"No Hurry to Recycle," Feature Article, May 2006

no hurry
to recycle

Some advocates believe no new nuclear power plants will be ordered until the question of long-term storage of waste has been addressed. But one proposed answer—fuel reprocessing—is too costly and complicated to embrace.

by Frank N. von Hippel

this spring the U.S. Department of Energy has been promoting a Global Nuclear Energy Partnership, a strategy for dealing with the accumulation of radioactive waste from power plants by reprocessing some of the spent fuel.

The primary domestic benefit of this initiative would be to reduce the quantity of plutonium and other transuranic waste that would have to be buried in Yucca Mountain, the Nevada site identified as the national depository for nuclear waste. Material would be separated and used to fuel a new generation of fast-neutron reactors. For the next two decades, the program would focus on the demonstration of "improved" spent-fuel reprocessing and fast-neutron reactor technologies.

This is not a new proposal. It has been proposed twice before and abandoned because of its enormous cost and adverse impact on U.S. nonproliferation efforts.

Before adopting the Global Nuclear Energy Partnership, then, it would be wise to ask a couple of questions: How much value is there in additional research and development in the field of waste reprocessing? And what are the proliferation implications of the United States abandoning its anti-reprocessing policy?

During the 1960s, nuclear energy policy was premised on two assumptions: The U.S. would be building 100 power reactors a year by the year 2000, and that, because of the scarcity of high-grade uranium ore, nuclear power based on light-water reactors could not be sustained beyond the turn of the century. The limitation of LWRs is that they exploit efficiently only the energy in the chain-reacting isotope U-235, which makes up 0.7 percent of natural uranium.

Another type of reactor had been invented on paper during World War II, one that could exploit the energy locked in U-238, which makes up the remaining 99.3 percent of natural uranium. The "breeder reactor" would convert the U-238 into chain-reacting plutonium. Most of the energy R&D investments of the leading industrial nations therefore went into the development of plutonium breeder reactors.

Situated in a sparsely settled Nevada desert, Yucca Mountain (below) could easily hold all the spent fuel produced so far by U.S. reactors.

During the 1950s, as part of its Atoms for Peace program, the U.S. declassified the PUREX reprocessing technology that it had developed to recover plutonium for weapons from irradiated uranium and encouraged other countries to master the technology as a building block for their own breeder-reactor R&D programs. One of the countries to which the U.S. transferred reprocessing technology was India. It was a shock, therefore, when India used the first plutonium that it separated to make the nuclear explosive that it tested in 1974. India's explanation that it was interested in peaceful uses for nuclear explosives did not receive much credence.

The Ford administration immediately reversed U.S. policy on the export of reprocessing technology and exerted maximum pressure on France, which was about to export reprocessing technology to Pakistan and South Korea; and on Germany, which had included reprocessing technology in a nuclear package deal with Brazil. Ultimately, none of these exports was consummated, and Japan is today the only non-weapons state that reprocesses spent fuel.

When Jimmy Carter became president in 1977, he launched a review of U.S. policy toward domestic reprocessing. On the basis of this review, he concluded that deploying breeder reactors would not be economical for many decades. (A 2003 study at the Massachusetts Institute of Technology supported this viewpoint. The researchers found that, even assuming a tripling of nuclear capacity, reprocessing won't be economically viable before 2050.) The Carter administration decided to discourage reprocessing domestically as well as in foreign non-weapons states.

That policy reversal was only fully accepted after the Reagan administration found that U.S. nuclear utilities were no longer interested in reprocessing because of its high costs. In response, Congress passed the Nuclear Waste Policy Act of 1982, under which the Department of Energy would, in exchange for a tax of 0.1 cent per kilowatt-hour of nuclear-generated electricity, take responsibility for disposing of the spent fuel in a geological repository.

In 1987, Congress selected Yucca Mountain to be the site of the nation's first spent-fuel repository. The Department of Energy proceeded to try to license the repository, but has suffered a series of reverses—most recently in 2004, when a federal court required that the repository must be designed to limit radiation doses to the public beyond the next 10,000 years.

Yucca Mountain Delayed

The Yucca Mountain repository was to be open by 1998, but it's unclear now when the site will begin accepting waste, if ever. As a result, nuclear power plant operators are buying on-site dry-cask storage for their accumulating spent fuel and many utilities are suing the federal government to recover their costs.

Technically, the fuel is safe at the reactor sites. As a study published last year by the American Physical Society, the professional society of American physicists, pointed out, "Even though Yucca Mountain may be delayed considerably, interim storage of spent fuel in dry casks, either at current reactor sites, or at a few regional facilities, or at a single national facility, is safe and affordable for a period of at least 50 years."

Politically, however, the situation is becoming hot. Rep. Dave Hobson of Ohio and Sen. Pete Domenici of New Mexico, the congressmen who chair the appropriations subcommittees that oversee the Department of Energy's budget, have become convinced that until it's clear that the spent fuel will not remain indefinitely on the reactor sites, no new nuclear power plants will be ordered. In his subcommittee's May 2005 report on its proposals for the Department of Energy's budget for fiscal year 2006, Hobson instructed the DOE to "begin accepting spent commercial fuel from the nuclear utilities and placing it in centralized interim storage at one or more DOE sites." The DOE was directed to select an advanced reprocessing technology and prepare to implement a spent fuel recycling plan by fiscal year 2007.

Six months later, in November 2005, when Hobson and Domenici issued their House-Senate conference report, however, the idea of interim storage had disappeared. Perhaps, the appropriators had realized that it would be bitterly opposed by states concerned that interim storage would turn into permanent storage if Yucca Mountain were not licensed.

These sites (not precisely shown) hold spent fuel and radioactive waste that need long-term storage.

The reprocessing initiative remained, however. Although reprocessing involves interim storage of high-level wastes and separated plutonium and other transuranic elements at the reprocessing site, it is made more attractive by thousands of jobs and the construction of multibillion-dollar facilities. Communities might well compete to host a reprocessing facility with the same zeal that they would fight to avoid becoming the home of an interim storage site.

The conference report laid out an explicit timeline: submit a detailed program plan by March 31, 2006; kick off a site selection competition by the end of June; select a site in fiscal year 2007; and initiate construction of the first waste recycling plant by 2010.

The DOE met the first target when it rolled out its proposed program plan, labeled a "Global Nuclear Energy Partnership," in February. The plan contains no near-term rationale for beginning to ship spent fuel off reactor sites, however. Instead, it is very heavy on long-term vision and research and development. The vision is built around fast-neutron reactors—basically, the same reactors that were being developed in the 1970s as plutonium-breeder reactors. In the current plan, however, they would be stripped of the plutonium-breeding uranium "blankets" surrounding their fissile-material consuming cores. They therefore would become "burner" reactors for fissioning the plutonium and the minor transuranic elements, neptunium, americium, and curium, recovered from spent fuel from light water reactors.

If such reactors were developed and built in sufficient numbers, reprocessing of U.S. spent fuel could be launched some decades hence and the recovered transuranics fabricated into fuel for their use. The transuranics remaining in the fuel discharged from these reactors would be recovered again, more transuranics from light-water-reactor fuel added, and the mix recycled again and again.

The objective of Global Nuclear Energy Partnership would be to fission all of the transuranics, aside from process losses. That would avoid sending most of the transuranics to Yucca Mountain and would, according to the DOE plan, make it possible to "defer the need for additional geologic nuclear waste repositories until the next century," even if there is a huge growth in U.S. nuclear generating capacity. In this scenario, if the nuclear industry and its supporters in the federal government can force through the licensing of Yucca Mountain, they will not have to face a repeat of this bruising political battle in their lifetimes.

Hot Rocks

Physically, the capacity of Yucca Mountain to store radioactive waste is limited by the temperature rise of the rock caused by the heat output of the spent fuel. Most of this heat would come from the decay of the transuranic elements. If these elements were fissioned, the mountain could take fission products from five times as much spent fuel before the temperature became a problem. If, as supporters of reprocessing advocate, two fission products with a 30-year halflife, cesium-137 and strontium-90, are separated and stored on the surface, the remaining fission products from perhaps 100 times as much spent fuel could be stored in the mountain.

This proposal, it turns out, also is not new. The Department of Energy's Argonne and Los Alamos National Laboratories had brought it forward in the early 1990s. Argonne was interested at that time—as it is today—in a mission for the fast-neutron reactors it had been developing for decades as its core mission. Los Alamos was interested in applications for high-current proton accelerators that it had developed for its weapons R&D program. The first Bush administration asked the National Academy of Sciences to carry out an assessment of their proposals.

The academy issued a massive report in 1996. Its conclusions were so discouraging that the Department of Energy dropped the subject for a decade. Today, DOE spokesmen for Global Nuclear Energy Partnership refuse to discuss the NAS study.

The main findings are indeed jarring. The academy found that it would take many decades or even centuries to significantly reduce the net amount of transuranic waste. Moreover, the reduction in the public exposure to radioactivity (compared to underground storage of unreprocessed spent fuel) would be too small to justify the cost, estimated to range from $50 billion to more than $100 billion for the disposal of some 62,000 tons of LWR spent fuel. And widespread implementation of reprocessing systems could increase the risk of nuclear weapons proliferation.

The conclusions of greatest concern today are those relating to cost and proliferation.

The NAS study scaled its cost estimate to 62,000 tons of spent fuel because that is approximately the amount of spent fuel that the Nuclear Waste Policy Act allows to be placed in Yucca Mountain before a second repository in another state is in operation. The current generation of U.S. power reactors will have discharged 62,000 tons of fuel by 2008 and DOE expects that these reactors will ultimately discharge approximately twice as much. This would still be significantly less than the estimated physical capacity of Yucca Mountain, however.

Surface storage of spent fuel in dry casks (top) is much simpler than reprocessing, which involves large industrial complexes, such as La Hague in France (middle) and Savannah River in South Carolina (above).

The cost estimate of $50 billion to $100-plus billion (in 1996 dollars) would therefore have to be approximately doubled to cope with the spent fuel from the current generation of U.S. power reactors. This would be approximately $1 billion to $2 billion for each of the 100 or so U.S. reactors—not a trivial amount. Two hundred billion dollars would be enough to build another 100 reactors. It would greatly exceed the approximately $40 billion that the federal government will raise with its tenth of a cent per kilowatt-hour tax on nuclear-generated electricity.

What about the impact of GNEP on U.S. nonproliferation policy? Since India's nuclear test in 1974, the U.S. has been very successful in discouraging other countries from reprocessing by telling them, in essence, "We don't reprocess and you don't need to either." The Department of Energy proposes to tell them now, "We need to reprocess, but you won't have to because we'll do it for you and keep your radioactive waste." (That's why the G in GNEP stands for "Global.")

This proposal was rejected with anger when President Bush made it to India. U.S. experience with the opposition in South Carolina in the early 1990s to the repatriation of a few tons of U.S. spent fuel containing weapon-grade uranium also suggests that it would be politically impossible to import tens of thousands of tons of foreign spent fuel. Russia has been able to implement such a policy, but only by suppressing massive domestic opposition.

Compared to reprocessing, interim storage looks quite attractive from economic and safety perspectives. The cost of dry cask storage of spent fuel is about one-tenth the cost that the National Academy of Sciences estimated for the GNEP. Moreover, the lethal gamma radiation field around spent fuel, because of the cesium-137 that it contains, would protect the plutonium in it for more than a century.

Given the fierce opposition to off-site interim storage of spent fuel, the path of least political resistance is likely to be continued interim storage at the current reactor sites. This seems reasonable at least as long as the reactors are operating. The risk from older spent fuel in dry casks is negligible in comparison to the risks from fuel while it is in a reactor core or in a spent-fuel storage pool.

This need not be the death knell for nuclear power in the United States. Congress has streamlined the reactor licensing process. The Energy Policy Act of 2005 extended through 2025 the Price-Anderson Act, which limits the liability of utilities from a catastrophic Chernobyl-type accident. The Energy Policy Act also provides $1.5 billion in insurance coverage against licensing delays for the first six new nuclear power plants ordered in the U.S. Finally, the availability of on-site interim storage removes the legal impediment caused by the delay in the licensing of Yucca Mountain.

In February, the trade journal Nuclear Fuel summarized the conclusions of Steve Kraft, director of used fuel management at the nuclear industry's Nuclear Energy Institute, as "NRC [Nuclear Regulatory Commission] is confident a repository will be operating by the end of 2025 and that spent fuel can be safely stored on the reactor sites in casks for at least 100 years. That confidence allows NRC to license new reactors and to renew the licenses of existing ones."

In this context, the huge cost of the Global Nuclear Energy Partnership would likely be more of a burden than a help to the future of nuclear power in the U.S.

Frank N. von Hippel is a nuclear physicist and a professor of public and international affairs at Princeton University. He served as assistant director for national security in the White House Office of Science and Technology Policy during the Clinton administration. Von Hippel is also a founding co-director of Princeton's program on science and global security and co-chair of the International Panel on Fissile Materials.