Centrifuge Enrichment and the Breakout Problem

Alexander Glaser Department of Mechanical and Aerospace Engineering and Woodrow Wilson School of Public and International Affairs Princeton University

Spring 2014 Meeting of the International Panel on Fissile Materials Rio de Janeiro, March 20, 2014

Revision 2 Background Global Enrichment Capacities, 2014 (16 operational plants in 10 countries, not including 2–4 military plants)

26,200

20,000

5,000 50

10 2,000

10 tSWU/yr are enough to make HEU for 2–4 weapons per year

tSWU/yr Total SWU-production in country/region A. Glaser, Centrifuge Enrichment and the Breakout Problem, March 2014 Global Enrichment Capacities, 2060 Based on the requirements for a (GCAM3) Policy Scenario in 14 World Regions

16,440

3,120 5,400 19,560 57,840 31,320 7,920 10,320 5,640 33,240

10,560 17,640

8,880

1,320

10 tSWU/yr are enough to make HEU for 2–4 weapons per year Countries in yellow have expressed interest in nuclear power program GCAM3 Policy Scenario envisions 1910 GWe for 2060

tSWU/yr Total SWU-production in country/region A. Glaser, Centrifuge Enrichment and the Breakout Problem, March 2014 Uranium Enrichment by Gas Centrifuge Enrichment Plants Used to be Gigantic (Gaseous Diffusion Plant K-25, Oak Ridge, TN, now demolished)

A.WWS/MAE Glaser, Centrifuge 353: From Enrichment Nuclear Weapons and the to Breakout Cyberwarfare Problem, March 2014 6 Why Centrifuges Are Different

Zippe Centrifuge, 1959

Characteristics of centrifuge technology relevant to nuclear proliferation Clandestine Option and Rapid Breakout

A.WWS/MAE Glaser, Centrifuge 353: From Enrichment Nuclear Weapons and the to Breakout Cyberwarfare Problem, March 2014 7 Clandestine Option Clandestine Option Sensitivity and Detectability of Different Enrichment Technologies

Detectability (Selected Criteria) Proliferation Sensitivity Size Energy Effluents

Calutron/EMIS (High) No Yes Yes Gaseous diffusion Low Yes Yes Yes Chemical exchange Very low (Yes) (No) (Yes)

Centrifuge High No No No

Laser High No No No

A. Glaser, Centrifuge Enrichment and the Breakout Problem, March 2014 9 200 meters

Fuel Enrichment Plant

(FEP)

The Natanz Site in Iran (2007) Google Earth, 34.885 N 50.995 E Breakout Option Minimal Breakout Times A.Q. Khan Cascade Scheme

C1/C2 Cascades HC-02 (456 machines) (3936 machines)

3.5% 20% 60% 90% HC-03 (128 machines) HC-01 (1312 machines)

67.5% of total enrichment capacity is used to produce 03.5%-enriched product 90.0% of total enrichment capacity is used to produce 20.0%-enriched product

A. Glaser, Characteristics of the Gas Centrifuge for Uranium Enrichment and Their Relevance for Nuclear Weapon Proliferation Science & Global Security, 16 (1–2), 2008, pp. 1-25

A. Glaser, Centrifuge Enrichment and the Breakout Problem, March 2014 13 Minimal Breakout Times Time to first significant quantity (25 kg of U-235 in weapon-grade HEU)

Declared facility that produces 3.5% enriched uranium prior to breakout 18,000 1st-generation machines, 1 SWU/yr (18,000 SWU/yr), in about 100 cascades using modified Khan scheme with valved down LEU cascades

Simulation Results

Without 3.5% LEU feedstock: about 6 months (vs 3 months) With sufficient stock of 3.5% LEU feedstock: about 2 months (vs 1 month)

Assumes that no significant extra time is required to valve down machines Numbers in red are based on simple SWU estimates

Numbers based on calculations by Patrick Migliorini (UVA) and Chuck Witt for Institute for Science and International Security, 2013

A. Glaser, Centrifuge Enrichment and the Breakout Problem, March 2014 14 What Can Be Done About It? Timeline of Centrifuge Programs Many countries have successfully developed viable centrifuge enrichment technology

A. Glaser, Centrifuge Enrichment and the Breakout Problem, March 2014 16 Preventing the Further Spread and Assuring Peaceful Use of Nuclear Technologies

Preventing • Tighten export controls (further) Further Spread • Delegitimize enrichment in today’s “non-enrichment” states • Increase the ability to detect undeclared facilities • Encourage multilateral approaches to the nuclear fuel cycle Assuring Increase the effectiveness of IAEA safeguards Peaceful • Use • Revisit alternative “proliferation-resistant” technologies • Devalue nuclear weapons

A.WWS/MAE Glaser, Centrifuge 353: From Enrichment Nuclear Weapons and the to Breakout Cyberwarfare Problem, March 2014 17 Multilateral Approaches to the Nuclear Fuel Cycle

Fuel Assurances

Joint Ownership of Enrichment Plants Construction of new facilities exclusively under multilateral control Conversion of existing facilities

A.WWS Glaser, 594q Centrifuge : Policy Analysis Enrichment - The and Future the ofBreakout Nuclear Problem, Energy - MarchPrinceton 2014 University - March 2, 2011 18 Dilemmas of Joint Ownership

Proliferation Can one share centrifuge technology without disseminating proliferation-sensitive information? Risk of premature deployment of sensitive nuclear technologies where they are not needed

Market Support of current technology holders needed (e.g. for new plants using “black-box” technology) Current (and mid-term future) enrichment demand already covered

Territoriality How effectively will the fact that a plant is multinationally owned reduce the risk of a “take over” by the host state?

A.WWS Glaser, 594q Centrifuge : Policy Analysis Enrichment - The and Future the ofBreakout Nuclear Problem, Energy - MarchPrinceton 2014 University - March 2, 2011 19 Can Multilateral Approaches Strengthen Nonproliferation and Disarmament Efforts?

Development and Peaceful Use of Nuclear Energy Sharing enrichment plant with partners in the region

Nuclear Nonproliferation Avoiding additional small-scale deployment of centrifuge technology under national control Possibility of implementing advanced safeguards approaches in new plants

Nuclear Disarmament Application of IAEA safeguards in plants even if located in NPT weapon states

A.WWS Glaser, 594q Centrifuge : Policy Analysis Enrichment - The and Future the ofBreakout Nuclear Problem, Energy - MarchPrinceton 2014 University - March 2, 2011 20