For immediate release May 27, 2023

NNSA's effort to restart plutonium warhead "pit" production will cost more than the Manhattan Project. Why is NNSA trying to build two pit factories at once -- one that is adequate, and one that is not?

Contact: Greg Mello, 505-265-1200 office, 505-577-8563 cell

Permalink * Prior press releases

Albuquerque -- The National Nuclear Security Administration (NNSA), the semi-autonomous agency within the Department of Energy (DOE) which designs and builds U.S. nuclear warheads, is attempting to restart production of plutonium warhead cores ("pits").

There is no objective near-term need for new pits for any existing U.S. nuclear warhead or weapon system (NNSA: early pit production a "hedge," not strictly necessary; is there a "pit gap?", May 3, 2022).

Yet NNSA has a statutory requirement to produce at least one high-quality ("war reserve," WR) pit this year, then at least 30 WR pits by 2026, and then at least 80 WR pits by 2030 with at least 80 WR pits per year (ppy) thereafter.

These deadlines were set somewhat arbitrarily by Congress, on the basis of representations from NNSA as to what was possible. ("NNSA, Pentagon ‘sorting out’ pit needs; new pits are ‘a hedge,’ NNSA administrator says," Exchange Monitor, Apr 29, 2022; "NNSA: early pit production a 'hedge,' not strictly necessary; is there a 'pit gap?'" LASG, May 3, 2022).

None of these deadlines will be met, but this will not affect the current nuclear stockpile. NNSA anticipates making its first WR pit in the second half of 2024 (remarks by Marvin Adams, Feb. 14, 2023, audio at 16:20ff). From statements in NNSA's FY24 congressional budget request (FY24 CBR), and from public remarks by NNSA Administrator Jill Hruby (Feb. 14, 2023, audio 54:28 to 56:19) and others, it now appears that LANL will not be able to start making 30 WR ppy until some time after 2030 and possibly not until 2036 , while SRS will not be able to make at least 50 WR ppy until at least 2036 (FY 24 CBR, pp. 210-211) and possibly not until 2039 (FY 24 CBR pp. 236-241).

These newly-expected delays are greater than those which became public in December 2022 (pp. 2156-2157) and early February ("Installation of "Base" Capability to Produce 30 Plutonium Warhead Cores ("Pits") at Los Alamos To Be Delayed 4 Years, to 2030," Feb 3, 2023). (See "Schedule for Nuclear Warhead Core ("Pit") Production Slipping, Costs Increasing: NNSA's Strategy is Failing," Mar 22, 2023).

NNSA's pit program is far and away the most expensive program in the agency's history, dwarfing all others. Yet at this point, five years into implementation of a strategy chosen in the previous administration to build not just one but -- uniquely in this program -- two factories, NNSA still does not have a detailed schedule or cost estimate for this endeavor ("NNSA Does Not Have a Comprehensive Schedule or Cost Estimate for Pit Production Capability," GAO-23-104661, Jan. 12, 2023; "GAO: NNSA's Huge Program to Build New Warhead Cores ("Pits") Lacks Detailed Schedule, Budget, and Scope of Work, press release (2 of 2), Jan 12, 2023).

In its recent report, the Government Accountability Office (GAO) used the best available information they could get as of some cut-off date in the fall of 2022. Since then, as noted, NNSA's FY 24 CBR has provided updated cost and schedule information. Using that information it has become clear ("Pit Production Startup Costs By Site and Year," May 11, 2023) that the agency's two-site strategy to reconstitute pit production can be expected to cost more, in constant dollars, than did the entire Manhattan Project.

In 2023 dollars, the entire Manhattan Project cost $31.3 billion through the end of 1945 (Atomic Audit, Steve Schwartz et. al., p. 60). Total start-up costs to reach at least 80 WR ppy in the 2036-2039 period (NNSA's current estimated schedule) now lie in the $47-$54 billion range, of which $10 billion has already been appropriated. Considering only future costs, the minimum costs through "reliable" startup at each site -- at Los Alamos National Laboratory (LANL), at least 30 WR ppy by 2033; at the Savannah River Site (SRS), at least 50 WR ppy by 2036 -- are, we estimate, $16 billion (at LANL), and $14 billion (at SRS). Additional costs will be incurred at other NNSA sites, which we expect to be in the range of an additional $1.5 billion by 2036.

In other words, the minimum future cost of starting up pit production at two sites is comparable to the entire cost of the Manhattan Project. These costs do not include the costs incurred once 80 WR ppy production is underway.

The new cost range roughly doubles GAO's estimate of just 4 months ago. This is occurring partly because NNSA's expected capital costs have increased and partly because of now-expected further dilation of startup schedules at each of the two main sites.

Details and references can be found in our most recent white paper on this subject, "Toward a viable plutonium pit production plan: part 1," May 18, 2023, the executive summary of which is copied below. Other recommended background includes "Los Alamos warhead "pit" production preparations begin dangerous 24/7 work in struggle to meet deadlines; STRATCOM: "unlimited money" would not be enough to meet 2030 deadline," Mar 10, 2022 and "NNSA pit production strategy: no clear goals, plans, or likelihood of success; Production at LANL has high risks and costs, few or no program benefits," Oct 1, 2020), making due allowance for out-of-date cost and schedule data. See this comprehensive web page for more.

The single best federal study of how to restart pit production remains NNSA's Pit Production Analysis of Alternatives (AoA), Oct 2017 (Executive Summary) and its subsequent briefing slides.

Is NNSA serious about finishing one adequate pit facility? We are nuclear "doves" (or "disarmament hawks"), but this is not a hawk or dove "thing." We have to partially agree with nuclear hawks who are questioning NNSA's strategy ("Sluggish plutonium-pit production tells allies and adversaries U.S. nuke modernization ‘not important,” expert says," Exchange Monitor, May 26, 2023). As our recent white paper explains, NNSA decided "out of the blue" to build two factories instead of one, contradicting decades of DOE and NNSA studies (see the first four studies in slide 9). No NNSA study supports this strategy, except insofar as the Savannah River Plutonium Processing Facility (SRPPF) may in the future make all required pits.

In our judgment, NNSA made a mistake in its abrupt volte-face of May 2018 from which it has never recovered.

For a number of reasons we do not believe LANL will succeed in the new pit mission assigned to it. In its 2017 AoA, NNSA formally ruled against using LANL's old plutonium facility for permanent pit production, for serious reasons. In 2019, independent reviewers strongly urged NNSA to avoid double production shifts in LANL's old, crowded facilities, again for serious reasons. As Mark Davis, Assoc. LANL Director for Weapons Production and Chief Operating Officer put it (Feb. 14, 2023 audio recording at 22:29), "At LANL, the best analogy I can come up with is that we are overhauling and upgrading a plane during flight with a load of passengers on board."

Davis did not mention that LANL has arguably failed at this mission four times already.

The "plane," he might have added, is overloaded with existing missions and is based in a main plutonium facility that is over 50 years old, with supporting facilities spread across topographically-dissected mesas made of weak rock with high seismic risk, at the end of what LANL Director Thom Mason has called "the world's longest cul-de-sac." The inadequate local labor force has limited ways to even get to the facility. As NNSA apparently discovered sometime after its fateful decision -- see the white paper referenced above -- LANL's small plutonium facility must be run on a 24/7 basis to reach even 20 ppy, let alone a "reliable" 30 ppy, requiring twice as many staff and considerable additional infrastructure.

Study Group Director Greg Mello:

"NNSA deserves credit for coming forth with these new cost and schedule estimates, the implications of which are profound. We ask ourselves: is this a cry for help?

"At a scale involving tens of billions of dollars, this is not 'just another cost overrun.' These fairly-screaming indicators are telling Congress and NNSA that they must fundamentally reexamine what they are doing in the pit program.

"We have a number of questions:

• Given that the LANL pit mission does not include the infrastructure that would enable enduring pit production, even if reliable pit production can be achieved, or pit production to DOE safety standards -- is NNSA spending $16-$19 billion extra to build and operate a small, temporary or provisional pit factory at LANL? Or is there more major LANL infrastructure being planned that NNSA hasn't mentioned yet, including one or more new plutonium facilities? Are we seeing just the infrastructure "down payment" on this folly?
  

• In its FY24 CBR, NNSA says its pit plans are being delayed by a lack of skilled engineers and gloveboxes at both main production sites. The two sites compete for scarce experienced designers and equipment. Could NNSA build a new, adequate pit facility faster without also building a second one at LANL? Could NNSA be more successful in this program by spending about half as much money?
  

• In our judgment, the LANL industrial pit mission has been driven by politics alone, and it impedes acquisition of an enduring 80 WR ppy production capacity. Will New Mexico greed -- the real origin of the two-site plan -- end up derailing NNSA's pit program?
  
  
• Is NNSA unconsciously "double-billing" taxpayers for pits, because of an "expedited delivery" promise from LANL that has already slipped by 7-9 years (from 2026 to 2033-2035)?

• What exactly is the "deterrent value" of the more expensive, smaller, more fragile, non-enduring LANL project? Is there any?
  

• Why are there TWO pit factories and only ONE factory for everything else important? Would making two factories for everything make all these other missions more resilient in any necessary or valuable way?
  
  
• Why is NNSA still trying to "permanentize" production in LANL's old, small facilities, now that NNSA is beginning to see the problems predicted 6 years ago?
  
  
• As NNSA admits, there is no actual need to produce pits at the present time. There are even enough of the relatively "safe" W87-0 warheads to populate not just the initial Sentinel deployments as planned but the entire Sentinel fleet. What, then, is really behind this crash program?
  

• Is the potential capacity of SRPPF limiting in some way? We think not. As far as we can tell, that complex will be adequate, timely enough, brand new, safer, and in a far more sustainable situation than LANL.
  

• Is NNSA pursuing two simultaneous pit factories at this gigantic combined cost because it perceives funding as no longer "cost-constrained" (p. 3)?

Here is the executive summary of our recent white paper, "Toward a viable plutonium pit production plan: part 1," May 18, 2023:

• Despite the experience and integrity of present National Nuclear Security Administration (NNSA) management, NNSA’s unquestioning, contractor-like “mission orientation,” combined with what it perceives to be “unconstrained” funding, leads the agency to cling to the pit production plan it inherited from the previous administration, a plan we believe is failing.

• Practical engineering realities and other realities on the ground, as well as past choices, strongly constrain NNSA’s pit program. All “good government” policies will incorporate these constraints. We believe the considerations offered here lead to the best pit production policy choices realistically available, independent of political view.

• Congress may not yet understand the scale and rapidly-expanding cost and schedule of NNSA’s pit production program. Under the current strategy, the effort to restart pit production is likely to cost at least as much as the entire Manhattan Project. More than half of this cost has been and will be at LANL, which has consistently failed to fulfill pit expectations and commitments for the past quarter-century.

• The full scope of the LANL pit production effort is undefined. Assuming reliable production is even possible at LANL, enduring production would require additional nuclear facilities, and also other facilities. Some of these may not be feasible at the site.

• In addition to uncertainties inherent in the early phases of design, NNSA cannot produce a fully-resourced Integrated Master Schedule (IMS) or Life Cycle Cost (LCC) for pit production without knowing the scope of LANL’s pit program.

• The dramatic delays and cost increases predicted in the FY24 Congressional Budget Request (CBR) threaten success at both main production sites. The split-production, “double moon-shot” approach was always a very risky long shot. However, for NNSA’s nuclear physics labs especially as well as nuclear weapons contractors more generally, and for interested politicians, payoffs are certain and substantial.

• In 2017 NNSA rejected: a) split production, with its very high costs, impacts, and its inattention to economies of scale; and b) trying to use LANL’s old PF-4 building as an enduring pit facility, which it will never be. In 2019, the Institute for Defense Analyses (IDA) warned that 24/7 work at PF-4 would be “very high risk.” No federal study supports the present approach at LANL, which threatens to sink the entire program.

• Production milestones resolve into four levels: technology demonstration and training; “base” production; “reliable” production, and “enduring” production. How much time and money could be saved by focusing LANL solely on demonstration and training, while NNSA focuses on building an enduring, adequate, reliable pit facility?

• LANL is still about a decade away from having a pit production capability, let alone a safe and reliable one. Emergency construction of a temporary pit facility at LANL sets up expensive, delay-inducing competition for scarce engineers, gloveboxes, and other resources. NNSA is needlessly “competing with itself.”

• It is not clear why repurposing a portion of LANL’s aging, inadequate facilities, with poor safety margins and little surge capacity or flexibility, would add resilience to pit production. Further legacy safety and reliability problems remain to be discovered. We question whether “reliable” or even “base” production can be established at LANL.

• It is not clear why the nation should support two pit factories when the Savannah River Plutonium Processing Facility (SRPPF) alone, which requires less funding to produce more pits, with greater flexibility, would suffice. We doubt the nation will do so.

• For reasons beyond its immediate control, NNSA cannot say how many pits it can produce by any given date. This uncertainty will persist indefinitely. Statutory pit requirements should be eliminated in favor of annual executive branch direction with annual congressional review, authorization, and funding.

• Pit production in the 2030s is not necessary for full deployment of any current or planned nuclear weapon system, or for the retention of about 80% of current “hedge” warheads and bombs (i.e. all but the W78 warhead as it is retired).

• If pit production is to be possible by the late 2030s, planning and construction of an adequate, reliable, enduring production facility must continue now. The operation of that facility can be adjusted then.  Whether to have such a facility must be decided now.

• Initially, Sentinel ballistic missiles are to be deployed with W87-0 warheads, which besides using insensitive high explosive (IHE) are also well-studied, accurate, and numerically sufficient for the entire Sentinel fleet, absent using multiple independent reentry vehicles (MIRVing). Why precisely is the W87-1 warhead needed now?

• A sufficient number of new pits were supposedly going to be available to produce W87-1 warheads in the early and mid-2030s. Now, new pits are unlikely to be available in sufficient numbers to support the current W87-1 production schedule, which therefore appears to be unrealistic, as well as early-to-need.

• It appears that the baseline SRPPF design, with a single production shift, can achieve a production rate of at least 80 War Reserve (WR) pits per year (ppy). Beyond this, the potential production capacity of SRPPF is not publicly known. Besides producing W87-1 pits in the 2030s, is there a rationale for LANL production?

• If the cost of producing pits is included, the cost of the W87-1 Life Extension Program (LEP) is several times higher than NNSA currently projects. The capability being established by the current program of record at LANL is not enduring, so why should its cost not be included, given that the W87-1 LEP is its only pre-SRPPF “customer”?

• Halting preparations for industrial pit production at LANL would save roughly $11-$21 billion (B), about as much as SRPPF construction will cost. Such a policy would almost certainly decrease the time necessary to create an enduring 80 ppy capacity, while also lowering SRPPF costs and risks.

• NNSA and Congress should therefore halt the Los Alamos Plutonium Pit Production Project (LAP4) and related infrastructure expansion, while creating a detailed “off-ramp” from LANL’s pit production ambitions. Hiring for a second shift should stop.

• Nearly all of NNSA’s work, including assessments of pit production policy, is done by contractors, placing government in a weak position. To strengthen the hand of “Madisonian” government (including NNSA itself), to illuminate policy choices, and to fulfill its oversight responsibilities, Congress should require NNSA to clarify these and related issues in detail, with as much public transparency as possible.

***ENDS***