By Danielle Fong
It’s 2013, and my work in compressed air energy storage has finally generated some interest. This raises a lot of questions. Why energy storage and why compressed air? Why now? In which century are we? Most of all, if now, why me?
I knew that providing abundant clean energy would be one of the greatest challenges of my generation, and as a technologist, I knew it was something I could contribute most meaningfully towards. So, I was incredibly excited to be accepted to a program in fusion technology at Princeton University. I was seduced by the same dream that many people were seduced by: abundant energy, forever.
Fusion is, in some form, the energy forever solution. But, as I soon discovered, mainline fusion technology development was slated to take an incredibly long period of time. And we don’t have that time. Most models predicting global surface warming, recommend that we should not exceed a 2oC warming over pre-industrial levels. Fusion technology in the mainline development wasn’t going to be fast enough. To make the kind of difference I wanted to make, I had to move faster. I switched to something that I thought could work a little faster: renewable energy.
It’s pretty simple. Add the cost of off-peak energy and energy storage, and compare it with the cost of peaker plants – the inefficient plants that can rapidly provide power on demand to the grid – plus the cost of grid upgrades, the extra bandwidth that you need in the network to ship that power from where it’s appropriate to burn the fuel to where it’s needed. If the energy from more efficient off-peak sources (either more efficient fossil fuel power plants, or better, nuclear or renewable energy), supported by energy storage is cheaper, then you’ve solved the energy storage problem, economically speaking.
One approach that had not been adequately explored in my opinion: compressed air. It’s incredibly cheap. The tanks are cheap and long lasting. They last 30 plus years which is incredibly important for infrastructure. And it’s extremely scalable. You’re not going to run out of any essential resource scaling this solution.
But the problem has been that compressed air is inefficient. This was, essentially, just received wisdom. People just quoted poor efficiency values. Why, one might ask, is compressed air inefficient? It turned out that almost all the inefficiency was concentrated in a single, thermodynamic loss channel. It’s really basic physics. The problem is that when you compress the air, as anyone who pumps up a bicycle knows, the air gets hot.
PV = nRT
That hotter air gives you higher pressures and high specific volumes for the same mass of air. You have to do more work to compress the air to store it in the tank than you otherwise would. When it cools off, you lose temperature and you lose pressure.
What you want is for the air to be as cold as possible during compression and as warm as possible during expansion. People had talked about this, but it was completely theoretical, because people assumed you would need a mess of stages and interstitial heat exchangers to actually cool the air off during compression and warm the air up during expansion.
We had a better idea: spray water in during compression. Water has an incredibly high heat capacity, about 3300 times that of air at atmospheric pressures, so you don’t need a lot to hold the temperature of the mixture nearly constant during compression. And an atomized spray has an incredible amount of surface area. That means it can transfer an incredible amount of heat. It’s like a heat exchanger that you can squish and regenerate inside the compressing volume.
You store the air and the heat. You take the same water that’s been warmed and you spray that back in during expansion. Or, even better, if you have any other access to waste heat or solar heat you can use that to heat up the air even further and that increases the pressure and increases the amount of energy you get out. We also figured out how to, just by changing valve timings, use the same system to compress and expand the air.
The efficiency that we think that we can hit, eventually, is 70% plus. Our industrial scale proof of principle has run for more than 500 hours now and hit our thermal efficiency targets. And by applying the rest of the technology that evolved from the automotive world, we can hit a much higher efficiency for an overall system, between 60 and 70%. We’re on track for delivery in 2014.
That’s the technology. But we need to do much more than that. To really make the difference that we want to make, we’d have to be a critical element of the world’s energy infrastructure in the 2030 time frame. That means terawatts of energy storage. And getting to terawatts is going to be hard. We’ll have to have extremely aggressive growth: approximately doubling in sales every year for 16 years. We also have to be the best investment every step of the way. Luckily there are markets that are structured where they’re in progressively greater scale, but we can be cheaper than providing peak power by conventional sources each step of the way. So we’ll start out with a pilot, then get to volume production, and then can use extremely large underground chambers to store the rest of the air.
But it might not actually solve the problem of fossil fuels being burnt, because in the worst case, when you provide cheaper renewable energy the cost of the fuel will just go down. In the worst case it’ll hit the cost of extraction and processing, and that’s incredibly low, probably three to five times lower.
Eventually, I discovered that the guns were loaded. The total mass of fossil fuel reserves just on the books for companies that sell fossil fuels — and the countries that act like companies that sell fossil fuels – exceeds the carbon budget to stay below the 2oC warming limit.
But there is good news. The environment can absorb a certain continuous amount of carbon emissions, about a fifth of its current value. So if we use fossil fuels only for their most high value applications – which, it must be pointed out — would appeal to the self interest of the fossil fuel companies and countries, then perhaps the consumption can go down far enough for us to avoid climatic destabilization. For example, air transport represents only 1.7% of global greenhouse gas emissions. We could handle that with fossil fuels.
I’ll try to demonstrate it most clearly by making an analogy. Our solutions need not be technological. They can also be psychological. Essentially, a country that takes its resources, fossil fuels or otherwise, extracts them from the ground and burns them in low value applications is just extracting money and setting it ablaze for a moment of light. And it’s our job to put it out before it’s too late.
Danielle Fong is the Chief Scientist and Co-Founder of LightSail Energy, a compressed air energy storage company in the Bay Area, backed by Khosla Ventures, Peter Thiel and Bill Gates