Donald Sadoway, professor, Department of Materials Science and Engineering
Bruce Gellerman: I’m Bruce Gellerman from WBUR, guest hosting this episode of the MIT Energy Initiative podcast. Today we’ll be pursuing the renewable and clean energy holy grail: storage. The ability to store solar, wind, and hydro energy and release it when the sun isn’t shining, the air is calm, and the water is still, promises to transform our electric power future. In this episode, we’ll explore the current state of storage technology, where it’s going, and how we’ll get there with two guests from MIT, Donald Sadoway and Yang Shao-Horn, professors in the Department of Materials Science and Engineering. Yang is also co-director of the MIT Energy Initiative Energy Storage Low-Carbon Energy Center. That’s quite a title. She’s also the Keck Professor of Energy at MIT. Don, Yang, welcome. Thanks for showing up, I really appreciate it.
Donald Sadoway: Pleasure to be here.
Yang Shao-Horn: Thank you for having me.
BG: I don’t know what your house is like. My house, I walk into my bedroom, it’s a tangle of plugs and recharging my devices. I’ve got a zillion devices. My kitchen is filled with them. You go to a parking garage, there’s electric plugs now you can plug in. It’s hard to avoid. Where are we? Where are we going with battery technology? Am I ever going to get rid of all those plugs and devices? Or is it going to be more?
DS: I think that you’re always going to have to be able to plug in to the wall. That’s not going away. But with the appropriate storage, you’ll be able to disconnect and be mobile. That’s the way things are going. The other thing I would say in response to your question is that I think you’re going to see more intensive electrification moving away from the use of fossil fuels wherever possible. There can be more power outlets and more cords.
YH: To add on that, as you know, with dropping in solar and wind electricity costs, we’re looking at two or three cents per kilowatt hour. We’re going to see more and more usage of electricity, solar, and wind in our daily life. Not only from electric vehicle transportation, but also into our homes. We’re not only plugging our devices and appliances into the wall; we’re potentially plugging appliances into our storage devices. We can envision our cars as a storage device. We can actually tap into more electricity from solar and wind instead of electricity from fossil fuels.
BG: The solar and wind, the cost for generating has come way down, by a fraction of, what, 10% of what it was?
YH: Ten times.
BG: In a decade. But how about the price of batteries, has that come down considerably in the last decade?
YH: Absolutely. If you look at the cost of lithium-ion batteries for transportation, it has come down ten times in the past decade as well. This is why it really has positioned lithium-ion batteries at the center for powering various vehicles, hybridize, so with a different degree of electric power.
BG: How does that compare to the energy generation as opposed to the storage? What’s the price per kilowatt hour?
YH: It’s about $100 per kilowatt hour. Many orders of magnitude more expensive than electricity from solar or wind.
BG: The storage is?
YH: The storage is.
BG: We’re still not any place near the same cost, right? I mean, storage is expensive.
DS: The $100 per kilowatt hour, that’s the capital cost of the battery. Now you have to imagine, how long is that battery going to be in use? If the battery poops out after four or five years, that’s not going to be practical. But if we give you a battery that will last for 25 years, and it costs $100 or $200 a kilowatt hour, and it has, let’s say, a round trip efficiency of 80%, you put in 100 units of electricity, you get back 80 units of electricity. When you do the calculation on that, cycling every day, it will be economically feasible. We don’t want a battery that can store electricity from the sun that then you can draw upon after sunset, with the result that in the day, as Yang said, it’s going to be coming in at, say, three cents a kilowatt hour, but after dark it’s 25 cents a kilowatt hour. That’s not going to make people very happy. That’s what the research effort here at MIT is directed at. It’s to give us batteries for stationary storage. It’s a totally different set of requirements.
BG: Meaning not for cars, not for mobile phones, not for those mobile devices.
YH: Right. Essentially, if you look at the cost requirements for portable electronic devices, the batteries can bear the highest cost. As you go to electric vehicles, we’re looking at $100, $150, $200 per kilowatt hour. That’s still expensive. That’s why electric vehicles are maybe $50,000 instead of $20,000 or $30,000. There’s tremendous need to further bring down the cost of batteries to make electric vehicles more viable for consumers. If you think about using batteries for stationary application in our homes, the costs need to come down even further. Then we’re really looking at not only cost but also the sustainability issue of what is inside a battery, in order to power the planet. I think for stationary applications, the cost and the sustainability or the earth abundance of the materials are really very, very critical.
BG: Right now, you’re talking about grid storage, utility size storage, is that what you’re referring to when you say stationary? Or are you talking about residential house?
DS: All of the above. You could imagine an individual homeowner who has, say, solar panels on the roof. That homeowner is going to need to be able to power after dark so it would make sense to have not a community storage facility but to have a storage facility within a single-family home. But then also there are other instances where you might put the storage at the level of the substation and service 200 homes in a subdivision. The notion of storing the entire grid, I think that’s a little bit far-fetched.
BG: When you talk this type of battery, we’re talking lithium-ion right now. Everybody’s familiar with that. They’re in my phone, they’re in my headphones, they’re in my laptop. Which ran out, by the way, when I was writing the script for this yesterday night, right in the middle. lithium-ion, is that the state of affairs right now? Is that the cream of the crop?
YH: Absolutely. I think technology is the elephant in the room. It’s the technology at scale. If you want to have, for example, various applications for stationary applications, look at the business opportunities, either frequency-shifting or shift the peak, for example, you store electricity during the day and use at night. If you want to source storage devices or systems. and there’s many companies where you can actually obtain lithium-ion batteries.
BG: Tesla comes to mind.
YH: Yeah. Of course, if you look at lithium-ion batteries, we use cobalt in our cell phone and we’re moving more and more away from cobalt and into nickel. Because we’re currently using roughly 40 or 50% of the cobalt we mine today in lithium-ion batteries. We have to be aware of the sustainability and also the costs of cobalt and utilize less expensive elements like nickel and manganese. Even going beyond lithium-ion trying to look for new chemistry that can essentially potentially offer much lower cost of storage technologies. Like some of the activities Don has been working on.
BG: Let’s get nerdy about materials. Because that’s basically the focus of both your research. Lithium-ion battery works, it’s a solid-state batter, right? Or is it a liquid battery?
YH: It’s a liquid battery. Current commercial battery technology is you have organic electrolyte. Electrodes are solids.
BG: Negative and positive.
YH: Negative and positive. You have lithium-ion moving through electrolyte, from one electron to the other. Electrons move through the axon or circuit.
BG: The electrons are what power of my device?
YH: That’s right. Now there are challenges with flammability of the organic electrolyte, and this is particularly challenging for the safety issues of large lithium-ion batteries. You probably have seen there’s quite a bit of activity in developing solid state lithium-ion batteries. This is in the works and still in development.
BG: What’s the role of cobalt in the lithium-ion battery?
YH: Essentially, we store each electron with each cobalt. If you want to store a million electrons, we need a million cobalt atoms.
BG: And that’s good? Or is that bad?
YH: That’s good, in terms of the efficiency is very good for lithium-ion batteries. But if you say, I want to, then the amount of energy I use is very much dependent on the availability of cobalt.
BG: It’s expensive, right?
YH: It’s expensive. We need to have it and that’s essentially two issues. One is availability. Where can we mine cobalt cheaply? Also, the recycling of lithium-ion batteries will become increasingly important as we use more and more lithium-ion batteries in vehicles and also in large systems for stationary applications.
BG: Cobalt, as I understand it, comes from Central Africa, right?
BG: And lithium, we don’t have any in the United States, right? Or we do?
DS: There is some lithium here but the cheap lithium is coming from either South America or China. But I’d like to return to the other piece about the cobalt. It’s correct that the majority of cobalt is coming from Democratic Republic of Congo and the conditions under which it is mined wouldn’t stand up to scrutiny by human rights people. A lot of this stuff is at the surface and it’s being harvested by child labor. The face that people put on this is they put the term “artisanal mining”. “Artisanal mining” is a cover for child labor. Companies understand that they’re trying to engineer cobalt out of the lithium-ion battery. But when they make a battery that’s cobalt-free and, Yang, correct me if I’m wrong, but my understanding is if you try to make a battery that’s cobalt-free, and I’m not talking about iron phosphate here, but the performance drops off precipitously. The amount of cobalt in many of the formulations is down around 10%. They’ve got a target of 5%. They would dearly love to get rid of cobalt dependence completely. But that’s another concern. There’s the sustainability piece and then there’s the human rights piece.
BG: You guys are professors from MIT. You know the periodic chart. What’s so hard about finding a substitute for cobalt?
YH: To add on Don’s comments, if you look at the electric vehicle sector and the chemistry that most researchers are working on and what technology is focusing on, it’s replacing cobalt with nickel. Their target formulation is 10% cobalt and roughly 80% nickel, maybe 10% manganese. The performance and the energy storage density are very good. There are some challenges with cycling but I think we now really have good solutions addressing the stability. We will have in the very near future largely nickel-based lithium-ion batteries.
BG: When we say the performance and the efficiency, meaning the density, how much power you can put into the battery?
YH: Essentially, it’s cycle life. If you want the batteries to last 10 years, and you want at the end of 10 years to still have 80% of the initial rated stored energy, over thousands of cycles, we would want to make sure that the energy you can get out remains largely the same.
BG: Problem solved. We use nickel with a little bit of cobalt, is that it?
YH: But it’s still expensive. If you really think about, let’s say, the electric vehicle sector, going beyond. Think about the amount of energy we use in the U.S. or China. We would really would want to look ahead to 2050 or 2100. We would want to power the planet with the richest elements we have. That’s basically the elements we have in the Earth’s crust, in our water, and in our air. Those elements are the richest and they turn out to be the cheapest.
BG: Donald Sadoway, I saw a TED talk where you said, “If you want something to be cheap as dirt, make it out of dirt.”
DS: That’s correct.
BG: I thought it was a great line. How does it fit in with batteries?
DS: I’ve been practicing that in my research, in looking for different chemistries. Chemistry is different, completely different from lithium-ion because it was my position going back 10 years ago that lithium-ion was absolutely the right choice for mobile devices. By that I don’t mean cars, I mean…
BG: … headphones, cell phones?
DS: Mobile phones and portable computers and so on. When you start getting to grid-level storage, the lithium-ion will perform but when you have a single lithium-ion cell inside a phone, it’s thermally managed just by surface cooling. If you put hundreds and thousands of these cells, in each one of them the workhorse cell is about twice the size of a double A battery. If you put, say, some thousands of them in close proximity, they’re not going to just naturally cool except the ones on the outside. The ones on the inside have neighbors that are also hot. That requires additional plumbing to have some kind of forced air or forced water to keep them cool because if they rise in temperature, lithium-ion is very temperamental. You get up to about 70 degrees Celsius and you’re going to have bloating of the cells and ultimately you could end up with fire. I started looking at something completely different. I began from the perspective of, choose something that is abundant and cheap and then turn around and make try to figure out how to make that work in a battery. That was sort of the direction that I took.
BG: You started off at the cost factor, the safety factor, and then worked backwards?
DS: That’s correct.
BG: What did you come up with?
DS: There’s a suite of batteries that I call liquid metal batteries. They have their plurality of choices for the negative electrode, things up in the northwest part of the periodic table, things like magnesium, lithium, sodium. For the bottom electrode, I chose things from the southeast part of the periodic table. In other words, pairing metals that are as far removed from each other as possible. That way they’ll generate voltage. That gives you things like antimony, lead, tin, bismuth.
BG: And it’ll blow up.
DS: No, it doesn’t blow up.
BG: Or generate that much heat that you lose the efficiencies in cooling things off?
DS: You want to have the heat, in this case. I took a liability and turned it into an asset. If you’re going to run a battery at 500 degrees Celsius, you want it to generate heat and you trap that heat. That’s how you make it stable at that environment. The last thing, Yang mentioned, the lithium-ion battery has an organic liquid as the electrolyte. That organic liquid is sort of a cousin of gasoline. It’ll burn and it’ll burn well. I chose a molten salt instead of salt dissolved in water as the electrolyte. Just take the salt and melt it. Now you have something that’s bulletproof against thermal runaway. That was the genesis of this. I want to be fair and acknowledge that there are people here at MIT that are working on flow batteries and these are gigantic tanks of allyl valent metals that change valence. There’s a plurality of options. There’s a temptation to take lithium-ion, which we know a lot about, it’s served us very well in phones and in computers, and try to scale up. As Yang mentioned, we don’t have any evidence of lithium-ion batteries lasting 10 plus years. Nobody has a 10-year-old phone in his pocket. If he or she does, most certainly they’re not running with the original battery. But the batteries that have to go onto the grid or go into homes, 10 years is a minimum. There’s a lot of work yet to go.
BG: Where’s the future? I know you’ve got a company called Ambri. It’s here in Massachusetts. It’s producing these batteries that are very high temperature. They’re like bricks. Do they work?
DS: They work. But the answer your question is yes and no. Yes, they work, but no, they’re not ready for release into customer hands. Because we have to not just manufacture them but we have to test them to the point that we are confident that they will pass reliability expectations. The last thing I want to do is to put a really cool battery chemistry in a customer’s hands and then have some unforeseen mishap occur. There are some people that would say to me, because grid operators are very, very conservative. They would say to me, you claim this thing will last 10 years without incident? Show me 10 years’ worth of data. I better get started this afternoon.
BG: The utilities in Massachusetts are starting to require storage, right? We just had the first request for proposal for offshore wind and they’ve got to store a considerable amount, like 1200 megawatts. That’s a chunk of energy, right? What would you use right now? Would you use Don’s battery?
YH: It’s not available for customers. I think this is where, whether conceptually lithium-ion batteries are the ideal technology for storage or not, it is available. I think this is where to put lithium-ion battery at the center of the stage because of the scale of the technology and because of the availability of the technology, not necessarily because it’s the ideal technology, because of this cost and because of availability of materials to scale.
DS: I think this is exactly what we’re seeing in automotive. I remember seeing a video clip of [Elon] Musk where he says the problem with batteries is that they suck. He wants to get an electric car on the road today so he’s going to take what is available, lithium-ion. But that car is agnostic when it comes to chemistry. All of the power electronics, all of the mechanics, everything is there. If better chemistry comes along tomorrow, they’ll jump to that. But they’re not going to wait until the better chemistry is discovered to delay the rollout of all electric vehicles. I think that’s the same thing here with stationary storage. If the battery that’s available right now, it’s not going to be economically viable. They’re going to have to put all kinds of safety measures in there but they’re going to get to use offshore wind with storage together and learn how this thing works. I predict that in addition to the known benefits, they’re going to discover unanticipated benefits, which is going to make it even more compelling a case. Then when the better chemistry comes along, we’ll push out the lithium-ion batteries which will already have lost some of their storage capacity. We’ll have gotten that much farther instead of just sitting around waiting for the perfect battery.
BG: Yang, you’re trying to perfect the perfect or more perfect battery, right?
YH: I think part of being a faculty member is we look ahead. Not only do we work with lithium-ion batteries, we look ahead and develop a concept for new storage mechanisms. What are the chemistries we should use for 2050 that can potentially power the planet?
BG: We’ve got to supply them with energy. They want to plug in their devices.
YH: Right. With 10 billion people in mind, what are the elements available to store energy? To capture the electrons from solar wind and use them on demand, we can think about lithium-ion batteries. The reason cobalt is useful in lithium-ion batteries, it can actually have the mechanism or the function of storing electrons, because cobalt is embedded in the oxygen matrix or oxygen lattice. Our concept is to take cobalt away and use oxygen to store the electrons to develop either hydrogen oxygen fuel cells or electrolytic cells or metal air batteries.
BG: If I understand you right, you’ve got a material, and you’re going to take in the lattice, in the molecular structure, you’re going to lace that with oxygen?
YH: In the lithium-ion batteries, what stores electrons is lithium cobalt oxide. What we are advocating is to develop hydrogen oxygen, or lithium oxygen, or aluminum oxygen. Essentially, remove transition metal elements in the chemistry to store electrons.
BG: When I hear you say hydrogen oxygen, I hear fuel cell, right? Chemical fuel cell?
YH: Yes. That’s the terminology but from a fundamental or physics point of view, it’s really the oxidizing power of oxygen. How to convert oxygen to water that essentially gave us energy.
BG: And it’s been used in space. That’s how they powered Apollo. It works, it’s just very expensive, right? And the power density is not tremendous?
YH: It was used first in Gemini. Also, recently, Toyota has launched fuel cell vehicles, priced at $70,000.
BG: What’s it called?
BG: Mirai. That’s it. It’s expensive. I went pricing clean energy cars the other day. Holy cow. They’re out of my field. They’re way out there.
YH: Absolutely. But this system can be scaled. Because you can actually decouple energy and power. You can always make the tank much larger and make the stack scale differently from the size of the fuel.
BG: Stack meaning the stack of batteries?
YH: The stack of fuel cells that generate power. For lithium-ion batteries, power and energy, they scale, where for open systems like fuel cell, like flow batteries, you can scale energy and power differently. Therefore, that’s one of the reasons potentially flow batteries can be developed for stationary applications. Because you can have…
BG: … a lot of them. They’re very big. You have to have tanks, right?
YH: Very large tanks where you store the energy, where you can have reasonable-sized flow battery stacks.
BG: There’s lots of different chemistries you can use for flow batteries, right? I know that down the street there’s a university of some renown, they were testing rhubarb and it seemed to work. Kind of weird.
DS: All you need is an aqueous solution and something that’ll shift valence. In the end it comes down to dollars per kilowatt hour.
YH: The advantage of utilizing hydrogen or lithium oxygen or aluminum oxygen is they can store much greater energy per unit weight relative to lithium-ion batteries. Typically we’re looking at a system level, for example, three times greater energy density on the graph, the metric scale. Where, versus some of the chemistry we currently know for flow batteries, they actually store less energy per unit weight than lithium-ion.
BG: But because they’re so big and you can have them stationary you don’t have to make the mobile. That’s not a tremendous factor, right? Or it’s not the defining factor? Or is it?
YH: I think in the end it’s really what is the cost of the system? If you have a very large system, maybe in the end it will boil down to the cost of chemicals. But for a reasonable small or medium size, all the supporting system to make the flow batteries or the fuel cell work are also a big contribution to the overall cost.
BG: Where is the breakthrough in battery technologies? Where’s the science? Where do we need to go? What are you doing in your laboratory? We’re on the MIT campus.
YH: We’re working on making lithium oxygen and metal air, or lithium air batteries, a viable option going forward. There are many challenges. The technology, as I mentioned, can store much more energy. The downside is they’re not as efficient as lithium-ion batteries in terms of energy efficiency. But imagine, if we can make the process energy-efficient or the process reversible then imagine that these types of technologies or chemistries can be used to power electrical planes. That’s also a very exciting area. We see many important steps or activities in this area.
BG: A battery powered airplane, right? I know there actually was one that was tried by the Air Force, or maybe it was DARPA. They had these flow batteries. But they were very unstable. They were very big.
YH: They’re heavy.
BG: They’re very heavy. Do you see a day when we’ll have electric airplanes?
YH: Yes. In the past few years, there are many more startup companies developing various electric taxi or electric-type vehicles that can potentially power a flying object that can deliver or transport people, like an electric taxi from an airport to downtown. These require very different storage chemistry. Because light weight and storing more energy is absolutely critical.
BG: I know there was a company here, it started here, it’s now in California, called Wright Aircraft, something like that. They were experimenting with batteries. It’s interesting because they started with a business model and worked backwards, like you did, Don. That is, they started with the idea that we want to have an electric plane, but it can’t be one of these jumbo jets, it has to be a small jet, a commuter jet that goes 200 miles, 150 miles, carries 60 people or 30 people, and they work the math backwards. Then they said, what’s the chemistry, what’s the engineering that we need to do for this? They’re not there yet, but it’s kind of a very tantalizing idea. Do you think we’ll have flying electric planes?
DS: I don’t know anything about aerospace. I’m going to dodge that question.
YH: I heard this week that the government in Norway is mandating that a certain fraction of the airplanes in Europe by a certain year in the near or distant future will be electric.
BG: We’re getting there, slowly but surely it seems. We have these mandates for offshore wind and solar. There’s the need to shift the demand, store it during the day when people aren’t needing it. Storing it in batteries and shifting the supply to when they need it, early evening, late evening, right? That’s the whole idea behind this utility size storage.
BG: So, when? How fast? I was listening to an article this week about Tesla. They said they can now charge the battery in 15 minutes. That sounds really incredible. That changes the whole thing, right?
BG: So, where are we? What’s your prediction?
DS: Niels Bohr said prediction is always difficult, especially with it’s about the future. I can’t tell you when the breakthrough is going to come. But there’s no question that the tipping point is going to be achieved by economics. If I gave you a battery that is going to get price of the car on the showroom floor, down to par with the same car pump powered by internal combustion engine, that’s when things take off. We’re waiting for that breakthrough. I can’t predict it. People are working. To go back to your earlier question, you said we have all the elements in the periodic table, what’s taking you so long? The answer is that, in spite of all of the advances in computational chemistry, the complexity of a battery is such that it’s not a matter of just running a computer program and then out pop the top list of candidates. We have to mix that with experimental evidence as well, and these experiments take time.
YH: To go back to your earlier question about breakthrough. I think for lithium-ion batteries, electric vehicles, or different degrees of hybridization in a vehicle, it’s happening. Various car companies have very ambitious plans to implement lithium-ion batteries in their vehicle technologies. I think it just matter of time for the lithium-ion battery powered vehicles to penetrate our fleet.
BG: Do you have an electric car?
YH: I’m waiting for the electric Mini to come out next year.
BG: And you, Don?
DS: I don’t own a car.
BG: [Laughs] You’re cheap.
DS: No, it’s not cheap. You want me to be behaving in a sustainable manner. I walk to work. [Laughs]
BG: Well, thank you both very much. I really enjoyed it. Thank you very much.
YH: Thank you. Glad to be here.
BG: That’s Donald Sadoway, he’s a professor at MIT in the Department of Materials Science and Engineering. Yang Shao-Horn is co-director of the MIT Energy Initiative Energy Storage Low-Carbon Energy Center. You’ll find show notes and links for this episode, they’re available at energy.mit.edu/podcast, and you can tweet the Energy Initiative @mitenergy, that’s one word, with any questions or comments. I’m Bruce Gellerman from WBUR. Thank you for listening.
Press inquiries: firstname.lastname@example.org