15 Questions for the Don of Liquid Metal Batteries

Donald Sadoway, a professor of materials chemistry at MIT, aims to deliver a “lifesaver” for renewable energy in the form of a stable, low-cost, large-scale battery. With one of the earliest grants awarded under the Department of Energy’s ARPA-E program, his team is working on a battery project that sandwiches molten salt between two layers of liquid metal.

The plan is to move the battery from “shot glass” scale, to hockey puck, to pizza, and eventually to ping-pong table-sized. Down the road, Sadoway tells us, he envisions possibly spinning off a company with some of his coworkers. Here are 15 questions and answers (excerpted and edited) from our interview with Sadoway:

Q. What are some of the benefits of using liquid metals for electrodes–why focus on this?

Sadoway: The thing that drew us to the use of liquid metals was a belief that a battery based on liquid metal electrodes would be stable, and scalable at an acceptably low cost for grid storage and renewables storage applications. That’s the number one problem with storage right now. We have batteries that can cycle and do all sorts of things that will meet the technical requirements of the application, but they’re far too costly.

We thought that by going to liquid metal, if you wanted to make a big battery, we would just have a large pool of liquid metal as opposed to thousands of individual cells, each of them the size of a soda can, for example. That difference in form factor is what we believe will allow us to reach the cost targets.

Q. What progress have you been able to make under the ARPA-E program?

Sadoway: The ARPA-E money allowed us to expand our operations to hire on more staff, students and post-docs, and to divide the technical challenges amongst a large team. We’ve looked at a lot of alloy systems that we couldn’t have without the ARPA-E funding. We’ve been able to address materials challenges, operation challenges — and by that I mean how you run one of these cells, what kind of charging currents, discharging currents one can sensibly expect — basic electrochemistry. We’ve been able to really move much more quickly than we could have before we got the ARPA-E funding.

Q. How do you envision this technology moving out of the lab and into a commercial product?

[Editor’s note: Sadoway explained that his team is currently working in the lab with 1-amp-hour cells in order to screen various alloy combinations. In parallel, they have also designed and built a cell at 20 amp-hours.]

Sadoway: There’s the 1-amp cell, which I call the shot glass, the 20-amp cell which I call the hockey puck, and there’s going to be the 200-amp cell, which will be about the size of a pizza. Once we can show that as we move from the shot glass to the hockey puck to the pizza, that things hold, or if anything the performance improves, I think at that point we’re ready to build a big cell. That would be about the size of a ping-pong table.

Q. Would that be full-scale?

Sadoway: I don’t know just yet. We might get to the size of the ping-pong table and decide it would be better to have something maybe four times that. Or we may decide, based on the performance, that the ping-pong table is sort of the right cell size. Instead of making a bigger battery, you could put a whole bunch of ping-pong table sized batteries together. That’s all part of this question of how do you build large-scale storage capacity cheaply? I don’t know. This is something we’re going to learn by scaling as it comes.

Q. What is the timeline you’re expecting for scaling up?

Sadoway: The ARPA-E grant runs its course about two years from now. By the end of that grant, we need to know if we’ve got something here or not. Because building the cell the size of a ping-pong table is going to cost far more money than building a cell the size of a hockey puck. Somebody’s going to have to put up that money, and they’re going to want to do so with some reasonable assurance that the risk is a sensible risk and not a Hail Mary kind of last-minute pass down the field.

Q. So after two years, the idea would be to get further investment in order to set up more manufacturing capacity?

Sadoway: Yes. Well, I think there’s an intermediate state there. You want to take something off-campus. Because at that point you’re going to be building cells that are too big to be operated on campus.

Q. Would you spin off a company, as opposed to licensing out the technology?

Sadoway: I want to make sure the technology doesn’t stumble, and I think by spinning off a company with some of my coworkers here, that heightens the chances of success. You’ve got people who are really competent with the technology and who are eager to see it succeed. When you license it to an organization that doesn’t have a stake in what’s going on, I have concerns about the level of commitment and what lengths people are willing to go to make something succeed.

ARPA-E wants to see these technologies in the marketplace. Either you license the technology to some big company, which then says “we’re going to build these things and sell them,” or you see the technology move off campus into a startup company, which has the added incentive of trying to build a business out of nothing except an idea. That would be a powerful incentive.

Q. Are you devoting a lot of energy and focus now to the idea of building a business out of this?

Sadoway: We’re definitely considering. The ARPA-E folks know what our plans are in that regard, and they support us because they want to see the technology mature to the point where it’s actually available for people to use.

Q. If this technology lives up to its potential, what might the future look like? What could this enable?

Sadoway: It will enable grid-level storage in the extreme. There are a variety of applications. You could imagine batteries about the size of a small refrigerator in the basement of every home, where people can take energy off the grid in the wee hours of the morning, then draw upon that stored energy throughout the day and maybe even sell it back to the grid during peak demand times.

You could imagine batteries the size of a small building acting as a ballast at the level of a subdivision. You could imagine batteries near power plants, central facilities being able to draw down huge amounts of electric power and then push that back in the middle of the day. And that would mean you don’t have to build more power plants.

When it comes to renewables, these batteries could be the lifesaver, because wind and solar both suffer from the fact that they’re intermittent. Nobody wants intermittent power. They want power they can count on. That means we’ve got to figure out a way to allow wind and solar to become part of base load, and the enabling device is a battery. And so if this battery is scalable and at an acceptable price, then you can imagine it being used in conjunction with either a wind farm or a solar farm to give us the ability to store that energy and provide continuous electricity.

We haven’t invented a battery; we’ve invented a battery platform. There’s a suite of chemistries that all have the common feature of liquid metal on top, liquid metal on bottom, molten salt in between. But the identities of the two metal layers and the molten salt layer, those identities can vary over a wide range of chemistries.

Q. What do expect to be the earliest applications for this platform; what’s the lowest hanging fruit?

Sadoway: I’m thinking maybe the unit for a single-family home. Maybe, but I’m not sure.

Q. What makes that seem the most feasible?

Sadoway: It may be the smallest scalable size. I say we’d like to be able to store the grid, but I just don’t know if we can build something that big and do it right the first time. So I’d like to start with something smaller. If it’s too small, it won’t be self-heating, so I think the single-family home is probably the right application for starters.

Q. What’s your target for the cost, and how far is your team from reaching that goal?

Sadoway: It’s got to be below $100 per kilowatt-hour. The materials cost are down around $20 a kilowatt-hour. The question is, if we start with something that just costs $20 a kilowatt-hour for starters, can we put it all together and ship it at say around $50 and have some head room for profit? I don’t know the answer to that.

Q. Who do you see as your biggest competitors, and how do your costs compare?

Sadoway: There really are no competitors. People are starting to use a little bit of sodium sulfur in stationary installations, and sodium sulfur, as best we know, is up around $600 a kilowatt-hour. Lithium-ion is almost $800 to $1,000 a kilowatt-hour.

Those batteries are just far too expensive. But people are trying to work with them anyway in the hope that there’s some kind of unforeseen benefits and ultimately a cheap battery will come along, so let’s get the electrical engineering in place. [This way,] when the right battery comes along we’ll have enough operational experience to know how to use a battery in boosting the productivity of a power generating facility.

Q. Would you explain why this platform is meant exclusively for stationary applications?

Sadoway: Because you’ve got three liquid layers, you can’t have motion. If you put this in a car and the car accelerates, you could end up in a situation where the contents of the battery slosh back and forth. You would just get mechanical mixing of the two alloys and at that point you can’t draw current from it.

Q. What would happen if one of these batteries was disturbed in some way, during an earthquake for example?

Sadoway: You would lose the composition of the battery. The top metal might alloy with the bottom metal, but then you could restart the battery. That’s very different from trying to put it in a car or train or something. Sometimes you need to slam on the brakes, and you could cause the contents to flip over. You don’t want to lose the battery while you’re driving. It’s elevated temperature, too. I’m not sure you want a battery at 700 degrees in your car, so we’re focused on the stationary applications. It’s a huge market and it’s an unmet need right now, so if we can make it work there, that will be a big service to society.

Image courtesy of Simon Strangaard.