Fusion and non-proliferation: assessing misuse of neutrons and associated signatures in fusion systems

Summary

This presentation examines theoretical non-proliferation risks and potential dual-use concerns associated with intense neutron fields in nuclear fusion systems, specifically the misuse of neutrons to breed special fissionable materials (Pu-239 or U-233) from fertile source materials. It evaluates the physical feasibility, minimum power thresholds, and sub-critical multiplication requirements for breeding a Significant Quantity (SQ) of fissile material per year. It also reviews multiple detection and verification methods—including radiological monitoring, thermohydraulic and satellite thermal imagery, environmental sampling, and design information verification—to support proportionate international safeguards without stifling commercial fusion development.

Slide 1: Title Slide

Fusion and non-proliferation: assessing misuse of neutrons and associated signatures in fusion systems Lee Packer, Steve Bradnam June 2025 IAEA DEMO Special Topic Discussion [email protected] UK Atomic Energy Authority

Slide 2: Acknowledgements

Acknowledgements

UKAEAIAEAPPPL • SRNL • MPIL • Several others

Some views are of the authors, not necessarily of UKAEA or other organisations listed.

IAEA DEMO workshop - non-proliferation special topic session (Page 1)

Slide 3: Background: some previous analysis on fusion and non-proliferation

Background: some previous analysis on fusion and non-proliferation

• A. Glaser and R.J. Goldston 2012 Nucl. Fusion 52 043004 • G. Franceschini, M. Englert, & W. Liebert, (2013). Nuclear Fusion Power for Weapons Purposes: An Exercise in Nuclear Proliferation Forecasting. The Nonproliferation Review, 20(3), 525–544. • M. Englert, The proliferation risks of intense neutron sources (2012) • G. Franceschini; M. Englert (2013): Safeguarding Fusion Reactors. A plea for a proliferation-resistant design of nuclear fusion, HSFK Report, No. 7 • P. Sauter, Safeguarding Nuclear Fusion – Nuclear Non-Proliferation Law in a Fusion-Powered Future (May 15, 2023). Max Planck Institute for Comparative Public Law & International Law (MPIL) Research Paper No. 2023-13 • J. L. Ball et al 2025 Nucl. Fusion 65 036038 • S. Desai et al., Building a path toward global deployment of fusion: Nonproliferation and export considerations, Atlantic Council issue brief, April 2025 • M. Diesendorf, D. Roser, & H. Washington, (2023). Analyzing the Nuclear Weapons Proliferation Risk Posed by a Mature Fusion Technology and Economy. Energies, 16(3), 1123 • IAEA Consultancy Meeting (held in 2013) • US Workshops, PPPL and SRNL - numerous contributors • Office of Technology Assessment at the German Bundestag (TAB) 2024 study. (examples, not a comprehensive literature review)

IAEA DEMO workshop - non-proliferation special topic session (Page 2)

Slide 4: Theoretical proliferation concerns

Theoretical proliferation concerns

• Misuse of neutrons to breed special fissionable material from source material (focus of this presentation) • Diversion of tritium or lithium-6 inventories and production technologies to use as boosters (Alex Somers, SRNL) • Aquiring fundamental inertial confinement fusion knowledge to inform the design of thermonuclear weapons

IAEA DEMO workshop - non-proliferation special topic session (Page 3)

Slide 5: Basic idea: misuse of neutrons

Basic idea: misuse of neutrons

Misuse scenario:

  • Neutrons directed towards source material in the neutron field: U-238 or Th-232
  • Extract irradiated target material
  • Reprocessing stage to extract Pu-239 or U-233

Key Points: • The breeding of weapons-grade fissile material in significant quantities, from source material, is theoretically possible using intense non-fission neutron sources • Fusion (DT) power plants and high-power accelerator-based neutron sources are within scope • Plant modifications would be necessary together with material processing/reprocessing capabilities • How to mitigate misuse concerns without unreasonably hindering progress in development of beneficial nuclear technologies to society?

IAEA DEMO workshop - non-proliferation special topic session (Page 4)

Slide 6: Fusion: dual use & proliferation concerns

Fusion: dual use & proliferation concerns

Fusion power plants, based on DT fuel; an intense neutron source • 1 GW fusion power ~ 3.56 x 10^20 n/s • Fusion power plants do not use source or special fissionable material. This distinguishes them from nuclear (fission) power plants - a clear advantage with regard to potential proliferation concerns • Hybrid fission-fusion machines, by definition, use source or special fissionable material - no advantage • Some breeding blanket designs include Be as a neutron multiplier; incidental production of Pu through U impurities in Be • The assessment of misuse risk depends on the specific design • Blanket technology and the plant’s potential to be modified are key aspects to consider • Material processing/reprocessing would also be required • Reasonable expectation that controls should be proportionate to the risks as fusion energy technology is developed

IAEA DEMO workshop - non-proliferation special topic session (Page 5)

Slide 7: Fusion’s place within the global non-proliferation regime?

Fusion’s place within the global non-proliferation regime?

• Treaty on the Non-Proliferation of Nuclear Weapons – ‘NPT’ • Entered into force 1970 • Distinctions between NWS and NNWS • Three pillars of the NPT: non-proliferation, peaceful uses of nuclear energy, and disarmament • As fusion technologies progress towards commercial demonstration and deployment, how will/should fusion plants be handled within the Non-Proliferation Treaty (NPT) and the global non-proliferation regime • In this discussion space there are some different national, private industry, public organisation perspectives • International collaborative activities needed to develop thinking in this area as the technology evolves e.g. through IAEA • Aim to collate views and establish some actions from the international DEMO community

IAEA DEMO workshop - non-proliferation special topic session (Page 6)

Slide 8: Scale of machine is an important consideration

Scale of machine is an important consideration

Some scenarios*:

  1. An undeclared small machine is built i.e. without the knowledge of the international community (clandestine scenario) • Small physical dimension and power profile to minimise the possibility of detection? • If sufficiently low fusion power, some or all tritium needs may be supplied externally • What’s the lowest fusion (and thermal) power machine that might trigger concern? (see later example analysis)
  2. Covert production in a declared GW scale facility without international community knowledge • Clandestine operation of DEMO (GW scale) machine widely considered to not be credible - (e.g. see Goldston and Glaser’s conclusion) • Balancing tritium self-sufficiency requirements with available neutrons for misuse
  3. Breakout scenario for a GW-scale machine

*Discussed in detail in: A. Glaser and R.J. Goldston 2012 Nucl. Fusion 52 043004

IAEA DEMO workshop - non-proliferation special topic session (Page 7)

Slide 9: Factors that might impact the level of concern

Factors that might impact the level of concern

• Neutron output rate and spectrum; availability of machine (capacity factor); Fraction of blanket that is utilisable • Time required to produce 1 SQ (smaller quantities could be considered) • Isotopic purity • Facility design information • Access and ease of installation of quantities of target material into blanket region. Access and ease of removal of materials, remote handling • Ancillary facilities e.g. hot cells, reprocessing facilities • Power handling systems and arrangements • Oversight body inspection frequency • Should be sufficient to allow an inspection regime to deter any misuse • Compare the speed with which an SQ could be bred and the time gap between inspections

TABLE II. SIGNIFICANT QUANTITIES (IAEA Safeguards Glossary):

  • Direct use nuclear material:
    • Pu (for Pu containing less than 80% 238Pu): 8 kg Pu
    • 233U: 8 kg 233U
    • HEU (235U ≥ 20%): 25 kg 235U
  • Indirect use nuclear material:
    • U (235U < 20%) (including low enriched, natural and depleted uranium): 75 kg 235U (or 10 t natural U or 20 t depleted U)
    • Th: 20 t Th

Significant quantity (SQ): the approximate amount of nuclear material for which the possibility of manufacturing a nuclear explosive device cannot be excluded. https://www.iaea.org/sites/default/files/iaea_safeguards_glossary.pdf

IAEA DEMO workshop - non-proliferation special topic session (Page 8)

Slide 10: Neutron multiplication in sub-critical systems

Neutron multiplication in sub-critical systems

• A sub-critical system contains source or fissile material but cannot sustain a nuclear chain reaction on its own, i.e. keff < 1 • An external source of neutrons can be used to drive the system: • Breed fissile material from source material • Induce fission reactions -> release energy • Subcritical systems with higher keff have (by definition) higher neutron multiplication

Homogenous vs. Heterogeneous Cases: • Neutron multiplication in these source material cases are different. If suitable moderator/coolant is used with geometric optimisation, then keff, hom < keff, het.

IAEA DEMO workshop - non-proliferation special topic session (Page 9)

Slide 11: What’s the lowest fusion power to achieve 1 SQ/y in sub-critical systems? An example case

What’s the lowest fusion power to achieve 1 SQ/y in sub-critical systems? An example case

• The answer is somewhat arbitrary, as it depends on keff (the effective neutron multiplication factor for the system) • Previous published work has, in most cases, studied misuse and production in homogenous blanket materials (keff not necessarily optimal) • Higher neutron multiplication systems -> more fission-like spectra in targets • Clearly, the limit of keff = 1 is a self-sustaining fission reactor • For subcritical systems, there are isotopic composition tradeoffs – fission versus (harder) fusion spectra • In our work, several sub-critical assembly case studies have been explored systematically, adopting a pin cell approach • Example case shown to determine optimum keff: • Natural and depleted U metal, Al cladding with light water moderator/coolant in a heterogeneous matrix • Parameterised pin cell arrangement (infinite lattice case: a (pitch), b, c, U metal, Al (or Zircalloy), Light water)

IAEA DEMO workshop - non-proliferation special topic session (Page 10)

Slide 12: What’s the lowest fusion power to achieve 1 SQ/y in a sub-critical system?

What’s the lowest fusion power to achieve 1 SQ/y in a sub-critical system?

• A simple 1D fusion study for the nat. U case indicates a 14.1 MeV neutron source output of ~5E16 n/s (equivalent to a continuous fusion power of ~140 kW, or ~1.5E24 emitted neutrons/y) could in principle achieve 1 SQ/y • Significant nuclear heating is unavoidable ~16.8 MW at full power per SQ/y Pu-239 • An important detection signature for small fusion systems • Other systems being investigated

Illustrative representation of 1D model:

  • Inboard build: Central Solenoid | TF Coil | Vacuum Vessel | Shield | Blanket | FW | Scrape-off layer
  • Plasma core
  • Outboard build: Scrape-off layer | FW | Blanket | Shield | Vacuum Vessel | TF Coil | Biological shield
  • Heterogeneous lattice of target material, cladding and light water coolant/moderator

IAEA DEMO workshop - non-proliferation special topic session (Page 11)

Slide 13: Hypothetical criteria to assess credibility of undeclared production: thermal power

Hypothetical criteria to assess credibility of undeclared production: thermal power

• Binford* and Bragin et al. studied fission research reactors and derived approximate quantitative methods that the IAEA could use to judge credibility of undeclared production of a significant quantity of fissile material annually • Relationship was developed based on the fission reactor thermal power (thermalised neutron spectra) • But not directly applicable to fast reactors • In the context of non-fission neutron sources, what if similar criteria based on neutron output (and other parameters e.g. spectral index) could be used for this purpose? • Fusion and accelerators can exhibit different neutron spectra to fission systems • Nevertheless, production of fissile material unavoidably produces some fission and releases significant energy

*Binford, F T. “Diversion assumptions for high-powered research reactors.”, Jan. 1984.

IAEA DEMO workshop - non-proliferation special topic session (Page 12)

Slide 14: Hypothetical criteria to assess credibility of undeclared production: neutron emission rate

Hypothetical criteria to assess credibility of undeclared production: neutron emission rate

• Neutron emission rate often reported, but the time integrated emission rate over a relevant time period e.g. 1 year is more relevant i.e. accounting for availability % • If we consider 1.5E24 n/y as a hypothetical neutron threshold? Then, what would fall in scope? • Many (probably all) DEMO concepts - ~3.5e20 n/s, availability ~30% • Materials testing facilities - ~1e19 n/s, availability ~80% What might fall in scope? • Some non-fission tritium production systems based on intense neutron source technology? • Some proposed private fusion prototypes? What would not fall in scope? (too low neutron fluence) • All current experimental fusion facilities (JET’s lifetime neutron budget was 2E21) • Current commercially available DT generator technology • Near-term fusion demonstrators (aiming for Q>1) would not produce sufficient neutron outputs due to low availability and fusion power

IAEA DEMO workshop - non-proliferation special topic session (Page 13)

Slide 15: Detection: technical application of monitoring techniques?

Detection: technical application of monitoring techniques?

IAEA DEMO workshop - non-proliferation special topic session (Page 14)

Slide 16: Detection: radiological signatures

Detection: radiological signatures

• Monitoring Coolant and Blanket Systems

  • Analyse particle flow of fertile materials, potential diversion pathways, radiological fields and optimal assay points • Evaluate background radiation levels (on/off-load) near coolant loops to establish detection baselines
  • Gamma Spectroscopy: For identifying abnormal isotopic signatures in materials and coolant systems
  • Neutron Detection: To monitor for abnormal neutron emissions suggestive of unauthorised materials or reactions (post operation) • Passive and active assay methods for monitoring incoming/outgoing materials • Antineutrino Detection (more exploratory):
  • Investigate potential background fields e.g. from Li-7(n,γ) → Li-8 → β⁻ + ν̄ₑ
  • Assess feasibility and sensitivity for monitoring fusion reactors.

Illustrations include: BEGe CS system, JET KN1 Neutron Diagnostic, and gamma/neutron count spectra.

IAEA DEMO workshop - non-proliferation special topic session (Page 15)

Slide 17: Detection: thermal signatures

Detection: thermal signatures

Producing fissile material from source material unavoidably generates excess heat, which could be detected

Thermohydraulic power monitoring: • https://www-pub.iaea.org/MTCD/Publications/PDF/nvs1_web.pdf • The advanced thermohydraulic power monitor (ATPM) may be used to monitor the power output of a research reactor and can verify that the output is consistent with the operator declared power level. This system monitors the temperature and water flow in the reactor’s primary cooling loop. These parameters are used to calculate the energy flow rate and the total energy produced in the reactor.

Satellite / thermal imagery: • Direct detection of heat sources • https://www.satellitevu.com/news/unlocking-hidden-insights-the-power-of-high-resolution-thermal-imagery • Example case study indicating thermal signature from reactors: https://www.38north.org/2024/01/north-koreas-pursuit-of-an-elwr-potential-power-in-nuclear-ambitions/

IAEA DEMO workshop - non-proliferation special topic session (Page 16)

Slide 18: Detection: environmental sampling

Detection: environmental sampling

Environmental sampling is used to verify absence of undeclared nuclear material

• Uses sensitive techniques e.g. mass spectrometry to detect nuclear activity residues • Focuses on swipe samples from nuclear sites and facilities under additional protocol. • Samples analysed in bulk (entire sample) or particle mode (individual microparticles). • Techniques include γ ray spectrometry, XRF, alpha/beta counting (alpha from Po-210 generated in LiPb) • Particle analysis: fission track method -> particles identified and mounted for e.g. thermal ionisation MS, SEM, SIMS • IAEA ES kit (Environmental Sampling kit) • Reference: https://www-pub.iaea.org/MTCD/Publications/PDF/nvs1_web.pdf

IAEA DEMO workshop - non-proliferation special topic session (Page 17)

Slide 19: Detection: other techniques

Detection: other techniques

• Camera systems (Reference: https://www-pub.iaea.org/MTCD/Publications/PDF/nvs1_web.pdf) • Open source information - publicly available information from external sources, including scientific literature, government reports, media, trade information and satellite imagery • Design information verification - e.g. facility design features, ancillary facilities, conduct engineering assessment of time needed to replace blanket modules and restart plant operations etc • Assessing physical properties - e.g. weight of blanket modules prior to installation

IAEA DEMO workshop - non-proliferation special topic session (Page 18)

Slide 20: Summary: fusion non-proliferation discussion

Summary: fusion non-proliferation discussion

Fusion plants that are not designed to and do not process, use, or produce nuclear material have the potential to realise a higher level of proliferation resistance than that of (fission) nuclear facilities • Existing international safeguards agreements apply to designs that process, use or produce nuclear material • For fusion designs that do not process, use or produce nuclear material, the IAEA does not currently apply any safeguards measures other than those needed to resolve questions, where applicable, relating to the accuracy and completeness of the information provided by Member States • As more information about fusion plant designs becomes available, further consideration is required to ascertain whether the scope of IAEA safeguards would apply more broadly to fusion plants • Key fusion technologies and materials (e.g., tritium, lithium-6, and related technologies to produce or handle these materials) are covered as dual-use items under existing export control guidelines • Further considerations may be required to strengthen export controls • Harnessing the benefits of fusion requires discussion on how to address non-proliferation without negatively impacting development or cooperation in the field of fusion • Safe and secure deployment of fusion energy technology is key to the UK’s international collaboration efforts. It includes the development of regulatory frameworks and an international non-proliferation regime • Close working with international partners on these topics is needed to contribute expertise • Such cooperation would ensure that controls are proportionate to the risks as fusion energy technology is developed

IAEA DEMO workshop - non-proliferation special topic session (Page 19)

Slide 21: Some questions for the community to consider

Some questions for the community to consider

• How can the fusion community support proportionate, effective non-proliferation strategies as technology evolves? • What are the key proliferation concerns for large-scale fusion machines? • What specific risks do small-scale fusion machines pose, especially in potential clandestine use? • How can the current regulatory framework support non-proliferation in widespread fusion adoption? • Near-term monitoring considerations – awareness in the community of how the IAEA is treating fusion now? i.e. under Complementary Access, where an Additional Protocol is in place • Are IAEA Additional Protocol measures sufficient for emerging fusion technologies? • How can fusion plant designs deter misuse or material diversion? • What R&D is needed to enhance misuse detection in fusion systems and how can international cooperation support it? • What improvements are needed in fuel inventory monitoring for fusion? • How should non-DT fusion systems be addressed in non-proliferation efforts? • How should the fusion community engage with the IAEA to meet non-proliferation objectives? What future IAEA initiatives (e.g., World Fusion Energy Group, technical meetings, coordinated research projects, INPRO, etc) would be most useful? • How can the IAEA best support the DEMO community in this context?

IAEA DEMO workshop - non-proliferation special topic session (Page 20)