Why a decision on a second US plutonium-pit-production factory should be delayed

By Frank von Hippel | June 12, 2020

The troubled Mixed-oxide Fuel Fabrication Facility project at the Savannah River Site is proposed to be transformed into a plutonium pit production facility. Photo (c) Timothy Mousseau, 2019.The troubled Mixed-oxide Fuel Fabrication Facility project at the Savannah River Site is proposed to be transformed into a plutonium pit production facility. Photo (c) Timothy Mousseau, 2019.

The National Nuclear Security Administration (NNSA), the organization within the Energy Department that is responsible for producing and maintaining US nuclear warheads, is moving forward with a plan to build a plutonium-pit-production factory at DOE’s Savannah River Site in South Carolina. “Pits” are the form of the plutonium in the fission trigger “primaries” of US two-stage nuclear warheads.

The primary motivation for this move is lack of confidence in the pit-production capacity at Los Alamos National Laboratory, which has been responsible for preserving US pit production expertise since production at the Rocky Flats Plant outside of Denver shut down at the end of the Cold War. There are also political motivations, including filling the jobs gap at the Savannah River Site resulting from the collapse of NNSA’s effort to build a Mixed-oxide Fuel Fabrication Facility there to process some of its excess Cold War plutonium pits into power reactor fuel.

NNSA’s rush forward may result in a debacle on top of a debacle. If the experts at Los Alamos can’t manage pit production there, why does NNSA think that they can design and train the staff to operate a pit-production facility at the Savannah River Site?

Also, the United States need for pits is unclear at the moment. In 2007, the pits produced at Rocky Flats—now 30 to 40 years old—were pronounced to be good for at least a century and, in 2012, the Lawrence Livermore National Laboratory upped the durability estimate to 150 years. NNSA did not support the necessary research to solidify this conclusion, however—an oversight that it now promises to remedy.

The NNSA also claims that it needs to produce new pits for two types of safer primaries for two new nuclear warheads, but there seem to be enough already-existing pits for one of the warheads, and the design for the second has not yet been decided.

Thus, there are multiple arguments for delaying a decision on the proposed second pit-production facility for a decade or so. By then, Los Alamos should have mastered the production of pits, the longevity of the legacy pits will be better established, and the need for pits not available in the legacy stockpile should be clarified.

Pits, the critical components. Each US nuclear warhead contains a miniature advanced version of the Nagasaki nuclear bomb weighing only about two percent of what the original Nagasaki bomb weighed. This “primary” is built around a hollow shell of plutonium, which is surrounded by chemical explosive.

If that explosive is triggered, the pit will be imploded rapidly into a spherical solid mass compressed to perhaps twice the normal density of plutonium. Near the point of maximum density, before the plutonium begins to bounce back to its normal density, a small neutron generator will spray it with a burst of neutrons that will initiate exponentially multiplying fission chain reactions. Within a microsecond, about 20 grams of the plutonium will fission, releasing energy equivalent to the explosion of about 300 tons of TNT, heating the material in the primary to about a million degrees Centigrade.

At that temperature, fusion reactions will occur in the several grams of tritium and deuterium that were injected into the hollow pit just before the implosion. Those fusion reactions will produce an intense burst of neutrons that fission hundreds more grams of plutonium, “boosting” the energy of the fission explosion to the equivalent of about 10,000 tons of TNT, half the power of the Nagasaki bomb.

At that point, the primary will be so hot that its glow will be mostly X-rays that will fill the “radiation case” surrounding the primary and a secondary nuclear explosive nested next to it. The X-rays will vaporize the outer layer of the “secondary,” imploding and heating it and igniting a mix of fusion and fission reactions that, depending on the warhead, will release from a few times to 25 times the energy of the Nagasaki bomb from a warhead about one-twentieth the weight of the 4.4 ton Nagasaki bomb.

Will aging pits still work? The last nuclear test of a US primary was conducted in 1992, when Congress imposed a moratorium on US nuclear testing, launching negotiations on the Comprehensive Test Ban Treaty, which the United States signed in 1996. The treaty has not come into force because eight countries, including the United States, have not ratified it; nevertheless, it appears to have moral if not legal force. No country other than North Korea has tested a nuclear weapon since 1998, and North Korea stopped in 2017.

Despite the lack of US testing, there is no doubt that the primaries in US nuclear warheads will implode or that, if the primary works, the secondary will explode. The issues that have been raised relate to whether the plutonium might become brittle and fragment as it implodes, and whether, if that happened, the tritium-deuterium boost gas would ignite. The NNSA spends billions of dollars each year investigating this question with ever-more-refined computer simulations of what happens inside a pit during its implosion and tests of the behavior of aging plutonium under shock, including in subcritical tests in tunnels deep under the former Nevada Test Site.

Every year, the three NNSA weapon labs—Livermore, Los Alamos, and Sandia—and the US Strategic Command, which would deliver most US warheads in a nuclear war, go through an elaborate review process of the condition of the warheads and of the simulations, including by “red teams” of skeptics, before they certify to Congress, via the secretaries of Defense and Energy and then the president, that they are confident the warheads would work and that no tests are required.

During the Cold War nuclear arms race, pits and other weapon components were replaced regularly as one generation of nuclear weapons succeeded another. That evolution stopped with the end of nuclear testing. Today, most of the US warheads have undergone or are going through a “life-extension” process, with most components being replaced. The pits themselves are not yet being replaced, however. The United States has not been making new pits in significant numbers since 1989 when the FBI and Environmental Protection Agency raided the Energy Department’s pit production plant at Rocky Flats outside Denver for environmental crimes. The plant was subsequently razed. (See former Bulletin editor Len Ackland’s 1999 book, Making a Real Killing: Rocky Flats and the Nuclear West.

That the US is not able to replace the pits has become a perennial point of anxiety for the US nuclear-weapon establishment and its congressional overseers. Proposals have been made, but no new pit production facilities have been built—in part because the pits made by Rocky Flats have proven remarkably durable. They are sealed and plated, and they have suffered virtually no corrosion. The main question is whether the emission of “alpha particles” (helium nuclei) by the slow decay of the plutonium is changing the mechanical properties of the material.

In 2005, Congress directed the NNSA’s administrator to commission an independent review of the efforts at the Los Alamos and Lawrence Livermore laboratories to estimate pit lifetimes. The review was carried out by JASON, a group of experts, mostly academics, who do summer studies on issues of interest to the Defense and Energy departments. An unclassified summary of the group’s findings was released in early 2007.

Los Alamos and Livermore had been analyzing the effects of plutonium aging on the functionality of US pits. They also had been doing accelerated-aging experiments on samples of the plutonium alloy used in US pits by spiking them with plutonium-238, which decays by alpha emission with a half-life of 88 years, versus the 24,000 years for the dominant isotope in weapon-grade plutonium, Pu-239.

The summary conclusion of the 2007 JASON report was:

We judge that the Los Alamos/Livermore assessment provides a scientifically valid framework for evaluating pit lifetimes. The assessment demonstrates that there is no degradation in performance of primaries of stockpile systems due to plutonium aging that would be cause for near-term concern regarding their safety and reliability. Most primary types have credible minimum lifetimes in excess of 100 years as regards aging of plutonium; those with assessed minimum lifetimes of 100 years or less have clear mitigation paths that are proposed and/or being implemented.

Another reason to cancel the Sentinel missile: the rising cost of its nuclear warhead

The JASON report also recommended additional research (pp. 17-18):

to gain experience with Pu that has suffered the equivalent of a century or more of aging (i.e., with accelerated aging), thereby allowing an interpolation rather than an extrapolation in estimating performance changes and degradation due to aging. In particular, one wants to know the modes of failure that will be among the first to appear, because these can inform the stockpile surveillance program in order to make it most sensitive to aging-induced degradation [and] ongoing study of the current accelerated-aging Pu samples, which are spiked with the rapidly-decaying 238Pu, as well as production of samples that have been aged by alternative means. In all of these cases, the objective is to get the equivalent of multi-century experience on aging phenomena, associated with decay (e.g., radiation damage) as well as with activated processes such as annealing.”

Some work on accelerated aging did continue and, in 2012, the Livermore lab reported, “no unexpected aging issues are appearing in plutonium that has been accelerated to an equivalent of ~ 150 years of age.”

Livermore’s deputy program leader for enhanced surveillance of pit aging stated, “In the near term, the nation can save tens of billions of dollars that might be required to build a new production facility.”

Pit-production problems at Los Alamos. In 1999, to maintain US pit-production expertise, the Energy Department instructed Los Alamos to establish pit production capacity of up to 20 pits per year within its large PF-4 plutonium facility. Surprisingly, however, despite the history of Rocky Flats having routinely produced 1,000 to 2,000 pits per year, Los Alamos has struggled to produce a small number of pits, even though the lab has spent billions of dollars on pit-production efforts.

In 1996, the Energy Department tasked Los Alamos to produce 31 “war reserve” pits for the W88, the high-yield warhead for US ballistic missile submarines, for an order that had not been completed because of the shutdown of the Rocky Flats plant. It took 16 years—until 2012—to fabricate the pits. Eleven were produced in 2007 but then a declining number annually thereafter.

The plan was to transition to producing pits for additional W87 warheads for the US Minuteman III intercontinental ballistic missile. In 2013, however, pit production at Los Alamos was shut down because of inadequate worker safety training and concerns about potential plutonium criticality accidents. Seven years later, pit production is still shut down at Los Alamos and the expectation is that this situation will continue until 2023.

NNSA’s budget submission for fiscal year 2021 states that Los Alamos is engaged in “activities to hire, train, qualify, and retain required pit production personnel, recapitalization of equipment needed to restore Plutonium Facility (PF)-4’s ability to produce War Reserve (WR) [pits,] towards producing the first WR pit during 2023 [and] manage capital acquisitions to increase production capability of PF-4 to produce 10 pits per year.”

The cost of the planned upgrades to PF-4 is estimated at $1.75 billion through fiscal 2025.

Eighty pits per year? In 2008, the departments of Defense and Energy decided that the United States needed a pit production capacity of 50 to 80 pits per year. This range was based on the very rough computation that it would take 30 to 90 years to replace the pits in a US stockpile of 2,500 to 4,500 warheads at that rate. It was assumed that the production could be carried out at Los Alamos.

But Los Alamos continued to flounder. In 2014, Congress backed the goal of 50 to 80 pits per year with a “sense of Congress” statement (which does not have the force of law) backing a requirement of a production capacity of 30 pits per year by 2026 and a demonstration over 90 days during the following year of a production rate of 80 pits per year. In 2019, the date of that goal slipped to 2030, but, in the Defense Authorization Act of 2020, Congress added the message that “any further delay to achieving a plutonium sustainment capability to support the planned stockpile life extension programs will result in an unacceptable capability gap to our deterrent posture.”

The Trump administration’s 2018 Nuclear Posture Review – which also does not have the force of law – turned these various assertions into a requirement for an “enduring capability and capacity to produce plutonium pits at a rate of no fewer than 80 pits per year by 2030.”

Two pit production facilities? After an initial expenditure of $7 billion on construction of a Mixed Oxide Fuel Fabrication Facility at the Savannah River Site to fabricate excess Cold War plutonium into reactor fuel, the cost continued to grow: It was estimated that the cost to completion of that NNSA project had climbed to $30 billion—for a program that Congress had originally been told would cost $2 billion. Congress voted to shut the project down and, in early 2018, the Trump Administration agreed, raising the question of what to do with the fortress structure that had been built and how to compensate the South Carolina delegation—especially its powerful leader, Sen. Lindsey Graham—for the loss of jobs the cancellation of the MOX plant entailed.

In this context, the idea was born to divide the pit-production mission between Los Alamos and Savannah River.

In 2018, the Defense Department and the NNSA issued a joint statement asserting that there would be two pit-production facilities: one at Los Alamos producing “at least 30 pits per year” and one at Savannah River producing “at least 50 pits per year.” In this way, the 2008 goal of demonstrating a production capability of 50 to 80 pits per year at Los Alamos had 10 years later become a minimum combined production rate of 80 pits per year at Los Alamos and Savannah River.

Updating the pit durability estimate. Meanwhile, the question remained as to what a decade of additional research at Los Alamos and Livermore had revealed about the lifetimes of the legacy pits.

In March 2018, the Senate Appropriations Committee, in its report on the Energy and Water Appropriations Act for fiscal 2019, directed the NNSA administrator to contract with JASON to do an update on its 2007 report on expectations for the longevity of the Rocky Flats legacy pits. The instruction was that JASON should be contracted to “assess the efforts of the NNSA to understand plutonium aging and the lifetime of plutonium pits in nuclear weapons [and] include recommendations of the study for improving the knowledge, understanding, and application of the fundamental and applied sciences related to the study of plutonium aging and pit lifetimes, an estimate of minimum and likely lifetimes for pits in current warheads, and the feasibility of reusing pits in modified nuclear weapons. The report shall be submitted in unclassified form but may include a classified annex.”

The Senate instructed the NNSA administrator to “make available all information that is necessary to successfully complete a meaningful study on a timely basis.”

A year and a half later, in November 2019, after a near-death experience at the hands of the Trump administration, JASON submitted a three-page letter report informing Congress that it could not update its previous estimate because, “in general, studies on Pu aging and its impacts on the performance of nuclear-weapon primaries have not been sufficiently prioritized over the past decade. A focused program of experiments, theory, and simulations is required to determine the timescales over which Pu aging may lead to an unacceptable degradation of primary performance.”

The JASON letter also suggested that, contrary to Congress’s instruction, NNSA had not cooperated adequately with the review: “The labs briefly presented their program to address Pu aging to JASON. The plan seemed sensible, but a detailed JASON assessment would require additional information about the program as well as technical details.”

Texas wildfires force major nuclear weapons facility to briefly pause operations

Laudably, NNSA was embarrassed and, in April 2020, administrator Lisa E. Gordon-Hagerty informed the chairman of the Senate Armed Services Committee Subcommittee on Strategic Forces that her agency planned to fund a second phase of the JASON study during the summer of 2020 to “[a]ssess the need for the full study, and if deemed necessary and timely, perform a more detailed, multi-year JASON study.”

The letter also stated that “NNSA has launched an enhanced program focused on understanding the potential effects of plutonium radioactive decay, or aging, on pit performance.”

Other needs for new pits? In addition to the potential need to replace pits because they are aging, NNSA also is advocating the new production facility to produce pits “with enhanced safety features to meet NNSA and DoD requirements.”

This quest goes back 30 years, to the launch of the Stockpile Stewardship Program by the Clinton Administration. At the time, the weapon labs proposed to replace the W78 ICBM warhead and the W76 and W88 submarine-launched ballistic missile (SLBM) warheads with new warheads containing “insensitive” high explosive. That proposal has been sustained over the decades since through a number of incarnations, including proposals for warheads that would be “interoperable” between the ICBMs and SLBMs.

The purpose of insensitive high explosive is not to reduce the probability of an accidental nuclear explosion. Other elements of the safety design are supposed to do that, and, to date, no warhead accident has resulted in a nuclear yield.

The benefit from the use of insensitive high explosive would be to reduce the number of accidents in which the chemical explosive around a pit is detonated and plutonium is dispersed. There were many such accidents involving aircraft-carried warheads prior to the decision not to fly nuclear-armed aircraft in peacetime. The most famous was the collision of a nuclear-armed B-52 strategic bomber with its refueling tanker over Spain in 1966, which resulted in a large area of plutonium contamination on the ground, requiring 1,600 US military personnel to be deployed for up to 12 weeks, working with minimal protection, gathering contaminated dirt and crops into barrels for shipment back to the US for burial on the Savannah River Site. The Navy has had no such accidents with its SLBM warheads, however, and believes that reducing the risk significantly would require redesigning its Trident missile as well as their warheads. It therefore has in the past not been willing to invest in adapting new insensitive high explosive warheads to its missiles, a process that would include flight tests.

It appears, however, that the Navy has finally acquiesced or been overruled on this matter, and the plan is to replace its two SLBM warheads, the W76 and W88, with new warheads that use insensitive high explosive.

The current proposal is to build two new warheads. The first is the W87-1, which would replace the W-78 on the “Ground-Based Strategic Deterrent,” the successor to the Minuteman III missile, and potentially also the W-88, the high-yield warhead on the Trident II submarine-launched ballistic missile. A second warhead, sometimes referred to as the W93, would replace the W76.

The pit of the W87-1 would be identical to the pit of the W87-0, which is currently deployed on the Minuteman III, and is to be used on the Ground-based Strategic Deterrent. The 400 Minuteman IIIs are to be replaced one-for-one with the new ICBM, which, like the Minuteman III, is to be deployed with only a single warhead per missile.

The Defense Department reportedly has 540 W87-0s in stock, of which 200 are deployed on the Minuteman III along with 200 W78s. Therefore, the W78s could be replaced with stored W87s. The department wishes, however, to preserve the option of loading more warheads onto the new Ground-Based Strategic Deterrent, in case of a breakdown in nuclear arms control with Russia. This has been called a “warhead upload hedge” since the Clinton administration. To load up 400 of the new missiles with three warheads each would require 1200 warheads, which would require more W87-1s and therefore more pits.

No realistic circumstance that would require uploading the US ICBMs has been suggested, however. The Joint Chiefs reportedly informed President Obama that they could cover all essential targets in potential adversary nations with one third fewer warheads than the 1,550 counted warheads that are allowed by New START. Further, Minuteman IIIs were downloaded to one warhead each after the end of the Cold War to make the deterrent relationship with Russia more stable. After the downloading, destroying one US warhead in a first strike would, on average, require more than one Russian warhead.

Beyond those arguments against uploading, many respected defense experts, including former Defense Secretary William Perry, argue that the US should abandon fixed land-based ICBMs altogether, because Strategic Command insists on keeping them in a dangerous launch-on-warning posture.

Little firm information has been made public about the design of the proposed W93 warhead for submarine-launched nuclear missiles. In fact, NNSA’s Fiscal Year 2020 Stockpile Stewardship and Management Plan describes “the Next Navy Warhead,” as “not yet an established program of record.

An anonymous “senior defense official” has offered the reassurance, however, that the W93 would be “previously nuclear-tested designs, it’s not going to require any nuclear testing.” This must mean that a previously tested insensitive high explosive primary would be used.

In 1990, in hearings before the Senate Appropriations Committee’s Subcommittee on Energy and Water, the Energy Department’s then Deputy Assistant Secretary of Energy for Military Applications listed all US nuclear warheads with insensitive high explosive, including warheads that had been produced and deployed and some that had been tested but not deployed as a result of the end of the Cold War. The candidate warhead that has been discussed for three decades is the W89, one of the warheads that was tested but not produced. The W89 was to use recycled pits. According to one report, the pits were to be from the W68, a previous-generation SLBM warhead for which thousands of pits were produced during the 1970s. If that option were pursued, and the W68 pits were found to be still functional, no new pit production would be required.

The decision on the second pit production facility can wait. NNSA could announce its decision to move forward on building a pit-production facility in South Carolina as early as September. Based on the above context, this decision should be delayed for a number of reasons:

1. Since the Savannah River Site staff has no experience with pit production, the facility would have to be designed and the staff trained by the Los Alamos group. But the Los Alamos group has not yet demonstrated that that it can design and staff its own pit production facility.
2. Within a decade, we should have a new lower limit on the functional lives of the legacy pits. If they will indeed last for at least 150 years, as the Livermore experts concluded, then there will be no need for a large production facility to replace them anytime soon. The Los Alamos facility, if it can be made operational, should be sufficient for some decades.
3. The argument for producing additional warheads with insensitive high explosive for the Minuteman III replacement is very weak, and the debate over the need to produce new pits for a warhead to replace the W-76, the most numerous warhead in the US operational stock (about 1,500) cannot be made until NNSA and Defense Department are ready to discuss what pit they would use in the W93.

We can wait for another decade before we decide on whether the United States requires two pit production facilities. Indeed, we can wait for another decade before we decide on whether we need any new pits at all.

Together, we make the world safer.

The Bulletin elevates expert voices above the noise. But as an independent nonprofit organization, our operations depend on the support of readers like you. Help us continue to deliver quality journalism that holds leaders accountable. Your support of our work at any level is important. In return, we promise our coverage will be understandable, influential, vigilant, solution-oriented, and fair-minded. Together we can make a difference.

Get alerts about this thread
Notify of
1 Comment
Newest Most Voted
Inline Feedbacks
View all comments
brian Whit
brian Whit
4 years ago

Let’s stop listening to military minds to direct national policy. We have ENOUGH warheads, some of which are sitting idle, only to be strapped to the correct rocket, to blow up the Earth hundreds of times over. We don’t need more, even as we are spending over a trillion to upgrade, and AFTER we have made what we have 3x more lethal. The biggest threat of nuclear war comes after 3 things: 1.Huge imbalance of power relationships between adversaries 2. Misunderstanding of one or both combatants of actual lethal result in limited nuclear war, first strike nuclear war, or all… Read more »


Receive Email