#29: The science of solar



Frank van Mierlo, CEO, 1366 Technologies



It turns out that solar is a kind of energy farming, and so your yield per acre is pretty important. And we have consistently increased the efficiency of solar panels over the last couple of decades, but we are now up against the fundamental limit.

Frank van Mierlo: Hello, I’m Frank Van Mierlo and I’m the CEO of 1366 Technologies.

JS: Frank, you know a lot about energy technologies and climate change but particularly solar. I want to start there. For people who are new to solar, can you talk about where it is today?

FV: Yes, absolutely. We’ve come a long way. Initially, when solar was first invented in the labs, it was around $10 per watt peak. Now, we’re very close to a cost that can deliver an electricity cost around five, four cents per kilowatt hour. We’ve seen a dramatic, dramatic cost decline over the last couple of decades.

JS: When you talk about solar you’re talking about traditional silicon solar? Is that right?

FV: Yes, solar has always been mostly silicon. There have been different types of silicon, Czochralski furnaces. But in solar, 80-90% of the market has always been silicon. There have been some thin-film technologies. The most notable there is a company called First Solar that has a product called CdTe. There was very inventive technology that came in parts out of NREL as well and has been commercialized by First Solar and they have factories in Ohio and other places.

JS: What you do at 1366, is that silicon or is that something else?

FV: Yes, we do silicon. We’ve always believed in silicon. For us the first piece we had printed out for the company was, don’t bet against rock, and it had a silicon symbol on the back. At the time it was actually not popular. There were close to 200 solar startups and many of them were doing alternative materials. We are one of the very few survivors of that particular batch and silicon absolutely stayed and we’re now more dominant than ever.

JS: I’ve heard you talk in other places about this idea of the next step for solar, as in solar only shines during the day. What happens when it’s not shining and how energy storage plays into that. Can you talk a little bit about your thoughts on that?

FV: You’re absolutely right. Clearly, if you’re going to use solar, and if that’s going to drive your economy, you need to have some way of powering things at night when the sun isn’t shining, so how do you store that energy? It turns out that there is actually a lot of very viable solutions available today. When the local nuclear plant was built in the middle of the last century, they had the opposite problem. Nuclear plants, they produce power steadily during the day and during the night, you can’t modulate that.

JS: One of the limitations seems like energy storage, the other one that I’ve heard you talk about is this idea of single junction versus tandem technology in solar. Can you just explain what that is?

FV: As you know, I recently gave a TED Talk on that topic. It turns out that solar is a kind of energy farming, and so your yield per acre is pretty important. And we have consistently increased the efficiency of solar panels over the last couple of decades, but we are now up against the fundamental limit. For a single junction, meaning one material, that limit is around 24% on a module. The theoretical limit is more like 30-31%—that’s the Shockley-Queisser limit—but in practice, because you always have some losses, you end up with around 24% in a module.

Clearly, it would be better if we could get a higher efficiency. If you introduce a second material, and just now that’s becoming possible, until now, that really has not been economically viable. We knew how to do this for space stations, but at a cost that was 100 times more expensive than regular solar. Because costs are coming down, because knowledge of materials is getting better, it’s now possible to make a module with two materials. Therefore you have one material harvesting the infrared light, where most of the photons are, but they don’t have that much energy. Then the other material harvesting the high energy photons, and then you can lift the total energy yield of the solar module.

Turns out that it’s pretty significant. If you think that you’re going to power all of the U.S. with solar, going to tandem, would save you an area of the state of Connecticut. It has a big impact. What the Green Revolution did for agriculture, tandem is going to do for solar.

JS: What is the second material that’s able to extend the band gap?

FV: There’s a couple of materials that people are working on. There are working prototypes in several top players. Basically, there’s about 10,000 different photovoltaic materials. Basically, you want something with a band gap of about 1.7 electron volts, so you have options there. The real breakthroughs are not so much in the selection of the material but more in the doing it in a cost-effective manner.

JS: Right, because this will only be viable for, you’re saying, space stations and really large scale things.

FV: That’s right. Until now, the multijunction panels were only possible on satellites where you don’t mind paying 100 times more than what you pay for a terrestrial panel.

JS: What’s making it possible now?

FV: Breakthroughs in the layers at the top, you need conductive electrodes, you need to have insulators that have really good transmission. Then you have to be able to apply all of that in a cost-effective way. It’s really breakthroughs in engineering, material, and manufacturing techniques.

JS: If the cost of solar is going down overall, is this new innovation going to further decrease the cost? Or is it going to increase it because you’re adding this new material on this new technology?

FV: It will absolutely further decrease it. Roughly, you can keep the cost per watt per panel the same, but you’re just simply going to get more watts per panel. Now the corresponding infrastructure gets leveraged over a more efficient panel. You need less lens, you need less labeling, you need less racking, and so the total installation cost is going to go down. This is what’s going to get us to the two cents per kilowatt hour. In most areas of the world today, we cannot achieve that, with this tandem, I think we can get there.

JS: What are the drawbacks to this?

FV: There never are drawbacks to this sort of thing. This is just simply a better technology coming online. There are no drawbacks. Spoken like a true MIT engineer, life just gets better when we invent better technology.

JS: You mention MIT, let’s go there a little bit. You graduated from MIT with a degree in mechanical engineering, is that right?

FV: Yes, a long, long, long time ago.

JS: Talk a little bit about your time there. Who or what influenced you the most during your time there and led you to where you are today?

FV: For me, MIT was a fantastic place. I had a lot of fun there. It was a fun place where I made a lot of friends. The key thing I learned at MIT was that I couldn’t get my problem sets done alone. They were too difficult. I became really good at organizing these sessions to do the problem sets. The way that played out is that I would find a smarter engineer, Audrey Hartman, Jim Bellingham, people like that. They would come and we would do the problem set together, I would serve tea, and everybody was happy.

In a way, I look back on that, and I think it was such great training to become a CEO. Because, first of all, it’s very important in life to recognize that other people are smarter. You get a lot more done if you get that basic concept in your head. Second of all, it is really important to get things done as a team. The MIT problem set structure really taught me that. If I had to pick a single thing that really worked well for me, it’s probably that experience.

JS: For people who didn’t go to MIT, explain what a problem set is.

FV: About 10% of your grade was homework that you had to do. This homework tended to be pretty difficult in some of the courses, and so a number of students would get together and together you’d try to solve a particular problem.

JS: What about your interest in energy? Did that come from MIT or did that come after?

FV: The first company I spun out of MIT is a company called Bluefin Robotics. That was a fantastic little venture, we pioneered autonomous underwater vehicles. It was really building on the work of Jim Bellingham, who I had met doing problem sets. He’d stayed at MIT and built up the Autonomous Underwater Vehicle Lab. We really pioneered artificial intelligence. Because there is no radio link to submarines, you have to endow these little vehicles with enough intelligence so they can execute their mission.

We did that, the company was always profitable, always grew double digits, and then was successfully sold in 2005. It’s currently part of General Dynamics. After selling Bluefin, I tried retirement for a little bit. My wife quickly informed me that it was pretty important I find a job outside of the house. I went back to MIT and at the time, energy was a big focus at MIT. I actually sat next to the then-President Susan Hockfield when she launched the MIT Energy Initiative in Kresge. That seemed like a good thing to get involved in. For about a year, I looked at all sorts of different technologies. I came to the conclusion that solar really was probably the right solution to tackle climate change. Then I started 1366 with a team of engineers from MIT and other people—people from RPI, and there was a local company at the time that was doing solar photovoltaics and we got a couple of their top engineers—and that was the beginning of the venture. Yes, absolutely both my companies have roots that started at MIT.

JS: I would like to dig in a little bit into the green hydrogen. Just because I don’t really understand what that is, so for anyone listening who maybe needs an explanation as well, can you just start at the top? What is green hydrogen and how does it play into solar?

FV: Today, if people use hydrogen the way that it’s normally made is you take methane, you crack it, and you get hydrogen. But that also gives you more CO2. A cleaner way to make hydrogen is to use electricity to split water. Water is H2O, of course. When you use electricity, you can split it into hydrogen and oxygen. If the cost of electricity is low enough—you need around two cents per kilowatt-hour—then you can actually produce hydrogen in a pretty cost-effective manner. Hydrogen is a fantastic fuel that can be the feedstock for many things. Once you have hydrogen, that’s such an energetic gas that you can use it to make steel, for example.

In a lot of places we are currently using things like coal and diesel and methane, hydrogen can replace that. One of the challenges with hydrogen is that it’s a small molecule and it’s not that easy to store or to ship. There tends to be leakage. If you want to store it for one time, you have to compress it or you have to make it extremely cold. A way to store hydrogen in a different molecule is to turn it into ammonia. This is what’s called the Haber-Bosch process. Then that ammonia is a fuel that’s a lot like diesel. You can use that for all sorts of things. Hydrogen really is the key to truly going away from carbon completely.

The first solution is really cheap electricity from solar and then you want to store it. Intermediate energy storage, short term energy storage, can happen with things like battery and pumped hydro. Then to transition the very energy-hungry industrial processes, and to do the seasonal storage, that’s where green hydrogen starts playing a role. All these pieces have to come together to completely change our energy mix. It’s a daunting task, right? This is not easy and nobody should ever tell you it’s easy. But it’s one that’s absolutely possible if we truly want it. If you would price carbon at about $75 per ton, the market would solve it for you. But you would need the political will to do that.

JS: So we’d need leadership that would maybe be enforcing that?

FV: Yes, and in order to get leadership, in a democracy, the majority of us need to believe that we actually have to solve this problem. These things are related. Ultimately, the problem is us. This is a problem that’s always difficult for human beings. It’s the same reason we have trouble controlling our weight, the same reason that we have trouble with our national debt. That is, problems that are in the future, if there’s tension between the future and now, human nature says that the now tends to win.

With climate change, that’s particularly the case because that’s a problem that is not even us, it’s our grandkids. But, boy, is it important we get it right. We really have to rise above ourselves. We need that psychological shift that we recognize this is a problem we want to solve. Then we’re eminently capable of solving it. We are an incredibly capable species and we have lots of great technology tools: tandem, green hydrogen, pumped saltwater hydro, great batteries, all of these things.

JS: It sounds also like maybe an issue of education. Maybe if more people knew the basics of how solar worked, or about green hydrogen, or how all of these pieces come together, they might be more likely to advocate for it or petition the government to be doing more.

FV: That’s where podcasts like the ones you are doing are so important. That’s where MIT clearly has a vital role to play. I think MIT and you are doing that and it’s an important task.

JS: If you had to send people away with one message or learning about anything—about climate change, about energy, about solar—that you think would benefit them, that you just wish more people knew about, what would that be?

FV: Really it’s the sense that tools have gotten much, much better in the last decade. We have solved a lot of problems. Now, it really comes down to a will to tackle climate change in a way that’s substantive. But the solutions are there. We are able to do this, that’s the message. We just have to now decide that we want to do it.

JS: When you say “we decide”, you mean politically? Because individually, there’s not much we can do.

FV: As a species, as a country, as a nation, the majority of us have to recognize that this is a problem want to solve. The other thing I like to point out is that if tomorrow you tax carbon, actually, that doesn’t cost any money. It’s just a policy decision. Tax you pay to yourself. For all the tax that you collect with taxing CO2, that revenue can be used to reduce the tax burden somewhere else. It’s more a smart policy than that it truly has to increase the burden in an unacceptable manner. But you need a price signal in the market that pollution isn’t free. You need to give people an incentive to deploy all of these solutions that are on the shelf and not continue with the way we are producing our energy at the moment.

JS: What’s the next challenge that you’re facing at your company specifically?

FV: Absolutely we have to make this tandem a success. We have to make sure that we build these gigawatt factories and deliver the quality and the reliability that we believe we can deliver. Then we need to continue to scale. My hope is that we create not a two gigawatt company or 10 gigawatt company, my hope is that we create a 50 gigawatt company that plays a big role in the next phase of solar and helps bring that cost down to two cents.

JS: Frank, I want to thank you for such a comprehensive discussion about solar and the surrounding technologies and your compelling vision for a better world. Thank you for the important work that you’re doing, and thanks for coming on the show.

FV: Thanks for having me.


Press inquiries: miteimedia@mit.edu

We're hiring! Learn more and apply