Areg Danagoulian Nuclear Disarmament Verifica1on via Resonant Phenomena (and other adventures in nuclear security) Areg Danagoulian 1
Areg Danagoulian Outline • What’s the big problem? (Nuclear Arms Reduction Treaties) • Why template verification and how does it work? • What is Nuclear Resonance Fluorescence (NRF)? • à NRF based verification • Epithermal neutron physics • à epithermal neutron verification 2
Areg Danagoulian Nuclear Arsenals • Significant Reduction since the Cold War • ,, Доверяй , но проверяй ! ’’ Но как ? How? Ինչպե՞ս : 3
Areg Danagoulian VERIFICATION • How do treaty partners verify that the other side is dismantling actual warheads and not fakes? They don’t. • Verification: delivery vehicles – eas ier to verify. • Problems: large leftover of non-deployed warheads • theft à nuclear terrorism, nuclear proliferation à Authenticate warheads, without revealing secret information 4 4
Areg Danagoulian Our Research: physics-based cryptography, template verification Authenticated template Candidate “ copy” of W88 copies, W88 Picked from a randomly Challenge : perform selected ICBM Is A 0 = A 1 ? ü checks while A 1 A 0 = A 2 ? ü • protecting secrets A 0 = A 3 ? ü • isotopicaly sensitive . . . A 2 à need A 0 cryptography - physical cryptography A 3 à need resonances! . . . 6 6 To dismantlement
Areg Danagoulian Analogy: NRF to Op0cal Spectroscopy Op1cal Spectroscopy Nuclear Spectroscopy Bremsstrahlung Back-scattered NRF Transmitted NRF Absorption lines, ~eV (W. Bertozzi) 7 NRF: unique fingerprint of isotopics
Areg Danagoulian Broad-spectrum source à NRF 235 U NRF cross sections (300K) U-235 NRF 600 spectrum 1815 keV 1769 keV Counts (arbitrary units) 400 1733 keV detector 200 HEU 46 keV background PHYSICAL REVIEW C , 041601(R) (2008) 0 0 1700 1800 1900 2000 gamma energy (keV) U-235 unique line spectra for U-235, U-238, Pu-239, Pu-240… Γ ~ meV à thermal motion à eV 8
Areg Danagoulian NRF Weapon authen0ca0on Concept Cryptographic Foil Bremsstrahlung NRF filtered brem (X-ray) Everything classified by the host Everything open Shielding Hosts : • provide the candidate warheads (to be authenticated) • Foil – thickness unknown to the inspectors, but of agreed upon isotopes Inspectors : • Detector, electronics (to be verified by hosts) • Visual access to the foil Joint : • Template (“golden copy”) 9 9
Areg Danagoulian NRF Weapon authen0ca0on Concept Weapon A: authenticated template Weapon B: candidate Cryptographic Foil Bremsstrahlung NRF filtered brem (X-ray) Everything classified by the host Everything open Shielding A B • Physical Cryptography: • No direct data from the weapon itself • SIGNAL = • Impossible to extract (Weapon) • Soundness and completeness: • Authenticated template A -- acquire S NRF (A) • Candidate weapon B -- acquire S NRF (B) 10 10 and compare
Areg Danagoulian What’s a bomb and how does it work? Plutonium or Uranium 11 (source: wikipedia)
Simulated 2.1 or 2.5 MeV bremsstrahlung beam > 1000 core hours for sufficient NRF sta1s1cs
Canonical hoax scenarios Template (black) vs hoax (red) Strongest counts per keV Hoax scenario discrepanc y ( σ ) WGPu à 107 U-238 WGPu à FGPu 14.6 1.7 1.9 2.1 2.3 2.5 2.7 energy E [MeV] R.S. Kemp, A. Danagoulian, R. Macdonald, J.Vavrek, Physical cryptographic verification 13 of nuclear warheads , PNAS 113 (2016) 31.
Areg Danagoulian NRF experimental setup � • Van de Graaff Accelerator • 2.5 MeV e-, DC beam • 20 m A HPGE � Cu radiator � g detector � DU target � photon beam axis � x-ray � imager � 14 Vacuum Tubes!!!
Areg Danagoulian Proxy Warhead • 3mm of 238 U • 0.5mm of 27 Al • 1.5” of plastic • “MIT Linear Implosion Design” 15
Areg Danagoulian Experimental setup 16
Areg Danagoulian Hoax to Genuine comparisons Uranium � Al � Target: � -- Lead Hoax -- genuine Plastic “explosive” � “Perfect Hoax” 17
Areg Danagoulian Hoax to Genuine comparisons 6mm Uranium � Pb � Al � 27 Al Half hoax discrepancy: 5 σ Target: � Full hoax discrepancy: 11 σ Plastic “explosive” � 238 U • 11 σ discrepancy in U lines • identical 18 counts in Al
Areg Danagoulian Extrapola1ons: the real bomb 5-10 σ in a (1+1)-hour proof-of-concept • “Black Sea” model: • 6X rate decrease • 25 uA à 2.5 mA beam current: 100x rate increase • 3 à 30 HPGe detectors: Black Sea Model 10x rate increase • à measurement 1mes of 2m ~minutes GammaSphere IBA TT100 Rhodotron J. R. Vavrek, B. S. Henderson, A. Danagoulian, “Experimental demonstration of an isotope-sensitive warhead verification technique using nuclear resonance fluorescence,” PNAS (2018), 201721278; DOI: 19 10.1073/pnas.1721278115
Areg Danagoulian Verification with Epithermal Resonant Assay (VERA) 20
Jayson Vavrek / Areg Danagoulian Epithermal Resonant Cryptographic Radiography • Epithermal neutron resonances in the 1-10 eV • Neutron Resonance Transmission Assay (NRTA) Chichester, D. L. & Sterbentz, J. W. Assessing the Feasibility of Using Neutron Resonance Transmission Analysis (NRTA) for Assaying Plutonium in Spent Fuel Assemblies. JNMM XL, 4 (2012). • Transmitted spectrum = isotopics geometry • cryptographic reciprocal mask Sample SAMMY fit 1 (TOF) 0.5 Transmission 0 -0.5 238 U 235 U 181 Ta 5 10 15 20 25 30 35 40 45 50 Energy [eV] Fig. 3. Example fit to normalized experimental data (60min • choose a resonance • à isotopic image 21
Jayson Vavrek / Areg Danagoulian Epithermal Resonant Cryptographic Radiography • Epithermal neutron resonances in the 1-10 eV • Transmitted spectrum = isotopics geometry. • cryptographic reciprocal mask cryptographic (TOF) ( ) • ~ flat image: no geom. information A recip = 1/ A object × const . • spectrum reveals nothing about the pit 22
Jayson Vavrek / Areg Danagoulian Simula1ons: Geometric hoax resistance template hoax (J. Hecla) 23
Jayson Vavrek / Areg Danagoulian Simula1ons: WGPu pit vs. a RGPu hoax + Different isotopics result in different transmission spectra + Only ~100k incident counts necessary for a 5 σ detection 24
Jayson Vavrek / Areg Danagoulian Geometric Informa1on Security + Compare the transmission image of the pit+reciprocal to that of a flat plate of the same total thickness + Images and spectra are identical – can’t differentiate, thus cannot infer any geometric information à geometric Zero Knowledge pit+reciprocal plate radial comparison 25
Jayson Vavrek / Areg Danagoulian Isotopic Informa1on Security + Protect the isotopics of the pit. + isotopics(pit+reciprocal) ≠ isotopics (pit) + MC simulations of three scenarios: - 70% 239Pu enriched pit, 98% enriched extension - 78% enriched pit and extension - 93% pit, 71% extension - the transmitted spectra are identical à isotopic Zero Knowledge • Jake’s MIT undergrad thesis Jake J. Hecla, Areg Danagoulian, “Nuclear Disarmament Verification via Resonant Phenomena, ” Nature Communications 9 , 2041-1723 (2018) 26
Jayson Vavrek / Areg Danagoulian POC Experiments: Rensselaer Polytechnic Ins1tute • Can we avoid simple imaging? Yes – single pixel tomography • no need for complicated reciprocals • simple detectors • Experimentally prove the feasibility of the concept • Proxies for “honest” template pit and “hoax” pit: • template: 90% Mo / 10% W (Mo ßà 239 Pu : W ßà 240 Pu) • isotopic hoax – different isotopic ratio • geometric hoax – perform rotations • Measurements: single pixel detector, 6Li glass, TOF Li glass detector à TOF à energy neutron beam Mo / W object encrypting foil, “unknown” composition • Work with PPPL on using smaller, precisely moderated DT sources for ~eV neutron beams. 27
Areg Danagoulian The Future • Perform epithermal experiments at to prove the epithermal concept • Collaborate • national labs • other countries • Russia • Need technological solutions for treaty verification à more ambitious, far reaching treaties • How can we, physicists, help solve major societal problems? 28
Areg Danagoulian The Team Postdoc: Dr. Brian Henderson Students: Undergrad S.M. PhD Jimmy Jayson Julie Will Ethan Ezra Ben Alumni: 29 Jake Hecla Dr. Buck O’Day Jill Rahon Bobby Nelson Jeremiah Collins
Recommend
More recommend
