Opinions Vary on Measures to Strengthen the Safety and Security of U.S. Nuclear Weapons
Titan II ICBM at the Titan Missile Museum in Arizona. [Steve Jurveston/Wikimedia Commons]
To improve the safety and security of nuclear weapons, specialists must weigh the risks and benefits of making intrinsic changes to the warheads (possibly degrading their performance) or pursuing external changes such as better access controls, according to experts at a workshop co-sponsored by AAAS and the Union of Concerned Scientists (UCS).
While improvements such as use of shock- and fire-resistant chemical explosives in the warheads could further decrease the risk of an accidental nuclear detonation or dispersal of plutonium, most of the experts at the workshop were not greatly concerned about the safety level of the current U.S. nuclear arsenal. There was more agreement that the security of the weapons — preventing them from falling into the wrong hands — is a "substantial issue" that deserves more attention.
The experts disagreed, however, on whether some intrinsic changes to the warheads — such as use of self-destruct technologies — were practical. It should be assumed, some suggested, that any stolen weapon can be used to cause a nuclear explosion. Others felt that detonating a stolen weapon would be difficult, given such features as permissive action links (PALs) and arming sequences. Still others held that intrinsic features could add a valuable additional layer of security, reducing the probability and consequences of nuclear use if a weapon was stolen by a terrorist.
report cover Read the AAAS summary report of the Workshop on U.S. Nuclear Weapons Safety and Security.
A new report summarizes the workshop, which was convened by the AAAS Center for Science, Technology and Security Policy (CSTSP) and the UCS on 12 December 2012. The workshop was conducted on an unclassified basis, under the Chatham House Rule, which encourages a frank exchange of ideas by assuring that participants will not be identified or quoted. Participants included active and retired scientists and engineers from Los Alamos National Laboratory, Lawrence Livermore National Laboratory, and Sandia National Laboratory; government representatives, including those from the (National Nuclear Security Administration (NNSA), the Department of Defense, and the State Department; independent scientists who are members of the JASON group that advises the government on nuclear weapons and other security issues; and experts from academia and nongovernmental organizations.
NNSA is seeking  to strengthen nuclear weapon safety and security, in particular by adding features to warheads and bombs as part of life-extension programs for these weapons.
One way to bolster the safety of nuclear weapons is to use insensitive high explosives, which are less susceptible to shock and fire, as triggering devices for the warheads rather than conventional explosives. But insensitive high explosives are less energetic than conventional explosives, the workshop report noted, requiring a larger mass and volume of such explosives in the warhead. The NNSA eventually would like to replace all conventional high explosives in the U.S. nuclear warhead arsenal with insensitive high explosives as one objective of the life extension programs, the report said.
While the high explosives and hundreds of other non-nuclear components of nuclear weapons can be tested to assess their reliability statistically, that has never been the case for the nuclear explosive package — the chamber containing the nuclear materials that produce a thermonuclear explosion. A key component of the package is the "primary," a shell of fissile material such as plutonium, called the pit, to be imploded by a surrounding layer of chemical high explosive. That explosion in turn triggers a secondary — and more powerful — fusion reaction.
Even when the United States conducted underground nuclear testing, there were not enough tests to provide statistically meaningful data on the reliability of the nuclear explosive package. Regular inspections of the nuclear components of the warhead can spot signs of damage or deterioration that may affect reliability, however.
Whenever intrinsic changes are made to the nuclear explosive package, it can be more difficult to certify that the weapon will work as intended. Some workshop participants argued that investments in the Department of Energy's Stockpile Stewardship Program — and the resulting increase in knowledge about the stockpile and how weapons work — allow confidence in making upgrades to the warheads, even without conducting nuclear explosion testing. (The United States, which has not conducted a nuclear test since 1992, signed the Comprehensive Nuclear-Test-Ban Treaty in 1996.)
"In this view," the report said, "technologies should not be frozen: as materials and issues are better understood it should be possible to incorporate changes in the nuclear explosive package."
But others favored a more conservative approach, arguing that more attention should be paid to pursuing safety and security measures outside the nuclear explosive package. Improvements outside the nuclear explosive package can be achieved much more quickly than warhead modifications, they said, and without creating concerns about the reliability of the stockpile.
However, that should not rule out consideration of some changes in the warheads to enhance security, some experts argued, including measures that would provide "self-protection" of the weapon in the event that other security measures fail. In this approach, the ultimate objective is a "designer proof" weapon, where even if a warhead fell into very technically capable hands, it could not be used. This could include disablement or "self-destruct" technologies to damage the plutonium pit at the heart of the warhead, for example.
Workshop participants stressed the value of "guns, guards and gates" to ensure that nuclear weapons do not fall into the wrong hands. The Department of Defense and the Department of Energy also have programs to undertake comprehensive screening and continual evaluation of personnel involved with nuclear weapons. Use controls also are in place, including a "two person" policy that requires simultaneous entry of codes at each of two terminals in each of two launch control centers to enable messages to reach nuclear-armed missiles.
Beyond such measures, there was discussion of further steps to prevent the theft of a nuclear weapon. One workshop participant proposed a beacon that would start transmitting the location of a weapon when coded signals indicate that the weapon is not where it should be. This technology might be shared with other nuclear nations.
Another threat comes from insiders who are presumed to be reliable but who cooperate with outsiders or act on their own, the report said. More broadly, some workshop participants said, "Those responsible for nuclear weapons need to find a way to discuss with the public in a serious but non-frightening way that the threat of terrorist access to nuclear weapons is real." It might also be beneficial to discuss these matters even with potential adversaries, the report said.
"There was broad agreement that the cybersecurity of nuclear command and control networks in the United States, Russia, and other states is of critical importance and warrants attention," the report said. "However, the high level of classification inherent to nuclear command and control procedures makes it difficult to have an in-depth understanding of the potential scope or severity of threats and of appropriate measures to counteract them."
The group discussed the interaction of nuclear weapon security, cybersecurity, and prompt alert operations, with one expert maintaining that "doubts about system integrity" is a reason to terminate "prompt alert" operations and lengthen the time required to launch a weapon during a crisis, especially with regard to the Russian system. The U.S. could discuss command and control network security with other nuclear weapon states, the report said.
Read the AAAS summary report  of the Workshop on U.S. Nuclear Weapons Safety and Security.