Page last updated: June 30, 2014 (10:00 a.m. ET)
Launched in 2011 with seed-funding from Global Zero and The Simons Foundation, our team has been working on warhead verification approach based on the template approach. This page summarizes the main challenges of nuclear disarmament verification and the concept of the template approach for warhead authentication, which is the basis for the Nature article from June 2014. The page also provides a brief overview of other verification projects currently underway and includes a list of useful readings.
Existing nuclear arms-control agreements between the United States and Russia place limits on the number of deployed strategic nuclear weapons. Verification of these agreements takes advantage of the fact that deployed weapons are associated with unique and easily accountable delivery platforms, that is, missile silos, submarines and strategic bombers, to which agreed numbers of warheads are attributed. The next round of nuclear arms-control agreements, however, may place limits on the total number of nuclear weapons and warheads in the arsenals.
How can an inspector be assured that containerized items are authentic without learning anything about them? One strategy is to use information barriers that process classified information but only display the result in a yes/no manner. We have proposed an alternative approach based on a zero-knowledge proof. Graphics: Tamara Patton.
There are two main new verification challenges associated with agreements limiting the total number of nuclear weapons: First, nuclear warheads in storage and warheads entering the dismantlement queue will have to be inspected and accounted for. This is a qualitatively new challenge because the design of nuclear weapons is highly classified information that cannot be exposed to international inspectors. A viable verification approach therefore has to resolve the tension between reliably verifying that the inspected warhead is authentic while avoiding disclosure of information about its design. A second challenge is to establish confidence in the completeness of a warhead declaration, i.e., to ensure that a weapon states retains warheads that were never part of the verification regime. This summary focuses only on the first challenge.
There are two established methods to authenticate nuclear warheads or warhead components: the attribute approach and the template approach.
Under the attribute approach, parties must first agree on one or more attributes that characterize the inspected item. For example, the presence of plutonium could be such an attribute, and the parties could then also agree on a minimum mass that must be present to pass the test in a yes/no manner. Other attributes could refer to the isotopics and chemical composition of the material or to the symmetry of the item itself. In order to be authenticated, an inspected item may have to pass a number of attribute tests. Most research and development efforts have so far been focused on attribute measurements.
The second approach to nuclear warhead verification is the template approach, under which inspected items are compared against a reference item (“golden warhead”). Template measurements do not seek to determine absolute or relative values of properties that characterize the item (such as plutonium mass or isotopics); instead, the template approach seeks to generate a unique “fingerprint” of the item and compares this signature against a recorded template previously generated with the reference item. It is generally accepted that the template approach is more robust against cheating (compared to the attribute approach), but it does require the availability of a reference item to generate the template.
The graphics below illustrates the main elements of an inspection using the template approach.
Artist’s conception of a warhead authentication procedure using the template method. The device shown is inspired by the fieldable nuclear material identification system (FNMIS) developed at Oak Ridge National Laboratory. Click for labeled high-resolution image. Graphics: Tamara Patton.
Prior to an inspection, one or more reference items are selected, for example, at a deployment site in order to have maximum confidence in their authenticity. These items are placed in storage containers and securely sealed. Reference items (highlighted here in red) are brought to a dedicated dismantlement facility, where the containerized warheads slated for dismantlement are also located. Dedicated inspection systems could use active neutron interrogation to generate unique radiation signatures of items containing nuclear material. In the standard approach, the reference item is used first to record the reference signature (template), which is then available for comparison with the signatures of inspected items processed later on. In an inspection setting, an information barrier would carry out the data analysis and display the result only in a pass/fail manner.
Our proposed template approach proceeds somewhat differently. In particular, for maximum security, classified information is never measured or stored. Prior to inspection, the host prepares arrays of neutron detectors that are specially preloaded with signal counts. These preloads should represent the number of neutrons that will be stopped by the tested item, at each angle, when irradiated. Critically, the inspector chooses which preloaded array is used with which item, so the host cannot use a special array to conceal a spoof warhead. One at a time, the containerized items are placed in the inspection system. Neutrons from a 14-MeV neutron source illuminate the item through its container, and the transmitted neutrons add to the signal count in the preloaded array. If the item is valid, its authenticity will be confirmed because the final signal count in all detectors will match the signal that would have been seen with no item present; if the warhead is a spoof, this will be evident as well. Since the signal and noise level expected with no item present are not secret, no secret information is divulged when true warheads are verified.
Several verification projects are currently underway. They are often based at national laboratories, especially in the United States and the United Kingdom, but also at universities and independent think tanks. Below is an incomplete list (to be expanded).
- 3D Virtual Reality To Support Nuclear Disarmament Verification Research: With the support of the Norwegian Ministry of Foreign Affairs, the Verification Research, Training and Information Centre (VERTIC) and the Vienna Center for Disarmament and Non-Proliferation (VCDNP) and are developing a fictional virtual nuclear warhead dismantlement facility geared to assist researchers in developing and refining dismantlement verification system options and chain of custody procedures. The work aims to develop and test the virtual platform’s ability to offer increased flexibility, decreased planning costs, and greater accessibility in a realistic environment.
- UK-Norway Initiative (UKNI): Since 2007 the United Kingdom and Norway have worked together on developing a gamma detection system with an information barrier for an attribute approach. The system is designed to detect the presence and approximate isotopic composition of plutonium components while protecting potentially sensitive information. Both hardware and software have been developed with a view to increasing host and inspector confidence in the measurements performed. Text adapted/quoted from 2013 INMM Paper.
- University of Hamburg, Germany: Disarmament verification research at the Carl Friedrich von Weizsäcker-Centre for Science and Peace Research (ZNF) at the University of Hamburg investigates neutron and gamma attribute measurement methods to authenticate nuclear warheads and components. It focuses in particular on the requirements of inspectors from non-nuclear weapons states to assess compliance, while balancing it with the needed protection of sensitive information. Measurement techniques are developed with respect to the boundary condition that little inspector knowledge is available on the inspected item and that measurement methods should therefore have low dependence on unknown properties such as shielding and item geometry. Research is carried out in close collaboration with the Institute for Peace Research and Security Policy (IFSH). In addition to research, disarmament verification simulation exercises are carried out with students in collaboration with the UK-Norway-Initiative.
The Consortium for Verification Technology (CVT) will be launched in September 2014. The CVT is sponsored by the U.S. Department of Energy, will be led by Sara Pozzi at the University of Michigan, and includes thirteen universities and eight national laboratories to provide the research, development, and training to address technology and policy issues in treaty-compliance monitoring. The project team will educate more than 60 Bachelors, Masters, and PhD students to meet the current and emerging challenges in this field. The CVT will address the major gaps and emerging challenges in treaty verification through six thrust areas: (i) treaty verification: characterizing existing gaps and emerging challenges, (ii) fundamental data and techniques, (iii) advanced safeguards tools for accessible facilities, (iv) detection of undeclared activities and inaccessible facilities, (v) disarmament verification, and (vi) education and outreach.
The university partners to the consortium are University of Michigan (UM), Massachusetts Institute of Technology (MIT), Princeton University, Columbia University, North Carolina State University (NCSU), University of Hawaii (UH), Pennsylvania State University (PSU), Duke University, University of Wisconsin (UW), University of Florida (UF), Oregon State University (OSU), Yale University, and University of Illinois at Urbana-Champaign (UIUC). The university participants will execute research projects in collaboration with the DOE national laboratories, including Los Alamos National Laboratory (LANL), Lawrence Livermore National Laboratory (LLNL), Sandia National Laboratory (SNL), Idaho National Laboratory (INL), Oak Ridge National Laboratory (ORNL), Pacific Northwest National Laboratory (PNNL), Lawrence Berkeley National Laboratory (LBNL), and Princeton Plasma Physics Laboratory (PPPL).
- C. Comley, et al., Confidence, Security & Verification: The Challenge of Global Nuclear Weapons Arms Control, Atomic Weapons Establishment, Aldermaston, 2000.
- D. Spears, Technology R&D for Arms Control, US Department of Energy, Office of Nonproliferation Research and Engineering, Washington DC, 2001.
- F. von Hippel, et al., Verified Warhead Dismantlement, Chapter 5 in Global Fissile Material Report 2009: A Path to Nuclear Disarmament, International Panel on Fissile Materials, Princeton, NJ, October 2009.
- D. Cliff, H. Elbahtimy and A. Persbo, Verifying Warhead Dismantlement: Past, Present, Future, Verification Matters, 9, September 2010.
- J. Fuller, Nuclear Archaeology and Warhead Verification Collaborations, Chapter 7 in Global Fissile Material Report 2013: Increasing Transparency of Nuclear Warhead and Fissile Material Stocks as a Step Toward Disarmament, International Panel on Fissile Materials, Princeton, NJ, October 2013.
In June 2014, Science & Global Security published a Special Issue on Approaches to Nuclear Warhead Verification (with guest co-editor Malte Göttsche), including the following articles:
- M. Göttsche and G. Kirchner, Measurement Techniques for Warhead Authentication with Attributes: Advantages and Limitations, Science & Global Security 22 (2), 2014, pp. 83-110.
- K. J. Bunch, M. Jones, P. Ramuhalli, J. Benz, L. Schmidt Denlinger, Supporting Technology for Chain of Custody of Nuclear Weapons and Materials Throughout the Dismantlement and Disposition Processes, Science & Global Security 22 (2), 2014, pp. 111-134.
- H. White, P. Daborn, P. Hayden, P. Ind, The Use of Modal Testing within Nuclear Weapon Dismantlement Verification, Science & Global Security 22 (2), 2014, pp. 135-159.
On zero-knowledge proofs:
- B. Chazelle, The Security of Knowing Nothing, Nature, 446, 26 April 2007, pp. 992-993.
- S. Goldwasser, S. Micali, and C. Rackoff, The Knowledge Complexity of Interactive Proof-Systems, SIAM Journal on Computing, 18 (1), pp. 186-208, February 1989.
- B. Fisch, D. Freund, and M. Naor, Physical Zero-knowledge Proofs of Physical Properties, Proceedings of CRYPTO 2014, forthcoming.