cooler.

Mark Wanlass: I think that’s potentially true, yes.

Matthew Wright: That’s pretty exciting and I guess you heard it first on Beyond Zero.

Mark Wanlass: I think I should add too that efficiency has a lot of value in the sense that in the U.S. here we think of deploying the stuff as utility scale power in places like Arizona, New Mexico, Nevada, California, Colorado for example, but there are other applications for example in Europe where you might want to be generating electric power and hot water in a sort of co-generation system on things like apartment buildings because they use electric power and hot water and that way you can really get a very high, overall system efficiency. The thing is with a building scenario like that you’ve got a constrained area so you want to produce as much power per unit area as you can so I think that’s where a higher efficiency solar cell can address that kind of application as well.

Matthew Wright: So you do see the potential that the price could come right down to a point where even if you pay your fair premium it’s something that apartment blocks could fitout on the roof maybe in ten years time or something?

Mark Wanlass: Yes. And I think the other thing too is that in those countries, because they’re not burning coal like we do, the cost per kilowatt hour is a lot higher so that makes these things even more attractive.

Matthew Wright: What sort of time period do you think before we see that application, because obviously the concentrating application is where it’s at at the moment.

Mark Wanlass: Well, that would also be a concentrator application. It would be some kind of a system that can be deployed in an attractive way on top of a building, say a flat-topped building for example. But I don’t produce products, I’m a scientist so it’s hard for me to say, but I think that the fact that Emcore is really getting on with this and if indeed if they get into production with this by the end of next year I think things will happen rapidly.

Scott Bilby: Mark, Could I just ask you another question? I’d like to know what else is going on at the National Center of Photovoltaics because I know that the division won another R&D 100 Award this year, could you tell us maybe about that or something else that’d going on there?

Mark Wanlass: Sure. The NCPV (National Center of Photovoltaics) has got really three major thrusts. It’s basically divided up into materials technologies. We’ve got things like amorphous silicon crystal and silicon cadmium telluride which is a thin film material, copper-indium-gallium selenide which is another thin film, and then the three fives which are high efficiency single crystal.

We’ve also got another part of the centre which is measurement and characterisation, so we do official solar cell measurements, fundamental studies of material [25:52] and? LAN? devices, we’ve got a whole array of charaterisation equipment that can look at dimensional and compositional aspects of multi-layered structures. We’ve got fine teams in each one of those areas; microscopy, surface analysis, things like secondary ion mass-spectroscopy, etc. etc.

So, we’re very well setup to characterise the materials that we grow and to measure the cells. We also make solar cells from the ground up, We grow all of our own crystals and then process them in a clean-room facility and then do measurements. Another part is outdoor testing, both cells and modules and we’ve kind of expanded over the last year into doing more reliability-related testing of materials and cells and modules as well.

Matthew Wright: So, really the research you’re doing is a really important feed-in to the commercialisation of solar by those commercial companies?

Mark Wanlass: Yes, especially in the last year. You know, our main customer here is the Department of Energy in the U.S. and they’re really interested in technology transfer, reliability of systems and kind of bridging the gap between doing basic research on photovoltaics and getting this stuff out of the lab and into the hands of industry.

Scott Bilby: And I guess a huge issue for them is energy security.

Mark Wanlass: Well, I think energy security is an issue for everyone around the world, but the global warming problem as you know is related to that and there’s a lot of concern. So, I think it’s not only the way we have been doing business, which we knew at some point we were going to run out of oil and fossil fuels in general, but it’s energy security and things related to that environmentally.

Scott Bilby: Dr. Wanlass I’m afraid we’ve run out of time but it has been fascinating speaking to you this morning and we’d like to congratulate you on the award you’ve won for your new cell.

Mark Wanlass: Thank you very much.

Matthew Wright: And keep up the good work because we’ll be excited to see the changeover of our power grid here in Australia which is also very dependent on coal to things like concentrating photovoltaic systems.

Mark Wanlass: Ok, very good. Thank you very much.

Links:
U.S. National Renewable Energy Laboratory
Info about the two R&D 100 Awards won by NREL’s National Center for Photovoltaics, including the Inverted Multi-junction cell