Episode 3 of our new podcast, Xconomy Voices, features a conversation about nuclear power with Nathan Myhrvold, the founder and CEO of Intellectual Ventures.
The former Microsoft chief technology officer is now vice chairman of TerraPower, a Bellevue, WA-based spinout of Intellectual Ventures that aims to revive commercial nuclear energy. The company is researching next-generation reactor designs, including the traveling wave reactor and the molten chloride reactor, that would be cheaper to build and operate than previous generations of reactors—as well as safer.
Myhrvold believes nuclear power has to be part of the answer to the challenge of greenhouse-gas-induced global warming, even if the U.S. government isn’t showing much leadership in clean energy R&D.
“The oil age will end when we invent something better, which we will do, we just haven’t yet,” Myhrvold says. “And the challenge with energy is that the current infrastructure is so cheap, and there’s no leadership from the government, or very little anyway, to help support nascent technologies before they’re economic. So we’re going to have to keep doing that ourselves.”
Here is the full transcript of our interview. (Only about half of this material made it into the podcast.)
Xconomy: Nathan Myhrvold, thank you so much for joining me today.
Nathan Myhrvold: It’s my pleasure.
Xconomy: So, what are you working on that has you excited these days?
NM: Well, I just finished my bread book, which is a 2,500-page book all about bread. I just opened an art gallery for my food pictures. And our nuclear company is doing really well.
Xconomy: OK. We could go in any number of directions, but I think nuclear power is incredibly interesting and incredibly important. So can you talk a little bit about that. You’re talking about TerraPower, right?
NM: Yes. So if we really want to make meaningful cuts in carbon dioxide in the atmosphere, we need a carbon-free source of baseload power. Baseload means stuff that runs 7-by-24. And today baseload power that is renewable doesn’t really exist. Hydroelectric is great, but hydroelectric does actually have some global warming consequences, at least for the first few hundred years after you build the dam. So new hydro actually is not a great thing from a global warming perspective. Old hydro is great.
Most baseload power is still fossil-fuel-based, so it’s coal. I think that nuclear is a terrific technology for this, but we have a problem. The problem is, in the 1970s the United States and a bunch of other places kind of freaked out about nuclear power. I say freaked out because although there are some possible dangers—there are some real dangers—the perception of the danger was wildly out of whack with the actual fact. A great example is that a couple of years ago there was a disaster in Fukushima. 15,000 people were killed by a tsunami. To date, zero people died because of the reactor [radiation]. But it’s portrayed as a horrible nuclear accident, and viewed by some as a reason we shouldn’t do nuclear.
Xconomy: All right. But to be fair, there’s a giant exclusion zone around Fukushima, and Chernobyl for that matter. And both of those incidents are expected to add measurably to the cancer load. So it’s not as if there was no impact. But you’re just talking about the visible sort of deaths and injuries?
NM: Well, it turns out that the cancer load from Chernobyl that was predicted hasn’t actually transpired. That’s largely because we don’t know what the impact of low radiation dose is. People have made worst-case predictions, and the worst case has not turned out to be true for Chernobyl.
The exclusion zone around Fukushima has higher radiation than it did originally, but it has lower radiation average than the state of Colorado or the country of France. And actually the reason is pretty simple. Mountains are radioactive, because rock is. And if you have the Rocky Mountains in the middle of your state, if you have the Alps go in your country, you’ve got more radioactivity than Fukushima does.
Now I don’t mean to minimize those things, but because no plants were built United States for the last 30 years, no one tried to use modern technology to sort of rethink what a nuclear plant could be. And that’s exactly what we’ve done in a company called TerraPower.
So TerraPower has rethought nuclear energy. It has some amazing designs that are much, much safer than the designs of the past. And they also have the property that really can scale to make a credible dent in our energy needs. In fact, you could actually run 80 percent of America’s energy needs—I pick 80 percent because that’s the percentage of energy that France gets from nuclear, so it’s certainly possible—we could run 80 percent of America for 125 years just based on nuclear waste we already have, using that waste as fuel. I think that’s pretty cool.
Xconomy: Is that one of the key innovations of TerraPower, building plants that can use spent nuclear fuel as their fuel?
NM: Yes. One of our key innovations is that we use the uranium very differently, and spent fuel that you couldn’t use in a conventional reactor looks like fresh fuel to us.
Xconomy: OK. What other kinds of innovations are embodied in TerraPower’s design? Because I think when people think of nuclear power, they probably are thinking of the post-World War II generation of light water reactors like the ones we built at Three Mile Island, and practically every other commercial reactor. They’re complex and they require a lot of redundant safety systems. And that’s part of the reason they fail once in a while. It’s the classic Charles Perrow argument about tight coupling. Is there a way to build an inherently safe nuclear reactor, and how hard are the people at TerraPower thinking about that particular problem?
NM: Well, that’s what we’ve done at TerraPower, is to make a much safer reactor. Here’s a good example. At Fukushima the type of reactor that they built really needs to have the power running constantly, and it gets into a bad way if power is cut off for as little as 15 minutes. Really stupid design. It was designed in the slide rule era. They should have replaced it a long time ago. They didn’t.
A state of the art, still under-construction reactor like the Westinghouse AP 1000 can go three days without power. If you have a disaster, you have three days to go back in and hook power back up so you can cool things down. And the TerraPower reactors—you can just turn the power off and walk out the door and nothing bad will happen. The reason for that is we designed it so that it will use natural convection of the atmosphere to cool down what’s called the afterheat.
The thing that really kills you in a reactor is that even after you turn it off there is something called afterheat that goes for a while and it’s that afterheat that can get you in trouble if you don’t have cooling pumps doing active cooling. I’m pretty sure we’re still going to have an atmosphere. And if we have an accident where we no longer have an atmosphere, nuclear power is the least of our worries. But that’s a good example of something where, because of a different design, you have very different margins for safety. So you don’t have to worry at all about having emergency power and duplicate, triplicate kinds of systems to ensure it doesn’t fail. It’s kind of fail-safe to begin with.
Xconomy: What stage is TerraPower at, in terms of the development of this design?
NM: Well, we are designing the main reactor. We’re negotiating with
