“Will we prevail? I hope so. But not without a lot of unnecessary suffering.” Steven Chu on the science and politics of renewable energy.
In Episode 29 of Cleaning Up, Michael Liebreich talked to Steven Chu, Nobel laureate, first Secretary of Energy under the Obama administration, and current Professor of Physics and Professor of Molecular and Cellular Physiology at Stanford University.
In a wide-ranging conversation, Michael and Steven talked about Professor Chu’s ongoing research into electrolysis and battery materials, the affordability of renewable energy, the economics of nuclear energy, and respect for science.
Below is an abridged transcript of the conversation, edited for clarity.
Michael Liebreich: Steven, thank you very much for joining us on Cleaning Up. Let’s start by talking about what's taking your time right now?
Steven Chu: Well, it's a mixture of things on the university research side. I'm working with a colleague of mine, Yi Cui, on lithium metal sulphur batteries. No one's been able to make a lithium sulphur battery because they can't keep the sulphur on the cathode side, so we're trying a new idea. We just published your paper this June on how to mine lithium from seawater using electrochemistry. If that works, lithium won't be a problem, at least for a couple of thousand years. There’s also work on a new way of doing ultrasound imaging, so-called nonlinear imaging, which turns out to be much more sensitive to tumours.
Finally we're doing new nanoparticle synthesis. I didn't know how to synthesise particles before. That's on my university side; on the other side, I advise some start-ups and some bigger companies. Royal Dutch Shell and Siemens is another, but also small start-up companies, carbon capture companies, a biotech company that is combining robotics with new powerful gene manipulation and machine learning.
ML: Last time we spoke, you also mentioned electrolysis. Nano structuring surfaces so that there aren’t the forces against bubble formation, one of the things that creates a parasitic energy requirement in electrolysis.
SC: That's continuing. I convinced Shell and Siemens about three or four years ago, to take electrolysis seriously, for the following reason. When electricity is $0.05/kWh, the energy cost is simply more than you can sell this stuff for, even if capex were zero. Shell thinks [renewable power] will be 1.0 to 1.5c c/kWh within a decade or so. All of a sudden, the cost of the energy is less than the third of everything, and you went from impossible to “let's think of more clever ways of doing it”.
It’s not only the bubble, creating internal resistance. Do I want to form a bubble? Or do I want to pull out the hydrogen 5000 times faster diffusion into the gas phase? So of course, that's the route you take, that clears out the hydrogen and similarly the oxygen side. And it’s catalyst-agnostic and has a lot of promise.
ML: You could use it also to reduce the costs and the energy requirement of direct air capture?
SC: Fundamentally, if you think about what we've been doing for the last 130 years, we take hydrocarbons, loaded with energy made by nature, and we go energy-downhill. Now, if you want to take water and carbon dioxide, and go back uphill, you've got to supply energy. The exciting thing is, in a world of 20 years from now, when we get really good at both generating energy, and we've got good at storing energy, then the move uphill becomes less and less expensive.
However, eventually we're going to have to take carbon dioxide from the air. It is almost a given now. We're at 415 parts per million, we're not going to stay below 450, that's in a decade or two, we're probably going to go over 550. And then when you get over 550, you're getting into pretty dangerous territory (although there's a little bit of time before the polar caps melt). So you're going to have to get CO2 out of the air, and put it back underground or mineralize it or do something, and we've got to figure out how to do that very inexpensively.
ML: The holy grail is of course a synthetic fuel, not starting with starch but with lignocellulose or something that nature produces in abundance. Or do you think we will go directly from CO2?
SC: That's tough to call. I think right now, it's more likely to go biomass, breaking down the lignocellulose much more efficiently. The hydrogen is going to be commercially viable with electrolysis within a decade.
Now the carbon. There's no way we go to zero unless we're grabbing stuff from the atmosphere, either by photosynthesis or direct air capture. I don't know if most of your listeners know, but the amount of carbon dioxide that's captured by the crops in grassland regrowth for grazing is more than the carbon emissions of the world. If you can take that natural photosynthesis, you can make substitute products: eventually fuel and most stuff. You will also generate some CO2, for example, in the fermentation process. If you can capture that, you can stick it underground, or mineralize it. We need some negative sources of CO2.
ML: You're wearing a SunShot t-shirt. At the Department of Energy, in 2011, you launched the SunShot initiative to reduce the total cost of solar energy by 75%, making it cost competitive with other forms of energy without subsidies by the turn of the decade. Were you surprised at the speed of development?
SC: Absolutely shocked. When we set up the programme, we had these waterfall charts and said, where could the cost reductions come from, and tried to really detail what the map roadmap might look like. And then in our first SunShot symposium, [PV industry veteran] Dick Swanson got up and said “when I first met with you, I thought you guys were nuts. You must be smoking something. But then we started talking, and I went back and thought about it and said, it's not crazy, it's an achievable goal.”
What we did not anticipate was the Chinese government-subsidised over-investment, which drove a lot of solar companies bankrupt, drove the biggest Chinese solar company bankrupt. But the Chinese, unlike the US said, no, we're not going to let you go under.
The next biggest challenge is, you've got solar, you've got wind, but they're not on all the time. You need energy storage, and it cannot all come from chemical batteries. In fact, the larger scale storage, storage for several days, will not come from chemical batteries, it's going to come from pumping water in existing dams, uphill pumped storage, which is 95% of all energy storage in the world today. Or it's going to be storage of heat in a new way, or mechanical.
ML: I am surprised because I thought you were going to say that the way to do [long-term storage] is by using some form of molecule, some kind of a “power to x”.
SC: That is going to be part of it for sure, and hydrogen. Hydrogen will play a role, possibly in certain types of mobility, but also in stationary storage. Hydrogen is great, but it doesn't ship very well, because liquefying it to 20 Kelvin is not a really good option. People are looking at all sorts of things including ammonia, which can be shipped at higher temperatures and a little bit of pressure.
I was very sceptical of hydrogen in my first year or two. I said hydrogen needs four miracles. First fuel cells have to be cheaper. Second, storage: you cannot store 10,000 psi in carbon tanks, that’s not practical for stationary applications. Third, you need a distribution system, right? And fourth, a source of cheap, clean hydrogen. So, I say, okay, you need four miracles, but that's okay, you get to be a saint with three miracles, so hydrogen is going to be part of the solution.
ML: What about nuclear? I believe the existing mega-projects have been tested to economic destruction. Am I wrong? And is there hope in the next generation, small modular reactors or new configurations and fuel types?
SC: I agree with you, the large gigawatt reactors, given all the requirements and what you need to do, and the fact that all the key components, down to key bolts, have to be traceable because of safety concerns. I don't think you're going to be able to whisk away concerns about safety, because public sentiment is what it is.
So how do you do this? Well, the hope is that you produce small modular reactors in a factory, where the quality control is achieved by stamping them out in their 1000s. We stamp out many, many cars, and aeroplanes. There are still safety concerns with cars and aeroplanes, but one can deal with it. It cannot be one-off designs: every time you build a reactor, you have to go back to the same thing.
ML: When you have to recall cars, 150,000 people are inconvenienced. If you have to recall a big chunk of our energy infrastructure, that's going to be pretty disruptive.
SC: You're not going to do a recall. It's been installed. It's a 50MW or 25MW or 100MW thing, so you've got to be able to do it on site. You've got to have spares. Think of our electricity systems, you have these massive substation transformers. Any sane, rational transmission distribution system has just a few standard sizes, and if one goes down, boom, a replacement is in, you're off the grid for a day or two, and then you're back.
ML: And you're optimistic about modular reactors, if they're made in that way, with the maintenance, replacement and spares system that you talked about - you think that they will be able to produce affordable power?
SC: The answer is no, I'm not optimistic. You asked us if they had a chance. If I were going to bet on anything, it would be some combination of chemical plus heat plus pumped storage, all those other things. Plus another factor of two in renewable energy getting less expensive, which is pretty well there - it's baked in. I would bet more on that than nuclear.
ML: Does that mean you would then starve the nuclear of funds and say, you're probably going to lose, so we shouldn't fund it? Or should we spend a few billion. I mean, what's a few billion in the modern economy, right?
SC: I agree with you completely. It's the same reason why we're spending billions on fusion. Now, if you ask me whether I think fusion will be economically viable, the answer is: even less likely.
I'm actually on the advisory committee of one [fusion start-up] and I watch the others. There's the most recent hype about one, which is a much more compact magnetic confinement design. Well it has a chance. But I tell you one thing that's not in the headlines, and that is whether it's going to be a commercial reactor.
ML: There's a group at the NASA Glenn Research Centre that's published a paper, about what I've spent most of my adult life thinking of as pathological science: lattice confined nuclear reaction - cold fusion or LENR. Have you looked at that?
SC: First, I haven't. So, I'm going to just speak from what I know from background physics. You can take an ion, and you can accelerate it, you can crash it into something, and you can create fusion this way. I can easily imagine having a lattice in which at the nanoscale, there could be some rapid last minute acceleration, just a nano scale accelerator that will create fusion and give you fusion signatures. But if it's a standard sort of fusion, deuterium-tritium, you get these neutrons that are just, you know, a real pain. They would destroy your delicate nano lattice acceleration.
If you don't see a pathway to real deployment, why bother? It's like small modular reactors, you know, on the scale of things [spending] a couple of billion a year is nothing. But don't hold your breath.
ML: You continued to do research while you were Secretary of Energy. I remember a paper on energy efficiency and standards.
SC: We looked at refrigerators, washing machines and air conditioning. What was surprising was that every time there was a new standard, capital cost started going down faster. It turned out the engineers took a blank sheet of paper, designed a better compressor, which means a smaller compressor, a more efficient compressor.
We thought this was a great discovery. It was a discovery. We submitted to the Journal of Science and got rejected. Two reviewers said “I don't care what these people say, it is impossible for regulation to make appliances cheaper.” And I said “Oh, so these economists don't believe in data. They believe in religion”.
ML: [Watching recent political events in the US] we were hearing about people who were ill with COVID, telling their doctors and nurses that COVID didn't exist. At the end of the Second World War, the Allies took German citizens into concentration camps to show them, knowing that if they did not do that, the existence of those camps would be denied in the future. Is that the sort of thing we need now ? Or some sort of Truth and Reconciliation Commission?
SC: Well, so what happened in Germany, the vast majority of people realised what they had been doing during Hitler's era. There was a national admission of guilt. You know, the former President of the United States has not admitted he did anything wrong.
ML: A final question. With the societal issues we've just touched on, but also the science and the engineering that we've talked about most of this conversation, are you optimistic or pessimistic about climate change?
SC: I am truly bimodal. Look, we got lithium-ion from seawater, there's going to be plenty of lithium! But we are going to go over 450ppm, we're going to go over 550, we may go over 600. The scientists and engineers will figure out something, it's just going to come a little later and be much more expensive.
There will be a lot of pain, suffering, death, and social disorder due to climate change. If 5 million refugees have made Europe much less stable politically, I don't know what 50 million or 500 million will do. But I'm pretty sure that's the scale of the climate refugees will be looking at in the coming half century.
In the end, you know, will we prevail? I hope so. But not without a lot of unnecessary suffering.