Solar superhero

Falling polysilicon and also c-Si module prices have been decisive factors in accelerating the path towards PV competitiveness, but at the same time have left a lot of destruction in their wake. What is your assessment of the market at the moment?

I think that the c-Si wafer technology has surprised everyone in how much of the cost has been taken out of the technology. I’m on record about ten years ago as saying, “getting the cost below US$1/W will be difficult for that technology.” It seems now that in the next few years we’ll be below US$0.50/W and even now when you look at opportunities for developments to go even past that, there is still clearly margin for further reduction below that.
These falling prices have made it very difficult for the competing technologies, so my view now is that competing technology can’t just solely target competition based on a potentially lower cost, it’s really got to have another bow to its arrow, and I think efficiency has got to be that. So I think for thin film technologies to mount a credible challenge they have to be looking at something like a tandem cell structure that can outperform silicon as well as that potential to sneaking lower, in terms of the optimum cost that you can achieve.

With the pressure on prices inevitably that puts pressure on R&D. Do you think there is still progress being made in terms of efficiency gains and further cost reductions?

Very much so. I think the survival strategy of many of the large players is to innovate quickly to help improve their manufacturing cost position, so I think it’s probably been an accelerator of the introduction of new cell technology and other ways of taking costs out of the manufacturing process.

What cell technologies are you most excited about?

We’ve been working on laser-based processes now for 20 or 30 years. And these always seem to us to have the most potential for allowing the efficiencies that we’ve demonstrated in the laboratory to be transferred into a production setting. So we have a laser-doped selective emitter process that we’re working with multiple companies to commercialize. Suntech in particular we’ve been working very closely with and they’ve introduced a Pluto technology that uses some of that technology – some of those ideas – as well as ideas that Suntech has itself contributed. Recently Suntech reported getting 20 percent efficiency with production material in a production setting with that technology, so I think that’s an important first step to take.
I think ultimately manufacturers will be able to reach efficiencies very close to those that we demonstrate in the laboratory, which is 25 percent. So I think we’ll see efficiencies close to 25 percent in production ultimately and these developments, such as [that which] Suntech has reported, are the necessary steps in that progression towards 25 percent efficiency.

Suntech is also one of the companies posting big losses of late, what is your take on this period of consolidation?

There are some very good aspects to it and some very bad aspects. The good aspects have been the reduction of prices which hopefully have opened up new markets – so maybe there is just a bit of delayed reaction, because present prices for photovoltaic modules have reached grid parity with retail electricity prices in most parts of the world. Even in Sydney where I come from, which is normally regarded as having cheap electricity, you can produce electricity on the roof of your home more cheaply than the electricity company wants to charge you.
There may just be some inertia in waiting for markets to realize that there is this new technology that has reduced its costs very rapidly and now is in a very competitive position than the better-established alternatives. So maybe it’s just a time of hanging in there until that realization kicks in.

Looking more closely at the cell level, what other technologies do you think have the potential to really deliver and do so cost effectively?

It seems to me that now we have this huge industry that has been very successful in taking the cost out of producing silicon wafers and the cells from them. And when you look to the future of how photovoltaics might develop, it seems that a path that was able to take advantage of the enormous amount of development within the industry would be the one most likely to prevail. So if you look at the standard silicon wafer cells, we now have monocrystalline and polycrystalline, but with the development of the directional solidification processes now producing quasi-mono material, I think by 2020 the market will be largely based on monocrystalline material of one kind or another.
I think efficiency is a key driver for photovoltaics and if you look at options for improving the efficiency of a silicon cell, you can get to 25 percent, like we have in the laboratory, but how can you go higher than that? One of the few ways is by stacking cells of different materials on top of silicon. It seems to me that that is the logical path to development for the industry. We’ll get to the stage where we’ll get these very high quality wafers that can be produced very cheaply and they would be an ideal template for growing a high performance crystalline cell on their surface.
###MARGINALIE_BEGIN###

###MARGINALIE_END###
Technically there are very severe challenges in doing that but if you look towards the long term development of the industry, that will be a path to development that someone is really going to nail at some stage – in the next 10 or 15 years. Some kind of tandem stack on silicon is my prediction as to where the industry might head, which might take your efficiency potential from 29 to something like 45 percent – which is what you can do with three or four cells stacked on top of silicon.
At the University of New South Wales (UNSW), we have initiated about five different programs, looking at different ways that you might be able to grow films on silicon and that seems to me to be a very sensible area for the concentration of research activities, for future generations for photovoltaic devices.

Is that work taking place elsewhere also?

There are a lot of groups that have been looking into it. In microelectronics there is a lot of interest in being able to grow III-V (pronounced “three-five”) semiconductors on silicon, so huge efforts have been put in from that industry. And we are collaborating with a group at Ohio State University that has grown gallium-phosphide on silicon because they have similar lattice spacing. So there are isolated groups that are working in that area, but I think it is probably an area that warrants more attention. You will need to find a way that you can grow these films very quickly and cheaply on silicon, but technically at least, it seems feasible.

If these technologies can potentially deliver efficiency gains in the future, what technologies will reduce costs?

There is ongoing process with equipment that is capable of higher throughputs, which points to the direction of lower and lower costs. You see that the size of ingots that are being cast is growing quickly, but there is no real reason that they can’t keep growing and it points to lower and lower costs for the mainstream c-Si technology in the future.
I’m expecting that the low costs being attained by the technology right now is exciting for everyone involved in the industry, so despite the present period of financial hardship, I think everyone has realized that the technology has got to the stage where it’s got a guaranteed future. So it’s just a case of hanging in there until we reach a stage where the market grows quickly enough to satisfy the manufacturing aspirations of the companies involved.

What then is your view of this “guaranteed future” for PV?

There was a study by the German advisory council on global change, which was published in 2003, and they looked at all the options that were available out there for sustainably generating future energy at the levels that the world is likely to require by 2100. And they came to the conclusion that the only viable one was one based on solar energy. And if there are no other options out there, then that’s probably the path that we’re going to follow.
Interestingly they made forecasts about the rate of installation of solar to meet their scenario whereby most of the world’s energy was provided by solar by 2100 and we’re actually about five years ahead of that scenario. So things have been going quite well in terms of the introduction of technology into the marketplace compared to what people were expecting only ten or so years ago.