by Taylor Loy*
Tritium, a radioactive isotope of hydrogen, is an essential component of most advanced nuclear weapons designs and has typically been produced in secure military facilities. Only a few grams of tritium per warhead are needed to boost nuclear weapon yields significantly. Tritium’s relatively short half-life means that it needs renewal more often than other weapons materials. With a long-term goal of introducing more transparency and promoting conditions for reduction and eventual elimination of nuclear weapons, it would make sense for ongoing tritium production to take place at civilian reactors.
There are some challenges to such an approach, but the advantages to non-proliferation standards, nuclear safety and cost savings should outweigh these difficulties. In 2004, the United States transitioned to civilian production of military tritium by leasing the services of a commercial reactor as the lowest cost alternative for replacing the capacity of the last military production reactor shutdown in 1988. The decision to produce tritium at the Watts Bar Nuclear (WBN) Power Station allowed for the decommissioning of a secretive military reactor with a questionable safety record. In contrast, the WBN site is publicly accountable to the U.S. Nuclear Regulatory Commission (NRC) and even subject to International Atomic Energy Agency (IAEA) inspections under the U.S. voluntary offer. Transitioning from aging facilities allows for both cost-savings and the safety benefit of reducing accident risks.
On March 18, 2024, the French Ministry of Armed Forces announced they would begin producing tritium for nuclear weapons in 2025 at two civilian Électricité de France (EDF) nuclear reactors at Civaux. Previously, France relied on dedicated military production reactors at the Marcoule site to maintain their estimated 290 nuclear warheads and ensure “strict sufficiency” for deterrence. By choosing to shut down unilaterally and decommission military production reactors, the United States and France have placed robust technological limits on their future fissile material production capabilities. These states also demonstrated more cost-efficient stewardship of existing stockpiles and increased transparency. Nuclear weapons states have been secretive about tritium production quantities, but when production shifts to commercial reactors this secrecy is no longer tenable. The U.S. decision to produce military tritium openly might remove some adversaries’ doubts of the durability of the U.S. deterrent and mitigate risks of a new nuclear arms race. In fact, the National Nuclear Security Administration (NNSA) Strategic Integrated Roadmap categorizes the 2025 goal to “Demonstrate enhanced tritium production capability” as meeting “Nuclear Deterrent” priorities. Significant challenges remain, however, in extending cessation of military tritium production to other nuclear states.
Shutdown of military fissile material production facilities should be verifiable. The simplest method for NWS to offer verification that their military reactors are irreversibly shut down is to invite international experts to observe the process, just as France did in 2008 with their facilities at Pierrelatte and Marcoule. Otherwise, open-source information (OSINT) such as commercial satellite imaging can verify non-operation of these sites. Non-invasive examination of gaseous releases from fuel re-processing sites could also signal any ongoing plutonium separation by the presence of the isotope krypton-85. Additionally, for sites suspected of illicit operation, evaluations of tritium content in waste streams can estimate tritium production, a mechanism used by Thomas Cochran, et al. to estimate U.S. production at the Savannah River Site in South Carolina.
Incentivizing nuclear-armed states such as India and Pakistan to shut down military production reactors would be a significant challenge, but they already have the requisite civilian nuclear infrastructure for a tritium production transition. If India and Pakistan agreed to make steps in this direction, the domestic challenges of operating outside the non-proliferation regime could be lessened. A possible model for this is the controversial 123 Agreement between the United States and India, which could be adapted to allow for limited tritium production. Israel and North Korea, on the other hand, would likely pose idiosyncratic and potentially insurmountable difficulties. Since the IAEA technically designates tritium a byproduct material, the organization is agnostic about its production. It would be worth pursuing tritium transparency in the IAEA context.
Another key provision of any such arrangement should be tied to upholding the nuclear norms of the Comprehensive Test Ban Treaty (CTBT). Because tritium is a vertical proliferation risk due to its unique contributions to weapon sophistication and yield, the nuclear testing taboo must be upheld to ensure this material is not leveraged to expand a nation’s nuclear capabilities. If this would enable the long-term reduction of fissile material production capacity, then such tradeoffs between principles and practicalities may prove to be worth making.
Eventual nuclear disarmament under strict and effective control will require the verifiable decommissioning of all military-production reactors unaccountable to international safeguards/inspections. Until that time, NWS still need to ensure the continued reliability, safety and security of nuclear weapons. Producing tritium for military use in declared civilian facilities will offer greater transparency to one of the more secretive dimensions of the nuclear weapon enterprise. With the added benefit of reducing global plutonium production capacity, this technological transition would provide novel approaches to verification that states could leverage for future arms control agreements once diplomatic conditions make these critical instruments for risk management and peace viable once again.
*Editor’s note: Writing for the Center’s new Next Up in Arms Control series, Taylor Loy is a Postdoctoral Associate at Virginia Tech working as a Co-PI under a Carnegie-funded grant to research the civilian/military boundary in nuclear infrastructures. He received his PhD in Science, Technology, & Society (STS) at Virginia Tech, and his dissertation focuses on tritium, a unique and ephemeral radioactive isotope.
DISCLAIMER: Next Up in Arms Control is a way for the Center to present an opportunity for dialogue and provocation through the thoughtful exchange of ideas and opinions on new or different ways to reduce and eventually eliminate nuclear threats. Opinions are the authors’ alone and not necessarily reflective of Center’s positions on the issues addressed.
