Transcript: Great Solar Business interview with Martin Green

This is a lightly edited transcript of the Nigel Morris interview with Dr. Martin Green: You can listen to the interview here on the Great Solar Business Podcast

Nigel Morris  00:30

Welcome back to Great Solar Business, my friends. This week, let’s talk about the cost of solar. The cost of solar has plummeted making solar more affordable than ever for consumers. Can solar costs go even lower? And is that a good thing for your solar business? This week we’ll try to answer the question how cheap can solar be? with our esteemed guest, Dr. Martin Green from the University of New South Wales. Martin, welcome to Great Solar Business.

Martin Green  00:58

Oh, thank you very much, Nigel,

Nigel Morris  01:00

You’re most welcome. It’s an honor to have you here. Now, if you’re in solar and you don’t know who Dr. Martin Green is, let me briefly illuminate you. Martin Green is a Scientia Professor at the University of New South Wales, Sydney; Director of the Australian Centre for Advanced Photovoltaics, involving several other universities from Australia and research groups. His groups’ contributions to photovoltaics are well known and include holding the record for silicon solar cell efficiency for 30 of the last 37 years, would you believe? Described as one of the top 10 milestones in the history of solar photovoltaics, they invented the PERC cell, have set countless records, and have won a host of prestigious international awards. The performance price and massive selection of solar manufacturers and solar panels we enjoy today is largely down to Martin and his team. Martin, I think we collectively owe you a beer, maybe a couple of beers.

Martin Green  02:01

Oh, thank you. I look forward to collecting them.

Nigel Morris  02:05

Now I’m going to go off script just for a second here because it’s many years since I was honored to share a podium with you, Martin, but all those years ago, you know, 20/30 years ago, did you ever imagine that PV prices would be where they are today?

Martin Green  02:32

I guess I always thought it would be cheap one day, but I think everything has happened a lot more quickly than I ever expected. So we got to where I hope we’d get to but a lot more quickly than I ever thought possible.

Nigel Morris  02:48

Well, you know, I think it’s caught everybody by surprise. Certainly, over the years all the forecasts that I looked at and were involved with, and I was given, no one ever guessed we’d be where we’re at today. But let’s get down to it. They’ve plummeted. Solar costs have plummeted over my time, and certainly your time in solar. What are the main things behind those cost reductions? Martin, you know, what are the, you know, top three things that have driven these massive cost reductions that we’ve seen?

Martin Green  03:22

Yeah, well, a very reliable one is just economies of scale. So as you make more of the cells, they get cheaper, you know, regardless of whether you improve the technology or not. So that’s been one steady factor in the background. And then there was a big change in the manufacturing industry around the turn of the century, you know, 20 – 15/20 years ago, when some of those students decided to start making the cells in China. And that just sort of entered, we entered a completely new era of cost reduction after that occurred. And then the third contributor is technology. And as you mentioned in the introduction PERC has now taken over the industry, and that is also helping to drive down costs, not only because of the improved efficiency, but the increased functionality of PERC, like you can do bifacial modules, and you can use much larger wafers in the processing and cut them down into smaller bits at the end. So there’s probably those three factors, I think the really important ones that have led us to where we are today with you know, in an era of extremely low costs.

Nigel Morris  04:35

It’s amazing to follow the trajectory, isn’t it, and, and to see these factors coming in and out. And again, I’m going to drift off script, I apologize. But, you know, I remember very, very early on in the days, when I was involved in PV manufacturing on the, I was never, I was never a technician, let’s be clear about that, I was always on the sales and marketing side, but I was very fortunate to visit lots of factories and always had, the factories had BP solar, bringing new technology out saying hey, we’ve got this new, we’ve got this new cell or we’re gonna build this new factory and along the way some of them didn’t survive Martin. Laser Groove Buried Contact being one of them,  the Sliver Dot Cells being another. You know what, why does some technologies not make it and others survive? Yeah, I think, you know, in photovoltaics, it may be a little bit unusual, but the whole industry seems to have to get on board with the technology. So with the Buried Contact Cell that BP marketed as its satin product for, you know, 15 years or something like that, you know, BP was the only one really selling that technology in volume. So all the developmental work and the specialized equipment you needed, and everything, had to be done by BP.

Whereas now with the PERC, the whole industry has decided that’s the way to go. So you’ve got, you know, dozens of equipment companies vying with each other to supply better and better equipment for making PERC and the material suppliers are trying to produce wafers and things that work well with PERC. So you’ve got the whole industry working with you. So you know, to, to go out on a limb with a unique technology can be a recipe for disaster within photovoltaics. You know, I think that’s what history shows. It’s, it’s, it’s a gamble. I was talking to a manufacturer recently, actually, and talking to him about some new lines they’d got, and, you know, he was sort of nervously saying, I hope we’ve got the right equipment, because this move, this industry moves so fast. You, you recently gave a really fascinating online presentation with predictions that blew my mind that costs were going to continue to fall to a staggering new low, in fact.

What’s this new low that you’re talking about? And how are we going to get there? Yeah, well, I keep track of what’s called the average selling price, which is really like a factory gate price in China. But people ring up the manufacturers every week and find out what they reckon they’re selling their gear for that. You know, and at the moment, it’s saying a PERC module sells for, you know, about 20 cents a watt. But if you just plot the history of those reported average selling prices, it turns out, they’re decreasing at about 20% a year. So that means in you know, three or four years, you’re down to half the price. So in three or four years, you know, if you just continue the trajectory that the industry has been on, for at least the eight years, and I think before then price reduction was even quicker, you know, we should get to 10 cents a watt within the next three or four years, unless something untoward happens. And there’s plenty of technology in the pipeline, I think with the introduction of PERC and all its attributes yet to be fully exploited, that’s going to help drive those costs down. So I feel quite confident we’ll get to 10 cents a watt, I certainly hope in the coming decade, but I think definitely within the first half of the decade that’s coming up. Holy smokes 10 cents a watt, folks, you heard it here from from the granddaddy of solar who knows stuff, so get ready for that. Do you predict, Martin, that there are going to be major technological changes?

We’ve seen the introduction, of course of PERC and that’s become, as you say, ubiquitous across the industry. Now we’ve started to see shingle products, we’ve seen large format products, you know, what, what’s coming down the pipeline, along with this 10 cents a watt ASP that we’re going to see that solar installers might need to be ready for? How’s how’s the technology gonna change? Yeah, well definitely, you know, PERC has opened up sort of flood gates of creativity within the manufacturing industry. So you know, we’re all of a sudden, we’re seeing these big modules and bifacial modules is the other one, but, you know, probably not so much for residential systems. But, you know, that’s something that PERC has made possible. But I’m quite intrigued by the possibility of these bigger wafers and bigger modules. So I guess the industry is diverging in that in the past, you know, one product suited all just to keep inventory cost down. So a residential module is pretty much the same as open rack mounted module for big multi megawatt installations. But I think we’re going to see diversification. So that there’s going, it’s already happening I believe, manufacturers now are concentrating on modules for residential use, and then different modules for open rack field installations. So I think that could lead to a big change.

So you know, for residential use, you probably really want a module that a couple of people can carry up a ladder to install or something like that. But for field mounting, you know, you might be looking at modules that grow enormously bigger than what we’ve seen today. We’ve seen, we’ve seen a bit of this in the past. It’s an interesting point that you make because I know, in some of my work with manufacturers over the years, when I was consulting, you know, I watched manufacturers desperate to steer solar installers towards certain types of products for certain types of applications. And it’s, it’s been a tough sell, if I’m honest. It’s been really hard for them to get installers to recognize that we’re going to have different products that are better suited to different applications. But that’s changing over time, I’m sure. The first solar panel I sold was 60 watts, and cost my customer $20 a watt. And, you know, today, we’re already in the realm of 20 cents a watt, you’re forecasting, we’re going to be at 10 cents a watt before long, but for a solar installer, doing a megawatt a year, they’ve got to sell more panels to get the same revenue, which is quite a challenge. And I’ve seen this happen over the years. As costs fall you need to do more of the same work, or same volume of work to get the same revenue into your business. And you got to adjust your margins along the way. How fast is this really gonna happen? You’re talking three, four years, you know, how solid is that Martin?  Are there risks around that? Could it take 10 years? Or could it take one year? Yeah, yeah, there are some risks.

Like at the moment, we’re having a shortage of, well, we just come through a period where there’s a shortage of glass for modules, and then the polysilicon a couple of fires in factories and things, wiped out 10% of the world’s production of polysilicon. So you’ve been in a period over the last three or four months where prices haven’t continued to come down, but have gone up, if anything. So there’s little hiccups like that. But the feeling is, they’re just temporary hiccups. And there’s dozens of companies reporting scaling up their polysilicon production volumes, and all these new companies entering the field so that, you know, we’re going to be swamped with polysilicon probably later on this year. Or next year. So you know, you get these fluctuations, but I think there’s enough improvements still in the pipeline to see the cost come down at least an additional factor of two to 10 cents a watt. You know, like, it seems a bit of a stretch to me to get down to five cents, but you’re like, I think if you get down to 10 cents quite comfortably over the first part of this decade, your five cents is definitely on the horizon is where you might be able to get to in the longer term. Good grief. It’s remarkable.

And so you’re talking about Moore’s law here, which is this perpetual cost reduction with scale and everything else. So 10 cents isn’t the floor, then? There’s scope to go further down than 10 cents, you’re saying, right? Yeah, I feel really quite confident we’ll get to 10 cents over the coming decade, and in probably over the first half of it. So this is sort of like headline figures, not what you could get for, what not what you’d pay for a module delivered here in Australia, for example, in, you know, moderate volumes. But still, it is a forerunner of what you will be able to get in smaller volumes in the future, I guess. But I think it will, we’ll see these headline figures of 10 cents a watt, in the first half of this decade. And definitely within the decade. Amazing. And again, forgive me, Martin, I’m going to drift off script for a moment because you’ve given me so many good ideas, but what’s what’s on your horizon? What what’s coming next in the labs? What are you working on that you think is going to contribute most significantly to the next generation of PV modules and the next leap in performance and the next leap in cost reduction?  You and your team have been so instrumental in helping drive these cost reductions and performance improvements, and you know, there was always this this battle when I was within BP Solar between “do we try and drive efficiency up so that we get more watts per square meter, which lowers the installed cost? Or do we just try and drive prices down by going for economies of scale and cheaper materials?” What’s  coming down the pipe from your guys?

Martin Green  14:47

Yeah, no, I think the really big problem to crack in photovoltaics and the most important research problem, you know, for us researchers to concentrate on is developing tandem cells ,and yeah, that’s where you stack a cell of higher band gap, you know, on to silicon. And I think if you stack it onto silicon, you’ve got a natural market entry. You know, the existing manufacturers can easily incorporate such a product into their existing manufacturing facilities, just a couple of add on steps to deposit another cell probably as a thin film layer on top of the silicon cell. So there’s a lot of interest in the moment, there’s probably most people involved with solar have heard of perovskites. So they’ve been quite remarkable…

Nigel Morris  15:33

Giles and I talked about perovskites, all the time, it’s held enormous promise. And there have been a lot of lot of records broken, a lot of sort of milestones made recently, including some fascinating ones. I can’t remember what it was, but there was some really innocuous and inert finding that helped to lower the degradation right of perovskites. And so is that where, is that the next big step do you think? Well, it would be great I think if it was, but  perovskites have a couple of problems. And one of them, the biggest one, I think, which is, you know, the real killer, is the stability. So if you look at reports of data coming from the field for fielded perovskite modules, you’re flat out finding any module that has maintained, you know, performance within 20% of its starting value, you know after a month in the field, or even a few weeks in the field.

So, you know, a lot of the testing results you see, you’re just testing under one parameter and things like that, rather than in actual field conditions. So there’s a really serious problem, it’s not just slightly unstable compared to where it needs to be, it’s orders of magnitude more unstable. So, you know, I was hoping that someone could find a way of fixing that problem. But, you know, there’s definitely been improvements. But they’re more like linear improvements, where you need exponential improvements, you need a major change in, in the properties of the material. So many people are working on it, so there’s still a chance that problem will be solved. But if you do solve that problem, no one’s found a way of getting the lead out of the perovskite cells. So there are about 30% lead in the, in the absorber material that you, that the perovskite compound is based on. So you know, you’ve got problems with lead. And I think eventually, you know, as solar is installed in larger and larger volumes, and more and more people become involved with them, you’re going to have to have a environmentally clean material. So I think the pressure to restrict yourself to clean materials is going to increase with time so that if you got manufacturing perovskites, you know, it might be a short loop to manufacturing future as your market closed down as different countries started banning their use. So it’s  not the perfect material. So we’re trying to find a material that, you know, works as well as the perovskites but doesn’t involve toxic materials and is stable. So that’s the two key criteria. Because we’ve seen this happen over time. I know again,  going back to my BP days, we we used to produce a CdTe, cadmium telluride thin film module.

We also had amorphous thin film modules coming out of different plants. We had the enormous thing, you know, polysilicon versus monosilicon battles that went on for years. And you know, when you look at the statistics now, what’s used around the world, the thin film modules have largely gone by the wayside it would seem,  and you know, despite the lower inherent efficiency of polysilicon cells, the PERC technology that you guys have developed is elevated them right to the fore. What happened to thin film?. Yeah, no, it’s it’s a good question.

I guess the two leading contenders, you know, cadmium telluride and the CIGS, the copper indium gallium diselinide both had a problem in that they involved scarce materials like tellurium in the case of cadmium, Telluride, and Indium in the case of CIGS. So I did, I had looked at this quite closely, you know, about a decade or so ago, but it seemed to me that there wasn’t enough Indium or tellurium available, you know, on the likely timescales they would be needed to do more than about 30 gigawatts, a year of photovoltaics, and many people think we’re going to hit 200 gigawatts a year this year in manufacturing volume.

So, you know, it would be those two materials be restricted to a small share of the market, even if they had been successfully commercialized. So, you know, of course, you don’t have that problem with silicon, and the introduction of PERC has shifted the emphasis almost entirely to monocrystalline, rather than the polycrystalline silicon. So, you know according to some market researchers I listened to recently, they say that the only operating polysilicon lines are now in India. The Chinese are completely dropped poly silicon module production. So, anything coming out of China that is polysilicon is old stock. Wow. Wow. I mean, it’s been so many years in the trenches, Martin, and  of course, when BP Solar and Solar- X merged Solar-X were renowned for their groundbreaking historical work in the polysilicon field and BP Solar had always focused on mono and we had to find a way to jam this product mix together. And it was very tricky because the entire market even to this day, I still hear sales people saying I know you got to go, you got to go poly because of this, or you got to go mano, because of that. But almost the entire world now is mono PERC, right?

Martin Green  20:55

Yeah, yeah. So according to one of the market companies, PV tech, who I listened to a talk by one of their guys recently, and their analysis is that 90.8% of the market last year was PERC and they maintained that 95% will be PERC this year. Monocrystalline, almost entirely of that. 95%. So that’s including the thin film, so the thin films and non PERC silicon technology had that other 5% to fight over.

Nigel Morris  21:28

Wow, wow, amazing. So let’s just, let’s go to the future a little bit and talk about, again, what I’m really trying to look for here is, you know, what are, what are solar companies needing to be aware of so we know where the price is likely to go, and they’re gonna have to look to generate revenues and generate profits potentially in other ways, because it isn’t going to come from solar cells, when they get to 10 cents a watt, they’re gonna have to find other ways to cover their overheads.

What’s your most controversial prediction of what’s going to happen? What’s going to change? In the next few years that if you’re a solar company, you should be prepared for Martin? Is it? You know, around the edges of of PV modules what else do you see coming down the pipeline that’s going to make big changes?

Martin Green  22:27

Yeah, well, I guess, you know, batteries, battery prices are coming down as quick as solar. So you know, I think batteries are going to play a more important role, definitely, in the, in mixed up with solar. So that’s a way of boosting revenue. And I think, you know, Tesla in the US are now only selling solar systems with a battery, so I guess that’s a model for installers to look at. And, you know, like bigger systems, you know, with the price coming down, like I’ve put in three different solar systems on my home’s.

You know, every 10 years, it seemed to be I upgrade, but my first one was one kilowatt, and that cost 20,000. That was back in 1999. The second one was three kilowatts, and that cost 20,000. That was in 2009. And now I’m looking at putting 14 kilowatts on my roof. And that’s gonna cost about that, too. I haven’t got the final figure yet. But I’m expecting that.

Nigel Morris  23:21

Wow. Wow.

Martin Green  23:22

So I guess everyone’s going to bigger systems now because it’s so affordable. So I guess that’s one way each job is going to be bigger. So rather than doing one kilowatt, like you’re doing at the turn of the century, you’ll be doing, you know, 10s of kilowatts.

Nigel Morris  23:39

I remember we used to have a half kilowatt system on our priceless back when I was at BP solar, which we thought was fabulous. But anyway, there you go! Don’t things change? Dr. Martin Green, it’s been an absolute honor and a pleasure having you as a guest on Solar Business this week. Thanks so much for joining us. Thank you, Nigel. It’s been great talking to you as well. Good o n you mate. Well, that’s a wrap folks. Thanks for listening. Thanks to our special guest, and thanks to our sponsors of course. My name is Nigel Morris, and I’m your host. Please tune in again in a fortnight for the next episode of Great Solar Business.

Closer  24:21

Great Solar Business was brought to you by Solar Juice. Australia’s leading solar distributor. Solar Juice aligns themselves with brands that share their values of service support, quality and value for money. Check out solarjuice.com.au and let Solar Juice help you become a great solar business.