Treaty Verification with Resonant Phenomena


Nations have been working to reduce the size of Cold War nuclear arsenals for more than thirty-five years. Because parity in force capability remains a central factor in strategic stability, reductions are possible only with intricately negotiated verification provisions that ensure neither side is cheating. Historically, verification was limited to delivery vehicles such as missiles and aircraft; it has not included checks on nuclear warheads directly. The limited approach allowed meaningful reductions in the level of deployed nuclear forces while avoiding the difficult challenge of having foreign governments inspect highly classified nuclear warheads. Ultimately, however, direct verification of warheads will be needed to facilitate warhead dismantlement, needed to eliminate the possibility of rapid rearmament, and to reduce the risk that “loose nukes” for the bloated US and Russian arsenals fall into the hands of terrorists. While many of the barriers to ambition, warhead-focused and verification-driven treaties are political, it is also clear that new technological research is also necessary to help solve this problem.

The Challenge

Despite decades of work, no protocol for direct warhead verification has been developed that is able simultaneously to provide high assurance that a purported warhead is real and protect the secrets of a warhead's design. A number of concepts have been suggested, e.g. the Trusted Radiation Identification System (TRIS). Such systems acquire data which contains information about the weapons, then a software/electronics "information barrier" module "sanitizes" the data in an operation that amounts to computational cryptography and only gives yes/no, clear/alarm answers. There is one fundamentally difficult problem with approach: the burden of the verification is now shifted to the instrument itself - one needs to verify that the instrument doesn't contain hacks which make it always say yes, and furthermore doesn't leak sensitive data via a backdoor exploit.

Can hack software. Can't hack physics.

So, why do this via electronics/software information barriers, when the similar goal can be achieved via physical cryptography? In physical cryptography the weapon-specific information is modified in a predictable but unknown way through the physics of the process. This means that by the time a theoretical detector acquires the data it is already instrinsically encrypted. In such a system we are not asking the participants to trust each other - no "trusted" devices are necessary. We only ask them to trust their own understanding of the physics of how the protocol works.

Physical Cryptography - with what physics?

There are many ways in which the above requirement of information security can be achieved. But in order to prevent hoax scenarios where the hosts could divert useful parts/materials from the weapons undergoing dismantlement, we need a physics signal which is not just element-specific, but even isotope-specific. While a simple radiography (with neutrons, photons, etc) may sound tempting, most interaction cross sections are very isotope- and even element-agnostic. We need a process which is highly isotope dependent.


Most bound states - a guitar string, a wine glass, an atom, a nucleus - have a very selective way in which they couple to external excitations. The guitar tuner, for example, emits sound of varying frequency, and when that frequency matches that of the string A2 the strings resonates. Your tuner knows it hit A2, and not E3. As the string resonates it emits its own acoustic waves - and the tuner listens to those. It will never mistake an E for an A.

The atoms and nuclei - states of electrically bound electrons, or nucleons bound by the strong force - are the quantum equivalents of the guitar string. When impinged upon by external excitations (photons, neutrons, electrons, etc etc) they selectively interact with those whose energies correspond to transitions from the ground state to some excited level. After they de-excite, the atoms and nuclei a secondary particle. The emitted or absorbed particle's energy can help us identify the precise type of the nucleus that was in the way. Just like the guitar tuner can identify the string.

This is called resonant phenomena.

There are a number of resonant processes involving nuclei:

Nuclear Resonance Fluorescence (NRF)

NRF involves the interaction of ~MeV photons with nuclei, causing very energy-specific interactions in most isotopes. By measuring these, one can achieve a physical cryptographic warhead verification. Learn more about how we perform treaty Verification with transmission NRF

Epithermal Zero Knowledge (epiZK)

Epithermal neutrons in the ~eV range exhibit sharp resonant interactions with most actinides. This behavior can be used to perform highly isotope-sensitive physical cryptographic warhead verification. Learn how we achieve treaty Verification with Epithermal Beams.