Nuclear Waste Storage: An Intermediate Option

Robert Bryce, a successful author on energy policy, recently wrote an op-ed in Politico encouraging the U.S. to store nuclear waste on government land.

His thesis is that, given the post-Fukushima Daiichi danger of storing spent nuclear fuel (SNF) on the grounds of reactors, the federal government should move it to regional collection centers on federal land, which is what people in the nuclear industry have been saying for awhile. This gets rid of the problems of moving the waste long-distance to Yucca Mountain (in the middle of the desert), which is a bad idea anyway, and would save the federal government billions of dollars in lawsuits.

I don't disagree with Bryce, but I want to clarify a couple of key points. First of all, the meltdown at Fukushima Daiici was exacerbated by spent nuclear fuel storage, yes. But there are two kinds of fuel storage. After being removed from the reactor core, nuclear fuel rods typically spend around five years in a pool of water, called the spent fuel pool, cooling off. After they're cool enough to handle, they're packed into giant casks and kept on the grounds of the plant from whence they came. Bryce's plan would fix the problem with the casks, which definitely needs fixing, but the pools are what went wrong at Fukushima and they're non-negotiable. There's not another practical way to cool down the waste, and there's not really another place to put it for the five years it needs to cool off. In this sense, his plan would lessen, but not remove, the danger of having waste on the grounds of each reactor.

Secondly, regional waste collection centers reduce the dangers of transporting waste by reducing the distance each cask has to travel, but they do not eliminate it. Any plan to relocate the waste from its current scattered state (at all 104 currently operating reactors, plus several other sites) has to take that into account. The casks are tested against falls, fires and floods, but they are not invulnerable (particularly to periods of extended heat; a truck crashing and catching on fire in a tunnel, for example) and should not be treated as such in the planning process.

Finally, waste collection centers are a first step, not a longest-term solution. The next step should be the construction of reprocessing plants to turn SNF into mixed-oxide fuels, which can be fed back into nuclear reactors and used to generate power. Because of the low price of uranium, there is little financial incentive to do this right now, but a reprocessing plant is the only known way to get rid of nuclear waste permanently. They will be expensive and hard to fund while the price of uranium remains low, but if nuclear power is still a part of the U.S.'s energy generation when the price rises, we will definitely need reprocessing technology. The time to make a start on that is now.

What We Should Do About Yucca Mountain

The last time I checked, there was no funding for the continued development, study of and construction of nuclear waste storage facilities at the much-debated Yucca Mountain in President Obama's fiscal year 2011 budget. The Department of Energy has chosen to try and shut down the site altogether, but as of this writing, it's being blocked in court by lawsuits from the states who would probably be next in line to have the nuclear waste repository in their territory if Yucca falls through.

The idea of a permanent nuclear waste repository is not only a bad one, it is something that humanity is fundamentally unable to build. Instead of trying to determine whether Yucca Mountain will be able to safely store nuclear waste for the next million years, we should be putting the time and money earmarked for Yucca into research into spent nuclear fuel (SNF) reprocessing technologies.

Here's my train of thought on the subject:

1. The U.S. government has spent north of $9 billion, as well as over three decades' worth of scientific reports and research, on studying Yucca Mountain and trying to determine whether it will be safe to store nuclear waste there for up to a million years in the future. Yucca was originally selected from a list of the nine most promising sites in the continental U.S., using the criteria of long-term tectonic stability, distance from major population centers, and political will. We have spent an amazing amount of time and money to learn as much as we can about Yucca Mountain, and we have found that it is probably unsuitable for the requirements we've asked of it (for example, the mountain is supposed to be dry and solid, but geologists' boreholes found it to be full of water and honeycombed with cracks). After $9 billion and 33 years of study on the best candidate for long-term storage, if this place isn't suitable for a repository, it's reasonable to ask if anywhere is.

2. Nowhere is. It is a crazy thing to think that we can build a facility that will last for a million years. The hubris of it is staggering.


We simply don't know how to build for that length of time. Most consumer goods aren't supposed to last more than five years. Buildings, levees, bridges and so forth rust, rot and decay within a hundred or so years. Even dams and seawalls will eventually break down if untended in, say, the 235 years that the U.S. has been a country.

Let me say this plainly: the idea that we can design something that will still exist and perform its original function 1,000,000 years in the future--4,255 times the lifespan to date of the nation that created it--is sheer folly. We just don't know how to compensate for a million years of erosion.

How exactly do you design for the Colorado River?

3. That's the thing about Yucca. It's not that it's unsuitable for so-called short-term storage (say, a mere fifty or a hundred or a thousand years). The plan for Yucca Mountain has always been what I like to call a fire-and-forget storage facility. We throw the waste in here and then we never have to worry about it ever again. It's the political and scientific equivalent of lifting up a carpet and sweeping the mess underneath. Oh, you've invested trillions of dollars in a form of electric power that produces an unavoidable and dangerously lethal by-product? Never mind. We can just dump it in this mountain and forget about it.

Also, be sure not to think about the fact that existing plants have produced enough waste since they came online in the 1950s and '60s to fill up the not-yet-built repository by 2014. The politicians who came up with the idea for the repository seem to be thinking in the longest term possible, but ironically, they're not thinking about what happens a mere fifty years in the future. What happens to the waste that's produced after Yucca is full? Do we go through this whole weary mess again, research and design and wrangle over and finally build another repository, or do we simply throw up our hands, cry "Fuck this!" and launch it into the sun?

No, seriously. That's one idea that Congress considered before Yucca was authorized.

4. Fine, a permanent repository isn't the answer. So what should we do?

My answer is twofold. First, instead of one centralized repository for all the waste in the U.S., build several regional storage facilities around the country. This would minimize the distance that the waste travels (since there are very real risks in transporting waste any distance) and, by moving the waste from 104+ separate sites to, say, 10 facilities, it would make the waste that much more secure against theft or terrorism. These facilities don't have to be inside mountains, for their purpose is not to store the waste indefinitely. Rather, they would store the waste until we have the technology to recycle it and turn it into power.

Sound fanciful? We could do it today if the proper facilities were set up. Reprocessed waste can yield up to 25% of the energy of the original fuel, through recycling of unused uranium-235 and plutonium-239 in an ordinary nuclear reactor (not to mention the U-238). This greatly reduces the volume of waste, leaving behind only volatile fission by-products. Even those could potentially be reused with the adoption of a closed nuclear fuel cycle.

So if it's that easy to reprocess waste, why haven't we done it? There are three main reasons. The first is President Carter's policy decision in the 1970s to curtail reprocessing technology in the U.S., in an effort to convince the world to do the same (which did not work). Every succeeding President has followed this policy. They did this because of the second reason; P-239, produced as a by-product in nuclear plants, can be used in nuclear weapons. However, inside the U.S., the risk of this is extremely low.

Reason III is good old economics. Right now, it is much cheaper to mine natural uranium than to reuse SNF, so there has been no real civilian push to build domestic reprocessing plants (although many foreign ones exist). But remember that $9 billion we've already spent on Yucca Mountain? The U.S. government has been willing to spend taxpayers' dollars to try and find a solution for the problem of nuclear waste. Let's start spending that money where it will do the most good: in further research of reprocessing technology and construction of reprocessing facilities. We can remove an environmental and national security threat by reprocessing the waste, we can generate power while we're doing it, and we can help find a permanent solution to the problem of waste. It's a win-win-win for everyone involved.

And the U.S. map can stop being pockmarked with nuclear zits.

Correcting the Post and the Journal: Mixed-Oxide Fuels

While the coverage provided by the Washington Post and other large media outlets has largely been very good at detailing the parts of the nuclear industry that need to be detailed, in the wake of the disaster in Japan, this morning I found a small but important error by omission that I think deserves some correction.

Both the Washington Post and the Wall Street Journal reported in their articles today that the danger of a nuclear accident at the stricken Fukushima Daiichi unit 3 could be heightened by that reactor's use of mixed-oxide fuel (MOX). The Journal went into some detail about how MOX tends to burn hotter than conventional nuclear fuel, so there would be more residual heat in the reactors, which might make it more dangerous to try and cool down.

This is all correct, as far as I know, but both papers omitted something about the making of MOX. The Post and the Journal cited a program called "Megatons to Megawatts", an ongoing effort between countries with large nuclear stockpiles to draw down their stockpiles and convert highly enriched uranum (HEU) to low-enriched uranium (LEU), the latter of which is suitable for commercial nuclear reactors.

Four Problems

1. The Journal cites M2M as part of an international effort to reduce nuclear stockpiles, and says that the MOX in the Fukushima reactor was produced in that program, while the Post is vague about the nature and scope of the program. As far as I know, the M2M program is solely between the U.S. and Russia. It's entirely possible that the fuel was resold to Japan by the U.S., but the articles imply that this is a global effort. As far as I know (again), Russia is the only one of the eight nuclear powers that's having its stockpile reduced in this manner.

2. Also according to the link up above from the United States Enrichment Corporation (which was started by the U.S. government but is now privately owned), the M2M process primarily involves turning HEU into LEU. The UNEC website's main page, FAQ and FAQ on the actual process don't mention plutonium. However, the Journal says that MOX is produced in part by "mixing low-enriched uranium with plutonium that has been recycled from a global stockpile of defunct nuclear weapons", while the Post says that MOX is a "mixture of plutonium oxide and uranium oxide, produced from recycled material from nuclear weapons"*. These do not appear to be supported by the facts of the process, although again, it's quite possible that I'm wrong here.

3. Here's the big omission: MOX can be produced commercially, through the use of a spent nuclear fuel (SNF) reprocessing plant. It doesn't have to come from old weapons. As I wrote in my term paper last semester, upon being removed from the reactor, nuclear waste is made up of around 0.8% U-235, 1% P-238, 93.2% U-238 and 5% other actinides, which aren't readily usable. If the fuel is reprocessed, however, the U-235, P-239 and U-238 can be recycled right back into the reactor. That mixture, now called MOX, will give you around a quarter of the original fuel's energy-not bad for recycled nuclear waste. Reprocessing isn't done in the US for a combination of political and economic reasons, but it is popular in several other countries, including Japan.

4. MOX is more common than one might think, if you consider an informal definition of the fuel. When I met with Nuclear Regulatory Commission (NRC) spokesman Scott Burnell in the course of my research last year, he told me the following:

"In a certain way, every reactor towards the end of a given fuel cycle, when it gets close to having to refuel, in a sense it’s running on MOX. Because plutonium builds up in the fuel during operation. And at the very end, I mean very end, the last few weeks of a given reactor’s operational run, it’s plutonium that in large part is powering the core. But again, it’s doing so in keeping with our regulations. So with MOX you’d simply be starting with plutonium from the get-go.”

 In other words, plutonium is transmuted from U-238 during the normal operation of a nuclear power plant, and builds up in the fuel on its own. It doesn't have to come from old weapons.

That's all I've got. Hopefully, both papers will make these corrections in their future reporting, and otherwise continue their exemplary reporting on the rest of the topic.

*The Post is technically correct in this statement--they could be referring to just the uranium oxide as having come from nuclear weapons--but they're being extremely vague about it, and could just as easily be referring to both the plutonium and the uranium.