Got Science? Podcast | Episode 66 HBO’s Chernobyl Sparks Questions About US Nuclear Power Safety

August 27, 2019

Physicist Ed Lyman discusses new safety threats to US nuclear reactors and why risks here are different than in Russia.

In this episode Ed talks about:

  • Nuclear power regulations and safety
  • What it takes for something to go wrong at a nuclear power plant
  • His visit to Chernobyl

Timing and cues:

  • Opener (0:00-0:56)
  • Intro (0:56-2:45)
  • Interview part 1 (2:45-12:57)
  • Break (12:57-14:05)
  • Interview part 2 (14:05-23:44)
  • Science FTW Throw (23:44-23:48)
  • Science FTW (23:48-26:28)
  • Outro (26:28-27:30)

Related content:

Full Transcript

Colleen: Ed, thanks for joining me on the podcast.

Ed: Thanks for inviting me. Great to be here again.

Colleen: So, Ed, did you watch the HBO miniseries "Chernobyl?"

Ed: I only managed to see two out of the five episodes, unfortunately, but I've heard a lot about it.

Colleen: I'm sure you've read a lot about it, there's a lot of conversation going on. What did you take away from it?

Ed: Overall, the big picture historical events were accurate, but I feel like some of the details were hyped up for dramatic effect. And possibly, the focus was shifted a little bit from where I think it needed to be. Overall, I'm certainly glad to see that it got people talking about nuclear safety. And I hope that the lessons of Chernobyl can be appreciated today, even in the different context here in the United States.

Colleen: Is there an easy way for you to explain what happened?

Ed: The accident was initiated by a safety test. The safety test was going to be run in a regime where the reactor was unstable, but the reactor operators didn't really appreciate that. So when they started the test, they didn't fully understand the conditions in the reactor or the instabilities of the reactor. And as a result, they took actions which caused the reactor to be more unstable and eventually to experience a runaway chain reaction that is essentially overheat so rapidly that it caused a massive steam explosion, blew the hole in the roof and dispersed radioactivity over a wide area. The ensuing graphite fire burned for many days continuing to release radioactivity until it was finally contained. So, that's the accident in a nutshell.

Colleen: Given the publicity of this mini-series I was really surprised to see here just in the news in the past couple of weeks the nuclear industry pushing for less oversight of nuclear power plants. And I'm just curious what's up with that?

Ed: Nuclear power plants in the country today are under great financial pressure, mostly due to the low cost of fossil fuels and their inability to compete. So, the owners of the reactors are looking for any way possible to cut their operating costs. And one expense that the operators see is due to the oversight of the Nuclear Regulatory Commission. The NRC conducts inspections that not only requires staff time at the reactors to prepare for those inspections, but it also could result in discovery of violations, which have to be fixed and that means spending money.

So if there are fewer inspections, if the inspections don't look as hard, they may miss problems. And the plant owners may have longer to address them because the regulators didn't catch them.

Colleen: What are some examples of types of oversight that they want to ease up on?

Ed: One concrete example is a type of inspection that the NRC conducts every two years. And this is a special kind of inspection where the NRC is actually inspecting how reactor owners catch problems at their own plants. So it's a very important inspection because years ago, the NRC delegated more responsibility to plant owners for programs where they would look for problems, and catch them, and correct them on an ongoing basis. Because NRC inspectors can't look everywhere all the time.

And so it's much better if the reactor owners are in a position to do their own reviews and inspections, but those have to be done properly. They have to be doing the right things, they have to be looking at the right frequency. And so as a result, to make sure they're doing the right thing, the NRC has to watch them. But one of the proposals now is to reduce the NRC's oversight of those self inspections.

So if you allow reactor owners to have more responsibility for catching problems as they arise, you have to make sure that they're doing the right thing. So it's very important to make sure those inspections of those programs are done at the right frequency. But the NRC has decided just to do them every three years instead of every two. And it's not clear that that's being based on any real issue that needs to be addressed.

And that's really the problem with the changes that are being proposed now. It's not clear that they're actually solving any problems. There's no real rationale for doing them except to reduce oversight of the industry. And in that particular case, there were substantial objections from some NRC staff about reducing the frequency of these inspections without first assessing what the impacts could be. In other words, doing a comprehensive analysis of what those inspections do, and how frequently do you really need to do them to make sure they're effective. That study has not been completed yet, yet the staff is going ahead and recommending that they reduce the inspection frequency anyway.

Colleen: Are there currently nuclear power plants that you are concerned about?

Ed: I'd say that every plant, you know, is unique and has its own concerns. Certainly, some make me worry more than others. For instance, the Indian Point nuclear plant in New York State, it's only 25 miles from the boundaries of New York City where I grew up. That plant should not have been located where it is because the number of people within 50 miles, last time I checked is over 16 million, is really too great.

If you're gonna have nuclear power, you should make sure that there's a sufficient region around every plant that's low population density. So that if evacuation or other emergency measures are needed, they can be carried out effectively. And by simply suburbanization and development, a lot of plants around the country that were originally sited in rural areas now find themselves in suburbs and the population's increasing.

And Indian point’s the poster child for that. It is shutting down in the next few years. But certainly, the potential impact of Indian Point, both from a safety and a security perspective has always been a concern. Then there are plants that are vulnerable to seismic events, that are vulnerable to flooding. And again, it's really highly dependent on the location of the plant and how it was designed in the first place. But I would say every plant has its own risks and they have to be considered in their own context.

Colleen: One of our listeners actually wrote in with a couple of questions. Let me run these by you. Given that our nuclear power plants are coming upwards of 40 years, what do current regulations say about when plants should be decommissioned?

Ed: Well, the current regulations allow for 40 year operating licenses for nuclear plants with the potential for a license renewal. Most plants have applied for and received 20-year license renewals. But that again is not an indication of the maximum lifetime of a plant. There is no regulation that specifies it. So some plants, I think there are six reactors now, have applied for what they call a subsequent license renewal that would be from 60 years to 80 years.

Colleen: Just keep changing that oil and, you know...

Ed: That's right.

Colleen: ...fixing the rust and it'll be fine.

Ed: Little facetious, but fundamentally, the idea is the same. The idea is that it doesn't really matter too much how old the plant is, as long as you can inspect and maintain those systems, structures, and components that are aging so that they stay within an acceptable range. Now there are certain things that can't be changed. For instance, the concrete and steel containment buildings around most plants, it's not something that's going to be replaced.

There's buried piping in a lot of plants, this piping was never intended to be replaced, but some of it is corroding. So, there may be an issue with how do you manage those structures that can't be replaced. And finally, the reactor vessels, these are the steel vessels that hold the nuclear fuel in reactors, they become brittled over time as they're bombarded with neutrons. And there is a risk that they could shatter like glass if they are sufficiently brittle and they undergo rapid cooling.

So that's one of the...what is called a time limiting aspects of a nuclear plant because those reactor vessels would be way too expensive to ever replace. So the NRC, essentially says those components that can be monitored and maintained, and replaced, if necessary, will be put under aging management programs. Those components that can't be replaced have to be assessed for how bad things will get by the end of the license period. So it's a long answer but basically, the NRC has no set lifetime for plants. But it is considering the first extended license renewal to 80 years.

[Break]

Colleen: So another question that came in was... He says, "We've read that some U.S. nuclear missile defense systems are still running software off of floppy disks, how up to date is the automation and training at U.S. nuclear power facilities?"

Ed: Well, here's the counter-intuitive thing. Most nuclear plants have analog instrumentation control systems for the most critical safety-related components. That's because the plants predate the digital era. And if you're going to upgrade those systems to digital, it introduces a wide range of issues that are hard to resolve. The most obvious is cybersecurity. So if you have a plant that has analog instrumentation controls, it's a lot harder for an off-site hacker to be able to affect those safety-related systems.

However, if you upgrade to digital instrumentation control, then you're potentially opening up a new avenue for cyber attack. So you have to proceed with caution in upgrading those to digital instrumentation control.

There's also an issue about really figuring out how those systems work in very complex scenarios because nuclear plants have a lot of interrelated safety systems. And being able to disentangle those and have a clear path between how safety systems work and how the digital instrumentation controls would interact with them turns out to be a difficult task. So it may sound like an obvious thing, those plants should be upgraded to digital, but that does introduce complications that need to be fully assessed.

Colleen: Ed, is it true that the next generation of nuclear power plants will be so safe that they can't meltdown?

Ed: It is not true. Any nuclear plant has vulnerabilities that could result in a serious accident or could be exploited. It is true that you can design greater safety into nuclear power, there are ways to reduce that risk. But by and large, you're always going to have these vulnerabilities and you can't depend on the design to save the day. It's always going to be a good design plus a well-run plant, plus well-trained operators, plus robust inspections and maintenance, and also robust security to prevent against sabotage attacks.

Colleen: How far-fetched is the idea that terrorists could attack a nuclear power plant? What would they be trying to do or to get?

Ed: For a commercial nuclear power plant in this country, the greatest concern is radiological sabotage. And that is a deliberate act that could destroy or disable enough of the safety systems and the backup safety systems that the reactor would meltdown and there would be very little that the plant operators could do about that. And it's a very real threat.

Because if there were a well-trained, paramilitary type terrorist attack at a nuclear reactor, without a robust security response, the attackers could essentially destroy enough equipment to cause a meltdown within minutes. So there is a very short time window for trying to respond if you have this type of event. The best thing to do is to prevent the attack from taking place.

So in order to verify that, the Nuclear Regulatory Commission runs what it calls force on force inspections where it actually leads a team including a mock adversary force, to nuclear power plants. Every plant is tested every three years. And they run scenarios to test the security force of the plant, see if they can actually effectively defend the plant against these attackers. That's a program that's been required since after the 9/11 attacks. However, it's undergone a lot of changes over the last few years.

The industry hates this type of exercise, especially if they lose. So they've been pressuring the NRC to make a lot of changes to the program that would make it harder for them to lose, and give them fewer chances to lose. So there used to be three different scenarios run at each plant during these inspections, that's been reduced now to one.

Colleen: If a disaster like Chernobyl happened here, how do you think that nuclear plant operators and the NRC would have dealt with it?

Ed: Well, to be clear, the U.S. does not have power plants of the Chernobyl type design, the RBMK. So the particular sequence of events that led to Chernobyl is not really plausible at the light-water reactors in the United States. However, there are other types of events that could result in similar consequences. And we saw that at Fukushima Daiichi, the accident in Japan in March 2011, where three nuclear reactors of a design very similar to designs that are deployed here in the United States, experienced meltdowns and large releases of radioactivity.

So it's a fallacy to think that Chernobyl was an event that was only due to Soviet incompetence and corruption and that that kind of thing couldn't happen here. Chernobyl couldn't happen here, but Fukushima could or something worse than Fukushima. So that possibility has to be addressed and protected against. And after Fukushima, the Nuclear Regulatory Commission did put into place requirements for plant operators to be able to cope with an event like Fukushima and hopefully to intervene effectively before the worst case where the reactor actually melts down.

But those plans require testing, they require vigilance. They require equipment that's well maintained and will be available. And they require thinking through all the what-ifs. You can't expect that you're going to anticipate everything that's going to happen, you have to be able to respond to events as they occur. And that's really the challenge is having assurance that you're going to be able to handle whatever comes.

Colleen: Would you take a trip to Chernobyl now?

Ed: Not only would I but I have.

Colleen: Really?

Ed: I was there in 2006. It was the 20th anniversary of the accident. I was in Ukraine for commemorations of the accident. They were being held in Kiev. And there was an organized tour of Chernobyl, so I did go. That was more than 10 years ago. So the dose rate was somewhat higher than it is today. We were all equipped with our own dosimeters because we didn't trust the Ukrainian Radiation Protection Authority. And so I kept an eye on the dose rates.

And if you've ever turned on a dosimeter when you're 30,000 feet in an aircraft, the dose rates are fairly alarming. And the dose rates generally in the Chernobyl exclusion zone were comparable to those you'd experience at 30,000 feet in an aircraft. So it's a little bit shocking to see those kinds of numbers when you're standing at sea level.

I did see...one of the funniest things, if you can say it's funny, was in the exclusion zone, we had stopped in a village within the exclusion zone, ruined, abandoned village with buildings, missing windows, and roofs, and doors, and totally overgrown. And so we're kind of just wandering around there like a ghost town. And then someone walks by walking a dog. There are people living there.

Colleen: That's fascinating. I know there's a whole big tourist industry around it and people...

Ed: Right, and I think maybe they need to give more attention to radiation protection for the visitors there. Like I said, when we went in, there was virtually nothing except we had to pass through whole-body counters at the perimeter. But again, I'm not sure I trusted those machines, I did my own surveys when I got home and didn't seem to find any contamination.

Colleen: Well, Ed, thanks for taking time to sit down and chat with me about "Chernobyl," the mini-series.

Colleen: Thank you. It was great.

Return to the top >

Credits: 

This Week in Science History: Katy Love
Editing: Omari Spears
Music: Brian Middleton
Research and writing: Jiayu Liang and Pamela Worth
Executive producer: Rich Hayes
Host: Colleen MacDonald