Bulletin 267: US plutonium pit production plans advance, part II: at least $25 billion in LANL life cycle costs omitted in past comparisons; more, 8 Feb 2020

Permalink for this bulletin. Please forward!
Home page; Press Releases; Bulletins; Letters; Pit production
To subscribe to our Friends listserve send a blank email here. To unsubscribe send a blank email here.
To subscribe to our Main listserve (less content, less frequent) send a blank email here. To unsubscribe send a blank email here.
Our blog (makeover coming): Remember your Humanity. Twitter: @TrishABQ.
Contribute. Volunteer. Contact us.

Previous Bulletin (1/31/20): US plutonium pit production plans advance, with new requirements (part I)

Bulletin 267: US plutonium pit production plans advance, part II: at least $25 billion in LANL life cycle costs omitted in past comparisons; more

February 8, 2020

Dear friends and colleagues –

This Bulletin continues the previous one. Portions of this Bulletin were published Feb. 4 on the blog of the International Panel on Fissile Materials. Next week, part III will discuss the Administration’s fiscal year 2020 (FY20) proposals.

This Bulletin will be relatively technical and is aimed at specialists. Key takeaways are:

A massive increase in pit production funding will be requested on Feb. 10.
At least $25 billion (B) in life cycle costs for pit production at Los Alamos National Laboratory (LANL) have been omitted in official cost comparisons, giving lawmakers a highly-distorted picture.
The Department of Energy (DOE) misunderstood and lowballed LANL pit production costs 24 years ago in a similar way, resulting in two decades of program failure. Fortunately, there has been – and still is – no need to make pits.
The JASON defense science group has written a strange “letter report” on plutonium warhead core (“pit”) aging. A longer discussion is needed, but pit aging has really nothing to do with the current drive for industrial pit production, which is needed solely for the proposed W87-1 warhead and for no other warhead between now and 2040.

Given 2. (and 3.), together with the enduring difficulties posed by the LANL site and facilities, the only realistic pit production policy available to the National Nuclear Security Administration (NNSA) within the current (wretched) national security consensus that is consistent with engineering realities, is to focus pit production investments on the large, new, empty plutonium processing facility at the Savannah River Site (SRS). This will sooner or later be necessary if new pits are to be built – an unwise path for the United States.

What pit production expertise LANL holds is inexpensively mobile. Attempting to build up LANL’s inherently-fragile production capacity will not add resilience or lasting capacity, will be far more expensive and risky than is currently understood, and is not necessary for training.

This is not a NIMBY position. It’s reality.

1. Massive increase expected in requested pit production funds this coming Monday

The Trump administration is expected to submit its fiscal year 2021 (FY21) budget request for the NNSA to Congress on February 10. Defense News reports that a Republican congressional delegation including twelve senators, has convinced President Trump to request $20 billion (B) for NNSA in FY21, far more than the $17.5 B that was pending and 20% more than the current NNSA appropriation of $16.7 B. It would be a 63% increase (in constant dollars 46%) over the FY17 NNSA appropriation of $12.9 B, the last one crafted in the Obama administration.

Within NNSA, the “Plutonium Sustainment” request – principally, for warhead core (“pit”) production – is likely to be roughly $1 B for FY21. Last year NNSA projected $977 million (M) for FY21 – five times what was projected in 2016 for FY21 and more than four times average spending in the 2000-2018 period, in constant dollars (historical charts). The FY20 request of $712 M was fully funded (e-p. 14).

There will be new details. Appropriators have required (e-p. 86) that the acquisition of new pit production capacity at LANL and SRS be placed in a more accountable “project” structure. Schedules and yearly costs are expected in the forthcoming budget request.

2. At least $25 billion (B) in life cycle costs for pit production at LANL have been omitted in official cost comparisons between pit production alternatives. As a result of political pressure from the New Mexico delegation these extra costs, which have already begun, have been quietly added to the baseline LANL pit program without comparison to cheaper, less-risky alternatives with fewer environmental impacts.

2a. Introduction

(An initial sketch of this issue was sent to key congressional staff on November 25, 2019.)

LANL and NNSA have variously said LANL needs to hire 1,000 to 1,500 additional permanent staff for the 30 pits per year (ppy) pit mission. Dr. Kelly Beierschmitt, Deputy LANL Director: “We expect hiring to continue at that pace [500 net hires per year] for the next five years and with the pit production mission we expect 1,400 additional staff.” According to Kelly Cummins and John Michael, NNSA’s Program Executive Officer for Strategic Materials and Director, Plutonium Program Office respectively, to reach 30 ppy LANL will need to operate PF-4 with multiple production shifts and hire an additional 1,000 to 1,500 plutonium-related staff to do so (11/14/19 briefing for South Carolina Nuclear Advisory Council, especially here). Congressional staff independently confirmed the need for multiple production shifts to reach 30 ppy to the Study Group on 1/8/20.

To make two shifts possible a third maintenance shift is required, which places PF-4, Technical Area (TA) 55 as a whole, and some supporting facilities and services on a round-the-clock schedule. To make such a “surge” posture possible requires in turn requires additional structures for training, parking, entry/egress, showers, meals, and so on, some which were shown to prospective contractors on 8/8/19 (slides 36-37).

None of the operational and capital costs to reach 30 ppy were part of NNSA’s comparisons of pit production alternatives (slide 3, p. 3-6). (NNSA’s 2018 Workforce Assessment was so heavily redacted upon release it is nearly useless.)

First consider operational costs. In its Engineering Assessment (EA), NNSA implicitly estimated that the 50-year discounted operations costs per pit production worker (costs, p. 3-2; number of workers, p. 3-7; divide the former by the latter) for each production alternative would be $14.4 million (M) at SRS (1,807 additional workers said to be required) and between $14.6 and $15.1 M at LANL (833 to 1,156 additional workers said to be required). The operational cost per worker for this mission, according to NNSA, falls into a narrow range across all sites, alternatives, and across a range from 833 to 1,807 workers.

Given the absence of any real LANL or NNSA plan with costs attached, we can use these same operational costs per worker ($14.8 M, midrange) for the “1,000 to 1,500” workers we now know are needed to reach 30 ppy. If we do this, we find that the discounted extra LANL operational costs over the same 50 years as NNSA used would be in the neighborhood of $14.8 to $22.2 B. These are operational costs only, and omit the additional buildings that will be needed to make these employees work possible.

Operational costs strongly dominate NNSA’s EA life cycle comparison (p. 3-18), comprising 87-94% of the total costs for the four pit production alternatives studied. Adding the $15-22 B in discounted costs we have just estimated (which only in the last 6 months have we learned are necessary to produce 30 ppy at LANL), to the cost of the three LANL alternatives in the EA (p. 3-18) makes all three LANL alternatives more expensive than the SRS alternative – dramatically so if we use Dr. Beierschmitt’s figure of “1,400 additional staff.”

But there is more.

In addition to these operational costs, the cost of replacing the 42-year-old PF-4, a very roughly $10 B project assuming it is feasible at all, has been omitted. (We cannot find a LANL location where such a building could be built.) If we add this $10 B to the extra operational costs associated with the additional staff, again leaving out the several new buildings already proposed for them, NNSA’s plan to install an industrial pit production program at LANL is putting taxpayers on the hook for an extra $25-32 B in discounted 50-year costs – beyond the costs reported to and used by Congress in comparing pit production strategies. If we use Dr. Beierschmitt’s “1,400” figure, our best estimate of those extra direct costs is $31 B, again exclusive of non-nuclear construction.

The bottom line is that producing an enduring 30 ppy at LANL is going to be vastly more expensive than Congress as yet understands – assuming it can be done at all, which we doubt.

Given the critical lack of housing in Los Alamos County and the congested roads leading to LANL, there will also be indirect costs associated with these extra workers. These costs would mostly be paid locally and we are not including them here.

2b. A little context

To more fully understand this a little background is needed.

NNSA has conducted two fairly recent detailed analyses of alternative pit production strategies: 1) a thorough October 2017 Analysis of Alternatives (AoA) (full report, Executive Summary, briefing slides), which was produced by NNSA itself, and 2) a subsequent April 2018 Engineering Analysis (EA) written relatively quickly for NNSA by Parsons Engineering (full report, briefing slides).

Although styled as consecutive and complementary reports, the 2018 EA was actually written with completely different assumptions than those of the 2017 AoA. In many ways it was a de novo analysis. The new direction occurred largely because of fierce protests over the AoA findings by New Mexico senators, especially Senator Heinrich.

Pushback started in September 2017 with a letter from Senate Armed Services leadership and is visible in this December 2017 letter from the New Mexico delegation as well as in the press (Santa Fe New Mexican, Albuquerque Journal). The misleading cost comparisons of the EA have been used frequently in the Senate (Heinrich, Armed Services, further answer by NNSA in the same hearing; Udall, Appropriations) and in the House (Cooper, Armed Services) to argue that LANL should be the one and only pit production site in the U.S. In 2018 – and 2019 and so far in 2020 – the New Mexico delegation (here are Udall, Heinrich, Luján, and Lujan Grisham, to take only one example) has been united in pushing to bring the entire pit production mission to New Mexico, based on absolutely zero sound analysis.

Subsequent studies completed by the Institute for Defense Analyses (IDA) for the Department of Defense (DoD) and NNSA in March 2019 – of which only the executive summaries (here and here) are available at this time – noted this new direction. As IDA said (p. iii), the “the options assessed by the EA team differed from the preferred alternatives identified by the AoA.”

In passing we should note these other important IDA conclusions:

Operating LANL’s main plutonium facility PF-4 in a multi-shift “surge” mode would be “very high risk.” That is consultant-speak for “don’t do it.” Yet an amendment sponsored by the New Mexico delegation, which became law in late 2018, required just that: LANL must “implement surge efforts to exceed 30 pits per year” and plan for possible “multiple labor shifts and additional equipment at PF–4” to produce 80 ppy.
“No available option can be expected to provide 80 ppy by 2030.” Yet in the following months, senators Heinrich and Graham successfully made the 80 ppy-by-2030 a legal requirement in the FY2020 National Defense Authorization Act (NDAA) (Section 3116). [lgh says that is not possible]
“Strategies identified by NNSA to shorten schedules [to meet the 2030 deadline] will increase the risks of schedule slip, cost growth, and cancellation.” Yet that is precisely what NNSA is doing and what senator Heinrich and Graham, and with them all of Congress, have now demanded.

Returning now to the 2017 AoA, NNSA had determined, among the AoA’s other findings, that:

Assuming design began immediately, a production rate of 80 ppy could not be achieved before 2033 “at the earliest” (AoA p. 2; pages in this section all refer to this document). Yet the EA found that the 2030 deadline was feasible after all. As noted above, IDA agreed with the AoA. NNSA continues to maintain that its 2030 deadline is feasible.
Splitting pit production between PF-4 (say, at 30 ppy) and another facility (say, at 50 ppy) would be impractically expensive (pp. 45-46). Pit production has very high economies of scale – a 50 ppy facility is almost indistinguishable in size and cost from an 80 ppy facility. NNSA’s AoA therefore eliminated this “split production” option. Yet, for political reasons, the Parsons EA made precisely this option – reliable production of 30+ ppy at PF-4 and 50+ ppy elsewhere – a common assumption in all four of its alternatives, while omitting all the costs associated with reaching 30+ ppy in PF-4. Those costs simply disappeared from the analysis. We now know that these costs would be great, as will be discussed further below. Contrary to NNSA staff and other expert opinion, in May of 2018 NNSA chose to split production between LANL and SRS in exactly this way. We believe LANL’s part in this split posture is driven by politics and is likely to prove temporary. It is needless, whether the pit production mission aims at 80 ppy, 30 ppy, or some other number.
Construction of two “modules” at TA-55 to augment PF-4 capacity would be inadequate, slow, expensive, and risky, without offsetting benefit (p. 46). The AoA concludes that seven, not two, modules would be required. Yet variations of this rejected “module” alternative comprised EA alternatives 2a and 2b (each with two large “modules”), and alternative 2c (with three underground modules).
Trying to make 80 ppy in PF-4 would be impossible (p. 44), yet this is required for years in the first phase of EA alternative 2c, the alternative LANL supplied.
As the AoA relates, NNSA had previously determined (in June, 2017) that “continuing to rely on PF-4 for the Nation’s enduring pit production capability presented unacceptably high mission risk” because of the building’s age. Doing so would jeopardize the plutonium “program of record” and because PF-4’s “[s]pace and capacity constraints” precluded it (pp. 47-48). Yet NNSA’s current plan, authorized and funded by Congress, relies heavily on PF-4 for pit production.
NNSA concluded that “after a new 80 WR ppy capability is established [in other facilities], PF-4 can return to the research and development mission for which it was built” (p. 2). That would be a good idea. In fact, it is the only idea that makes any sense at all.
Splitting the production process itself – the flowsheet – between PF-4 and another facility or facilities would be costly, slow, and have high risk (pp. 43-44). Yet this rejected idea forms the basis of all EA alternatives for LANL.
Given these findings, NNSA said in the AoA that the only viable LANL option for reliable, enduring pit production involved construction of a new, built-to-purpose pit production facility for 80+ ppy. The AoA found this option would be expensive, slower, and have a higher risk of failure than remodeling the MOX Fuel Fabrication Facility (MFFF), now called the Savannah River Plutonium Processing Facility (SRPPF). Yet the EA did not include a greenfield LANL alternative. The EA did find that all three LANL options would mostly have higher capital cost, take longer, and – most important – have anywhere from greater risk, to far greater risk of failure than SRPPF.

3. DOE misunderstood and lowballed LANL pit production costs 24 years ago in structurally similar ways, resulting in program failure. Fortunately, there is still no need to make pits.

A sketch of this history was provided to Congress by this organization on May 30, 2019. The misleading 1996 study which led to LANL gaining the pit mission is on my desk but not scanned or posted.

4. Mystifying JASON “letter report” released (h/t Federation of American Scientists)

The JASON study of pit aging required by Senate appropriators (p. 104) is underway but as of December 20, 2019, had not been received. Meanwhile a “letter report” paid for by NNSA on related topics and dated November 23, 2019 has been released.

The JASON letter adds little new information to the open literature. The letter affirms JASON’s 2006 conclusions, endorsed by DoD and DOE, that “the present assessments of aging do not indicate any impending issues for the stockpile.” They add that “studies on Pu aging and its impacts on the performance of nuclear-weapon primaries have not been sufficiently prioritized over the past decade.”

That NNSA dropped the ball on this is puzzling. In this context it is very much worth noting that in a December 2012 letter to congressional staff LANL wrote,

Pit production to replace pits in the deployed stockpile due to plutonium aging is not required, nor is it planned to occur. (emphasis added)