2
The Past: Before the 1992 Nuclear Explosion Testing Moratorium
There are two distinct periods in the history of the nuclear weapon laboratories: the Cold War period, during which nuclear explosion tests were conducted, and the period since 1992, when the nuclear explosion testing moratorium went into effect and new weapons development ceased. The vast majority of the warheads currently in the nuclear weapon stockpile were designed and tested during the Cold War period, which is the focus of this chapter.
The nuclear weapons era began in 1942. The Manhattan Project brought together many of the finest scientific minds in the United States and allied countries. This created a culture appreciative of intellectual excellence, which has continued to this day at all three laboratories.
The Manhattan Project began in academic settings, but in 1943 a weapons design laboratory was sited at Los Alamos, where the world’s first nuclear devices were created within about 2 years. The Manhattan Project had a sense of urgency brought about by the war and by the perception that Germany was working on an atomic bomb. Those involved felt that a race was on and the consequences of Nazi Germany being first to possess a nuclear weapon were unacceptable—and indeed appalling.
The Z Division of Los Alamos Laboratory was created in 1945 to provide military liaison, ordnance engineering, and surveillance in storage sites and to manage non-nuclear testing. This organization was physically located on Sandia Base near Albuquerque, New Mexico, to facilitate better interactions with the military. Z Division also began the development of test ranges on Sandia Base and in California during this period.
It was soon recognized that management of all those functions from Los Alamos was inefficient and that local management with industrial connections could be more effective. As a result, Sandia Corporation was formed in 1949 to oversee staff and facilities on Sandia Base, and AT&T agreed to be the first industrial manager of the Sandia Corporation.
A second nuclear explosive package (NEP) design laboratory was established in Livermore, California, in 1952 to provide competition to Los Alamos. It provided intellectually stimulating and productive competition, which expanded the potential design space for nuclear weapons. The Lawrence Livermore Laboratory, in California, was chartered to work on designs that were deliberately different from, but competitive with, the designs being pursued at Los Alamos. This step was in part a response to the understanding that excellence and creativity in any endeavor are enhanced when the individuals involved are challenged by peers who also possess a recognized high level of skill and in part because it was felt by Edward Teller and others that Los Alamos Laboratory was not doing enough to develop a hydrogen device. The establishment of a second NEP design laboratory was also a risk-reduction step, broadening the portion of design space considered and reducing the risk of groupthink or blind spots among the designers.
The new Livermore laboratory immediately demonstrated that it was taking a fresh look at weapons design, as evidenced by the fact that its first few designs, tested in 1953 and 1954, failed. By 1955, Livermore designs began to succeed in tests and in meeting new warhead requirements from the U.S. military. The goal of exploring fresh approaches led to a distinct and different technical culture and approach at Livermore.
After the establishment of the Livermore laboratory in 1952, it was decided to also establish a branch of Sandia at Livermore. This site was set up to have an independent system design capability that would draw on the science and engineering technical base that already existed at Sandia in Albuquerque. The Sandia-Livermore site became operational in 1956.
Sandia National Laboratories (SNL) is responsible for the non-nuclear components of the warhead such as the radar and the neutron generator; the arming, fuzing, and firing system; and use controls, as well as integration of the entire package with the delivery system. Los Alamos National Laboratory (LANL) and Lawrence Livermore National Laboratory (LLNL) are responsible for the NEP. Because SNL components could be tested much more extensively, SNL is treated separately from LANL and LLNL in the discussion below.
TESTING, PEER REVIEW, AND DESIGN COMPETITION IN THE NNSA NATIONAL SECURITY LABORATORIES
Los Alamos National Laboratory and Lawrence Livermore National Laboratory
The definitive feedback about the quality and result of the designers’ work during this period came from the direct validation enabled by nuclear explosion testing. According to the Comprehensive Nuclear-Test-Ban Treaty Organization (CTBTO),1 from 1945 to 1992, the United States carried out 1,032 nuclear explosion tests, 815 of which took place underground, primarily at the Nevada Test Site.2 The tests generally involved the NEP but not the entire warhead as an integrated system. In many cases, it is probably more accurate to call those tests “nuclear explosion experiments,” since many new design concepts were explored, and a number did not perform as expected when actually tested. Both the successes and the failures helped the scientists better understand the interplay between design and functionality of the NEP. In addition, the nuclear explosion experiments provided data to help validate and improve the computational models of the warheads that were put into the nuclear stockpile. Those computational models could then be used to estimate weapon performance in conditions that differ from those represented in underground testing.
Sandia National Laboratories
As noted above, Sandia is responsible for the non-nuclear components of the warhead. As such, testing of Sandia components could occur more readily in aboveground facilities, and this testing was not substantially inhibited by the 1992 nuclear explosion testing moratorium. Indeed, even before the moratorium, Sandia had moved away from reliance on underground nuclear explosion tests.
______________
1 See http://www.ctbto.org/fileadmin/user_upload/pdf/Sipri_table12b.pdf.
2 U.K. nuclear tests after 1962 were conducted jointly with the United States at the Nevada Test Site and are not included in this number.
Los Alamos National Laboratory and Lawrence Livermore National Laboratory
Each NEP laboratory did its best, using allocated resources, to develop and check its design, and the nuclear system test itself showed whether the developers were correct. Internal and external reviews of the science base, computer codes, and design work were conducted to ensure that the science base and the physics models in the code were the best that could be created with the knowledge available at that time. Reviews and extensive tests were also performed on electrical and mechanical systems that supported the functioning of the weapon. Interlaboratory peer review was less frequent and less formal than that which exists today, but it did occur on occasion, especially after nuclear test failures. In one famous case, a concept for a bomb-pumped x-ray laser that underwent development and preliminary testing by researchers at LLNL was shown by LANL to be infeasible.
Sandia National Laboratories
In the 1980s, Sandia initiated the independent Surety Assessment Center, which features a full- time team that conducts and contributes to the independent assessment of safety, use control, significant finding investigation (SFI) analysis and resolution, reliability analysis founded on certification and surveillance, and assurance of quality control. The team reported directly to the laboratory deputy director and director, which allowed maintaining significant independence from the design teams.
In the 1990s, Sandia relied on the independence between its New Mexico and California sites for effective independent technical peer review. Although the two sites were nominally part of the same laboratory, independence was gained through an organizational structure that separated nuclear weapons work at each site and geographically separated the nuclear weapons teams.3
Historically, Sandia has had a close relationship with the commercial electronics industry, which has stimulated intellectual cross-fertilization and also provided a kind of informal peer review of Sandia concepts and products.
______________
3 Gary Sanders, Sandia National Laboratories, “National Academy of Sciences Peer Review Study,” presentation to the committee on June 10, 2014.
Los Alamos National Laboratory and Lawrence Livermore National Laboratory
Within Los Alamos itself, design competition existed from the start of the Manhattan Project and continued there until Lawrence Livermore was established. Scientists and engineers involved in the Manhattan Project pursued two design concepts in parallel, including all associated necessary functions spanning fissile material production, weapon design, and weapon production—in essence, this was the original nuclear weapon design competition.
The subsequent decades of the Cold War years were characterized by an intense design competition between the laboratories in Los Alamos and Albuquerque, New Mexico, and Livermore, California, which resulted in the robust and reliable U.S. nuclear deterrent. The impact of this independent, interlaboratory NEP design competition was profound. For example, significant reductions in size and weight of a thermonuclear weapon were achieved. The new designs could fit not only on intercontinental ballistic missiles (ICBMs), but even on missiles that could be carried on submarines. In an example from the later years of the Cold War, the design competition for the W88 warhead particularly emphasized the need to reduce the size of the warhead while reliably attaining a desired yield. Through this very active competition, both NEP laboratories achieved notable size reductions, again opening up new military options. In the later years of the Cold War, safety and security systems also benefited significantly from the differences in approaches that resulted from design competition, such as the development of a new, insensitive high-explosive system.
Sandia National Laboratories
During this era, the nation had two distinctly separate and competing nuclear weapon design capabilities: the Los Alamos/Sandia Albuquerque (LANL/SNLA) design collaboration team and the Lawrence Livermore/Sandia Livermore (LLNL/SNLL) design collaboration team. The Livermore branch of Sandia took on much of the culture of LLNL in that it took pride in designing non-nuclear systems for nuclear weapons that were different from those that its colleagues in Albuquerque were producing for Los Alamos while still relying on many component design teams (e.g., for the radar and neutron generator) in Albuquerque. Similar to the case of design competition between Los Alamos and Lawrence Livermore, design competition in the non-nuclear components and systems of the nuclear weapon produced significant advances in safety and security sys-
tems, such as the detonator safety technologies, that exist in the stockpile today. The Livermore site of Sandia continues to be active, but its work is now part of a distributed program across both the Albuquerque and the Livermore sites.4
More than 50 formal design competitions were held between the two NEP design laboratories during the Cold War period. These competitions resulted in the deployment of several new weapon types in the nation’s stockpile and a stronger understanding of the relationship between different designs for NEPs and the consequent performance and reliability of a weapon type. When tests failed to meet design expectations, the results were shared among the three laboratories to identify and resolve design flaws.
It is important to note that these design competitions exercised all necessary aspects of the nuclear weapons complex, from design to engineering, fabrication, and testing of a prototype; each of these latter steps also provided feedback to all participants in the process, strengthening their judgment and insight. For the designs that were downselected for military systems, the production capability of the complex was also fully exercised. This work enabled the United States to meet the constantly evolving national security threat effectively throughout the Cold War era.
Based on discussions with laboratory experts at its meetings and on the accumulated experience of its members, the committee concluded that the following were key elements of the nuclear weapon design activities during the period of nuclear explosion testing (1945-1992):
- Scientific and technical staff of high competence and good judgment.
- Overlapping expertise between the two NEP design laboratories in Los Alamos and Livermore and between the two SNL sites, and the fact that the teams from the two locations used different technical approaches, which in combination made it possible to pursue genuine competition between two distinct teams with comparable competencies.
- A culture in each of the laboratories that encouraged extensive internal technical reviews to check the work being done.
______________
4 In some cases, Sandia took action to assign work to the Livermore branch that was previously conducted in Albuquerque because it was recognized that the volume of design work assigned to the Livermore branch at that time was not sufficient to maintain the design expertise.
- Vigorous competitions leading to successful weapon designs that were produced, tested, and fielded. As recounted in this chapter, design competitions led to significant tangible benefits both to the NEP and to system designs. These noteworthy successes led to the committee’s strong endorsement of design competition, discussed later in this report.
- A strong sense of an international nuclear threat.
- Nuclear explosive testing, which provided definitive validation and feedback to the weapons designers and engineers.
While the last of these elements is not a viable option today, the others are largely available as contributors to the nuclear deterrent.
