Extreme Electrochemistry for a Sustainable Future

Transcript

Michael Liebreich

Hello, I'm Michael Liebreich and this is Cleaning Up. One of the threads that runs through the net zero transition is the shift from thermal chemistry to electrochemistry, from molecules to electrons, from burning stuff to generating, storing and using electricity. My guest today is one of the leading figures behind that shift, both in terms of technological innovation and as an educator. Professor Donald Sadoway retired in 2022, after 44 years at MIT, most recently as John F. Elliott Professor of Materials Chemistry. He has one too many awards to mention, and his online course was called 'the best chemistry lessons anywhere' by Bill Gates. He is still actively working on radical breakthroughs in what he calls "extreme electrochemistry". Please welcome Donald Sadoway to Cleaning Up.

Before we start, if you're enjoying Cleaning Up, please make sure that you like, subscribe and leave a review, and tell all your friends about us. To make sure you never miss an episode, subscribe to us on YouTube or your favourite podcast platform. And follow us on Twitter, LinkedIn, or Instagram. Over the holidays, we move the cleaning up newsletter to Substack where you can find it on mlcleaningup.substack.com, that's mlcleaningup.substack.com. And don't forget there were over 170 hours of conversations with extraordinary climate leaders on cleaningup.live, that's cleaningup.live.

Cleaning Up is brought to you by our lead supporter, Capricorn Investment Group, as well as by the Liebreich Foundation, the Gilardini Foundation, and our newest supporter, EcoPragma Capital.

Donald, Professor Sadoway, welcome to Cleaning Up.

Donald Sadoway

Thank you. Pleasure to be here.

Let's start, if we might, with the very short version of what are you doing today? Sort of who are you, name, rank and serial number?

Okay. I'm Donald Sadoway. For 44 years I was a professor of material science at MIT specialising in applied electrochemistry. My two major strains of research: one was electro metallurgy, production of metals, aluminium, magnesium, titanium, etc. And then I started the second- went on a second spur, let's say, in the mid 90s started working in batteries. So those are the two major areas. I retired from MIT in 2022. And I started three companies here in the Boston area. I'm right now sitting at one of them, Pure Lithium. There was a second one called Avanti, which was the aluminium sulphur battery, ad the third one is my skunkworks, because I knew I will be giving up my laboratory space at MIT. And so I have my own labs in Watertown, and that's called Sadoway Labs Foundation. So that's what I'm doing these days.

Thank you very much. And I think there's going to be two themes to our conversation, two big macro themes here. One is the technology side of things, so the electrochemistry, the solid state chemistry, you've driven that in two directions which is around metals and around batteries and I have no doubt that we'll dive in there. But the second theme, I think, is around influence, change, getting things done, changing the world, improving the world, because I think you've really got, you know, sort of incredibly important insights on both of those. So, you know, the audience may not know exactly the sort of the scale of influence that you have had, you know, you can list the, I think it's six companies, those are the three you've just spoken about, there's others going back earlier, but they may not know about your teaching career and the amount of influence that you've had, and then the influence that you had through also this TED talk that you did. Talk a little bit about that; you ran this legendary course the 3.091 - I don't know how you describe 3.091 or 3091. That became just this absolute institution. Where- did you build that yourself?

No, actually, it was started in around 1968, 69 by a predecessor of mine, man by the name of John Wolfe. And there's a requirement that every MIT undergrad graduate take two physics, there's Newtonian mechanics, electricity and magnetism, they take two versions of mathematics, there's calculus - multivariable calculus, and then they have to take chemistry class. And back in the 60s, the chemistry culture was such that whoever was teaching that class, taught it in a way that most students found inaccessible. And so there was a moment there where John Wolfe stepped in and offered a class: it was basically physical chemistry, metallurgy, but he called it Introduction to Solid State Chemistry. And about 25 students took it and then petitioned to use the credit from that class in lieu of the chemistry offering, and it was approved, and next year 50 students took it and then 100 students took it. And finally the committee that's supposed to rule on these petitions said, "let's just grant this class equivalent status to that of the chemistry offerings." The chemistry department was outraged. But the committee said, "you people had the brand, you blew it, and we're gonna let these people-." So there were several people in between 1969, and I inherited it from my immediate predecessor, a man by the name of Gus Witt, he was from Austria. And he wasn't stiff and formal, he was a very warm human being. I would sit in the back of the class - I was one of his sub-teachers, we call them recitation instructors, we would have Q&A sessions, and I inherited from him. And when- just 1994, something like this, I decided that the day that I'm going to teach hydrogen bonding. I'll play a passage from Handel's Water Music. And my wife at the time said to me, "why only that day?" And then the bells rang for me and I realised, well, I can do anything I want as long as it's in good taste. I can't spew vulgarities and things like that. So this is a classroom that seats about 425 students. And every hour, there's a change of the occupants. So the lecture is really 50 minutes, 5 minutes to allow people to evacuate, and then 5 minutes to allow the next crowd to come in. So you start at 5 minutes after the houg. So I was thinking, "well, wait a minute, why don't I play music while they're filing in, and why don't I play music while they're filing out, but make that music thematically-linked to the content of the lecture?" So if I'm- if I'm teaching polymers, well, I could play something like Aretha Franklin singing Chain of Fools and and then when I- also with polymers, I took a film clip from The Graduate where the the young fellow played by Dustin Hoffman, he's called poolside and one of his parents' friends says, "I only have one word for you: plastics" and you know, bringing all of this in. So I realised that it was no longer a chemistry class but it was a chemistry-centred class and trying to weave in music, art, literature, and so on. The other thing about it was that I myself do no chemistry because I was a metallurgist, and a itinerant teacher from Canada. And I knew no chemistry, because in metallurgy curricula, you're not taught chemistry, you're taught immediately thermodynamics, because everything's prescribed. If you're going to make a steel, you're going to start with iron ore, reduce it with carbon. If you're going to make aluminium, you're going to start with bauxite converted to aluminium oxide, and then do electrolysis. Why do you have to learn chemistry? And so when I came to MIT, and I was assigned to support Professor Witt in this class, I was going to his lecture saying, "I don't know anything about electronic structure and bonding," and so on. So I was teaching myself. When I became the lecturer in charge, I was- I like to call myself an "arriviste", or an "intellectual immigrant". And I had, still in my mind, the journey I went on as I was learning, and so I was able to revise the delivery of the subject matter so as to engage with the first time learner. The textbooks are written as retrospectives for people who already know the material, and they put it in canonical formulations and so on. But that's not the way the first time learner learns: the baby in the crib listens and tries to make sense out of noise and turn it into an ordered sound string. The baby in the crib doesn't start with declensions and conjugations... this is stupid! So I had that. And then, lo and behold, the class grew because I was competing with the offering from the chemistry department. By 2002, I had 630 students, which is two thirds of the freshman class, and no class would fit. I had to give parallel lectures - they were recording the lectures and playing them on closed circuit TV. And finally, around 2004, there were a couple of fellows from Electrical Engineering who went to the president of MIT, and they said, "e want to pitch an idea to you." He says, "go ahead." They say, "we think we should take everything that's taught at MIT and put it on the internet and broadcast it for free to the entire world." And to his credit, Charles Vest, who was President at the time said, "that sounds interesting. Let's try." And so they came to me and said, "would you consent to have your chemistry lectures broadcast? And would you give us your exams and your homework assignments and your model solutions and all of that?" I said, "sure, I'm a teacher, why should I be teaching 600 when I could be teaching 6000? Let's go." And so by 2004, with the internet had streaming video capability, then this stuff was going out there, and people were watching it. And, you know, it was- my colleagues said, "you're crazy. If they can watch it online, no one's going to come to lecture." Wrong. They came to lecture because it was live theatre. They didn't know what might happen. I mean, it wasn't comedy, because downstream, they have to use this material as prerequisites for whatever subject they major in. And if they were impaired, compared to the students who took the chemistry variant of this, that would whisper through the dormitories and people would say, "don't take Sadoway's class. He's nothing but a stand up comic, you're wasting your time." But it didn't happen that way. So yeah, that was good. And then on the TED talk, I got a call from Chris Anderson saying, "we'd like you to give a lecture on this liquid metal battery you're working on." I said, "when?" This was October of 2011. He said, "well, February 2012." I said, "well, I just started the company. Why don't we wait another year or two? I'll have more data." He sayss, "no, I want to get it now, I want to get it fresh." And okay, fine. And he says to me, "so what else do you do at MIT?" And I said, "well, I teach this big chemistry class, it's got about- the classroom seats 425." He says, "oh, you must use fantastic audio-visual, and all this kind of stuff." And I said, "no, it's just me at the front of a room, a big auditorium with a piece of chalk." He says, "you use chalk at a chalkboard?!" I said, "yeah." I said, "if I can't reduce the idea to the simplicity that allows me to sketch it on the board in real time, then I haven't done my homework. I'm not prepared to teach. I could flash things faster than the persistence of vision on the screen. That doesn't mean that just because they can download the PDF doesn't mean that they got anything out of it. I have to take the time and draw it." He says, "would you use a chalkboard in your TED Talk?" Now I'm from Canada, you know, I'm polite, and I said, "yeah, okay, fine." And so that's how come there's the chalkboard in the TED Talk. And, you know, and then there's, of course, some spontaneity there, I brought my own coloured chalk from MIT because I wanted to do the thing in colour, and so on. And then all of a sudden, at one point, I just on the spur of the moment, I turn to the audience, I was halfway through the sketching on the chalkboard, and I said, "you know one of the greatest things about being a professor? Coloured chalk." And it- which just, this just came instantly to me. So yeah, the whole teaching was important.

This really resonates with me for a couple of reasons. One is my chemistry sort of background, but the other is sort of me as a communicator, or a storyteller around, you know, what I do now. My chemistry, that resonates because the last time I studied actual chemistry, I came out of a class of- a class of 20, I came 17th, right. And this was a class of classicists at a very good school that I went to, so I came 17 out of 20. Roll forwards a couple of years, I'm doing thermodynamics at Cambridge, and I'm knocking it out of the park and I was the best in the year in thermodynamics in undergrad in Cambridge. So I hear you on how chemistry, the way it's taught, can be just so kind of inaccessible and so dull and so mechan- you know mechanistic, and thermodynamics is just so riveting and so incredible and such a great space for sort of problem-solving and creativity and so on. So I'm with you on that. And then obviously, you know, what I do is: yes, I do a little bit of kind of engineering and analysis and modelling but mainly, I see myself as in the narratives business. I'm actually trying to get that "aha" moment in people's minds, when they say, "ah, we could put this and that together, and then we could build a system which will get through a storm or will-" you know, that's what I'm trying to do is I'm trying to have influence. And I think, you know, you've bridged that absolutely beautifully in your own way.

Well, you hit the magic word, storytelling. So that's the way I teach the class. You know, I start with Democritus and you know, the ancient conceptualization of what is an atom, and I even write in Greek, I write a - the alpha - and then tomi, which is to slice. So this is a- it's non sliceable, it's indivisible. And that's where we get atom from. And then fast forward to then Archimedes and on and all this crazy stuff where they have the four essences and how the science goes backwards. And then finally we jump to JJ Thompson and we spent a lot of time in England, and then we then we move over to the continent, and there's some French, some German, there's some Swedish, and we're bopping around. And I like to think of it as a 14 week soap opera, and you come back for the next instalment because you're going to find out what happens next. And it's a story written by human beings.

And how many students took that course under you over the years? There's a figure out there on the internet of 10,000. Is that possible?

Yeah, that's probably correct, yeah.

And so these are people who have gone there, MIT students, these are the best of the best, or amongst the best of the best in engineering and science and technology now, and they've got that little piece of their brain, which you're gonna live in forever.

Yeah, that's correct. I just met somebody at a- I was at a big conference in Arabia in Riyadh, it's Future Minerals Forum, and there's a young woman, she's now writing for The Economist, and she took 3.091 from me probably around 2000- I don't know 2009, something like that. And I was at dinner with her and she was regaling me with "I said this, I said this, I said, this, these these," these little contextualizations they stay with them.

The hook and the tricks- I love it. So, I think that's really important that you are this kind of educator-communicator. But let's talk about some of the breakthroughs. So, the first one: I first met you, you were- and it was at a Sustainable Energy for All event in, I believe in Brooklyn, in about 2016, something like that, 2015, 2016, and I hadn't heard of you- was it- when was it? You'll know, you'll remember this better than me.

We met earlier, and it was in Lisbon. It was a Sustainable- remember that one?

That's right, we shared a stage at the one in Brooklyn, but we met in Lisbon. But you see, you are this kind of quite intimidating figure because you have this incredible battery company, this liquid metal battery company, and you were this deep-tech electrochemist, what were you doing at a Sustainable Energy For All conference talking about development and bringing, you know, sort of solar power to rooftops across Africa? That was- I couldn't kind of- it didn't compute, I'd never met a technology professor, deep-science professor at any of those events before.

Well, I went because I was invited. Somebody in the planning architecture of the conference reasoned that there might be some synergy having me, and so I accepted the invitation. And then I had the good fortune to meet you.

And clearly, you know, batteries, obviously, at that point solar was taking off, even you know, in the developing world in the Global South. And it was very clear that in order to make it more useful, in order to give not just kind of the reading lantern for the kid doing the homework or whatever, but actually to turn this into an energy source that could empower people to become economically active, batteries were going to be necessary. So I kind of see the logic of the invitation. But still, you know, kudos to you for turning up, for showing up and engaging in those conversations. But talk to me about that company, liquid metal battery, which became Ambri. What does it do?

So it's- first of all, I have to preface, it's still on the path to full commercialization, and somebody may say, "what's taking them so long? It's been 13 years." And newsflash: this is tough tech. This is not like writing code. And the- I just heard Dave Danielson last year, gave a talk. Dave Danielson was employee number one at Breakthrough Energy Ventures, and he said that their reasoning is that in this area of tough tech, heavy industry, innovation, time to commercial market is probably 20 years. It's not 20 months: it's 20 years. And so Ambri is on the path of upscaling. They've made- Microsoft has adopted the technology, but it's still going to take some time before Ambri builds out the manufacturing facilities, It's sort of a chicken and egg; you don't build out a manufacturing facility for a massive market when you don't have the customers lined up and the customers don't want to line up until they've seen the thing operate at scale. And that's why you end up in this, this state of paralysis. And you can say the same thing about Boston Metal for the green steel and on and on and on. People want to see it work at scale. And I say, "well, how are we going to get to scale if we don't get funded?" "Well, we'll fund you when we see it work at scale,." I say, "well, I guess this conversation is going nowhere."

So I just had- I just published something, I've created a Substack, which is basically just a publishing area for my my bits and pieces that I write. And I just wrote something on nuclear fusion, where I looked at: how quickly could it become- could it serve 1% of global electricity? What is the kind of- if you redline every single stage from breakthrough to creating a pilot to the first-of-a-kind plants to scaling it up, creating the supply chain, creating the regulatory framework, creating the final- I think it's 2050 for 1%. And people were outraged. I've had people say, "don't be silly, we really need this... and the climate... and it's gonna go much faster than you think... and I'm sitting here in California." And I'm like, "well, guess what, I'm an engineer. And I think that, you know, when it when it really works, 20 years still. I mean, when it works in the lab, 20 years, before it becomes meaningful, is absolutely the right metric." So that's why I'm looking at 2050 for them, and they're not happy.

You can change the laws of man, you can't change the laws of physics. It takes a long time to tame that beast.

Yeah, and I think it's the laws of physics plus the laws of certifying a supply chain, and the laws of finance, and the laws of quality control, and the laws of regulators getting out of bed in the morning and doing nothing for, you know, for months on end, and all those sorts of things. It's all of the above added together, but it's very hard to change. I don't see it changing. I think that you're- I think from what you're saying, you don't either?

It's a long journey, a long journey from lab bench to market.

So Ambri, you spun out, it was- and I remember though, it has a few speed bumps as well, though. I mean, to be fair, it has not- it could have gone a little faster, right? Or am I wrong?

No, it could have gone faster if we already had the scar tissue from the experience in advance, then we would have jumped over that- we would have missed that speed bump. Yes.

What was the speed bump specifically? Actually take a step back. What is it that it does that's so clever? Because you have this thesis that if you want to make things that are dirt cheap, you make them out of dirt, right? So that's something that you've said many times. So what does what does Ambri- what is Ambry? Why is it so clever? And then we can come back to the speed bump, which is a sort of hopefully going to be just a short sub-chapter on its on its pathway.

Yeah. So Ambri is about the liquid metal battery. So we have- the electrodes and the electrolyte are all liquid. There are solid boundaries, there are current collectors and so on, but the the energy is stored in liquid metals. So we have a low density liquid metal on the top, it's the anode, and a high density liquid metal on the bottom, that's the cathode. And we have a molten salt in between, because we're operating at around 500 degrees Celsius. And so when the cell discharges, the metal on the top dissolves as ions and goes to the bottom and alloys with the metal on the bottom. And then we charge the battery, we assess force current through it, and essentially electro-refine the anodic metal out of the cathode on the bottom and send it back to the top. But in doing so, we're actually electro-refining the battery, we're purifying the metal, we're purifying the salt and restoring the battery to its pristine initial condition. We have data on this battery running years and years and years: full depth of discharge with zero capacity fades, still retaining 99+ percent of its net nameplate capacity. So it's really attractive. But it's, you know- the metal on top is from the northwest part of the periodic table, alkali metals, alkaline earth metals, and those things are pretty vicious. And so what it means is that you're going to have some challenges with the feed throughs, the seals, the containment, and so on, and everything has to work. It can't be 99% there: it has to be 100% there. That 1% can make the battery fail prematurely. And that's what's taken us so long.

So these three layers, they just float one upon the other, there's nothing that keeps them apart. They're just simply floating there by that densities because of their-

Density difference, exactly. Think of it as salad oil and vinegar, only instead, we have we have three layers. And you're right, Michael, there's no membranes, no separators, they just know where they're supposed to go, and the metal is immiscible in the salt and the salt is immiscible in the metal. It's just preordained. It's correct.

And all three are cheap materials. There's no neodymium or vanadium-

No, no.

-there's nothing funky in there that's going to drive up the fundamental- the costs before you've even started?

That's correct. Because I had this axiom that I use withmy group, I said that we have to think about cost on day one. So I coined the idea of "cost-informed discovery". People will say cost is important when you're commercialising, and I said, "in this sector that's too late. You have to think about cost on day 1, not on day 10,001." So I take the periodic table, and I overlay it with the economics of the elements there. So there are vast swaths of the periodic table I won't let my students touch because it just won't scale. For example, if you go back about 10 years ago, there's a lot of interest in Telluride materials for solar cells, right-

Right, CdTe solar cells, right?

CdTe, exactly. But Tellurium is as Earth-abundant as gold. So I don't care if you even get it for free; it just doesn't exist in quantities enough - to your earlier point about how much time it would take to make fusion deliver what percentage of the global need for electric power, I mean, you're just not going to scale it. I could make you a gorgeous battery out of lithium and Tellurium and stuff like that: don't even bother. And so we had built in, Michael, the notion that we would choose Earth-abundant- and, you know, I said this in my TED Talk, which was February 28 - noteworthy day, February 29, 2012 - and I said, "it's got to be dirt cheap, make it- if you want to dirt cheap, make it out of dirt," and then I added, "dirt that's locally sourced." Now, nobody was paying attention to supply chains in those days, but I was right, that we're not going to make it out of stuff that comes from halfway around the globe. And I said, "batteries in Africa should be built by Africans in Africa using African resources." That's a secure supply chain, plus they become authors of their own future, it just makes so much sense.

So let's just get back to that speed bump. Was that around containment, leakage, corrosion? And is it now solved? You're excited about Ambri again?

Yes, you've identified, Michael, it was the- it's called a seal. You have to have a feed-through, you have to get current in and current out. You think of this as a box, the box contains the liquid metals in the molten salt, but you gotta get- gotta get electricity out of that box, and that means you have to have a hermetic seal, and as it has to be such that you can go from room temperature up to 500, 600 degrees Celsius and undergo that expansion, and yet have something, as I said on the top, you've got something that's- oh, we could have something like calcium or magnesium and these metals will attack almost anything. The feed through has to be electrically insulated as well, it can't conduct current, so you can't weld it, you have to have a feed through with a washer or some such thing. And that obviously has to be a ceramic because it has to be something that is not an electronic conductor. It's too hot to be using polymers. You can have compression fittings, O-rings and things like this. So we had seals that would work and then after several days, they would crack- the gas would get out, gas would get in and we had to redesign, and on and on and on. And there was nobody out there to help us; you can't sit at your computer and say, "liquid metal battery seals.com," there is no such website. So yeah, it took us a long time and it nearly crashed the company - this is around 2015, had to layoff a large number of people, retrench and so on. So- and this is something that we didn't encounter when we were at the lab. At lab scale, we could use feedthroughs with compression fittings and O-rings, and so on. But once we put the battery together commercially, we're going to put hundreds of these cells inside something the size of a shipping container, and everything has to be able to endure temperatures of 600 degrees Celsius. And that never been done. Before we went even to Corning- I went to Corning, and I talked to people there and they said, "what temperature are you running at?" And I said, "well, 500 degrees." "Oh 500, that's nothing, that's nothing. I'm sure we can give you something, probably even a glass- it'll be be easy." I said, "we have lithium metal in the headspace. There's lithium vapour in the headspace." He said, "lithium vapour in the headspace?" I said, "yeah, lithium vapour in the headspace." "Oh, we can't help you, we can't help you. Lithium vapour will chew through any of our ceramics." "Okay, so thank you, thank you for the help." And they're good people. But this was like- it's unprecedented.

So what I think is fascinating about this is: these are the real challenges that they kind of don't appear on the spreadsheets of the- you know, I look at people and you know, I'm going to be the first person to use the word "hydrogen" on this episode, but I look at the people who are pushing and promoting hydrogen: they're absolutely convinced that you can do this and you can do that and you can reduce steel- and you know, whenever any of those people come to me and they want to have a chat, the first thing I do is I have a look at their educational background. And it is almost always econ, poli-sci, you know, they just don't come out of science and economics- science and engineering and physics and chemistry backgrounds. Because if they do, they're much more circumspect about what they think is going to be achieved, when.

Right. Well, Huxley's said "there's nothing like slaying a beautiful theory with an ugly fact."

So- now, Ambri, I'm assuming, from the way you talk about the speed bump as being in the past tense, that we should- we should now be expecting some great things from Ambri. But there's another company which arose out of your research, out of that, I think you call it the first strain of research, the second one being batteries, the first one being metallurgy, and that is Boston Metal. And that's direct electrical reduction of- turning rust into steel, right? Rust into iron, or iron ore- talk us through that one.

So, when I first came to MIT, I got involved in searching for an inert anode, carbon-free anode for aluminium. That was a topic that was of great interest in the 80s, not because people were worried about CO2 emissions - nobody cared about CO2 back then - but they were trying to reduce the energy consumption of aluminium. It's, you know, in the vicinity- about 12, 13 kilowatt hours per kilogram to make aluminium, very energy consumptive. And during that time, I had to free my brain and figure out how to really invent and have audacious ideas, and it would mean not just changing the carbon anode for another anode, but it means changing even the electrolyte chemistry which was sodium-aluminium fluoride, it was cryolite. Then I started to think, "well, why only aluminium? It's an important metal, but the most intensively used metal on the planet is iron. Steel is 100 times larger than aluminium by tonnage. So why don't we go for that one?" And that's when I started thinking about electrolysis of iron oxide. And I said, "well, like dissolves like," because I was already working in chemistry, and, "why are we putting an oxide into a fluoride solvent? Why don't we put an oxide into an oxide solvent?" And now I said, "well, okay, well, we'll use silicate slags and we'll turn them into electrolytes." And I did it, and we started- we were using carbon electrodes and passing current and reducing iron oxide. And then nobody wanted to fund it. The metals industry didn't have any interest in this. The legacy steel makers wouldn't touch this with a bargepole. But around 2000, NASA was interested in generating oxygen on the moon. And I said, "well look at the composition of the surface of the moon. It's called regolith." I said, "well, regolith looks a lot like the slag that I'm using for steelmaking." I was worried about what's going on at the at the cathode, the liquid iron: they're interested in what's going on at the anode? Well, I said, "well, we could do this." And so we used precious metals, platinum group metals because it's NASA. If you're going a quarter of a million miles from home, it costs you $100,000, $200,000 per kilogramme shipping charge. So it doesn't matter if you're shipping platinum or you're shipping carbon

Donald, let's just back up and make sure people have understood because I'm sort of keeping up. just about. What you're talking there is, you're talking about a rock that is rich in iron oxides, and you've been working on how to get the iron out, but of course, what's happening at the other end is oxygen is being given off, and that NASA is interested in oxygen, because it's kind of useful for astronauts and for other things. For plants, you want to do experiments in space, all that sort of stuff. So what- you're talking about then taking that process, and then using it in a sense, focusing in on the byproduct, which is the oxygen, but of course, you know, it's still producing the iron.

That's correct. And NASA at that time, wasn't interested in what's happening at the cathode, they didn't care about the iron. And it also makes silicon, because regolith is iron oxide and silicon oxide, which gives me the composition of the electrolyte. But we wanted to make iron. And we were using platinum or iridium, various platinum group metals as the anode. And we were able to demonstrate in my lab that this thing works. And it's compact, because the molten salts are very-

Has Elon Musk been in touch about all of this?

No.

Because I mean, he wants to go off to Mars, he's going to need oxygen, he's going to need to do this right? He's going to need some Sadoway special chemistry to do that!

Yes. So the interesting thing about Mars and the Moon: they're both virtually the same in terms of the regolith. The difference is that the moon is a highly reducing atmosphere, it's like 10 to the minus 10 atmospheres. So all the iron there is two plus. And so iron oxide, iron two plus oxide is white, and the Moon is white. But Mars has a CO2 atmosphere. If we keep going the way we're going, the Earth will look like Mars, but let's not be so depressed.

A nice reddish colour.

Yeah. And why does it have a reddish colour? Because on Mars, all the iron is three plus. And so Fe2O3 is red, that's hematite, same cognate is blood, right? So haemoglobin, hematite, same thing. So- well, I would tell my students in 3.091, "with the naked eye, you can tell the difference between one electron, Fe2+ versus Fe3+, at a distance of two thirds of a billion kilometres. Mars is red, the Moon is white. It's just one electron difference.

And I can see why your students are loved you. And so you- fast forwards you've created a company, Boston Metal. And another- we just brought in Elon Musk, but that's by-the-by because a different billionaire has become your, sort of, your major backer on that one: that's Bill Gates.

That's correct. So Bill came in at the very beginning with Ambri, and I met Bill again- it's a 3.091 story. Somehow- I was waiting at a bus stop one day and I get a text message from some person who says that he's at the annual summit of Microsoft, and they have this big powwow. And Bill Gates, of course, comes to the microphone and addresses the entire ensemble. And at some point in his remarks, he mentions my chemistry lectures. And he says- he urges everybody to watch them. He says, "I've watched them, all 35 of them, and you should do likewise." So this man writes me and says, "you should be pretty proud of this." I looked at it, I shrugged my shoulders, I said, "I guess that's nice but so what?" So now it comes to August 2009, I'm already trying to get funding for the liquid metal battery research. And I got an email from a woman who claims that she's Bill Gates executive secretary and he just stepped down as CEO of Microsoft. And she writes, "he's coming to Boston at the end of September, he'd really like to meet with you for 90 minutes. Would be able to do so." Well, I looked at the thing and I thought the students had hacked into my account. I ignored the email, and about 10 days later, she writes again, "maybe you didn't see this email, but Mr. Gates would really like to see you." And I showed it to some colleagues, and they said, "I think it's- I think this is real, you should probably reply." So I did and said, "may I ask him what it's all about?" And then he writes me, and it's very detailed, it's four talking points, and so on, and so on. And so he came, and we sat, it was on a Friday, I'd lectured 3.091 and then came to my office. We sat in my office for 90 minutes, we talked about computers and education, distanced-learning, all these different things. And towards the end of that 90 minutes, he said, "so what are you working on these days in the lab?" And I sketched on the whiteboard this very vague conceptualization of the liquid metal battery, he looked at it, he understood it immediately. And he said, "if you ever decide to spin that out, let me know, I'd be willing to put some money into it." And so I met Bill Gates, not because I wrote a piece for the Financial Times or for The Wall Street Journal, I met Bill Gates, because I was dutifully teaching a first year chemistry class. That's a nice story.

That's great. So I'm still waiting for my summons by Bill Gates, although I have been summonsed by his team, because at one of the Bloomberg New Energy Finance summits back in, oh I'm going to say 2016 or something like that, I was- you know, I do this kind of state of the state of the industry keynote, and I was perhaps not fully respectful of Mr. Gates, because he was- he was very disparaging about the chances of wind and solar and renewables of achieving anything much in energy terms, and I reminded people that this was the guy who missed, nearly missed, almost missed the internet, because all of this kind of random chaotic stuff in people's homes, you know, didn't sort of seem serious compared to the Microsoft Network, the big centralised solution. So he was off investing in nuclear and I poked a little bit of fun at him, and his people summonsed me to explain what a very smart man he was, and that I should be more respectfulm, to which I kind of shrugged my shoulders. I'm still waiting for my call, I'd love to have that sort of interaction and influence that you've got. So he funded Ambri, and then Boston Metal as well, through what is now a more formalised venture structure, the breakthrough ventures- [Breakthrough Energy Ventures] funded a lot of very interesting things, and some of them looking very promising, some of them maybe less so. But Boston Metal, and that's now run by Tadeu Carneiro down in Brazil, who's a real kind of steel-industry maven, decades and decades, how is Boston Metal doing? I know it just raised another big slug of money?

so again, this is a little long journey. And so the key missing piece- we didn't start Boston Metal, "we", that was I, along with a couple of my previous students, postdocs, we started Boston Metal only when we made a major discovery in 2011, which was an inert anode, a practical intert anode. There's no way we would make steel on planet Earth if we were making anodes out of iridium or platinum. That would just be too costly. So we made a discovery, it was an iron-chromium alloy that forms a protective surface oxide film, and that's practical. The cost of that is in harmony with people's expectations for steel. And so with the advent of the inert anode, to be used in a molten oxide, you know, this is running at 1650 degrees Celsius because we want to make liquid metal, we want to be continuously producing liquid metal tapping out of the reactor, and making oxygen is the byproduct, which by the way, it's industrial oxygen, so you're making two products. And so we started the company and that was formally started around 2013. So it's over 10 years old. And it's still just on the trajectory of building cells of larger and larger size. The cells that we made at MIT were externally heated, we got to 1650 degrees with electric resistance elements, things like molybdenum disulfide, or even graphite blocks, and so on and so forth. But this industrial cell has to be self heating in the same way the aluminium cell is self heating. You don't heat an aluminium cell, you cool an aluminium. So the action of electric current generates dual heats in sufficient intensity to keep the cell at the operating temperature, and we have to do the same thing here. Well, no one's ever built such a thing and we can't even stand on the shoulders of the aluminium industry because aluminium was made in a fluoride melt at 960 degrees; we're in an oxide melt at 1660 degrees. This is white heat, blinding white heat. And we have to do this. And so we're now at a point where we have some self-heating cells, we're up at around 20,000 amperes. But this thing has to run for years without interruption and we're not there yet. So yeah, and then the Brazil piece is Tadeu decided he wanted to make sure that the investors didn't grow weary and say, "we're going to hit the pause button at this moment." And so he started a parallel effort in Brazil to use some of the Brazilian ores. Some of them are rich in tantalum, niobium, vanadium, to make these high value-added ferro-alloys. And at the moment, they're using a carbon anode, so it's almost like a aluminium cell, but it's, sort of, think of it as training wheels. So that's good, but if what the world is counting on is green steel, there's still some work to be done.

Okay, so they are able to get value out of the intellectual property, the approach, but it's not the real thing. The real thing is, is this steel, the bulk steel, if it works, or, you know, if I'm going to be generous- gracious to my guest, when it works, how does it compare to the hydrogen direct reduction process? I mean, is it kind of a neck-and-neck race? Or if it works, is it game over for hydrogen? Or where are we on that?

If it works, it will be game over for hydrogen, because, going back to the cost informed discovery, I totally reject the idea of subsidies and all kinds of credits and so on and so forth. So, molten oxide electrolysis to make steel from ore is designed to make better metal than legacy metal, at a lower price point. And so, the thing that people don't appreciate it at Boston Metal is we've demonstrated that we can start, unlike with hydrogen - hydrogen has to start with high purity iron oxide, because iron ore has silicon oxide and aluminium oxide and various other metal oxides that would be viewed as impurities. And so they they can't purify. If you want to have high purity metal, you need high purity feedstock. And surprise, in the Boston Metal process, the molten silicate electrolyte is serving as a slag as well. And so we can charge the cell with iron ore, unprocessed iron ore, and all the silicon and aluminium stays behind. And we produce high purity iron with oxygen as the byproduct. And continuously, you know, because in hydrogen, they make solid pellets, they start with pellets of iron oxide that have been processed, that's an added unit of operation, and then these things are falling down a counter current reactor with hydrogen blowing in the opposite direction, and then they end up with solid pellets of iron. Well, the downstream operations, whether they're casting, or even continuous casting, you make sheet plate and so on, requires liquid iron. So they're going to take the solid iron pellets and then put them into an electric arc furnace and melt them. And we don't do that. Ore goes in the front and liquid iron comes out the back, and that goes right into the cash shop or whatever. So the throughput is good, the capital cost, the operating costs are all competitive, so that we're not telling steel makers to go and take a wrecking ball to the blast furnace, but when they have to retire the blast furnace because it's lived its full service life, instead of replacing it with another blast furnace, they'll consider putting in molten oxide electrolysis. Now the big caveat here, Michael, and it applies to the hydrogen as well, the assumption is that the hydrogen is green hydrogen, not brown hydrogen. So green hydrogen means you need to have electrolysis, electrolytic decomposition of water. Okay, we know how to do that. We've known how to do that for over 100 years. That means you need to have green electrons too. Well, if you have access to green electrons, why would you go and make hydrogen, which is an intermediate and then have to have an electric arc furnace and on and on and on. Why don't just take those green electrons and put them through my cell and go from ore to metal in one step? I mean, if you put the flowchart, the the block diagram on the wall for monoxide electrolysis and for hydrogen reduction, the hydrogen reduction has all these extra steps. An eight year old child could look at those two and say, "I know which one is easier to work with, it's the one on the left that just go in, one unit blackbox, out comes liquid metal."

Yeah, I mean, there's a lot of- I think it's fascinating and I'm not going to disagree in any way. I think that the challenges, though, would be that your electricity for the- for your process has to be 24/7, I don't know 365? I mean, once you start this, you don't want to switch it off, which renewables are not very good at. Whereas hydrogen, in theory, these electrolyzers could load-follow more. And I think the other challenge is: when you talk about making pellets and then having to separately re-melt them to do anything, politically, not economically, not from the chemistry or thermodynamics perspective, politically, that might be more attractive, because then you can do- you can make the steel over in Mauritania or Australia or India or wherever, but you can retain a steel industry, taking those iron and steel pellets, either, I don't know, sponge iron or the pellets or whatever, and then you can upgrade them so you can retain some jobs in places like Japan, South Korea, Germany, which is incredibly politically- because otherwise, the whole lot just goes to wherever the- wherever the renewables and the ore is, right?

That's correct. So, we have to be smart about this. You know, as we want to electrify everything, and we electrify with green electricity, then we'll be able to cite smelters where we have green electricity. And to your other point, you're absolutely correct. The aluminium smelter works 24/7, 365, because that was the expectation. But if I came along and said to Charles Martin Hall and Paul Heroult, "I'll give you electricity free for 14 hours a day, and I'll give it to you for free." What's the response to that? Well, I'm going to design a cell that can idle for six or seven hours without totally freezing up, and then it can load-follow.

Right, and this is something that I talked about with a previous guest, there's Alex Grant of Magrathea Metals, who's doing magnesium from salts. And he's, from the start, his team has tried to make sure that it can load-follow, that it does maybe not idle fully, but at least use lots of electricity when it's very cheap, and then use less when it's not available and it's more expensive.

Well, this was done in the 1930s by Henry Dow, because he built a magnesium smelter in Michigan, and this was the early days of rural electrification in the United States, and they had interruptible power delivery. So he designed the Dow Cell. And remember, at one point, Dow Magnesium was the largest producer of magnesium on the planet. They had 98,000 tonnes per year in Freeport, Texas. And, but originally, they started in Michigan. And when the power went off, the cells stopped working. And he actually had some gas burners that would just very gently prevent the cell from totally freezing up. It could start to freeze from the outside-in, but as long as it wasn't totally frozen, when the power came back on, they would start the electricity flowing, and then the dual heat would melt back and so on. So you just design for whatever the available power is.

Which just emphasises how much more we still have to learn, how much creativity, how much design we still have to do to optimise around the modern sources of energy and power, versus the ones we're used to using. I want to just cover- a couple of other things I want to do. One is: you've got some other- you've got Pure Lithium, you've got Avanti I want you to say a few words then about Sadoway Labs, and I'd love to do a kind of rapid fire, just to finish off, with your views on a few things that are out there like- but let's get to that, let's cover Pure Lithium. So what is it? Just give us the short version.

It extracts lithium metal from brine, and in one step removes, lithium ions from the brine which has got the magnesium, calcium, sodium, potassium, and in one step, deposits it on copper and manufactures the lithium metal anode. And as you know, there's a lot of talk these days about a lithium metal battery overtaking lithium ion, but when you look at the cost, the costs of lithium metal are just way out of line with expectations of the market, so you don't see anything in the marketplace. Pure Lithium could break through that impasse. Avanti is commercialising aluminium sulphur, which was a technology that I invented at MIT around 2015 or so, 2016. And then Sadoway Labs: this is my skunkworks. So- and there, we have funding from Eric Schmidt who was the CEO of Google for a number of years, he and his wife set up a foundation. And they are funding me to do whatever I want under the umbrella of climate change mitigation technology. So I'm working on still even more radical batteries. I have one that I'm working on right now that will operate, charge, discharge at temperatures as low as minus 80, minus eight zero degrees Celsius, and will not burn or catch fire at temperatures as high as plus 80 degrees Celsius. I call it the "whole earth battery". And the last audacious project we're working on is the destruction of CO2, because contrary to predictions, we're not net-zero by 2050. We're going to be burning hydrocarbons well into beyond mid century. And so we just destroy CO2 before it leaves the chimney. And I call it "the Zapper", the CO2 zapper, or maybe if people want to a more literary name, I call it "immaculate combustion". And this would also decarbonize cement, because-

Well let me ask on that one, this is the- you call this "the zapper", right? And when you say destroy CO2, I mean, this is not an- this is not- "you're not decomposing it atomically, so, I mean, this is going to produce carbon and oxygen and that's still going to be there. What does it turn the CO2 into?

It could be carbon and oxygen, or-

So biochar plus oxygen?

Yeah, it could be, could be, we're still- we're still researching. It's not something that I sat down and I calculated; we're doing this sort of by inspection. So we have an idea. It's all- everything comes under the the domain of extreme electrochemistry. If it's a Sadoway project, it's going to be electrochemistry. Not water; water is great for bathing, ,it's great for cooking, I love to swim, but I don't- I don't believe industrial electrochemistry happens in water. So it's molten salts and a whole class of molecular liquids.

Now, let's get to the rapid fire. That's a great segue, because one of the companies- I had Mateo Jaramillo from Form Energy who came on the show and talked about his battery, which is a- which is essentially, it's an iron-air battery. It extracts the electrical charge from the rusting processes, as I've kind of understood it, and then reversibly does that. But that would be- that would not- that's not extreme electrochemistry, to your definition, is it?

Correct, it's not.

And is it going to work? Is he going to be able to get that working at a- I mean, from the perspective of what is it made of, it is made from something pretty cheap, right, which is iron oxide. Doesn't it- why doesn't it tick the Sadoway boxes?

Well, it has to have the performance because in the end, these things: it's price-performance ratio. And if you look at Form's pronouncements, they boast that they can run for over 100 hours, maybe 150 hours. And that's true, but the power intensity is low. And then, as you know, the batteries are used to help balance the grids. There are times when you have a super-abundance of generation from sun or wind. And if supply exceeds demand, the quality of electricity is bad. Can you imagine, if every time you went to plug in an appliance, the question you've got to ask yourself is, "do you feel lucky?" because the voltage is wrong, the frequency is wrong you'll, blow up the motor, you'll blow up the device. So- but a metal-air battery can't participate in balancing with severe current surges, but-

Is it that it's too slow to charge? So you get this sort of moment of free electricity and lots of it, and you can only kind of sip it through a straw, because it takes you 100 hours to do anything?

Yeah, remember the battery has both electrodes, and the chain is as strong as its - in this case - weaker link. And the weaker link is the air electrode, which is really not an air electrode, it's an oxygen electrode. Nitrogen- you know, air is 80% nitrogen, 20% oxygen. And so that air-electrode: there have been so many attempts to make metal-air batteries. There was aluminium-air, there was even lithium-air, zinc-air, and they've all never made it to market. And now Form is trying to make iron-air work. And what do all of those chemistries have in common? The air electrode. So I'm not going to disparage anybody. If they can make this thing work and hit the price point expectations of the market, we need it. I don't view people like Form as competitors, I view them as allies: we're all trying to do the same thing and stabilise interruptible power generation. But for me, I don't have any great ideas on how to make that thing work, and so I'd rather than work on the others.

Okay, and then what about- thank you. What about, if we look at sort of vehicles, cars, because your batteries, I mean, when they work - when they work, there we go, I'm gonna be gracious again - when they work, they are great big things made out of dirt, stabilising the grid. Have you got any thoughts about the kind of, you know, we hear a lot about solid state batteries, we hear a lot about all sorts of funky chemistries that are taking over and they're going to allow cars to drive 1000 miles. Is that just straight out of the marketing department, or is that out of some of those, you know, 10,000 students who have actually listened to your lectures and been taught by you, and it's real chemistry?

The car that's going to go 1,000 miles, I think we're going to be waiting a long time for that one. Because all of this is under price constraints. If you wanted 1,000 mile battery for a car, I could give it to you, but it would come in at a NASA price point. And that's what makes this problem so, so difficult. It has to be done well. And the auto makers are going to be slow to adopt. They don't like risk. They're dragged kicking and screaming into the electric vehicle age because they're really good at manufacturing cars with internal combustion engines. Why would you throw all that away for some uncertain outcome. So there's room for innovation, maybe batteries that can't catch fire, that would be a good one, safety is something I don't think anybody would argue against, but without compromising the economics. That's why we don't see pervasive adoption of electric vehicles. You're being asked to pay more for less, and nobody does that. You can put subsidies in there; even with the subsidies, you're still being asked to pay more for less. We don't have the charging infrastructure. Besides here in Massachusetts, all the electricity is generated by burning natural gas. If I've drove an electric car, all I'm doing is taking tailpipe emissions and sending them to the power plant. And there's some calculations that indicate you might even be raising the CO2 emissions per mile of driving. So it's a very, very difficult problem, and there are no simple solutions. But you can't win if you don't play the game.

Right. And I mean, it's an interesting one, because you seem like an optimistic person. And certainly when you, when you and I have met and you know, we haven't, you certainly haven't sort of sat on your hands and said, "well, you know, this thing is just too big of a problem." You've done the opposite, you've really gotten stuck in. But even if you then your- your batteries help to decarbonise the grid, what you've just said is it probably won't help decarbonise transport. Now, I personally think that we will get that with transport but- I mean, I suppose that's almost a good place to end up: are you optimistic, pessimistic or something in between about decarbonising our economies?

I'm optimistic about decarbonising. I know that we have the mental capacity to invent. But I am not going to put on a fast track, you know, really good technology that's going to not bankrupt us. I want sustainable development, but I want profitable sustainability. I don't want a fork in the road that says," carbon future with the standard of living, we're used to, decarbonised future accompanied by poverty." Nobody wants that. So we have to be smart. It's a really good challenge, and I'm up for it.

Right. And I think that is the absolute perfect place to draw this to close. Thank you so much, Professor Sadoway, Donald, for joining us on Cleaning Up here today, giving us an enormous amount to think about.

My pleasure. It's always a pleasure to converse with you, Michael. And I hope that we get a chance to do so again in the not too distant future.

Absolutely. And hopefully, in person. Thank you.

Yes. Okay. Be well.

So that was Professor Donald Sadoway, Emeritus Professor of Materials Chemistry at MIT, founder of multiple startups, including battery company Ambri, Boston Metal, and his most recent: Sadoway Labs. As always, we'll put links into the show notes to episodes of Cleaning Up mentioned during our conversation. So that would be Alex Grant, Episode 142:f From Mining to Brining, and Mateo Jaramillo, Episode 144, Iron-Air Man. And of course, we'll also include links to Professor Sadoway's TED Talk, his online course: Introduction to Solid State Chemistry, and to his various companies.

If you've enjoyed today's conversation, please remember to like, share, and subscribe to Cleaning Up or leave us a review on your favourite podcast platform. And do please spread the word tell your friends and colleagues. And if you want more from cleaning up sign up for our free newsletter on the publishing platforms substack at mlcleaningup.substack.com, hat's mlcleaningup.substack.com, or visit us on cleaningup.live, that's cleaningup.live.

Extreme Electrochemistry for a Sustainable Future - Ep155: Prof Donald Sadoway

Listen On

Featured Episodes

Recent Episodes