According to Silicon Valley visionary and Sun Microsystems co-founder Bill Joy, much about the future of energy and climate change hinges on a simple question: How soon can the world’s electric grid get its battery pack?
As a partner in the Menlo Park, Calif. venture capital company Kleiner Perkins Caufield and Byers from 2005 to 2014, Joy says he focused on what they called the “big hard aggressive goals” or BHAGs. And from climate to water desalination to electric grid stability, the real challenges of BHAGs often circled back to cheap, safe, scalable, and reliable grid-scale energy storage.
Last month, Joy was in Woburn, Mass. for the ribbon cutting of new facilities for a solid state battery company he’s an advisor to and investor in, Ionic Materials. Ionic Materials has developed a plastic electrolyte—the medium in a battery through which energy-storing ions flow between cathode and anode and back again. This specially designed plastic conducts ions like a metal conducts electrons. And switching from today’s flammable liquid electrolytes to flame-retardant solid electrolytes, the company says, could bring about a new generation of lighter, more powerful, non-flammable rechargeable devices.
Joy sat down with IEEE Spectrum for a conversation about the promise of solid-state lithium (and rechargeable alkaline) batteries, the “Internet of energy,” and the coming electrification of airplanes.
IEEE Spectrum: As you have said, the English language doesn’t have a single word for the kind of battery breakthrough that’s needed—a solid conductor of ions, a solid electrolyte.
Joy: There’s this evolution of language. Like we had telephone, and then we had the touch-tone phone. And then the phones that weren’t touch-tone suddenly were rotary phones. And then you called on cellphones. But now when you say “phone” you mean a mobile phone, and if you want to talk about the other phone you need to say “landline.”
Now here’s this thing we’ve never had a name for. It’s like when you apply for a patent at the Patent Office, and they say, “No prior art.” It’s really a rare thing.
IEEE Spectrum: You called this new kind of solid electrolyte an “ionyl.”
Joy: I’m just trying to say that it could be like “metal.” I mean, you can have a liquid metal. But “metal” normally means a solid.
IEEE Spectrum: One battery expert I spoke with said that you can make all kinds of new materials, but what really matters is making the battery—proving that your new materials really work as the basis for a new generation of battery technology.
Joy: Making batteries has [traditionally] been hard, because they had liquid in them. Especially if the liquid needs to have no water in it, and there’s moisture in the air. To make something at scale involves a lot of investment in manufacturing equipment, and batteries made with our polymer require less capital equipment. But Ionic [Materials] is not building a battery factory. We can make batteries in the building. But the truth is for the high volume of manufacturing, we’re working with a bunch of different companies. And the different companies will have different techniques they want to use, different materials they want to use. We’re not a battery company. We’re a materials company and an IP company.
IEEE Spectrum: You’ve done some spreadsheet calculations that finds a plastic “ionyl” battery is cheaper than today’s lithium-ion batteries.
Joy: If you look at just the active materials in a battery, we can imagine getting down to the order of $10 per kilowatt-hour. That’s not the battery cost, that’s the cost of the ingredients… If I look at a lithium-ion battery, that would be graphite plus a separator plus a cathode plus the liquid electrolyte. So you’d need all four of those. And that’s about $100 per kilowatt-hour or more, depending on the current cost of cobalt. But we can make things that are as cheap as or cheaper than the stuff you can get in an alkaline battery at a hardware store. Because it’s basically zinc and manganese, which are way cheaper.
The other thing is I also found [energy storage in "ionyl" batteries would cost] less than a penny per kilowatt-hour. Which is a different measure, because that imagines putting the batteries on the grid. So for example if the batteries cost $10 per kilowatt-hour, to get to a penny, they’d have to do a thousand cycles. Because $10 is a thousand cents… So I just need to have a cycle life that’s a couple thousand, and I can imagine getting that cost of just the battery component down. But we’re not there today, because that’s high volume that we haven’t fully commercialized yet. But that’s where we are aiming to get to with our technology. And that gets… the grid to 100 percent renewables. Because wind and solar, if paired with inexpensive solid state grid storage, would be cheaper than [fossil] fuels. And they no longer have the variability, they would be dispatchable. All you have to do is buy the wind turbine or the solar panels and batteries. You don’t have the variable fuel cost.
IEEE Spectrum: You had referenced the cycling of these “ionyl” batteries. Are there any indications of how well they’ll run after 100 or 1,000 or more cycles?
Joy: It’s not generally the electrolyte that’s the problem. Think of it this way: I give you some anode powder and cathode powder, and I put them in a liquid. And they’re not completely pure. Anything that’s junk might be soluble in the liquid. It’s in solution now. And it might bump into something else that’s floating around. And they might stick together due to electrostatic charge. You can end up with dendrites, which look like stalactites or stalagmites… In other words, in the real world, the ingredients aren’t pure, and in a liquid [electrolyte] you get all kinds of other things going on. But if you take your powder and instead of soaking it in a liquid, you disperse it in a polymer, it’s still in dry form. If there’s some impurity in there, it’s not mobile. It’s not an ion, it’s not an electron.
The polymer helps remove some of the accidental chemical reactions… But say there’s silicon [in the cathode or electrode] that’s shrinking and expanding and falling to bits. Every time, that’s a stress-strain. It’s like taking a paper clip and bending it enough times that it eventually falls apart. If there are some of those breakdown kind of things going on either chemically or electrically, those will be cycle-limiting. But if you can knock those down, you can get thousands of cycles.
So for example, if you could get 5,000 cycles, and you do one cycle per day, that’s 10 to 15 years. We have other things in our lives that are solid polymers, and there are still limitations to their lifetimes. But it’s much, much easier to achieve long lifetime with a solid polymer than without it.
IEEE Spectrum: Technologies that are transformative—not incremental—sometimes spawn entirely new industries and applications that no one ever anticipated. Have you speculated whether these solid polymer electrolytes are like that? And if so, what do you think those new industries and applications are?
Joy: We've done some prototypes—we’ve just done it enough to know that it works. And it could always be the case that there’s something that makes it hard. But as far as we know there’s [a battery "ionyl" electrolytes could make more possible] called lithium-sulfur. And lithium-sulfur is really light. So you can see a huge step forward toward electric air transportation. Now the thing is I don’t know how to do electric jets. Jets are based on burning a gas and transferring that heat to pressure. What I have is electricity, which can turn an electric motor. That’s more like a turboprop. So, someone smarter than me might have a way to do an equivalent of a 747 that’s electric. I don’t know how to do that. But I think regional jets and commuter jets and drones will be radically improved by it.
And then for the grid, the major thing is [considering ] if the cost of storing and retrieving a kilowatt-hour are less than a penny. Then if I have wind at two cents… the question is how often do I need to store it? Some percentage of it will be used immediately. That won’t need to be stored. Some if it I can send the electricity to go around a congestion bottleneck in the transmission grid. I can move it at night. I can take an asset that was idle and would have been wasted—it’s like an empty airplane seat—and it’s still valuable to move energy around. So if by moving it at night, in a place where there would otherwise have been congestion, I avoid the need to build and depreciate a new grid.
Really cheap storage causes you to not have a power grid but an energy grid. And an energy grid is just not the same as a power grid. It’s totally amazing. Because you can buy and sell energy and use it later. I can warehouse energy and sell it back to people.
The third one would be if I could make batteries for $10 and I can sell them for $20 [per kilowatt hour]. In an electric car you have, say, 50 kilowatt-hours. That’s $1,000 for the batteries. That’s less than the cost of the engine by a lot. The cost of ownership of that battery is way less than the cost of ownership of the engine. But I don’t really need 50 kilowatt-hours of batteries for most places around the world. In most places I might need 15. So now you’re talking $300 for the batteries for a car. That’s mature prices, because in the end I have to buy zinc and aluminum. I have to buy things that there’s a mining cost. … So I don’t expect the cost of zinc or aluminum to come down. But, really, I do have the advantage that I’m not going to run out of them.