George McClellan’s background in solar panel production technology and his current role at REC

Kerim: Hi, everyone. This is Kerim Baran with SolarAcademy. I am here today with George McClellan, of REC Group. Hi, George, good to have you.

George: Good afternoon. Thanks for having me out, Kerim. This is a wonderful opportunity.

Kerim: Thank you, for being here. Today, we’re going to talk about REC’s HJT technology in relation to its new offerings, but before we go there, George, I want to talk a little bit about you and your background. I’ve looked you up on LinkedIn. You very interestingly have studied aeronautics in college first, and you’ve worked in a number of jobs. But you’ve been in solar for close to probably two decades if I’m not mistaken. You’ve worked deep in solar module production and technology. Can you tell us a little bit about that background, as far as you want to go, and how you ended up in REC?

George: Yeah. Absolutely. You have done a little bit of homework. I appreciate that. I actually spent years in the Air Force, spent seven years in the Air Force, and that’s where my aeronautics degree came from, once I left. I got that while I was on active duty.

I got out of the Air Force and was working within semiconductor production and I moved on to Sandia National Labs, where I was in fundamental device design, research, and development and which was really cool stuff at Sandia Labs. You have the best toys and a bunch of just absolutely brilliant people. 

Kerim: After having worked at Intel, right?

George: This is, so I moved from Sandia to Intel. So Sandia was great, but I needed to move and do some other stuff. So moved into it, worked at Intel in process development and once I’ve been working there for a while, my daughter had gone off to college, I was an empty nester, and some friends of mine had gone out to this new solar startup. This is in the early 2000s, got out to this new solar startup. And they said, “Hey, you know, I think we’ve got a position for you out here. Why don’t you come out and check it out?”

And you think about the early 2000s where you got people to work in solar, there were no schools, there was no training on it. So you just found someone who knew how to work. So this is solar panel development, solar cell development. You find someone who knows about processing silicon. And so we just grabbed a bunch of people from the semiconductor area.

So Intel, some of the manufacturers, that’s where it came. So I went to work for a company called Advent Solar, which was great. It was a good opportunity. The product we were doing was a back contact cell on polysilicon, which is very similar to what SunPower was doing at the time, only they were doing it on mono. 

Because we are doing it on poly, we couldn’t do it with the same efficiencies. And because we were a startup, we certainly couldn’t do it at the same cost. So our factory gate costs were like $4 a watt or something like that. So it was not the best technology, but it was great and good experience. I learned a whole lot there. I learned how to build a factory. We did some great stuff, but I moved from there over to REC in 2009. 

The good part was there were lots of opportunities in solar at the time. But if you wanted to go somewhere and do something, you could probably find a company that you would work with.

REC had a very good reputation. They were just coming to the United States, and looked like they had a good product. They had a good backing. So if you think about the early 2000 in 2008, 2009, and 2010, you’d think about all the companies that were out there that are gone now, whereas REC is still stuck around. 

Kerim: Yeah.

George: So they’re doing some things right. It was a Norwegian-based company, but the Norwegians have got their geniuses when it comes to strategic long-term plans. So I realized that was probably going to be a good thing. I started working with them and became their technical manager for the United States.

There’s one of me in every country, not every country, but all of our regions. 

Kerim: Major regions.

George: You’ve got them in APAC, you’ve got them in Australia, you’ve got European, you’ve got Eastern Europe. The good part about this is we answer it, so with the technical department, we answer questions. That’s our job. And we interface with the customers that have questions. It can be everything from why my warranty is taking so long? What are your dopant regions? What are the dopant levels in your cells? It leads to a very broad knowledge base, and you have to be able to share that with the customers in a way that they can understand.

Kerim: And these customers that ask these questions, are they utility-scale developers? Are they residential solar installers? Anything in between?

George: They’re just about everything. It’s everything from a homeowner, who has a 5-KW system on their rooftop to a CTO of a multinational company that’s just bought 600 megawatts and wants to discuss the finer points of a cell adhesion or something.

So you have to be able to talk the language of just about all of these people. Mostly it’s working with installers who are trying to understand either mounting or installing products. What are some of the issues there? I work a lot with the sales teams trying to figure out what are the finer points? What are the unique selling points of our product? A lot of performance modeling, and understanding and looking at the competition. What’s out there, and how do we compare?

There’s a lot of good stuff out there as well. So we’re not going to say we’re the only game in town, but we’re a very good product.

Kerim: Understanding what their products are, how they are designed, perhaps, reverse engineer some of that and incorporate it.

George: Absolutely. There’s a lot of good stuff out there. 

Kerim: Yeah. It makes sense.

A primer on the evolution of solar panel technologies over the recent decades

Before we get into REC and REC’s unique offerings and technology, I wanted to take this opportunity, George, to talk to someone who’s been in this industry for close to two decades, plus or minus, and who’s been deeply in the technical aspects of solar modules.

Maybe talk a little bit about the evolution of solar panels, over the past decades. 

George: Yeah. Absolutely. 

Kerim: Can you give us a primer on that? I remember going as far back as 2009. There were thin film companies like First Solar, and there were multi-crystalline options, like many of the other companies in the market, like REC and QCells and Canadian Solar.

I know the basic differences between them efficiency-wise and price-wise, but so much has evolved since those days. Maybe, can you give us categorically how these things have evolved and where are we are in history? 

George: Yeah. Absolutely.

Kerim: Obviously, the Swanson effect is in full force, in effect, right now. Prices keep coming down, and panel prices have gotten what? Probably 7-10 times cheaper in the last 14 years or 15 years that I’ve been following. Let’s dive into that a little bit.

George: So yes, it’s been very interesting because I’ve got to witness an incredible evolution of an industry. If I think back to when I first started, we were doing 5-inch wafers. Our big goal – do we make a 50-cell or a 60-cell module? We didn’t even know and that was the 2004 timeframe. But we were excited because we were making a 135-watt panel, 60-cell, 135 watts. We were spiking the football. We had done it. 

There are a couple of different things that you look at. There’s the technology and then there’s the format. So one, is there cell technology? Then how are you gathering? So the cell is going to be where you’re gathering your charge, but then how you are gathering that charge, there’s a couple of differences there.

So we’ve gone from, you mentioned earlier, so we’ve gone from polycrystalline to – with the monocrystalline we’ve moved so the polycrystalline took us up to about 18%, 19% cell efficiency. Then what you do is you take the material system as far as you can go. Now you’ve run out of headroom. You have to change that material system. So what do you do? So you jump up to the next technology, which is mono, and you’ve got a P-Mono and N-Mono. And that gave us a good jump up to 23%, 24%. 

Kerim: That jump happened. Which years are we talking about?

George: So the industry was still trying to work. You have to stay as long as you can on your current material system because you have to. Nobody likes change, especially in the process areas. So you’re driving a lot on poly, poly, poly. You start doing some tricks with poly things like doing some backside coating. So your PERC came in on poly cells about 2016, 2017. The industry really started to push into monocrystalline. So now you’ve got monocrystalline, which gives you better efficiencies. But if you press the poly, if you do all of your tricks on the poly, you can get close to that monocrystalline efficiency.

So that’s why you were seeing things like the big push was going from, up to a 300-watt module. So REC was making a 290-watt module with polycells, with a PERC background, but the industry had shifted up to, “Oh, we need 300 watts.” There’s always these little, you know, I mentioned earlier, 135 watts. Yay. That’s great. But there are always these imaginary lines that you have to cross to get acceptance in the industry. 

So this talks about it there. So, yeah, we’re in the second section there where there’s 20% to 22%. So you’re going into P-Poly, P-Mono. So we made a poly PERC product, which got us up close to 300 Watts for our modules, but we just couldn’t get over that 300 Watts.

So then to get up there, you have to move to a newer type of mono. And that’s when you move to N-Mono. So this is moving into N-Mono, which started happening a lot in the industry, 2020 timeframe, and then the whole industry was really pressing towards that. Along with N-Mono, so there was standard N-Mono PERT, but then they introduced TOPCon a couple of years ago, as well as HJT.

And so that takes us up to kind of where we are currently. There is a very big push on getting everything we can out of these N-Mono platforms, TOPCon, and HJT because the next level of technology is introducing tandem junction and perovskites and things like that, which will change, so those are notoriously difficult to work with.

Difficult means expensive. And as you mentioned earlier, the prices for solar modules are just going down. So nobody wants them to go up, but everybody wants more power.

Kerim: On that note, I’m going to plug in our SolarAcademy chart section here and show, the first chart we put in the chart section is the Swanson effect. There it is, 1977 price of $76 per watt. This is the cell pricing, which is now this chart shows only to 2013 at 74 cents, which is now probably in the 20 cents, if not, I mean, depending on the market plus or minus probably. But yeah, that’s at the cell pricing. Of course, not the final product, but still, incredible, incredible hundreds of times cheaper.

So I’ll take that away now. Yeah.

George: Yeah, nobody wants anything to get more expensive, but everybody wants more power and wants everything to stay exactly the same. Now, the interesting piece about this evolution, as I mentioned, is not just about the technology, but it’s your current gathering and your cell architecture has also played a big piece in this. So in 2014, 2015 timeframe, half-cut cells were introduced by REC.

Kerim: Yeah. 

George: The reason you do half-cut cells is to reduce your resistive losses. Your resistive losses come down as the square… 

Kerim: Do you have a slide on that over here? Should I pull anything? No?

George: I don’t have any of the half-cut cell stuff because that was – this is just saying that the timing of all of this also coincided with technological changes. So you’ve got not, only you’re introducing N-Mono and P-Mono, but then you’re doing half-cut cells. You started increasing the amount of busbars that run on the front side of your cells. You started doing your backside passivation. So there are a lot of things that are going on, not just material systems technology. All of this is in response to the customer’s insatiable demand for more watts peak at STC. So all of these things combined are driving up cell efficiency, but also module efficiency and driving up your peak power of these panels.

Right now, we’re up in the 470-watt, which is what our current residential offering is, which is just, I think about what we’ve started out with, and it’s just amazing.

Kerim: Is that the same format in terms of height and width, or length and width? In the good old early 2010s I remember, like, we had 200-watt panels or 225 was the biggest wattage back then.

George: So the interesting piece is solar panels have always been rectangular in a kind of a pretty much a 1×2 aspect ratio. And from about 2010 to about 2016, we were so close to standardizing. We had 60-cell panels, and 72-cell panels, and that was it. And they were all 990 millimeters wide by 1776 millimeters.

Kerim: Yeah.

George: Just everything was within a couple of millimeters. It was so great. That’s one of the things that I really could not understand in this industry when I moved to it, there’s no standardization. And if you think, if you go back to when, how did computers get so cheap? I mean, the very first computers I bought for my daughters were $5,000. So then that was back in the 1998 timeframe, or 1996 timeframe. So how did we get cheaper to where you’ve got these fantastic computers for $600? It’s all about standardization. So volume and standardization. But here everyone’s just a little bit different.

What are the forces driving larger-format solar panels? What can we expect in the near future?

As the industry decided to change and the industry is driven by the customers, we’ve gone to demand-pull instead of technology-push. It used to be here’s what you get, buy it. Now the customer is saying, no, this is what I want. And so we’re actually fulfilling the customer’s need with the demand-pull.

So they want more power. We’re trying to give them certain ways of getting more power. That’s where half-cut cells came in. That’s where N-Mono came in. And we still stayed with 60- and 72-cell formats until probably two years ago. And then things just started going a little bit crazy where everybody wanted a 400-watt panel.

So they just started adding cells, which made your modules bigger. “Hey, that’s not technology. That’s just, you know, adding more, you know, X’s and Y’s.”

Kerim: Do you think that is being driven because the sales guy at the installer wants to say, “Hey, we got the bigger”, it’s like that…

George: Absolutely. So I like to…

Kerim: That’s just a little bit sad in a way because…

George: It’s very sad because what matters is kilowatt-hours per kilowatt peak. I know in the customer’s invoice, it says kilowatts, but what the customer is buying is kilowatt-hours.

Kerim: And maybe kilowatt-hours per square foot or square meter of panel space you put on the roof or on a specific space.

George: Absolutely. Yes. 

Kerim: Yes. I totally agree with having developed solar farms and put solar on my home. I get the math. 

George: Yeah. So people are looking at, “I just want more kilowatts” because their neighbors got an 8 KW system. They needed 8.5 KW and it’s kind of like –

Kerim: There are a few layers of high school physics that people need to remember to do that. And even my friends with advanced masters and PhD degrees, I see them just not putting in the time to understand that extra level of calculation and math. And so that’s why the panels are getting to be slightly bigger in size, and we’re not standardizing on a specific format.

George: And it’s going to change even more in the next 18 months to two years. What I liken it to is, when people are buying a car, they look at horsepower.

Kerim: Right.

George: Horsepower doesn’t really matter. It’s miles per gallon that matters. But everyone wants, “Oh no, I’ve got, you know, I’ve got, you know, three…” 

Kerim: Ask that from my 17-year-old boy.

George: Exactly. 

Kerim: Big difference. 

George: I know. But what really matters in the end is miles per gallon. 

Kerim: Absolutely, absolutely. 

George: But you hit on a really interesting piece. In the past couple of years, things have started to get a little bit crazy with module formats. So one of the things is solar cell sizes are changing. We’ve moved slowly from 156. So we’ve moved from M4 to M6, M6 plus. So we moved up 162-millimeter, and the cell manufacturers have realized that to make a solar cell, it costs the exact same amount of money to make a 162- millimeter cell as it does to make a 210-millimeter cell.

Kerim: Is that because of the way it’s cut in circles? 

George: So yeah, you’re pulling an ingot, you’re drawing out a boule, and the size of that boule really doesn’t matter. The cost of pulling that boule, a little bit larger, is not as significant. It’s a second-order effect when it comes to the cost. So now they can pull these bigger boules. They can make the G12 cells or the M10 cells at the same price as making an M6 plus cell.

Kerim: Yeah. 

George: So why would they make an M6 plus cell when they can sell a G12 for more profit? So what’s going to happen is those M6 plus cells are not going to be manufactured anymore. And that’s why there’s this preponderance of G12s and G12R and M10 cells on the market right now, which are bigger. They’re 210-millimeter wide cells. That changes the format. That changes your electrical layout, and it changes your physical layout. That’s changing everything in the industry. Our new products we’ve got a 44-cell product right now. That’s the –

Kerim: What’s the wattage on that?

George: So we have a 44-cell product that is 470 watts. That’s the Alpha Pure RX series. What’s interesting about the Alpha Pure RX is that because of the cell size, we have to run the cell strings towards the short side, parallel to the short side, as opposed to most modules, you can see in the background here of my video, they’re run parallel to the long side, which changes the location of your junction boxes.

But lots of companies out there are getting very creative on how they are going to put their product together, to try to be on maximized power, but also have something that can be installed. Our new product is actually shorter and wider than most products. It’s 44 inches wide or 48 inches wide, which is six or eight inches wider than most panels, but it’s also shorter.

It comes in right at 50 pounds, just under 50 pounds. I mean, there are lots of things you can do. You can have a 700-watt module. It’s just going to be really big. 

Kerim: Yeah. 

George: There’s going to be a lot of cycling around between all of the manufacturers as they move to G12 and M10 cells over the next 18 months, what classes will go up, but you’re going to get to a point where it’s just going to be so with utility-scale, I think it’s going to be brutal. They’ll have 8-foot modules out there. But they’ll also have 750-watt modules. It’s crazy. So that’s almost one horsepower. If you think about your high school physics, it’s almost one horsepower in a single module. That’s crazy. 

Kerim: Yeah. Wow. 

George: That’s pretty cool. But anyway, that’s what I see is going to be happening with the evolution of product right now because there’s a lot of things changing and they’re not changing because we want them to. They change because the customer demands more power, more power, and lower costs. It’s the triangle there. You can’t change one side. Everything else changes as well.

Kerim: I mean, you have been so involved in this segment of the value chain, I would love to do a side conversation one day with you on what your personal views are on whether we should bring standardization and how, but that’s a conversation for another day, maybe a future solar conversation.

George: I’ve got some unique views on that.

Kerim: Yeah. Well, maybe we can tag it onto the back of this one. 

REC’s current offerings, HJT panels, and their benefits

But for now, I want to go into a little bit about where REC is in its current innovation curve offerings. And let’s talk a little bit about the new HJT technology and the benefits it brings to the market today.

George: Absolutely. So, yeah, as you mentioned, we’ve got a couple of products in the market right now. We are still manufacturing an N-Mono TOPCon product. That’s the N Peak 3 series. Great product coming in about 400 watts, 410 watts, and 420 watts. But what we are really standing on is the HJT product, header junction product, and that’s the Alpha series. So that’s the Alpha, Alpha Pure, Alpha Pure2,  and Pure R series and Pure RX. So those products are all HJT. They are on two different size cells. So the Alpha and the Alpha Pure series are M6 plus cells. The Alpha Pure R and the Alpha Pure RX are actually on the G12 cells, which again, causes a shift in the way that the cells are laid out.

It’s a very functional product. It’s great. It works well for residential as well as C&I because it does have up to 470 watts, so it’s a good player for the C&I. It’s also not huge, but there’s a lot of products out there that are 500 watts, but they’re 7 feet long. That messes up ground coverage ratios pretty bad.

Kerim: Of your total offerings, how many main SKUs do you currently have from the resi to C&I to utility offerings? And how do the various technologies play into that? So HJT is not a utility solution, right, these days. I assume there must be a reason for that. 

George: That’s a very good question right now. In order to reduce the amount of SKUs in people’s warehouses, we moved to a minus 0 plus 10 watts binning strategy. So we just manufacture 400, 410, 420 in say, our Alpha Pure series is 400, 410 and 420. That’s it. We don’t do 400, 405, 410, 415, which has been accepted and preferred by the installers. It used to be taht everybody always wanted, “Just give me that extra 5 watts. I can settle on that.” But it’s easier for them to just say, “Nope, it’s going to be a 410 or a 420.” And that’s what you get. That’s across all of our Alpha Pure and Alpha Pure RX products. We’ve actually manufactured the last one of those. 

We’re moving up to the Alpha Pure 2. So that’s one product line with three SKUs. The Alpha Pure R will “end of life” soon. It will move to the Alpha Pure RX. Again we’ll have 450, 460, and 470 watts.

So again, three SKUs for that. Later this year, we will be introducing the Alpha Pro M series. And that is 640-watt. It’s G12 half-cut cells in a different layout, so it’s actually a larger module, but that will be coming. Basically, we’re planning for that for C&I.

Some people may want to stick it on a rooftop, a resi rooftop. It’s a bit bigger. Well, it’s a lot bigger, but that will be our 4AN to C&I as well as it can be used for small utility, but it is not a bifacial panel. It doesn’t have a clear backsheet. That will be coming out probably within the next 18 months, a clear backsheet as we move back into the utility-scale space.

Further explanation of PERC/PERT, P-Mono vs N-Mono and TOPCon panels

Kerim: So talking about these various different technologies, I’m going to go share my screen again to summarize the differences between all these different products that we’re talking about. So going back here, George, if you can give us a little bit of a summary on the various different technologies that are out there right now, what are the differences between, for example, P-Mono and N-Mono offerings, as well as TOPCon, and the HJT. And maybe compare PERC and PERT and go up.

George: Things have all been so that we’re always looking for a better way to do things in an industrialized setting. A lot of this stuff has been around for a very long time, but it’s really, how do you commercialize these things? So the easiest to commercialize was a polycrystalline product. So that’s a P-type polycrystalline product with an aluminum backside field. That’s what the BSF is for. That’s your standard product that was made from the seventies up to the mid-2000s. We started moving into monocrystalline products, which is a more purified single crystal product wafe in the 2010 timeframe. 

We also started introducing backside passivation processes. That was the introduction of PERC, which is for the passive emitter rear cell. And then the follow on to that was PERT, P- E-R-T, passive emitter rear totally diffused. So similar products, but what that does is that increases the efficiency of the cell by using a backside treatment of the cell. 

Now, as we got more sophisticated, we moved from P-Mono to N-Mono, and the difference there is the bulk material is either positive or negative. You’ve got a P, the bulk material is charged one way and then you put your PN junction on top of that and that’s where you generate your current across that PN junction.

So if your bulk material is doped with a positive dopant, so silicon is a semiconductor, it’s neither positive or negative. You can turn it one way or the other based on the dopant or the impurities that you put in it. So if you put Boron in a silicon ingot, you’ll end up with positive or P-Mono.

The problem with P-Mono is that it is subject to light-induced degradation. So that’s LID, which we’ve all heard about, but that’s your degradation that happens in the first 50 cumulative kilowatt-hours of exposure to the sun, and that is caused by Boron-Oxygen defects. So we moved from using Boron as a dopant to a negative dopant, which is Phosphorus. And now that gives you an N-Mono product. If there’s no Boron, there could be no Boron-Oxygen defects, which means there could be no LID, which now you’ve gathered another one and a half points of efficiency. You’ve reduced the degradation of your product just by moving over to this material system, this N-Mono. 

Kerim: Wow.

George: So now you have your N-Mono product, which is going to be more efficient and is not subject to LID. So now you’ve gained efficiency and you’ve gained power, without having to spend too much money because you can make super efficient stuff. It just costs lots and lots of money.

So the goal is to keep the cost down. So then we started doing backside treatments on our mono products. We introduced PERC and PERT to mono, but a few years back, tunnel oxide passivation, so that the rear coating on the back of the cell was done by a tunnel oxide, which is even more efficient than your PERC and PERT.

Kerim: So that’s the TOPCon technology.

George: That’s TOPCon. And TOPCon is very good, but it still is not changing things. So it’s similar in that it’s a backside treatment. 

The technology behind HJT and comparison of HJT, TOPCon, and PERC Solar Panels

We moved to HJT. Basically, this is a –

Kerim: That’s the latest and the newest and the most current – 

George: This is the most advanced. Well, the interesting piece about HJT is that HJT has been around since the early 2000s. It was originally manufactured by Sanyo. That was their HIT product. They copy wrote it as HIT. So that’s why no one says HIT. The Sanyo product was using HJT as well as the product from – Panasonic bought Sanyo.

If you remember, they used to be very powerful products, but they were also twice as expensive as everything else. REC figured out a way to make the HJT cell without the excessive costs and also came up with a new type of interconnect. This again, is for our desire for more watts peak, but you can see what it comes down to is it’s basically a sandwich.

You’ve got an N-Mono cell in the middle there. And then you’ve got a silicon layer that goes on top of the N-Mono on the top and the bottom of that cell, which now increases the spectral response of that cell. So the cell now can see, is responsive at the regular N-Mono wavelengths, but also the amorphous silicon on the bottom and the top of the cell, it also responds to those wavelengths. 

So you’ve increased the spectrum that your cell, your overall cell is responsive to. One of the good parts about this is that it has increased the sensitivity on the long end of the spectrum. So around 1200 nanometers that’s where your red light is.

And it’s turning that red light into energy, watts, not temperature. So it’s not dropping this. So what it does is that infrared light comes in, hits the cell, they turn that into current, as opposed to heat, which brings down this – 

Kerim: Is this the difference that it results in?

George: And this is one of the differences. So this is an annual production of an array. This was modeled. This is an 8-KW array in Palm Springs. But you can see that the energy yield, because we’ve got more energy, is being transferred into, more photons are turning into current as opposed to heat. And so, yeah, this shows a higher production because we have a better thermal coefficient for our product. And that’s one of the benefits of HJT. 

So if you look and compare this to other products out there, lots of people are looking at, well, what is the bifacial rate? If you look at the way that the product is made, you can see it’s like almost a perfect sandwich. So the bifaciality rate of an HJT product is almost 95%, where if you compare that to TOPCon, and even PERC, if they try to make a bifacial product, they will work as bifacial products, but the bifaciality is lower for these other products.

And that’s just cell architecture. Your efficiency is significantly higher, the degradation again, by protecting your products, also having an N-Mono product, your first year degradation is going to be very low. Annual power degradation –

Kerim: That’s incredible. So I’m looking at that. That’s 0. 25%. 

George: Right.

Kerim: So do you guys then write warranties that promise like 93% plus at year 25?

George: 90%-92 percent

Kerim: 92%? Okay. Wow. 

George: 92% at 25 years, which is just, I mean, if you think about that, 92% of nameplate value after 25 years is just incredible. I mean, what’s out there at this point to last 25 years, and also be able to provide 92% of that nameplate power after 25 years? It’s just incredible. 

And we have things coming in from the field that are showing. So this has been one of the benefits of being an early adopter of HJT. We started making it in 2019. We have got products out there in the field that we do comparative energy yield analyses and we’ve got products in the field and they’re meeting this degrading significantly less than this 0.25%. 

So just a really fantastic product out there with respect to annual degradation. And part of that is due, again, as we mentioned earlier, to this temperature coefficient because it’s turning more photons into energy, as opposed to heat. So you’re keeping your product cooler.

And everyone, I think we all know that as solar panels have a negative temperature coefficient, as the temperature goes up, their performance goes down. But the contrary to that is also true. As the temperature goes down, performance goes up. So they ultimately will operate very well on a cool, clear day.

Kerim: One more thing I want to make sure we touch on is the micro cracks issue. I think this was the slide relating to that. Right? Can you tell us a little bit about the benefits of HJT in the context of micro cracks? 

George: This is a really good image. I’m really glad that we can share this because I thought there was some secret sauce here, but I guess it’s understood by the industry. 

The way that HJT cells or the advanced interconnect cells – some people call it smart wire, but smart wire was trademarked by Meyer Berger. So everyone just says advanced interconnect, but connecting the cells, the cells are not connected by wires anymore. They used to be connected by Y interconnect ribbon. They started out with two ribbons, then went to three, then went to five. And it really started getting kind of all over the place. We’ve moved away from that and gone to very, very fine  conductor wires that are embedded in a carrier film.

Kerim: Is this slide related to what you’re talking about?

George: Yes, this is exactly. So you can see, standard connections use these five bus bars.

Kerim: Yeah.

George: Whereas, the REC Alpha Pure cells used these very thin wires. They’re in a carrier film. That carrier film is very thin, almost a cellophane. It’s a TCO, a transparent conductive oxide. It’s very thin. It’s like the stuff that you use, the film that would wrap up DVD packets or an old cigarette pack for those that smoke cigarettes. 

So these wires are carried in that, but they’re all embedded in that. They’re laid on the top of the cell and then they’re bonded electrically during lamination. It’s a very low-temp bonding. 

Kerim: Yeah.

George: So the thing here is you’ve got this extra layer of this transparent conductive oxide that goes over the front and the back of the cell. So that’s providing another layer of protection. So you’ve got your cell, then you’ve got your TCO layer, and then you’ve got your encapsulant on top of that. And then you’ve got your glass on top of that. So you have an extra layer of protection there. 

What that does is provide a stronger bond between everything. So now it’s a very strong lamination. The other piece is that the cells are thinner and because they’re thinner, they will bend more. They don’t end up cracking. So what happens there is that you’re more protected against any loads, such as an installer walking across your rooftop or hail or things like that. 

Microcracks in solar panels and how HJT addresses them

These panels will actually survive better in that environment because they are better protected. It’s a stronger, stiffer platform, but it’s also, if it does flex, it will flex and not cause micro cracks.

And micro cracks are a real concern because when they happen, they don’t show up. They don’t show up immediately. The problem that happens is after a few years, after this thing is cracked through thermal cycling, this micro crack grows into an actual real crack, separates the conductors, and now you’ve lost a cell or maybe a panel.

So micro cracks are a ticking time bomb. I don’t know if I’m allowed to say that, but they are a future problem, not a problem right now. So those are the –

Kerim: Why do you think they are like a future problem? Do they get worse over time?

George: Yeah.

Kerim: Do they even cause fire or things like that?

George: Yeah. What happens with a micro crack is that when the damage happens, you don’t know it. During an installation, if someone’s standing on a panel or if the panel got dropped, you can still check that array. It will still check out as being full power. And so the installer walks away. Says, “Yep, I’m done.” 

But during that, over the time because of thermal cycling the temperature going up, coming down, going down, coming up, your panel’s going to shrink and grow, shrink and grow. And what will happen there is that crack will turn into a full size crack,  breaking the conductors or causing a hot spot and hot spots are not going to cause a fire as much as it will cause damage to the system. It’ll cause a hot spot and that will cause damage to the back sheet and open up, it could burn or melt the back sheet, basically. And if it compromises the back sheet, now you have a path for moisture incursion. And moisture incursion is bad because you’ll start shorting out your system. You’ll get arc fault, which if you’ve troubleshot an arc fault problem on an array, they’re really hard to do because once they get wet, they’d short out. But if they dry out, they don’t short out anymore. 

So your system performance is going to drop off. You’ll lose modules. It’s difficult to troubleshoot that type of thing. And the bad part is that damage may have happened three years ago. And so everyone’s just pointing fingers. “No, it’s your problem.” “It’s my problem.” And you know –

Kerim: I’ve recently heard of a startup, just taking those fleets of residential solar assets and changing the back sheet on them, for a certain fraction of the price of a new solar module. 

George: Absolutely. Yeah, we’ve learned a lot. So if you think about what we’ve learned, how many gigawatts we’ve installed since, the modern era, so say 2010, how many gigawatts have been installed? So if you think about what we did from 2010 to 2014, there are some pretty bad modules being manufactured, I’m not gonna lie.

So the prices, if you think about price per watt, so that was the overseas invasion, price per watt just absolutely tanked. How did we get there? Is that economies of scale? Or were there some things happening there that may bite us in the end?

And we’re finding out. We’re seeing back sheets fail. We’re seeing EVA brown. We’re seeing bad connection points and all of this had to do with the move from manual assembly, which a lot of people were doing up till about 2010-2012, to now where everyone is fully automated and has very good process control. But there’s a bunch of stuff out there that we’re going to be having some problem with over the next 5 or 10 years. It’s an opportunity.

Is HJT technology coming to utility-scale solar?

Kerim: Do you think the HJT technology is going to come to utility-scale solar as well? Or is it already? 

George: Absolutely. So that’s a really good point. So utility scale the reason that a lot of companies aren’t selling that into utility scales, utility scale is a very low, low margin business. So how do you –

Kerim: And you guys do utility too, right? I mean, it’s probably, that a significant percentage of your business is utility as well, right? 

George: We used to do utility. We stopped doing utility in 2018 because we were capacity-limited. We only had 1.2 gigawatts of capacity, and we were selling all of that into residential and C&I.

Kerim: In Europe and US mostly, I assume, right?

George: Yeah. So there was there was no need to sell into utility scale because you sell into utility scale project, you’ve just sold 400 megawatts. So with us, that would be a third of our capacity, annual capacity at very low margins. Now, if you’re making 7, 8, 10 gigawatts, you’ve got extra, you’ve got capacity that you can sell off into that. Those margins aren’t going to kill your bottom line. 

So we are moving into producing utility-scale products within the next 18 months, but in our factory in Jamnagar, India. But the question, will HJT make it into utility scale? I believe it will. It becomes a question of can you make it affordable enough for the utility scale? Utility scale really like having more power. And if you remember, if you go back to the slide about the benefits, our bifaciality, the bifaciality of an HJT product is higher than TOPCon or any of the other HJT products out there.

So if you’ve got a 95% bifacial rate, you’ll be getting more energy. Ultimately –

Kerim: Five to ten percent more energy than the regular.

George: Right. And that’s huge in utility scale. So then it just becomes an LC, a levelized cost of energy question. So you’re going to be getting more energy, but what is the cost of that energy? It becomes a tradeoff.

Kerim: Yeah.

George: There’s a lot of work that needs to be done on bifaciality. Understanding bifaciality, how we can optimize bifaciality for utility scale. And once that I believe that is figured out, the HJT product will become more used. I don’t want to say it’s prevalent, but I believe there will be use. I mean, we’re banking on it. It will be happening. We will be selling it, that’s for sure. 

I think there’s going to be a good market for it. But can you fight with some of the prices of some of these products coming out? I don’t know, you know. We’ll see. I mean, if I’m buying something that’s going to be a 25-year relationship, now they’re moving out to 30 years, 40 year relationships, I’m not going to go out and buy the cheapest product out there. There’s a couple of things you don’t want bargains on. You don’t want to bargain on brain surgery and tattoos. I’m not looking for the lowest cost on those things.

Kerim: Yeah. And you having worked at REC for 14 years now, if I were buying from REC, I would feel good that somebody knows why certain decisions were made the way they were. And that comes through in this conversation. So thank you for that. 

REC’s new product roadmap, perovskite innovations on the horizon.

With that can I ask you a little bit about what is to come in the REC offerings, over the next months, quarters, and years?

George: Yeah, absolutely.

Kerim: Is that something we can talk about in this conversation?

George: I like talking about those. So as I mentioned earlier, we’ve got the Alpha Pro M series. That is a product that is going to be coming into the US, probably December. It is a 640-watt product.

That’s going to be targeting C&I and small utility scale. We see that in up to 2 megawatts ground mounts targeting AG, things like that. And then also C&I rooftops. It’s a monofacial product because you don’t really need bifacial in C&I space.

People like to have it, but I mean, really what are you getting with a bifacial product, that is mounted coincident with the roof and really not getting much bifacial gain? So we feel that we’ll have good penetration with the Alpha Pure M series, in C&I and then some small utility-scale ground mounts where bifacial gain really isn’t that important.

We’ll be introducing the the Alpha Pro L series, which is also an HJT product, using half-cut cells, and that’s going to be up to, I believe, that’s 740 watts product, a bigger product. It’s 2.3 meters tall. So yes, that’s big. But that’s what you get when you string all these things together.

But that is focused clearly on utility scale. And that’s got lots of things. So it’ll be a glass on glass product. But it’s 1.9 millimeter glass, so it’s not going to be as heavy. So that’s the big dig on glass on glass products right now is that, yes, they’ve got a very good warranty because they’ve got no backsheet on them, but man, they’re heavy.

So you reduce the weight of that by doing thinner glass. If you’re mounting on a single-axis tracker, you’re mounting on the torque too, and so now you’ve got a longer module, it becomes big problems with the drive motors and things like that.

But the people out there at Nextracker and ATI, there’s a lot of very smart people out there that are learning, that will be there. They’ll have a solution in time. But yeah, we expect those to be coming out, moving farther down the line. I think lots of companies are looking at perovskites.

Perovskites will be a tandem junction. So there’ll be two PN junctions on the cell. That will give you, again, it’s our lust for more power, more watts peak at STC. So that’ll give us even higher, 850-watt panels, that type of thing. But I think a lot of people in the industry understand the limitations of perovskites. So they’re not as reliable, or I don’t want to say reliable. They’re not as robust as we would like for them to see.

Kerim: Yeah. Yes.

George: They degrade in sunlight. They degrade when they’re outdoors and in sunlight. Well, that’s just not a good match for solar panels, but they’re working. There are some very smart people working on that. And just about everybody has a solar perovskite plan in their future for when that happens. 

In the short-term, what I see is happening is where the arrays will be built with the perovskites on them. The perovskites will add 10%-12% power to that panel, but they’ll only last for 10 years.

Kerim: 10% to the 23%, so like another 2%. 

George: 10% in watts peak. So it’ll turn an 800-watt panel to a 900-watt panel.

Kerim: Oh, wow. Okay.

George: So it will drive that up to 900 watts, but it only lasts for 10 years.  That perovskite layer will only last for 10 years. So after 10 years, it degrades down to an 800-watt panel again, which is a lot of power.

Kerim: Yeah.

George: I think that’s the way that perovskites are going to make their entrance into solar because I don’t believe that we’re going to be able to make something that’s going to last 25 years right now, but that’s coming. It’s a perovskite-like film that we can add to that. I think everyone’s looking at perovskites. REC certainly is. I would expect that’s going to follow HJT. And then there’s – who knows after that? There are lots of good stuff out there.

Kerim: We’ve already done a conversation with Scott Graybeal, of Caelux, which is a company associated with REC. I believe you guys are an investor, if not the parent company of Caelux.

George: Yeah, it’s common knowledge that we’ve got an investment in Caelux, and they’re doing great work down there out of Pasadena. 

They’ve just moved to a new facility and they’re doing what’s called a 4-terminal, 4T. So what they do is they put perovskites on the front glass. That’s one solar panel and then they bond it with the other solar panel. So now you’ve got two panels, but you’ve got four (it’s 4T) terminals and you just join those two together. That’s your output. Pretty cool stuff. 

I believe 4T is probably going to be the most effective, as opposed to putting perovskites directly onto solar cells. That’s a bit more elegant, but it’s also showing to be a bit more problematic.

What makes REC unique among its peers?

Kerim: We’ve come close to the end now. Maybe I should ask you just generally what you think of REC. You’ve been there for 14 years. You must have some good thoughts of the company that you know it’s rare to find someone who stays at a company this long these days. So maybe you can tell us a little bit about what makes REC special and unique for you.

George: I would appreciate that. I like the idea to let you know a little bit about REC. So the interesting piece with REC is, I mentioned earlier that they do have a very good strategic vision and that’s part of their Norwegian heritage. 

I’ve spent a lot of time in the Singapore factory, so we manufacture at Singapore. I spent a lot of time over there providing help with the factory, doing factory tours with our customers and things like that, providing audits.

The team over there is very, very focused on quality. And that’s one of the things that comes through in our product. I always hear people say, “Well, you’ve got a great product, but how do you prove that?” Well, I’ll tell you how we prove it. Look at our infield failure rates, things like that, which are second to none.

We’ve got infield failure rate of about 60 parts per million, which is about one module for every 15,000 is failing out in the field. But this all starts at the lowest level. So you can go into our factory. One of the reasons we’re in Singapore is there is a fantastic workforce there that is very focused on quality and just doing – Singapore is interesting. The group there, they’re kind of fatalists. Everything has to be going wrong for them to feel good. And so they love to solve problems, always to make sure, “Oh, if it’s going good, there’s got to be something wrong.”

So it’s pretty funny, but everyone really focuses on just making things a little bit better, a little bit better. They’re not out to hit homeruns. But they got a lot of base hits, a lot of singles. Bang, bang, bang, bang. Just turn the crank. It’s nuts and bolts engineering. Like I said, earlier, I came from Intel Corporation and they’re fantastic at just incremental improvement.

And that’s what REC does. They’re not looking for the white knight. They’re not looking for the homerun. It’s just turn the crank, get it done, get it done. And it’s really interesting because I was over there once. I was doing a factory audit with Clean Energy Associates. They were coming over to do an audit of our factory.

And I’m walking around with the inspector and he goes, “Well, George, this looks really good, but are you sugarcoating anything or that type of stuff?” I’m like, “No, that what you see is what you get.” He says, “Okay.” 

He went to one of the technicians, and I can’t remember what it was. I think it was solder temperature, solder gun temperature or something. He asked the technician and goes, “Well, how do you know that solder gun temperature is correct?” And the guy was so excited. He’s like, “Let me tell you.” And he went right to the book. He went to the exact page in the book. “Here’s my calibration procedures. Here’s the timeline. Duh, duh, duh, duh.” And they just love, everyone to a person in that factory, just loves pointing out what they do and how good they do it, and it’s just something they are very proud of. 

And to see that where you go into some, you know, I’ve been into places where they don’t really care what’s going on. They’re just getting a paycheck. 

Kerim: Yeah.

George: You know, the factory could be burning down, but as long as they can clock out, they don’t care. But these people are just really, so that the group there – and that goes from the lowest technician up to the CEO – I mean, it’s just crazy. I’ve had some of the most interesting discussions, we’ve had a handful of CEOs, but each one, you have these discussions like, wow, these guys really know what’s going on. So it’s not just about, “Oh, we’ve got to take care of our customers or our investors and our stock.” 

It’s like, no, he’s very focused on, “What about this? You know, the way this machine is set up. Are you sure this is the most optimal?” So everyone there is very focused on putting a good product out. 

Now, the second piece of that is that we have grown this team in the US. So we grew it from, you know, we started in, we came to the United States in 2008 and there were four of us here.

I think we sold 64 megawatts, something like that the very first year. And now 64 megawatts is Tuesday for us. We’ve got a very good team. It’s a very high-performance team. We’re still a small team. We were 15-16 people up until a year and a half ago.

And we’re selling yeah, a gigawatt of product, which no one else is doing. But the only way that that happens, is if you’re all high performers on that team, and it’s a true team. So that’s one of the things we’ve put together here at REC. We’ve got the right people, we’ve got the right management, and the right product.

And so that’s what keeps me coming back. It’s real easy to wake up in the morning and go out and support this product because it’s great people to work with, great product. There is altruism in this industry. I mean, we’re doing a good thing.

“We’re saving the planet one day at a time” is what I say to my wife as I’m walking upstairs to my office, if I don’t go into our real office in San Luis Obispo. I go, “Save the planet, baby.” But it is place to work with. And it’s been good for the past 15 years.

Kerim: Thanks for sharing that, George. Thank you.