Technology R&D for Arms Control

Home , Arms control

Arms Control and Nonproliferation Technologies Office of Nonproliferation Research and Engineering Spring 2001 TechnologyTechnology R&DR&D forfor ArmsArms ControlControl • • • • • • • • • • • • • • • • SERVICES • • FACILITIES and INSTRUMENTATION using x-ray generators from 30 Fixed facility and field digital radiography computed tomography systems data acquisition system-Calibrated PIN diodes-Internet 11 connection Supporting Instrumentation-Calibrated 20-GHz sampling scope-Multi-parameter D/T Neutron Generator-Sodern Genie 16 Tandetron-High-Voltage Engineering, Inc. 1.5 MeV/amu Two positive ion Van de Graaffs-High-Voltage Engineering, Inc.-2-MV potential – – 18-MeV Linac (Varian Associates, Inc.) 3000 R/min (@ 1 m) widths: 500 ns to 4 Varitron Accelerator (Varian Associates Inc.)-2 Two 4-MeV Linacs-Field radiography/neutron source capability experimental needs and applicable resources. Single point-of-contact: simplified coordination between researchers and modifications Customized accelerator performance Experimental verification of predictions and objectives Customized radiation detection system development Photon and neutron dosimetry Photon, electron, and neutron transport calculations for system applications Customized user support Wide range of accelerator types available to researchers 0.5 space 21,000 sq. feet of laboratory energy spread-Long-pulse mode: pulse width: 4 Pulse widths: 15 ns to 2 Beam energy and current analysis Short-pulse mode: pulse widths: 30 Repetition rate: single shot to 240 Hz-Three beam ports – 25 MeV electron Linac µ s Beam energy and current analysis-Capable of up to µ s – 450 kV – 0p,1 nC/pulse, 0.5% beam 50 ps, 10 – 12-MeV energy range-Pulse µ s, 2,000 nC/pulse- www.iac.isu.edu Pocatello, Idaho 83209 1500 Alvin Ricken Drive Idaho Accelerator Center Email: [email protected] Fax: 208.282.4538 Phone: 208.282.2902 Pocatello, Idaho 83209 Campus Box 8060 Idaho State University Associate Director Dr. Jay Kunze Email: [email protected] Fax: 208.526.5208 Phone: 208.526.1730 Idaho Falls, 83415-2802 PO Box 1625 (INEEL) Environmental Laboratory Idaho National Engineering and Associate Director Dr. James L. Jones Email: [email protected] Fax: 208.282.5878 Phone: 208.282.5877 Pocatello, Idaho 83209 Campus Box 8263 Idaho State University Director Dr. Frank Harmon CONTACTS See reverse side for an explanation of this chart Uranium mining Plutonium production reactors Civilian reactors uranium Enrichment Natural LEU fabrication Fuel processing Uranium Production Reprocessing Materials

P Spent U

l Strategic

Reprocessing Reserve u storage

Naval HEU r

t fuel o a

n n

i i u u

m m or LEU Natural

Military use Military

Strategic Strategic Reserve

fabrication Pu Post-1997, naval fuel, commercial Fuel

Pre-1997, weapons-grade Pre-1997 fabrication fabrication Enriched Uranium Pu parts Highly Parts Military use Military Assembly Weapons HEU weapons component Pu weapons component Assembly

l U

u r t a

o n

i i u Warhead

u storage/

staging m m Trilateral Initiative U.S.–Russian Plutonium Disposition Agreement Plutonium Production Reactor Agreement (PPRA) Mayak Transparency Material Protection, Control, and Accounting (MPC&A) HEU Transparency Cooperative Threat Reduction (CTR) Weapons stockpile Weapons Stockpile De-Mated warhead storage P

l u

r t a

o n n i

u i

u m m Disassembly Weapons Disassembly Strategic Strategic Storage disassembly and Storage Storage Reserve Reserve For an electronic copy of this file, contact the ACNT Production Office at Lawrence Livermore National Laboratory. excess Pu component HEU component disassembly and conversion conversion trg Disposition Storage downblending Processing/

Storage

excess P

u U

t r

o a

n n

i i

u u m m fuel production Pu conversion and blending Mixed-oxide fabrication Fuel fuel rods

Uranium

Material Storage immobilization Pu waste Civilian reactors Civilian reactors reactor Naval rods High-level storage waste Spent fuel NN-50’s Materials Protection, Control, and Accounting (MPC&A) program focuses on enhancing the physical protection and material-accountancy infrastructure at former Soviet Union sites containing weapons-usable nuclear material. This work is augmented by the program’s involve- Accelerator Applications and ment with enhancements to the regulatory framework within Russia, national-level material Radiation Science accounting, nuclear-material transportation, as well as other related activities. The program’s goal is to reduce the risk to U.S. national security of an undetected theft of weapons-usable material from the former Soviet sites by enhancing their MPC&A capabilities. The placement of a yellow box around a given type of facility on the diagram signifies that the program is involved, or is planning to be involved, in infrastructure upgrades at these sites.

The Idaho Accelerator Center (IAC), located at Idaho State University, has operated since 1994 in partnership with the Idaho National Engineering and Environmental Laboratory and the Department of Energy RESEARCH Conducting fundamental and applied research in low-energy nuclear science using accelerator-produced radiation

IAC provides university, government, and industrial scientists and engineers unparalleled research opportunities: Radiography, tomography, and nuclear techniques for nondestructive evaluation and nondestructive assay; industrial and agricultural applications of accelerator-produced radiation; ion and photon- beam analysis for environmental and mineral extraction needs; radiation science in medicine; radioisotope production; accelerator-based neutron therapy; radiation effects testing for semiconductor devices; instrument and radiation-detector testing for weapons surety studies and fundamental nuclear physics research.

EDUCATION Offering undergraduate and graduate degrees in physics, health physics, engineering, and applied science

IAC supports educational activities at all levels of Idaho State University’s aca- demic areas, including physics, health physics, engineering, waste management, geology, biological sciences, and health sciences. The University offers bachelor’s and master’s degrees in physics and health physics, and in cooperation with the College of Engineering, doctorates in engineering and applied science. Students participate in all IAC research and development. The research staff has published more than 100 peer-reviewed publications over the past five years.

PARTNERSHIPS AND COLLABORATIONS Teaming with university government and commercial partners on applied research, testing and technology deployment.

IAC offers technical expertise and state-of-the-art facilities for collaborations with universities, government agencies, and commercial and industrial organi- zations. These partnerships promote practical advances in nuclear and radiation science and facilitate the transfer of technology to the private sector. IAC collaborations range from agricultural applications of accelerator-produced radiation to nondestructive examination through gamma-ray spectroscopy, radiography, and tomography. IAC partners conduct research and testing on applications in environmental remediation, waste management, chemical-weapons verification, contraband detection, and radiation effects. Published by Focus on Transparency U.S. Department of Energy National Nuclear Security Administration Arms control treaties and agreements directly Defense Nuclear Nonproliferation Programs influence the national security of the United Kenneth Baker, Acting Director States. Security decisions made today are based on that influence. Strategic Arms Reduction DOE/ACNT Project Manager Treaty (START) I began a reduction of nuclear Robert Waldron Office of Nonproliferation Research and Engineering weapons. Fortunately, START I is easily verified using straightforward methods—literally tape Issue Editor measures and plumb bobs coupled with David Spears National Technical Means and direct observation. As we enter into new negotiations, there is Scientific Editor Arden Dougan a desire to expand the traditional approach by focusing on the warheads themselves rather Managing Editor than on their delivery systems. More modern Gorgiana M. Alonzo treaties, involving both nuclear warheads and delivery vehicles, are not as simple to verify. Art/Copyediting/Design/Layout John Danielson Within the last few years, and in the foreseeable Dwight Jennings future, emphasis has increased on “transparen- Sylvia McDaniel cy measures” that will open “windows” on the Kelly Spruiell nuclear-weapons activities of the signatories to such agreements—including the U.S. These Reviewers Thomas Gosnell transparency measures cannot be verified in the James Morgan strict START I sense, but all of the proposed ini- Robert Scarlett tiatives are based on high-technology measure- James Woodard ments to provide the necessary windows.

Editorial Assistance Catherine Anderson, Los Alamos National Laboratory We define transparency as measures and Francis Poda, Savannah River Technology Center procedures that increase our confidence in a George L. Cajigal, Threat Reduction Support Center negotiated activity taking place as required. Verification, then, is defined as measurements Production Office and procedures that prove a negotiated activity Arms Control and Nonproliferation Technologies Lawrence Livermore National Laboratory is taking place. 7000 East Ave. (L-185) Livermore, CA 94550 To support research and development (R&D) in this area, the Departments of Energy and Correspondence Defense developed the Joint Integrated Gorgiana M. Alonzo Phone 925.424.6100 Technology Implementation Plan. Technical Fax 925.423.9091 experts from both communities contribute to E-mail [email protected] the “Integrated” Plan, supporting R&D in arms control and maximizing the research dollar by Disclaimer: avoiding duplication of effort. Reproduction of this document requires the written consent of the originator, his/her successor, or higher authority. This report was prepared as an account of work sponsored by Contact: David Spears the United States Government. Neither the United States NNSA/Office of Nonproliferation Research and Engineering Government nor the United States Department of Energy Phone: 202.586.1313 nor any of their employees makes any warranty, express or Fax: 202.586.2612 implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of any information, Email: [email protected] apparatus, product, or process disclosed, or represents that its use would not infringe privately owned rights. Reference herein to any specific commercial product, process, or service by trade name, trademark, manufacturer, or otherwise, does The purpose of Arms not necessarily constitute or imply its endorsement, recom- mendation, or favoring by the United States Government. Control and Nonproliferation The views and opinions of authors expressed herein do Technologies is to enhance communication not necessarily state or reflect those of the United States between the technologists in the NNSA Government and shall not be used for advertising or community who develop means to verify product endorsement purposes. compliance with agreements and the policymakers who negotiate agreements.

ACNT • Technology R&D for Arms Control • Spring 2001 Contents Transparency and Verification Transparency and Verification Issues Associated with Arms Control ...... 2 Verification Methods: Attributes vs. Templates ...... 4 Information Barriers ...... 6 Certification and Authentication ...... 7 Treaties, Agreements, and Initiatives Strategic Arms Reduction Treaties (START): I and II ...... 8 The INF Treaty ...... 9 The MRI Agreement ...... 9 Transparency Measures for the U.S.–Russia HEU Purchase Agreement ...... 10 Converting Weapons-Grade HEU (90% enriched) to LEU ...... 11 Mayak Transparency ...... 12 Fissile Material Cut-off Treaty ...... 13 Fissile Material Transparency Technology Demonstration ...... 14 Attribute Measurement System with Information Barrier Technology ...... 15 The Trilateral Initiative: Attributes Verification ...... 16 U.S.–Russia Plutonium Production Reactor Agreement ...... 18 A Core-Discharge Monitor ...... 19 Future Arms Reduction Initiatives ...... 20 The Comprehensive Test Ban Treaty ...... 21 Facilities The Pantex Plant ...... 22 Savannah River Site’s K Area Material Storage Project ...... 23 TA-18 at Los Alamos National Laboratory ...... 24 The Radiation Measurement Facility at Lawrence Livermore National Laboratory ...... 26 Overview to Arms Control Technology About the Atomic Nucleus, Radioactivity, Fissile Materials, and Transparency ...... 28 Plutonium Using Low-Resolution, Gamma-Ray Spectroscopy To Detect the Presence of Plutonium ...... 30 Measuring Plutonium Isotopes and Age with MGA ...... 31 Pu300, Pu600, and Pu900 Systems ...... 32 Absence of Oxide—Neutron Multiplicity Counting ...... 35 Estimating Plutonium Mass via Singles Neutron Counting ...... 36 Measuring Plutonium Mass by Neutron Multiplicity Counting ...... 37 Minimum-Mass Estimates of Plutonium ...... 38 Symmetry Measurements of Plutonium Objects ...... 40 Gamma-Ray Imaging To Determine Plutonium Shape ...... 42 Autoradiography Adaptation Using Optically Stimulated Luminescence ...... 43 Attributes and Templates from Active Measurements with NMIS ...... 44 Uranium Determining the Presence of HEU with Passive Detection Methods ...... 46 Active Interrogation for Monitoring HEU ...... 48 Blend-Down Monitoring System for HEU ...... 50 Blend-Down Enrichment Monitor for HEU ...... 51 Portable Equipment for Uranium Enrichment Measurements ...... 52 Determining Uranium Enrichment ...... 53 Chain-of-Custody Chain-of-Custody Monitoring of Warhead Dismantlement ...... 54 Fission-Product Tagging ...... 56 Nondestructive Assay of Special Nuclear Materials ...... 57 Alternative Signatures Identifying Weapon Components in Sealed Containers by Electromagnetic Induction ...... 58 Thermal Infrared Signatures of Containers ...... 60 Tags and Seals Tags and Seals for a Potential Strategic Arms Control Monitoring Regime ...... 62 Seals for Transparency and Treaty Monitoring ...... 64 Detection of Nonnuclear Components Chemical-Explosive Detection by Neutron Interrogation ...... 66 Points of Contact ...... 68

ACNT • Technology R&D for Arms Control • Spring 2001 1 Transparency and Veriﬁcation

Transparency and Veriﬁcation Issues Associated with Arms Control

he overarching problem for a large The attribute approach to transparency number of current transparency activi- measurements involves determining a small Tties—the Highly Enriched Uranium number of measurable quantities character- (HEU) Purchase Agreement, the Trilateral istic of the object under inspection—be it Initiative, the Plutonium Production warhead, component, or fissile material. Reactor Agreement (PPRA), and Mayak These attributes are determined from first Transparency—revolves around identifying principles as being a necessary element of the material in a closed container as either a that object. Possible attributes might warhead, a weapon component, or fissile include— material from a dismantled nuclear weapon. • the presence of radiation Further complicating the problem is that • the presence of plutonium some or all of the information concerning • the presence of weapons-grade plutonium the material in the container is considered • plutonium mass classified by at least one party to the agree- • uranium enrichment level. Along with the determination of the attribute itself, a threshold value for that attribute must be established that separates acceptable and unacceptable material. Although most of the attributes being considered at the present involve radiation measurements, this is not the only approach. Attributes such as acoustic signatures, chemical emanations, and eddy-current effects have been considered in the past and may yet prove fruitful in some situations. The history of attribute measurements can be traced back at least to the START I negotiations. The attribute of interest then was the neutron radiation, or lack thereof, emanating from containers of a certain size found in a weapons storage area. In this case, the attribute contained no classified information. The attribute measurement technique really came to the forefront with the Joint Statement on Mutual Reciprocal Inspections U.S. and Russian experts engage ment. Two solutions to this problem have (MRIs). Following the signing of this agree- in joint experiments for the been proposed: template-matching and ment, weapons experts from both sides met determination of appropriate attribute measurements (see page 4). Most for three week-long sessions in Moscow to plutonium attributes and mea- of the current negotiations seem to be head- determine the attributes characteristic of surement methods for the MRI ed toward the attribute approach because it material removed from a dismantled nuclear Agreement (1996). does not require the retention of classified weapon. It was finally agreed that the useful data obtained during an inspection mea- attributes for this agreement were— surement. However, the Russians have • presence of weapons-grade plutonium recently proposed some novel applications • mass above a certain threshold of template-matching that invite further • shape of the object in the container. study.

2 ACNT • Technology R&D for Arms Control • Spring 2001 Transparency and Veriﬁcation

Because the signing of an Agreement for U.S. and Russian experts discuss Cooperation with the Russians (which a shape measurement of an would have allowed the exchange of classi- item inside a Russian storage fied information for arms control purposes) container for special nuclear seemed imminent, the issue of the classified materials. nature of some of the attributes was ignored. Although the actual MRI agreement eventually foundered on the shoals of the classification issue, the procedure developed for determining the useful attributes has persisted to the current era. The nature of the specific agreement determines the complexity of the attribute set appropriate for that negotiation. These can run from the very simple for the HEU Purchase and PPRA to the highly complex Transparency: for Mayak Transparency negotiations. For Measurements the HEU Purchase, essentially one attribute and procedures is important, namely the uranium enrich- which increase ment prior to the blenddown of material confidence that a removed from dismantled nuclear weapons. negotiated activity For the PPRA, the interesting attributes are is taking place as the presence of weapons-grade plutonium required. and the time (age) since chemical purifica- tion of the plutonium in the container. On Verification: the other hand, for the Mayak agreement, Trilateral Initiative negotiations, attribute Measurements the attribute set is highly complex, driven techniques were given a new importance. and procedures by the need to develop confidence that the In these techniques, the measured value of that prove a material, possibly in an unclassified shape, a given attribute is compared to an agreed, negotiated activity is derived from dismantled nuclear unclassified threshold. The system then is taking place as weapons. reports whether or not the measurement required. The linchpin that makes the attribute falls above or below the threshold value as method attractive for modern negotiations a pass/fail (“red light/green light”). With has been the information barrier concept the development of attribute measurement (see page 6). As noted, one side or the other systems that encompass the information considers the values of some or all of the barrier idea, the future for attribute tech- relevant attributes to be classified from its niques in arms control agreements seems security point of view. With the collapse of assured. the negotiations on an Agreement for Cooperation in 1996, it appeared that James Morgan attribute methods would no longer be use- Lawrence Livermore National Laboratory ful. However, with the development—and Russian acceptance—of the information barrier concept within the context of the

ACNT • Technology R&D for Arms Control • Spring 2001 3 Transparency and Veriﬁcation

Veriﬁcation Methods: Attributes vs. Templates

wo fundamentally different approaches The template approach identifies an item can provide confidence that declara- by comparing a measurement with an Ttions concerning items in a nuclear- empirical template for the declared item. If weapon arms control regime are true. The the item is classified, then the template is attribute approach is based on the intrinsic also classified and is never viewed by characteristics of nuclear weapons and their inspectors. This requires an automated pro- components. The template approach com- cedure to certify that the measurement pares the radiation signature from an inspect- agrees with the template within specified ed item with a known standard for a weapon uncertainty limits. Template comparisons or component of the same type. Both also require secure storage and certification approaches—the centerpiece of technology of a classified database, but no a priori R&D for several years—have been investigat- assumptions are made regarding the charac- ed in parallel with the treaty negotiations. teristics of the inspected item. An issue Attribute methods increase confidence in with the template approach is the need for the authenticity of an inspected item by a “trusted” item of each weapon and com- demonstrating that the item possesses the ponent type to use as template sources. characteristics of a nuclear weapon. The arms control treaty dictates which Although several characteristics might be verification method is most appropriate. considered, two are illustrative: (1) the mass Procedures associated with attribute mea- of plutonium must exceed a specified thresh- surements are simpler because a classified The Trusted Radiation Attributes old and (2) the ratio of 240Pu to 239Pu must database is not required. There is also a Demonstration System, or be less than a declared maximum. Though degree of comfort in knowing that each TRADS, uses a high-purity ger- classified data are collected to confirm these inspected item exhibits a clearly defined set manium detector to confirm two attributes, it is not necessary to store of characteristics. In contrast, template attributes of the inspected item, classified information because the attributes comparisons rely on the abstraction that an a W84 warhead in this photo- themselves are unclassified. A key feature of item is essentially the same as one measured graph. The detector is mounted the atribute approach is the ability to previously. Template comparisons are the inside the cart and the “trusted” authenticate the measurement system using only practical solution if the objective is processor and electronic com- an unclassified standard. demonstrating that two or more weapons or ponents are on top of the cart. components are of the same type. Depending on the requirements, either method or a combination of the two might be used. Attribute Approach Several attribute-measurement methods use both passive and active (where the inspected item is irradiated by an external radiation source) techniques. The Trusted Radiation Attribute Demonstration System (TRADS) is an example of a passive system. The only detector required to confirm the attributes of weapons-grade plutonium and highly enriched uranium (HEU) is a high- purity germanium spectrometer. The system uses the Minimum–Mass Estimate method (see page 38) to confirm the isotopic composition and the 239Pu mass threshold. The

4 ACNT • Technology R&D for Arms Control • Spring 2001 Transparency and Veriﬁcation

presence of HEU is confirmed indirectly based on the 2,614-keV emission from the isotope 232U, which is almost always present in HEU. Measurements are completed after a 10-minute counting time. The front of the sensor is located one meter from the axis of the inspected item. There is no size limit for the inspected items. The analysis algorithm is sufficiently robust to accom- modate the effects of intervening materials, so items ranging from small components to complete weapons can be inspected. The TRADS uses a “trusted processor” to acquire and analyze data and to display unclassified messages. The trusted processor employs a divided architecture and software technology was derived from systems current- The Radiation Inspection System design that protects sensitive information. ly used for domestic safeguards to confirm (RIS) uses a sodium iodide The needs of the inspecting party are the identities of pits in containers. detector to measure the addressed by several features including easi- gamma-ray spectrum of an ly inspected components, a tamper-indicat- Use of Templates in Arms Control and inspected item. Identity is con- ing enclosure, and a secure hash algorithm Domestic Safeguards firmed if the measurement for software authentication. The greatest difference between arms con- matches the certified template trol and domestic safeguards applications is for another item of the same Template Approach the way in which the template database is type. In a measurement at the The fissile materials in nuclear weapons created. Inspectors in international applica- Pantex plant in Amarillo, Texas, emit gamma rays with spectral distributions tions cannot view spectra when the database RIS identified the item in only characteristic of the isotopes contained in is created, so authenticity must be certified in 30 seconds. The system is the materials. Because gamma rays are scat- another way. A certain degree of confidence portable and battery-powered. tered and absorbed by intervening materi- is obtained by allowing inspectors to ran- The detector operates at room als, the gamma-ray distribution is also domly select items used in the benchmark temperature. affected by non-emitting materials. The measurements that create the database. resulting spectra are sufficiently distinctive Confidence is augmented when the database to identify items by comparing a measured is established by confirming attributes using spectrum with the template for the declared a system such as TRADS. This contrasts with type. The template is created by measuring domestic applications where trusted individ- one or more items certified to be authentic. uals view the spectra, when measurements (The Russians use a similar approach but are recorded to ensure that the characteristics use the term “passport” to describe what we represent what are expected for the items. generally call a template.) The Radiation Inspection System (RIS) is Dean J. Mitchell an example of a template system. This sys- Sandia National Laboratories tem, originally developed to confirm the identities of weapons and weapon components in a dismantlement scenario, measures spectra with a sodium iodide detector. The

ACNT • Technology R&D for Arms Control • Spring 2001 5 Transparency and Veriﬁcation

Information Barriers

nformation barriers are being studied in developers for a wide variety of fissile mate- the United States as well as in Russia (for rial and warhead reduction agreements Iexample, under the Warhead Safety and between the United States and the Russian Security Exchange agreement laboratory-to- Federation. laboratory program). These studies are most Efforts had been underway to establish a often conducted in conjunction with radia- U.S.–Russia Agreement for Cooperation, tion measurement systems to help solve that would have enabled the exchange of potential monitoring problems with nuclear selected restricted data for the purpose of warheads, warhead components, and the monitoring nuclear weapons and materials fissile material associated with warheads. A agreements. Absent such an Agreement, radiation detection system information bar- information barriers probably offer the rier consists of technology and procedures only effective solution for the use of tech- that prevent the release of sensitive nuclear nical measures, such as radiation measure- information during a joint inspection of a ment systems, on sensitive items and sensitive item, and it provides confidence materials. Because the ionizing radiation that the measurement system functions emanations from a nuclear warhead consti- exactly as designed and constructed. The tute classified information in both the U.S. U.S. Government has been studying infor- and Russia, relevant detection technology mation barriers in a coordinated manner will only be allowed—will only be useful— since January 1997. Under the auspices of if information barriers have been engi- the Joint DOE–DOD Information Barrier neered into the measurement system. It is Working Group, a set of fundamental strongly believed by U.S. technical special- design criteria have been developed and ists working in this area, and probably peer reviewed by security specialists for the most U.S. policymakers as well, that the purpose of guiding measurement system successful implementation of information barrier technology can only result from a cooperative joint development effort involving all the parties associated with any particular negotiation. Significant effort has been undertaken recently to demonstrate to the Russian Federation and the International Atomic Energy Agency the feasibility of integrating information barriers into radiation measurement systems. Attribute measurement systems incorporating information barriers were demonstrated for a U.S.–Russian Federation–International Atomic Energy Agency audience in June 1999 and for a U.S.–Russian Federation audience in August 2000. In addition, the concept was part of the Plutonium Production Reactor Agreement Workshop in November 2000.

A rack shields the electronic components as part of an information barrier incorporated into the Fissile James Fuller Material Transparency Technology Demonstration, held at Los Alamos National Laboratory in August 2000. Paciﬁc Northwest National Laboratory

6 ACNT • Technology R&D for Arms Control • Spring 2001 Transparency and Veriﬁcation

Certiﬁcation and Authentication

Often mistaken for one another, certification and authentication are complementary processes. Certification of monitoring equipment by the host country establishes trust that no classified information is revealed during arms control or transparency measurements on sensitive items. An independent agency known as a certifying authority provides an unbiased assessment. Certifying authorities in the U.S. and Russia have stated independently that to be certified, a piece of equipment must be supplied by the host. For the monitoring party in turn to trust measurements using host-supplied equipment, the opportunity to examine the equipment in detail and to witness its operation on non- sensitive substitutes for the controlled items must be granted. These operations, along with exhaus- tive evaluations of equipment design and construction, comprise authentication by the monitoring party.

The certification process is well defined in both the U.S. and Russia and includes— • Security and vulnerability testing of hardware with information barriers • Testing and evaluating computers and software for processing classified data.

Requirements for authentication, however, are still being established as of this writing. Successful authentication will ensure the monitor that accurate and reliable information is provided by a measurement system and that irregularities, including hidden features, are detected. NNSA and DoD researchers are crafting a model U.S. position for authentication that includes the following elements: • Functional tests using calibration sources • Evaluation of design documents and comparison to systems “as-built” • Evaluation of hardware and software • Random selection of equipment by monitoring party • Tamper-indicating devices, including tags, seals, and secure video recordings • Detailed human procedures accounting for all monitoring activities.

Certiﬁcation and authentication share a concern for system reliability. Clearly, a system that mal- functions affects both the integrity of the measurement and the security of any classiﬁed information within it. This argues for clarity of design and the sharing of design guidance between the host’s certifying authority and the monitor’s authenticating authority. The authorities are also interested in facility decisions that can greatly affect the ease and cost of building the trust necessary for a system to be accepted by both sides.

Richard T. Kouzes Paciﬁc Northwest National Laboratory

James K. Wolford, Jr. Lawrence Livermore National Laboratory

ACNT • Technology R&D for Arms Control • Spring 2001 7 Treaties, Agreements, and Initiatives

Strategic Arms Reduction Treaties (START): I and II

TART I is a treaty between the U.S. by mutual agreement, and may be extended and the U.S.S.R. on the reduction and for five-year intervals by agreement. Selimination of strategic weapons START II, built on the provisions of signed in July 1991. Following the dissolu- START I, was signed in January 1993. It has tion of the U.S.S.R., Soviet START I obliga- not yet been ratified by Russia—a U.S. con- tions were assumed by Russia, Ukraine, dition for beginning future negotiations. Belarus, and Kazakhstan—formerly Soviet Both sides have subsequently agreed to shift republics with strategic nuclear weapons the deadline for the completion of START II A Russian inspector examines a on their territories—under the Lisbon reductions by five years to December 2007. cruise missile. (Photo courtesy of Agreement of May 1992. Under START II, the U.S. and Russia can DTRA) Among its many provisions, START I lim- deploy no more than 3,000 to 3,500 strategic its the U.S. and Russia (as the sole inheritor nuclear warheads on ICBMs, SLBMs, and of Soviet nuclear warheads) to 1,600 heavy bombers. No more than 1,700 to deployed ballistic missiles (both 1,750 warheads can be deployed on SLBMs, Intercontinental Ballistic Missiles [ICBMs] and all Multiple Independently Targeted and Submarine-launched Ballistic Missiles Reentry Vehicle (MIRV) payloads must be [SLBMs]) and heavy bombers for each coun- eliminated on land-based missiles. MIRV try. The treaty also limits each side to 6,000 payloads on ICBMs and SLBMs may be “accountable” deployed warheads of which “downloaded” to achieve a reduction in no more than 4,900 deployed warheads as well as ICBM “de- may be on ballistic MIRVing.” The number of warheads on missiles, 1,540 on heavy bombers will no longer be discounted heavy ICBMs (the as they were in START I. In addition, up to Soviet SS-18), or 100 bombers that have never been equipped 1,100 on mobile for long-range nuclear Altitude-launched ICBMs. Complicated Cruise missiles (ALCMs) can be shifted to counting rules dis- conventional roles and will not be counted count the numbers in the overall START limits. of bombs and missiles carried by heavy Radiation Detection Equipment in bombers. A separate, Monitoring START politically binding Annexes to the START I Inspection agreement limits Protocol describe what and how radiation each side to 880 detection equipment may be used for on- long-range (greater than 600 kilometers) site inspections. In general, such equipment Submarine-launched Cruise Missiles confirms that some inspected items are not (SLCMs). nuclear. Specifically, a neutron source and Other provisions of START I address war- detection equipment used by U.S. inspec- head-downloading from ballistic missiles, tors may distinguish between long-range new types of ICBMs and SLBMs, mobile nuclear and non-nuclear Russian ALCMs ICBMs, non-deployed missiles, exemptions mounted on heavy bombers, and confirm from treaty limits, verification, data denial, that containers do not hold long-range and treaty duration. nuclear-armed ALCMs. Weapon reductions are scheduled to be completed by December 2001, seven years Ronald L. Ott after the treaty entered into force. START I Lawrence Livermore National Laboratory has a duration of 15 years, unless changed

8 ACNT • Technology R&D for Arms Control • Spring 2001 Treaties, Agreements, and Initiatives

The INF Treaty

The treaty between the U.S. and the Soviet Union on the elimination of their intermediate-range and shorter-range missiles was signed on December 8, 1987. Known as the Intermediate Nuclear Forces (INF) treaty, it was the first treaty to allow one party to measure radiation from the nuclear warheads of another party. The treaty eliminated missiles and launchers but not warheads. During the negotiations, general rules for conducting inspections were considered for a former missile site. Objects large enough to contain a treaty-limited item would be subject to inspection. The inspection team would be permit- ted to bring documents and equipment, including radiation-detection devices. Because the Soviets planned to use SS-25 missiles at former SS-20 bases, and because the launch canisters of the SS-25 are large enough to contain an SS-20 missile, the SS-25 missiles at former SS-20 bases would be subject to inspection. A special Verification Commission produced a Memorandum of Agreement (MOA) on the implementation of the verification provisions of the treaty. The result was that a neutron detector would measure the neutron intensity pattern in the vicinity of the warhead section of the missile launch canister, and for certain cases, the end cap would be removed for visual observation of the warhead section. Benchmark measurements were made on SS-20 and SS-25 missiles by a U.S. team in the summer of 1989. This confirmed significant differences between the warheads of the two missile systems. The MOA, signed on December 21, 1989, contains a detailed description of the inspection equipment and procedures for its use.

Keith W. Marlow and Dean J. Mitchell Sandia National Laboratories

The MRI Agreement

In March 1994, the then-heads of the Department of Energy, Hazel O’Leary, and Russian Ministry of Atomic Energy (Minatom), Viktor Mikhailov, signed an agreement to work jointly toward “mutual reciprocal inspections” (MRIs) of ﬁssile materials removed from dismantled nuclear weapons. In pur- suit of this agreement, U.S. and Russian technical experts engaged in exchange visits during which technological components of MRI were demonstrated, discussed, and ultimately, jointly researched. Demonstrations of and joint experiments with possible MRI technology occurred over the next two years at Rocky Flats, Lawrence Livermore National Laboratory (twice), and the Siberian Chemical Combine located at the closed Russian city of Seversk (Tomsk-7). No enduring inspection regime has yet resulted from the MRI agreement, but the technology research and discussions have been seminal to nearly all of the other initiatives discussed here. The attributes approach to conﬁdence-building measurements draws heavily upon the MRI experience. Issues associated with protection of sensitive information through administrative and technical means were a regular theme in MRI exchanges. Finally, the MRI visits accomplished their objective of engag- ing U.S. and Russian technical experts to address the knotty problems now coming to light as the more structured transparency regimes develop. In this regard, the MRI exchanges can certainly be viewed retrospectively as successful.

M. William Johnson Los Alamos National Laboratory

ACNT • Technology R&D for Arms Control • Spring 2001 9 Treaties, Agreements, and Initiatives

Transparency Measures for the U.S.–Russia HEU Purchase Agreement

n accordance with the 1993 U.S.–Russia Also according to the Agreement, a per- HEU Purchase Agreement, 500 metric manent monitoring office was established Itons of highly enriched uranium (HEU) in 1996 at the Ural Electrochemical from dismantled Soviet nuclear weapons Integrated Enterprise (UEIE) in Novouralsk. will be down-blended into low enriched Transparency monitoring began in 1996 uranium (LEU) hexafluoride over a 20-year with visits to UEIE and the Siberian period. This LEU hexafluoride will be sold Chemical Enterprises (SChE) in Seversk. to the U.S. as fuel for commercial power Monitoring at the Electro-Chemical Plant reactors. The Agreement also includes trans- (ECP) and the Mayak Production parency monitoring by both parties to Association began in 1997 and 1998, affirm the nonproliferation objectives of the respectively. Agreement. U.S. monitors access storage and process Russian deliveries began in 1995 with areas. They inspect containers of HEU 186 metric tons of LEU containing the weapons components, HEU metal chips, equivalent of six metric tons of weapons- HEU oxide, HEU hexafluoride, and LEU grade HEU while the transparency-monitor- hexafluoride. They witness the burning of ing arrangements were being worked out. HEU metal chips to HEU oxide and observe Under the Agreement, the U.S. has the the input and output of processes of purifi- right to send technical experts to the four cation of HEU oxide and conversion of HEU Russian plants that process HEU to LEU oxide to LEU hexafluoride. They also (see map). Up to six 5-day monitoring visits inspect equipment where HEU (90% are allowed each year at each Russian plant. enriched) is down-blended (as gaseous uranium hexafluoride) with 1.5%-enriched LEU to produce LEU in power-reactor enrich- ments from 3.6% to 4.95%. Visual observation by technical experts is key to transparency monitoring. The experts’ effectiveness is enhanced significantly by using U.S. monitoring equipment in the Russian plants (see pages 51 and 52). Since 1997, portable nondestructive assay equipment has confirmed the enrichment of HEU (in its various forms used in Russian processing) by measuring the intensity of gamma rays from 235U. The U.S.-developed Blend-Down Monitoring System (BDMS), which confirms the flow and enrichment of uranium hexafluoride in the down-blending process, continuously monitored the blending of The Blend-Down Monitoring System (BDMS) consists of a flow monitor and an enrichment monitor for eight metric tons of HEU at UEIE during each pipeline in the Russian down-blending facilities. 1999. An enrichment monitor compares

10 ACNT • Technology R&D for Arms Control • Spring 2001 Treaties, Agreements, and Initiatives

Located in formerly secret cities of the Soviet To U.S. nuclear-weapons complex, four plants process weapons-grade HEU to LEU for nuclear-reactor fuel: St. Petersburg the Siberian Chemical Enterprises (SChE) at LEU RUSSIA Seversk, the Ural Electrochemical Integrated MOSCOW HEU LEU LEU Enterprise (UEIE) at Novouralsk (also referred to as Novouralsk (UEIE) LEU the Urals Electrochemical Integrated Plant, UEIP), Zelenogorsk (ECP) HEU HEU LEU the Mayak Production Association (MPA) at Ozersk, Seversk HEU (SChE) Ozersk HEU LEU and the Electro-Chemical Plant (ECP) at (Mayak) HEU Oxidation HEU HEU Oxidation Zelenogorsk. and HEU LEU

Siemens Power Corporation Richland, WWashingtonashington

LEU LEU Oxide

Portsmouth, Ohio WASHINGTON, D.C. UNITED STATES LEU Framatome Cogema Fuels ABB/Combustion Engineering Lynchburg, Virginia Hematite, Missouri GE Nuclear Energy Wilmington, North Carolina Westinghouse Fuel Fabrication Facility Columbia, South Carolina Converting Weapons-Grade HEU (90% enriched) to LEU 235U gamma rays with the attenuation of gamma rays from a 57Co source. A flow monitor uses a modulated 252Cf neutron source to induce fission of 235U atoms flowing in the pipe and a downstream gamma-ray detector to measure the time delay for the fission products arriving at the detector (see page 50). The BDMS, for the first time, directly measures the quantity • The HEU-metal component is removed of HEU blended at the UEIE. Similar equip- from a nuclear weapon. ment will be installed at ECP and SChE. • The component is machined into metal Through February 2000, 81.3 metric tons shavings. of Russian HEU has been down-blended to • The metal shavings are heated and LEU and shipped to the U.S. An additional converted to an oxide. 30 metric tons are scheduled for delivery in • Any contaminants are chemically removed 2000, and at least 30 metric tons in each from the HEU-oxide. subsequent year until the agreed amount is • The HEU-oxide is converted chemically reached. Actual quantities each year are into uranium-hexafluoride gas. determined by a contract between the • The uranium-hexafluoride gas is diluted United States Enrichment Corporation with a much lower enrichment level of ura- and Techsnabexport, the commercial nium-hexafluoride gas, producing an LEU- arm of Minatom. hexafluoride gas for nuclear-fuel fabrication. • Cylinders are filled with the Douglas A. Leich LEU-hexafluoride gas. Lawrence Livermore National Laboratory • The cylinders are shipped to the U.S., where they are delivered to nuclear-fuel Ed Mastal manufacturers to make fuel rods for Department of Energy, Germantown, Maryland commercial nuclear-power plants.

ACNT • Technology R&D for Arms Control • Spring 2001 11 Treaties, Agreements, and Initiatives

Mayak Transparency

n 1992, Russia requested U.S. assistance is not used for nuclear weapons. The U.S. for the construction of a fissile material and Russia have agreed on these objectives Istorage facility, explaining that a lack of as well as the specific procedures necessary adequate storage capacity was delaying the to meet the second and third objectives. Russian nuclear-warhead dismantlement Negotiations continue on procedures to process. In response to this request, the meet the first objective. The challenge to Department of Defense, under its reaching complete agreement is ensuring Cooperative Threat Reduction (CTR) pro- adequate protection of sensitive nuclear- gram, committed to assist Russia in con- weapons information. structing a facility at Mayak to store fissile Russia is concerned that any measure- material from dismantled nuclear warheads. ments confirming the derivation of stored When completed in 2002, the initial fissile material from nuclear weapons could 12,500-container-capacity facility will be reveal sensitive information about its capable of storing plutonium from more nuclear weapons. To address this concern, than 6,250 nuclear weapons. the U.S. has proposed a suite of attributes Ongoing negotiations on a transparency that employ threshold measurements and regime for Mayak are intended to provide information barriers designed to protect confidence that: (1) the material stored at sensitive information (see page 6). Mayak is from dismantled nuclear weapons; (2) the stored material is safe and secure; Jessica Amber Kehl and (3) any material withdrawn from Mayak Office of the Secretary of Defense

Primary Storage Area

Artist’s concept of completed facility. Mayak Fissile Material Storage Facility under construction at Ozersk, Russia. (photo courtesy of DTRA).

12 ACNT • Technology R&D for Arms Control • Spring 2001 Treaties, Agreements, and Initiatives

Fissile Material Cut-off Treaty

The UN General Assembly, the Conference on Disarmament, and the 1995 and 2000 Non-Proliferation Treaty Review Conference have all endorsed the negotiation of a ban on the production of fissile material for nuclear weapons and other nuclear-explosive devices, but these negotiations are unlikely to begin in the near future. FMCT States Parties would be prohibited from producing highly enriched uranium and plutonium for nuclear weapons. They would also be obligated to accept measures to verify that all fissile material produced after the cut-off date is not used for nuclear weapons or other nuclear-explosive devices and to detect undeclared production. These obligations would be non- discriminatory. The U.S. has identified verification arrangements that it believes would be effective and efficient. Under these arrangements, “fissile material” would be defined as plutonium-239 and uranium enriched to 20% or greater in the isotopes 235U and 233U, separately or in combination, and any material containing one or more of the foregoing. “Produced” would be defined as the separation of irradiated nuclear material and fission products, or increasing by a process of isotopic separation the abundance of 235U or 233U in uranium or 239Pu in plutonium. (Incidental isotopic or fission-product separation resulting from chemical In the Mayak Storage Facility, a crane lowers the fissile material into a processes would not be considered production.) “nest,” a cylindrical space several meters in length in which the The verification regime would consist of routine monitoring of AT-400Rs are stacked. declared production facilities to verify State Party’s declarations, envisioned to be carried out by the International Atomic Energy Agency; consultation and fact-finding to resolve questions about correctness and completeness of, or inconsistencies related to, information provided by a State Party; and nonroutine inspections to resolve questions regarding possible undeclared production.

ACNT • Technology R&D for Arms Control • Spring 2001 13 Treaties, Agreements, and Initiatives

Fissile Material Transparency Technology Demonstration

delegation of Russian officials visited System with Information Barrier technology Los Alamos National Laboratory from (AMS/IB). When fully developed, this sys- AAugust 14–17, 2000 to observe a suc- tem will protect sensitive information while cessful demonstration of a new technology providing increased confidence that U.S. for monitoring nuclear materials removed and Russian fissile materials have been from military programs. By combining properly certified, packaged, and stored. innovative data barriers and a simple yes/no The technology being developed measures display, U.S. scientists assured the Russian the radiation emitted by the fissile materials. delegation that the nuclear-material sample These radiation signatures give observers being tested had the declared bomb-grade confidence that the packages contain fissile characteristics. material. The attributes evaluated included The Russian delegation, headed by a rep- the presence of plutonium, plutonium iso- resentative of the Ministry of Atomic topic ratio, mass, absence of plutonium Energy, observed the Attribute Measurement oxide, symmetry, and age of the plutonium. During the demonstration, these attributes Open Mode were measured with commercially available, high-resolution gamma and neutron detec- Sample Isotopics? Mass? No Oxide? Pu Present? Symmetry? Age? tors assembled by a multi-laboratory team. ZPPR plates in Data from the detectors were sent to a com- “dumb bell” putational block where they were analyzed configuration and compared to threshold values. A pass/fail signal was sent through information protec- Large oxide tion technology to a display with a series of sample red and green lights. No sensitive data were upright emitted from the measurement system. Secure Mode The Departments of Energy and Defense and the Defense Threat Reduction Agency Weapon are jointly developing the technology to component ensure the safe and secure storage of excess Large oxide fissile material from the Russian nuclear- sample on weapons program. U.S. and Russian officials its side discussed the joint development of a similar system in Russia. The technology demonstrated to the Russian delegation at Los The output lights from the various tests are simple to read. The easiest way to interpret the results is to Alamos was part of the Cooperative Threat ask the following questions. Does this fissile material sample meet the isotopic ratio criteria for weapons- Reduction program in conjunction with the grade plutonium? Does this sample mass exceed an agreed-upon threshold? Is plutonium oxide absent? Mayak Fissile-Material Storage Facility in Is plutonium present? Is the plutonium in a symmetrical shape? Is the plutonium older than an agreed- Ozersk, Russia. upon date? A green light indicates a positive answer to the question while a red light indicates a negative The U.S. and the Russian Federation answer. The samples tested were Zero Power Physics Reactor (ZPPR) plates; fuel-grade plutonium metal; share a common interest in maintaining 1.75 kilograms of plutonium oxide; and a nuclear-weapon pit. The equipment performed flawlessly, giv- and improving the safety and security of fis- ing all the proper responses in each test. sile material. Related work being carried out under the Cooperative Threat Reduction program with Russia includes the provision of over 12,000 transportation and storage containers for Russian fissile material and construction of the storage facility in

14 ACNT • Technology R&D for Arms Control • Spring 2001 Treaties, Agreements, and Initiatives

Ozersk. When complete, the storage facility follow-on to this initiative could be will be capable of the safe, secure, and envi- installed in the storage facility as an integral ronmentally sound long-term storage of at part of a joint monitoring system. least 30 metric tons of ﬁssile material excessed from Russia’s weapons program. Larry Avens An AMS/IB system jointly developed as a Los Alamos National Laboratory Both photos show the Attribute Measurement System with an Information Barrier used in the Fissile Material Transparency Technology Demonstration, August 2000. The nuclear-material package is contained in the blue Neutron Multiplicity Counter (right photo, center). The shielded electronics rack contains the computational equipment that analyzes the data (left photo). The readout display is mounted on top of the electronics rack (see illustra- tion, previous page.

Attribute Measurement System with Information Barrier Technology

Many of the attributes discussed in the “Treaties, Agreements, and Initiatives” section can be measured using traditional, nondestructive assay methods. Although these measurement techniques are well established, they become problematical if the item being measured is classified. Because useful radiation data generated from a classified item is generally classified, the data must be protected and not displayed directly during a measurement. An information barrier (IB) that protects the classified information must perform two functions:

1. The IB prevents the release (either accidental or intentional) of classified information. 2. The IB, at the same time, provides confidence that the measurement systems are functioning correctly and that the unclassified display reflects the true state of the measured item. (This is often referred to as the “authentication problem.”)

An Attribute Measurement Systems incorporating an IB was shown to a Russian Federation audience in August 2000 at the Fissile Material Transparency Technology Demonstration. In this demonstration, hardware and software combined with procedures addressed both requirements of the IB. The IB was designed with the needs of authentication in mind. Each element of the measurement system (including the IB) should be simple and easy to inspect and should not have any extraneous functions. If the measurement system is composed of simple building blocks, or modules, then the function of each element can be well deﬁned. Similarly, if each of the protective features is simple, then it is straightforward to verify that the protective functions of the IB are operating as speciﬁed.

Duncan MacArthur Los Alamos National Laboratory

ACNT • Technology R&D for Arms Control • Spring 2001 15 Treaties, Agreements, and Initiatives

The Trilateral Initiative: Attributes Veriﬁcation

n September 1996, the Secretary of years, technologies and procedures to per- Energy, the Russian Federation’s mit inspections of these materials can IMinister of Atomic Energy, and the make an important contribution to excess Director General of the International materials verification. Atomic Energy Agency (IAEA) came togeth- The challenges presented by the Trilateral er under the Trilateral Initiative to explore Initiative are in fact common to many the technical, legal, and financial issues potential arms control and arms reduction surrounding IAEA inspections of nuclear treaties and agreements, including future materials removed from defense programs. arrangements that might involve warhead The technical work has focused primarily dismantlement transparency, verification of on developing approaches that would per- plutonium disposition in the U.S. and mit the IAEA to conduct its inspections Russia, and U.S. inspections of Russian without violating Article I of the Non- materials to be stored in the Mayak Fissile Proliferation Treaty (NPT), which forbids Materials Storage Facility (see page 12). the sharing of nuclear weapons informa- The Department of Energy’s International tion with nonnuclear weapons states. Safeguards Division was tasked with sup- Opposing this absolute requirement porting a Trilateral Initiative working group. (which is also codified in the U.S. Atomic A team with members from Lawrence Energy Act) is the need for the IAEA to Livermore, Los Alamos, Pacific Northwest, conduct credible, independent inspections and Sandia National Laboratories held a to assure the world that excess nuclear number of workshops and technical meet- materials removed from defense programs ings with Russian and IAEA scientists to are not returned to nuclear weapons. develop an approach to inspecting excess Because much of the excess materials are nuclear materials with their inherent classi- currently classified and will not be con- fied characteristics. Key to the concepts that verted to unclassified forms for many emerged from these meetings was the idea of an information barrier concept (see page 6). The challenge was to take this information barrier and turn it into a workable instrument that satisfies very stringent security requirements. The Prototype Inspection System with an Information Barrier was the first realiza- tion in hardware and software of a measurement system with an information barrier for fissile-material transparency measurements. The working group focused on a system capable of determining plutonium mass and the presence of weapons- quality plutonium in sealed storage containers. Plutonium mass is measured using neutron-multiplicity coincidence counters (see page 37). Gamma-ray spectrometry determines the ratio of 240Pu to 239Pu. Both sets of these classified measurements

The Trilateral Initiative working group has participated in several workshops and technical demonstrations in an effort to identify challenges and difﬁculties in the IAEA inspections.

16 ACNT • Technology R&D for Arms Control • Spring 2001 Treaties, Agreements, and Initiatives

are then compared with unclassified The team, working with a larger support- threshold values, determining if containers ing cast from DOE/NNSA, successfully of classified materials should be accepted demonstrated its information barrier con- into IAEA verification. This “attributes” cepts, and as a result, has begun the next verification approach results in simple one- phase of the Trilateral Initiative’s technical bit information (yes/no) for the inspectors efforts: the development of technical through an information barrier. requirements and specifications for the The conventional wisdom of most partici- actual inspection systems. It is noteworthy pants early on was that an information bar- that in the Trilateral Initiative’s consulta- rier would be implemented in software; tions, the Russian representatives have however, authentication of software to repeatedly pointed to the success of the ensure both security and IAEA independence technical working group and have supported was recognized as a formidable challenge. ongoing development of the next phase of The DOE/NNSA drew on the expertise of information barrier technology. their U.S. colleagues, the IAEA, and the Russian nuclear institutes (All-Russian James Tape Scientific Research Institute of Experimental Los Alamos National Laboratory Physics, All-Russian Scientific Research Institute of Technical Physics, and the Institute for Physics and Power Engineering). The team developed a modular concept that clearly separates data acquisition from “computational block” (where measurements are compared to the attribute thresh- olds). This modular concept includes a data barrier implemented in hardware. This ensures that only yes/no information is transmitted outside the shielded enclosure containing all of the electronics and computers (and their classified information). Other concepts included “volatile” memory, a “security watchdog” that removes power and thus erases data if the enclosure is opened or if tampering is detected, and the potential to operate the system in “secure” and “authentication” modes. In the authentication mode, the IAEA can calibrate with unclassified nuclear materials to ensure that the instrument will perform properly in the secure mode, permitting independent The prototype Inspection System with an Information Barrier was demonstrated at Los Alamos National authentication of the instrument by the Laobratory to technical delegations from the Russian Federation and the IAEA in late June 1999. On the IAEA. The totality of these ideas is unique, left is a neutron multiplicity counter to determine plutonium mass. On the right, in an anodized shielded providing the required information barrier enclosure, is a high-resolution gamma-ray detector to determine the presence of “weapons-quality” that is both flexible and relatively straight- plutonium. In the center background is an equipment rack containing the data-aquisition equipment and forward to implement. the computers that controlled data aquisition and analyzed the data. In the center foreground is the small box with red and green lights that provided the pass/fail indications for the items measured during the demonstration.

ACNT • Technology R&D for Arms Control • Spring 2001 17 Treaties, Agreements, and Initiatives

U.S.–Russia Plutonium Production Reactor Agreement

hen-Vice-President Gore and then- take place twice a year after the first Russian Prime Minister Chernomyrdin signed declaration. Tthe Agreement Between the Government Twenty-four shut-down production reac- of the United States of America and the tors are covered under the PPRA at three Government of the Russian Federation sites in Russia (Ozersk, Seversk, and Concerning Cooperation Regarding Plutonium Zheleznogorsk) and at two sites in the U.S. Production Reactors on September 23, 1997. (Hanford, Washington and Savannah River, The Plutonium Production Reactor South Carolina). Currently in the second Agreement (PPRA) requires the implementa- year of monitoring, the monitoring mea- tion of measures to ensure that plutonium sures at the shut-down reactors include the production nuclear reactors currently shut installation of seals or visual monitoring at down in both countries do not resume oper- the reactors not deemed to have been irre- ation. Additionally, the last three Russian versibly dismantled. Annual monitoring vis- production reactors will be converted to an its are made to the shut-down reactors to operating mode that does not produce ensure that they remain shut down. weapons-grade plutonium. Plutonium oxide The PPRA estimates that between 4.5 and The U.S. is exploring the use of produced in the interim from the operating 9 metric tons of plutonium oxide will be the Russian Greenstar data- reactors’ spent fuel will be monitored to monitored. Monitoring provisions for the acquisition card as part of the ensure that it is not used in weapons. plutonium oxide include checking tags and radiation-monitoring equipment According to the Agreement, measurements seals on containers in storage, as well as for the PPRA. measurements on a random sample of the containers, to determine if the mass of the container is as declared and whether the material is “weapons-grade” (i.e., comes from low-burnup fuel) and has been newly reprocessed. To date, no monitoring of plutonium in storage has occurred. The material is considered weapons-grade if the plutonium isotopic ratio of 240Pu/239Pu is less than 0.1. Determining the elapsed time (age) since the plutonium was chemically purified tells us whether the material is newly processed (see page 31). Both attributes will be measured with a high-purity germanium gamma-ray detector and analyzed with a standard isotopics code. It has been proposed to measure the mass of the plutonium with a neutron multiplicity counter (see page 37). The isotopics of plutonium is classified information to the Russians; therefore, this has resulted in the concept of information barriers to collect—but not reveal—classified instrument data, process those data, and pass an unclassified yet meaningful result to

18 ACNT • Technology R&D for Arms Control • Spring 2001 Treaties, Agreements, and Initiatives

A Core-Discharge Monitor

The core-discharge monitor, pictured, determines when nuclear material moves from one place to another. This material is characterized as irradiated fuel pellets or some other material, providing data for nuclear an inspector (see page 6). Measurements safeguards or control and accountability. The result in yes/no answers for declared mass, core-discharge monitor is based on a subset isotopics, and age since chemical separation. of hardware used by DOE and the Monitoring—after the reactors cease pro- International Atomic Energy Agency for secu- duction of weapons-grade plutonium— rity and safeguards applications, respectively. would confirm the composition of fuel The redundant system runs in an loaded and verify that fuel is not discharged autonomous, unattended mode for the early. This is accomplished through the inspection period. measurement of random samples of fresh This application quantifies the amount of fuel and the installation of a monitoring gross neutron or gross gamma radiation, but it device in the fuel discharge area to detect does not give quantitative information on the reactor fuel discharges. The reactors will be nuclear material present. Ratios of gross radia- shut down after fossil-fuel plant replace- tion determine the type of material flowing ments are operational or no later than at past the detector, e.g., irradiated fuel or poi- the end of their normal lifetimes, consistent son slugs. Sampling time is in fractions of sec- with prudent safety considerations and onds. A cadmium–telluride detector provides a amendment of the PPRA. medium-resolution gamma spectrum of the A Joint Implementation and Compliance material as it flows by. It does not measure the Commission (JICC) oversees implementa- isotope content, but rather determines its tion of the PPRA’s provisions, resolves any presence. This is helpful in determining differ- issues that may arise, and considers addi- ences among the various types of materials. tional measures to promote the objectives In this application, the speed at which the of the Agreement. The JICC has met four fuel moves is a challenge to the design. Field times: December 1997, October 1998, experience on second-generation equipment February 2000, and September 2000. exceeds four centuries of continuous system Technical discussions to work out some operation with fewer than five documented operational details were held at Los Alamos cases of loss of safeguards’ continuity. National Laboratory in November 2000. A PPRA-specific demonstration is being pro- 1024-ch posed at Lawrence Livermore National Multichannel Laboratory. Integrated detectors that are eas- Reactor Analyzer Channel Integrated ier to replace and easier to authenticate will Circuit be demonstrated. The isotopic ratio of 3He 240 239 Pu/ Pu, age since chemical purifica- CdTe Fuel tion, and mass of plutonium will be mea- Detector sured.

Michele Smith U.S. Department of Energy Mini-GRAND Mini-GRAND Zachary Koenig Lawrence Livermore National Laboratory

James Halbig Los Alamos National Laboratory

ACNT • Technology R&D for Arms Control • Spring 2001 19 Treaties, Agreements, and Initiatives

Future Arms Reduction Initiatives

he United States government is cur- on issues related to the transparency of rently considering a broad range of strategic-nuclear-warhead inventories, the Tpossible nuclear arms reduction mea- storage and handling of nuclear materials, sures consistent with our evolving national and the destruction of strategic nuclear war- and international security context. heads. The details and technologies to be Scientists and engineers have been develop- employed for many of these arms reduction ing a flexible and robust set of monitoring and transparency measures have yet to be technologies to support the measures based worked out or are under active negotiation. on guidelines derived from previous accords It is anticipated that these measures will and statements. require the development of innovative tech- Previous agreements and accords nological solutions. A new type of require- accounted for deployed warheads as attrib- ment associated with many of these mea- uted to their delivery systems. Thus, deliv- sures and agreements is the simultaneous ery-system reductions were the primary need for information barriers to protect the focus of U.S.–Russia verification activities. host country’s sensitive information and While future reductions may continue to authentication technologies to provide the focus on warhead-delivery systems, there is monitoring party with the confidence that an interest in accounting for the destruction measured data can be trusted. The major of both the warhead and the nuclear mate- areas of consideration for these applications rials associated with that warhead. If the include warhead and special nuclear materi- nuclear materials are not mechanically or al identification based on radiation detec- chemically altered, they would be placed in tion, warhead and material monitoring, long-term, monitored storage facilities tamper-indicating devices such as tags and designed to ensure that the materials are seals, and technological alternatives to radi- not re-used for defense-related purposes. ation detection. Different technologies need In an effort to promote the irreversibility to be developed to support measurement of reductions and ensure the international instrumentation ranging from field size, community that the U.S. is meeting its point-of-use equipment to large stationary Nuclear Non-Proliferation Treaty (NPT) installations. commitments, several U.S. and joint Carolyn Pura U.S.–Russia programs are currently focused Sandia National Laboratories

U.S. and Russian scientists discuss the Fissile Material Transparency Technology Demonstration at Los Alamos National Laboratory in August 2000.

20 ACNT • Technology R&D for Arms Control • Spring 2001 Treaties, Agreements, and Initiatives

The Comprehensive Test Ban Treaty The Comprehensive Test Ban Treaty (CTBT) intended to mask the signals) and to distin- bans -12any nuclear explosion anywhere in the guish them from other sources. The treaty calls world. The product of four decades of multilateral for networks of atmospheric, underground, effort, it was opened for signature in September and oceanic sensors integrated into an 1996. To enter into force, the CTBT has to be rati- International Monitoring System. Data flows ﬁed by the 44 nuclear-capable states that formally continuously between National Data Centers participated in the 1996 Conference on (NDCs) and an International Data Center Disarmament who possess nuclear power and (IDC). The center at Patrick Air Force Base ana- research reactors as listed in the treaty. As of lyzes the U.S. data. The IDC processes the data February 2000, the CTBT has been signed by to produce event bulletins and to screen out 155 nations and ratiﬁed by 53. Although the U.S. events that are very unlikely to be nuclear Senate voted in 1999 not to ratify the treaty, it explosions. The NDCs are responsible for remains on the Senate calendar and could be determining if an event violates the treaty. voted on again at any time. Former Secretary of Monitoring is complicated by similarities State Madeleine Albright recently announced the between the effects from nuclear explosions and appointment of General John Shalikashvili to the effects produced by non-nuclear sources. “construct a path that will bridge any differences Also, signals are distorted or blurred as they pass and ultimately obtain Senate advice and consent through geologic structures. To meet the chal- to the treaty.” lenge, work is needed on data analysis techniques The challenges in CTBT monitoring are to that will ensure timely assessments of events and detect very-low-yield nuclear explosions (as data collection that will calibrate the sensor net- well as any conducted under conditions works to account for geologic structures.

Jay Zucca Lawrence Livermore National Laboratory

ACNT • Technology R&D for Arms Control • Spring 2001 21 Facilities

The Pantex Plant

The Pantex Plant is America’s only jected to electrical and/or explosive testing. nuclear-weapons assembly and disassembly Evaluation of warheads using this disassem- facility. Located 17 miles northeast of bly and inspection process is a part of Amarillo, Texas, Pantex is centered on a NNSA’s Stockpile Stewardship strategy 16,000-acre site. The plant’s primary mission designed to ensure high confidence in the is stockpile stewardship of U.S. nuclear safety and reliability of the weapons stock- weapons. Operations include assembly, disas- pile without nuclear testing. sembly, refurbishment, maintenance, modifi- With the U.S and Russia reducing their cation, and evaluation of nuclear weapons, nuclear-weapons stockpiles in accordance plus interim storage of plutonium pits. with Presidential declarations, Pantex plays a Pantex was originally a conventional vital role in this effort. Most of the weapons bomb plant for the U.S. Army during World sent to Pantex for disassembly were original- War II. Ten months after Pearl Harbor, the ly assembled at Pantex. The dismantlement first bomb came off the assembly line. As of nuclear weapons is not a new process as with many World War II-era munitions NNSA has disassembled approximately plants, Pantex was deactivated after the war 60,000 nuclear weapons over the years ended. In 1951, at the request of the Atomic through all its predecessor agencies. Energy Commission (now NNSA), the Army reclaimed the plant for use as a nuclear- Interim Plutonium Storage weapons production facility. By 1975, The bulk of nuclear-material parts disas- Pantex was the only facility for the assembly sembled from nuclear weapons is traditional- and disassembly of nuclear weapons. In ly returned to the NNSA plants that original- September 1991, the last new nuclear ly manufactured the parts. However, due to weapon was assembled. the end of plutonium processing at the To maintain the reliability of the nation’s Rocky Flats facility in Colorado, Pantex weapons stockpile, a number of randomly serves as the storage site for the plutonium selected warheads from all active systems "pits." A pit—the core of a nuclear weapon— are removed from the stockpile and contains plutonium hermetically scaled in a returned to Pantex each year for surveillance, metallic shell. testing, and evaluation to determine if the As an assembly-and-disassembly facility, components are in good working order. Pantex has long had the capacity to stage pits The weapons are disassembled and as they were coming and going. As the stor- inspected. Then certain components are age site, Pantex will store all U.S. pits in assembled into test configurations and sub- excess of 20 years or until final disposition for the pits is determined. Arms Control It is conceivable that a future arms control treaty will directly affect Pantex. In preparation for this possibility, Pantex has hosted a series of NNSA-sponsored initiatives to evaluate technologies that have been designed by NNSA’s national laboratories that could be used in future arms control treaties.

Leigh Bratcher Pantex Plant

22 ACNT • Technology R&D for Arms Control • Spring 2001 Facilities

Savannah River Site’s K Area Material Storage Project

he Savannah River Site (SRS) is ready- The KAMS project was chosen as the ing a storage facility for storing national solution for several reasons: Tweapons-grade plutonium from dis- • The facility underwent stringent, well- mantled nuclear weapons. The K Area documented earthquake and structural Material Storage (KAMS) project provides a upgrades during a restart campaign in the place to store excess plutonium in the years early 1990s. before new disposition facilities come on • It is a robust building, constructed of line at SRS. The Rocky Flats Environmental concrete walls many feet thick. Technology Site, for instance, can save mil- • Much of the security infrastructure is lions of taxpayer dollars if it ships its pluto- already in place. Security in 105-K has nium out in the near future, rather than waiting for the new facilities. That material been enhanced, and access to the build- can be held in KAMS until the new disposi- ing is limited to essential personnel only. tion facilities are ready. • Necessary modifications were relatively In January 2000, Secretary of Energy Bill minor, compared to the cost benefit. Richardson issued a Record of Decision on Security is ensured through a system the Surplus Plutonium Disposition called a radio-frequency tamper-indicating Environmental Impact Statement designat- device (RFTID), developed by Sandia ing the SRS as the recipient of three new National Laboratories. This system uses plutonium disposition facilities. This makes thin, fiber-optic wires, used with storage SRS the nation’s cornerstone for excess plu- drum arrays, which indicate possible tam- tonium disposition. Shipments will start in pering with drum storage. 2001. The total capacity when completed Also, material balance and accounting is will be 3,000 shipping containers. handled via neutron multiplicity counters KAMS is the first step in preparing for the developed by Los Alamos National new disposition facilities. It is located in Laboratory for neutron detection. Operators 105-K, the building that formerly housed confirm the container record by a neutron the K Reactor, which produced nuclear multiplicity counter, built by Canberra. materials during the Cold War for nearly The transition from reactor production to four decades. It was the United States’ last plutonium storage closes a circle that began operating production reactor, shutting in the early 1950s, when K and four other down for the last time in 1992. reactors at Savannah River began producing plutonium and tritium for the national defense. Now, the nation’s plutonium from weapons dismantlement will be disposed of through three new operations planned for SRS—the mixed-oxide fuel fabrication facility, the plutonium immobilization facility, and the pit disassembly and conversion facility.

Frances Poda Westinghouse Savannah River Company Savannah River Technology Center

ACNT • Technology R&D for Arms Control • Spring 2001 23 Facilities

TA-18 at Los Alamos National Laboratory

he unique capabilities for handling quantities of nuclear materials. Further nuclear materials at Technical Area 18 required are the development and evalua- (TA-18), coupled with Los Alamos tion of new nuclear measurement and T detection technology. TA-18 can respond to National Laboratory’s expertise in experimental measurements and sensor develop- these requirements by providing a test bed ment, are vital to threat reduction programs to advance the technology for automated at every level. Future treaty-verification facility monitoring. Beyond materials and efforts and nonproliferation missions can- equipment, the training programs for law not credibly be conducted without such enforcement and other government person- facilities and expertise. Fundamentally, the nel are an essential part of the TA-18 mis- arms control initiatives will result in a sig- sion, as are criticality safety training and nificant buildup of excess nuclear weapons first-responder training. materials, leading to more densely config- Equipment for arms control, nuclear ured storage of these materials, i.e., to con- materials disposition, and waste manage- figurations of greater concern with regard to ment receive realistic tests and objective criticality. validation through the Los Alamos Critical Thus, as a first requirement, both arms Experiments Facility (LACEF) and other control treaties in particular and nonprolif- TA-18 capabilities. Some concepts and eration in general require the capability to equipment are developed at other DOE fabricate and characterize realistic nuclear locations and brought to TA-18 for evalua- assemblies constructed from Category I tion; others are developed at TA-18 with special nuclear materials. Among the contri-

“Trust but verify.” By using a simulated Soviet SS-20, instrumentation is developed for treaty veriﬁcation at the TA-18 facility at Los Alamos National Laboratory.

24 ACNT • Technology R&D for Arms Control • Spring 2001 Facilities

butions in this arena are the invention, own storage vault. An additional laboratory development, and technology transfer of is available for short-term use of special portal monitors, waste management, and nuclear materials for hands-on measure- hand-held radiation detection and analysis ments. The facility also provides linear equipment, as well as the development and accelerators, x-ray generators, and neutron evaluation of transparency regimes. generators, as well as the necessary expertise LACEF operations are centered around to assist experimenters. three specially designed laboratory build- ings that house critical assemblies with Richard Malenfant remote capabilities. Each building has its Los Alamos National Laboratory

Observations with sophisticated instruments on a surrogate nuclear weapon are necessary to develop mutual trust while preserving security.

ACNT • Technology R&D for Arms Control • Spring 2001 25 Facilities

The Radiation Measurement Facility at Lawrence Livermore National Laboratory

or over a quarter-century, as a result of like assemblies (NELAs) containing plutoni- concern for the health and safety of um and uranium parts and inert high those who handle nuclear weapons explosives, although its use is not limited to F those materials. and components, Lawrence Livermore National Laboratory has been active in mea- The standard instrument suite within the suring the intrinsic nuclear radiation emit- RMF includes high-purity germanium and ted by such assemblies. During this period, sodium iodide gamma-ray detectors; a high- Livermore has had a succession of specially accuracy dePangher Long Counter for neu- designed facilities wholly dedicated to high- tron flux; Bonner spheres for neutron spec- quality radiation measurements of compo- tra; and a variety of neutron- and gamma- nents and assemblies containing fissile dose measuring detectors. A wide variety of materials. Each of these has improved and commercial and NIST-standardized neutron expanded these capabilities. The latest of and gamma-ray calibration sources are these specially designed measurement facili- located in a shielded well in the bay floor. ties was opened for business in 1988. The Visiting experimenters do, of course, bring Radiation Measurement Facility (RMF) is their own equipment as well. unique in the United States and is comple- Programmatic areas of RMF use have mented by Livermore’s long experience in included nuclear-weapon assembly and disas- nuclear-radiation measurements. sembly, and other national security activities, Within the RMF, we can make high-quali- as well as analogous activities within the ty neutron and gamma radiation measure- Department of Defense. In the area of ments of sources containing fissile material nuclear arms control, the RMF has seen fre- with little corruption from room return or quent use by Livermore and visiting scien- background. Located in Livermore’s tists from other national laboratories as an Superblock, the RMF is specially designed to experimental facility for the development of permit "free-field" measurements of spectral verification and transparency technologies. A and dose fields around nuclear explosive- wide variety of items is available for experi-

The RMF’s main bay is 50 x 50 x 30 feet high. In the center of the bay, rising 12 feet above the floor, is a 20 x 20-foot platform. This low-scatter platform, with its aluminum-grating floor, keeps the radiation source and the various radiation detectors as far as possible from the concrete walls, floor, and ceiling of the bay. The structure of the platform minimizes the amount of reflected radiation seen by the detectors. The floor of the bay is available for measurements when radiation scattering and background are of less concern.

26 ACNT • Technology R&D for Arms Control • Spring 2001 Facilities

menters, including unclassified uranium and Plutonium Facility were used to demonstrate plutonium items as well as nuclear weapon candidate verification technologies for the components and NELAs. Trilateral Initiative. In this latter demonstra- The RMF has also hosted a number of tion, 60 participants and observers, includ- international experiments and demonstra- ing delegations from the Russia and the tions with technical experts from the IAEA, observed a variety of measurements of Russian Federation and the International plutonium attributes using unclassified Atomic Energy Agency (IAEA). In late 1994 sources of weapons-grade plutonium. and 1996, the RMF was the facility for joint U.S.–Russia experiments to explore candi- Don Goldman date transparency technologies using unclas- Lawrence Livermore National Laboratory sified sources of weapons-grade plutonium. In late 1997, the facility and the adjacent

Adjacent to the bay on the level of the measurement platform is a large control room Not all transparency measurements made in the RMF are nuclear. Workers from Paciﬁc where data collection and processing can take place. It also contains areas for mechani- Northwest National Laboratory perform early measurements using their electromagnet- cal and electrical bench-top work and small-group meetings. The platform and the ic induction method to determine the presence of plutonium in this AL-R8 ﬁssile mater- measurement region are easily observed from the control room. In this photo, ial storage container (see page 58). observers at the 1997 Trilateral Initiative Technical Workshop are gathered in the control room for refreshments and side conversations.

ACNT • Technology R&D for Arms Control • Spring 2001 27 Overview to Arms Control Technology About the Atomic Nucleus, Radioactivity, Fissile Materials, and Transparency

The Atomic Nucleus which the nucleus splits into two large frag- atter is composed of atoms. Atoms ments of nearly equal size accompanied by comprise a tiny nucleus with a posi- the emission of several energetic neutrons. Mtive electrical charge surrounded by Fission can also be induced. This common- a cloud of negatively charged electrons. The ly occurs when a neutron emitted from a fis- nucleus contains two types of particles of sioning nucleus interacts with another nucle- roughly equal mass—neutrons and protons. us to induce another fission and release more Neutrons have no electrical charge and pro- neutrons. The production of neutrons from tons have a positive electrical charge equal to successive fission processes of this kind is the electron charge. The number of protons called neutron multiplication. Fission neu- in the nucleus (called its atomic number, or trons are quite penetrating and can escape Z) determines the chemical element. For from plutonium in storage containers where example, all hydrogen atoms have one pro- they can be observed with a neutron detector. ton, all iron atoms have 26 protons, and all Another form of radioactive decay com- uranium atoms have 92 protons. Except for mon among the heavy elements is the emis- the very lightest elements, the nucleus has sion of alpha particles. An alpha particle is a about twice as many neutrons as protons. tightly bound unit containing two neutrons The nuclei of all chemical elements can vary and two protons. Alpha particles have short in their number of neutrons. These variations ranges and cannot escape from storage con- are called isotopes. tainers but are of interest because the inter- Isotopes are written with a superscript actions of these particles with light impurity α number indicating the total number of neu- elements [called alpha-n or ( ,n) reactions] trons and protons in the nucleus, followed by also produce energetic neutrons. the chemical letter abbreviation of the ele- All these neutrons are quite penetrating ment that identifies its atomic number. 235U and can escape from the plutonium in a is an important isotope of uranium (Z = 92) nuclear weapon or component, or from that has a total of 235 neutrons and protons. containers of bulk plutonium. Appropriately constructed radiation detectors can deter- Radioactivity mine the amount of plutonium present and For a nucleus to remain stable, it must other attributes by counting the neutrons. have a proper balance of neutrons and pro- Another radioactive emission of particu- tons. A nucleus with too many or too few lar interest to arms control is gamma rays. neutrons is unstable and seeks stability by Following radioactive decay, the resulting emitting particles—the process of radioac- nucleus is usually left with excess energy. tive decay, or radioactivity. All of the iso- This energy is typically released through the topes of the heavy elements above bismuth emission of gamma rays. Gamma rays are (Z = 83) are radioactive. electromagnetic radiation, like x rays, but A wide variety of subatomic particles is even more penetrating. Moreover, they have emitted during radioactive decay. Because of sharply defined energies and the pattern of their ability to escape from the interior of gamma-ray emissions is unique for each nuclear weapons or items in thick storage radioisotope—providing a nuclear signature. containers and be observed by external radi- As seen in this issue of ACNT, gamma-ray ation detectors, two of these emissions are signatures reveal a wealth of information relevant to arms control applications: neu- about the materials emitting them. This trons and gamma rays. includes such attributes as isotopic compo- For heavy elements, spontaneous fission sition, the amount of time that has elapsed is of considerable interest to arms control, since plutonium was last purified, and the particularly that of plutonium (Z = 94). presence of chemical impurities. Spontaneous fission is a decay process in

28 ACNT • Technology R&D for Arms Control • Spring 2001 Overview to Arms Control Technology

Fissile Materials experience with many items will build con- Nuclear fission can be stimulated by flood- fidence over time that the signatories are ing heavy elements with neutrons. Fissile living up to the agreements (see figure.) materials are defined as those rich in heavy Transparency measurements based on radi- nuclei that undergo nuclear fission when ation signatures are quite intrusive, as they exposed to slow neutrons. Materials with this can potentially reveal aspects of a nuclear property are the essential ingredients of weapon’s design. With this in mind, novel nuclear explosives and therefore of keen inter- approaches to making these measurements est to the arms control and nonproliferation are necessary. These include robust measure- communities. The two fissile materials of the ment techniques that function regardless of greatest interest are highly enriched uranium an item’s configuration and information bar- (HEU) and weapons-grade plutonium. In the riers that protect classified information, yet U.S., HEU is defined as uranium enriched to allow the inspecting party to authenticate greater than 20% in the fissile uranium iso- that measurement results are genuine. tope, 235U. The greatest interest is in HEU enriched to greater than 90% in 235U. Thomas B. Gosnell Weapons-grade plutonium contains more than Lawrence Livermore National Laboratory 90% of the fissile isotope, 239Pu. Technical methods to identify the presence and quantity of fissile materials are essential components in Weapons-grade proposed arms control regimes. plutonium Transparency Early efforts at nuclear arms control relied on “surrogates.” First, there were the test ban treaties verified with satellite Correct Correct mass imagery and seismic monitoring. Newer All of the age Weapons- Possible treaties, such as START I, focused on deliv- inspected weapons- grade ery vehicles and were verified with satellite party's origin plutonium material imagery and direct observation (tape mea- plutonium sures and plumb bobs). Even newer agreements involve nuclear Correct shape warheads as well as delivery vehicles and are not as easy to verify. One reason is that nuclear weapons, their components, and bulk fissile material are relatively small and easily moved or concealed. Another reason is that details associated with the design of A hypothetical example of how a transparency regime could gain confidence that the material in a sealed nuclear weapons are among the most close- storage container is plutonium removed from a dismantled nuclear weapon. To do this, we can conduct a ly guarded secrets of nuclear-weapons states. series of measurements that considerably reduce the likelihood that the material came from another As a result, the emphasis is now on “trans- source. The first measurement, represented by the large blue circle, identifies the material in the container parency measures” based on the inspection as plutonium. A second measurement, represented by the inner dark blue circle, identifies the plutonium of the weapon’s fissile materials, typically by as having weapons-grade isotopic abundances. To further increase confidence, three more measurements examination of their radiation signatures. are made. First, a significant mass of weapons-grade plutonium must be in the container—one or two Transparency measurements provide win- grams won’t do (green circle). Second, the material must be old enough to have plausibly been in a dows on the activities of the nuclear- nuclear weapon—we’re not interested in newly produced plutonium (purple circle). Third, the plutonium weapons establishments of the treaty signa- must be formed in the correct shape (yellow circle). The intersection of these three circles is the region of tories. While the agreements are not verifi- possible weapons-origin. Finally, if these measurements are extended to thousands of containers, the like- able in the strict START I sense, inspection lihood that a significant quantity of this material is not from weapons is further significantly reduced.

ACNT • Technology R&D for Arms Control • Spring 2001 29 Plutonium Using Low-Resolution, Gamma-Ray Spectroscopy To Detect the Presence of Plutonium

he simplest method for detecting pluto- Quantitative information regarding mass, nium non-invasively uses an energy etc., is beyond the scope of the technique. Tspectrum of the gamma rays emitted by Algorithms allow instruments to assert that plutonium. The technology is both simple the mass present has a lower bound, that is, and mature: a detector made of a scintillating greater than some value; however, that value crystal (commonly sodium iodide) is connect- may be much less than the actual amount ed to a photomultiplier that feeds signals to present. Furthermore, the method does not analog electronics and a multichannel analy- specify the isotopic composition of the pluto- zer. The result is a “low-resolution” spec- nium, except in a few, restricted applications. trum—many details of the photon energies For quantitative estimates of mass, neu- are washed out by the limitations of the scin- tron methods are preferred, despite their rel- tillator. The analyzer in turn passes informa- ative complexity and cost. To determine iso- tion to software that determines whether the topic composition, high-resolution gamma- energies of gamma rays are consistent with ray spectroscopy is preferred, despite its those of plutonium, again within the limita- logistical difficulties (notably cryogenic sup- tions of the low-resolution detector. port). Finally, situations in which plutoni- Commercially available instruments exist um is shielded by high-Z materials between for a wide range of isotope-identification the plutonium and detector pose problems functions, although most must be cus- of data acquisition and analysis. tomized for plutonium detection. One Research continues on detector materials example is the RANGER-PLUS system, modi- that will improve the energy resolution fied to incorporate information-barrier fea- attainable with portable, room-temperature tures, used in the fall 1999 measurement equipment. The payoff is increased confi- tests at Pantex (see page 6). This instru- dence that the material present is indeed plu- ment, originally devel- tonium, coupled with shorter counting times oped by Los Alamos and greater robustness in the face of shield- National Laboratory, was ing. DOE has supported research on cadmium transferred to Quantrad, a zinc telluride, a higher-resolution detector private company. material currently being incorporated into a This method is limited successor to the RANGER-PLUS system. in transparency regimes. Only the detection of plu- M. William Johnson tonium is possible. Los Alamos National Laboratory

A RANGER-PLUS system (lower right) and its accessories in its rugged transport case. This system is a low-resolution spectrometer that can detect and identify plutonium.

30 ACNT • Technology R&D for Arms Control • Spring 2001 Plutonium

Measuring Plutonium Isotopes and Age with MGA

lutonium metal parts, as well as chemi- The gamma rays from americium and cal compounds, are often transferred plutonium penetrating the container can Pand stored in stainless-steel containers. be analyzed to determine the age. MGA These containers are about 5 centimeters was designed to accurately count gamma thick. To identify the material in such con- rays and this determines the plutonium tainers, it is sometimes necessary to deter- isotopic content. mine the isotopic content. A standard way Several commercial companies sell is to analyze the gamma rays escaping from gamma-ray measurement systems that use the container. A gamma-ray spectrometer MGA. We are currently verifying the ability acquires a gamma-ray spectrum, which is of two of these systems (from different then analyzed with a computer program to American companies) to accurately deter- determine the isotopic composition. One of mine plutonium age. In our experiments the most accurate programs is Multi-Group with “old” (27-year-old) plutonium, we Analysis (MGA). Because MGA analyzes rela- have determined that the age of plutonium tively low-energy gamma rays, its usefulness in stainless-steel containers as thick as 1.27 is limited by the thickness of the container centimeters can be accurately determined. enclosing the plutonium. To verify the ability of MGA to accurate- A 1.6-kilogram plutonium sample was ly determine the age of freshly purified plu- analyzed with a commercially available tonium, we are currently preparing such a detector to see how accurately and rapidly sample. We will periodically measure it MGA could measure the isotopes of plutoni- over a period of months to verify the accu- um stored in stainless-steel containers of dif- racy of the age determined by MGA, as well ferent thicknesses. The plutonium was mea- as the maximum allowable thickness of sured at varying distances (0–2 meters) and stainless steel. count times (10 seconds–30 minutes). To determine the maximum allowable contain- Rodney J. Dougan er thickness, we measured a 0.4-gram pluto- Lawrence Livermore National Laboratory nium source, with containers ranging from 31.8 centimeters to 2.5 centimeters thick. Plutonium isotopes can be quickly and accurately determined for a large plutonium sample inside of a 1.27-centimeter-thick stainless-steel container. Counting times as low as three minutes were used to determine the 240Pu/239Pu ratio to within 5%. Plutonium Age It is sometimes necessary to know the age of plutonium (how long ago the plutonium was chemically purified) to determine when it was made, or when it was last chemically purified (for example, reprocessed reactor fuel). A signature of plutonium age is the 241Am content. Freshly purified plutonium has no americium, but as the plutonium ages, the 241Am content increases in a predictable way.

ACNT • Technology R&D for Arms Control • Spring 2001 31 Plutonium

Pu300, Pu600, and Pu900 Systems

eveloped especially for plutonium Determining plutonium age transparency, the Pu300, 600, and The key attribute measured by Pu300 is D900 systems measure attributes of age, defined as the amount of time elapsed classified plutonium items in heavy, closed since the plutonium was chemically puri- storage containers. They exploit higher ener- fied. Age is based on the decay characteris- gy gamma-ray emissions from plutonium tics of 241Pu and its daughters, 237U and than are traditionally used in nuclear safe- 241Am. We use the highest energy region of guards—particularly for the important attrib- the gamma-ray spectrum where a high-con- utes of plutonium age and weapons-quality fidence age measurement can be made— plutonium. The high-resolution spectrum between 325 and 350 keV. from a plutonium item is rich with informa- Freshly separated plutonium contains sev- tion and much can be learned about the eral plutonium isotopes, including 241Pu but item that produced it. For transparency mea- no 237U or 241Am. 241Pu is not stable and surements, we therefore only acquire data decays through two paths. The alpha decay from the narrow bands of the spectrum nec- branch goes to 237U, and the beta decay essary to determine the attributes. branch goes to 241Am. Both the 237U and the The Pu300, Pu600, and Pu900 systems 241Am subsequently decay to two identical include gamma-ray detectors with high- states in 237Np by gamma-ray emission, energy resolution, data-acquisition electron- including two intense gamma rays with ener- ics, simple computers for instrument con- gies between 325 and 350 keV (see figure trol and data analysis, and elements of an below). The plutonium decay gamma rays, information barrier (see page 6). These sys- including gamma rays from 239Pu, are collect- High-resolution gamma-ray tems were integrated into the attribute mea- ed with a high-purity germanium detector spectrum of an unclassified surement system for the recent Fissile and analyzed with the Pu300 code that item shows the spectral regions Material Transparency Technology resolves the peaks from 241Am and 237U from from which data are acquired Demonstration (FMTTD) (see page 14). the 239Pu peaks in the region. The number of for the Pu300, Pu600, and counts in these peaks gives us the emission Pu900 analyses.

106

105

104

103 Counts/channel 102

101 0 100 200 300 400 500 600 700 800 900 1000

Energy (keV) Pu300—Pu age, presence

Pu600—weapons-quality Pu, presence

Pu900—absence of PuO2

32 ACNT • Technology R&D for Arms Control • Spring 2001 Plutonium

rates of these gamma rays. The branching Determining the presence of plutonium 241 decay of Pu causes the levels that emit the Determining the somewhat redundant 332.4- and 335.4-keV gamma rays to be pop- “presence of plutonium” attribute is a by- ulated at different rates. Because these rates product of the Pu300 and Pu600 analyses. are a well-known function of time, they allow For the FMTTD, we required that the soft- us to uniquely determine the age. ware determine the presence of 239Pu peaks at 345.0 keV from the Pu300 analysis and Determining the presence of weapons- 645.0 and 658.9 keV from the Pu600 analy- grade plutonium sis. To declare plutonium’s presence, we The Pu600 method examines a narrow required that all of these peaks exceed the energy region of the plutonium spectrum underlying continuum by at least ﬁve stan- containing a complex multiplet of gamma- dard deviations. ray lines between 630–670 keV. Pu600 measures the relative amounts of the isotopes Determining the absence 240Pu and 239Pu. Because more than 99% of of plutonium oxide the mass of weapons-grade plutonium is The “absence of plutonium oxide (PuO )” 240 239 2 due to Pu and Pu, a low value (<0.1) attribute is a surrogate for the truly desired 240 239 of the Pu/ Pu ratio indicates the pres- attribute—the “presence of plutonium ence of weapons-quality plutonium. metal.” The consensus among several nation- The Pu600 analysis uses a variant of the al laboratories was that determining the pres- MGA code to determine peak areas in the ence of plutonium metal in a sealed contain- 630–670-keV energy region (see page 31). er is not technically feasible. Instead, noting 240 239 The Pu/ Pu ratio is proportional to the that the most probable alternate form would 240 areas of the Pu peak at 642.5 keV and the be PuO , and that detecting the presence of 239 2 Pu peak at 646.0 keV. oxygen seemed possible, it was decided to substitute an “absence of PuO2” attribute using the Pu900 system.

241Pu (14.35 y) β-(99.998+%) α 10 (0.00245%) 241Am

160.0148.6 103.777.1 (432.2 y) α(100%)

- 237U β (100%) (6.75 d) 64.8 368.6335.4 370.9 5 332.4 267.5208.8164.6 333/335-keV emission rate 59.5

237 0 Np 0.1 1 10 50 (2.14 × 106 y) Plutonium age (y) The branching decay of 241Pu shows the gamma-ray transitions The age of plutonium can be determined by the relative emission measured by the Pu300 method. rates of the 332.4- and 335.4-keV gamma rays.

ACNT • Technology R&D for Arms Control • Spring 2001 33 Plutonium

We focused initially on a 870.7-keV with nitrogen impurities in the oxide, gamma-ray photopeak that is absent when 14N(α,p). Subsequent work, first at Pacific the sample is a metal. The 870.7-keV Northwest and then corroborated by gamma ray is emitted during de-excitation Lawrence Livermore, involving PuO2 sam- of the first excited state of 17O. It was ini- ples of varying degrees of chemical purifica- tially thought that the mechanism that pro- tion, indicates that the latter mechanis duces this excited state was due to alpha dominates. Nevertheless, the presence of particles from the decay of plutonium, the 870.7-keV peak unequivocally indicates interacting with oxygen by coulomb excita- non-metallic plutonium. Because the PuO2 tion, an inelastic process, 17O(α,α’). It was sample used in the FMTTD exhibited a pointed out by workers at Pacific Northwest strong 870.7-keV peak, it was decided, due National Laboratory that another mecha- to time constraints, to use this indicator. nism is possible—alpha particle reactions Meanwhile, measurements at Lawrence Livermore demonstrated a possible alternative, the 2438.0- and 2788.8-keV peaks from 18 21 105 the O(α,n) Ne reaction. Unlike the 21 240Pu 239Pu 241Am 870.7-keV peak, Ne appears to be unam- biguously due to oxygen. The 21Ne lines are 4 10 weakly emitted, requiring care in the selection of measurement geometry and the use 103 of digital data-acquisition equipment to obtain a signal of adequate strength for an arms control regime. At publication time, Counts 102 the issues surrounding measurement of the presence of PuO2 by gamma-ray spectrometry were still unresolved. 101

Dan Archer, Thomas Gosnell, John Luke, and Les Nakae 0 10 Lawrence Livermore National Laboratory 635 640 645 650 655 660665 670

The 630 to 670-keV region of the gamma-ray spectrum shows its resolution by nonlinear regression into its isotopic constituents for the Pu600 method.

34 ACNT • Technology R&D for Arms Control • Spring 2001 Plutonium

Absence of Oxide— Neutron Multiplicity Counting

hen oxygen is present in a sample are present in the plutonium metal, (α,n) of plutonium, the alpha particles neutrons are also produced, and alpha is Wemitted by the plutonium react not zero even if no oxygen is present. Thus, with the oxygen nuclei to produce neutrons. using measurements of alpha to verify the These neutrons can be detected by neutron “absence of oxide” can be ambiguous. If multiplicity counting, which measures the alpha is near zero, we can be confident ratio of these (α,n) neutrons to the neutrons there is no oxygen present. However, if spontaneously emitted by the plutonium. alpha is non-zero, oxygen may or may not This ratio is called simply “alpha.” For pure be present. plutonium metal, alpha is zero. If oxygen is Greater confidence in the absence of present, alpha is non-zero. A near-zero alpha oxide is achieved by analyzing the gamma measurement indicates the “absence of rays emitted by the plutonium sample and oxide.” A measurement of alpha will never requiring that, in addition to alpha being be exactly zero because of the statistical close to zero, there is no measurable nature of neutron measurement. gamma-ray line attributable to the presence A problem with using alpha to indicate of oxygen. the absence of oxide is that other materials also cause excess neutrons to be produced. Diana Langner If elements of low atomic number (fluorine, Los Alamos National Laboratory boron, beryllium, magnesium, or chlorine)

Neutron multiplicity counting was demonstrated at a Trilateral Initiative Technical Workshop in 1997. Participants included the Russian Federation and the International Atomic Energy Agency as well as several DOE national laboratories. This neutron multiplicity counter is one of a pair developed to measure plutonium items in storage containers.

ACNT • Technology R&D for Arms Control • Spring 2001 35 Plutonium

Estimating Plutonium Mass via Singles Neutron Counting

very crude estimate of the plutonium neutrons escaping a object, which may con- mass in an object can be measured— tain neutron-absorbing materials, leading to Asubject to some extremely important an underestimate of the plutonium mass. limitations—by simply counting the gross More significant effects produce overesti- number of neutrons emitted by the object. mates of the mass; the most important are Instruments such as the INF detector (see neutron multiplication (always present sidebar, page 9) or its Russian analog count when chain-reacting material is present in neutrons at a fixed distance from the object. quantity) and emission of neutrons (pro- Following corrections for geometry and duced when plutonium is in intimate con- detector efficiency, the count rate estimates tact with light elements such as oxygen or a plutonium mass based on the plutonium beryllium). Neutrons are a particular prob- isotopic composition (known or assumed) lem in measuring plutonium oxide, particu- and the fact that the isotope 240Pu emits larly impure material, in which singles approximately 980 neutrons per gram per counting might overestimate the mass by as second, owing to spontaneous fission. much as a factor of 10. Any mass estimate thus derived, however, Finally, and perhaps most importantly, must be considered with caution, because isotopic sources such as 252Cf also emit neu- physics interferes with the direct relation- trons that cannot be distinguished from ship between the neutron count rate and those emitted by plutonium solely by sin- 240Pu mass. Detectors can only register the gle-neutron counting. Because of these concerns, singles counting cannot provide quantitative information on plutonium mass, or even definitive proof of the presence of plutonium, without corroboration by some other technique; the much more powerful neutron-coincidence counting is preferred for quantitative measurements. A very simple form of singles counting was demonstrated during the initial Mutual Reciprocal Inspections (MRIs) (see sidebar, page 9) exchanges, where the goal was merely to show that “a lot” of plutonium was present in a container. The instrument used in that demonstration was the NAVI-2 system, which incorporated a small neutron detector and also a gamma-ray detector to provide some evidence that the neutron- emitting material was indeed plutonium. This instrument has been enhanced considerably since the 1994 demonstrations. The NAVI-2 is used today in applications where general confirmatory measurements and not detailed mass information are required. Vayacheslav Yanov of the Russian Research Institute of Pulsed Techniques (center) prepares to make a plutonium mass estimate using a SRPS7 neutron detector during the 1994 joint experi- M. William Johnson ments with Los Alamos and Lawrence Livermore National Laboratories. Los Alamos National Laboratory

36 ACNT • Technology R&D for Arms Control • Spring 2001 Plutonium

Measuring Plutonium Mass by Neutron Multiplicity Counting

eutron multiplicity counting (NMC) detector’s performance parameters to plutoni- is a rapid, nondestructive assay um are adjusted using an unclassified pluto- Nmethod to passively measure plutonium standard. Provided that neither the nium. The International Atomic Energy neutronics of the sample packaging nor the Agency (IAEA) uses NMC to measure pluto- detector’s electronic configuration changes nium in Japan and to measure excess substantially, the detector is calibrated “for weapons plutonium in the U.S. It is also life.” For well-known detector designs, even used for domestic safeguards programs in the last step of measuring plutonium is avoid- the U.S., Europe, and Russia. ed, and Monte Carlo calculations can adjust Recently, NMC has been proposed for the detector’s performance parameters to plu- determining if the mass of a plutonium- tonium. If sample packaging changes and the bearing object is above or below a specified package geometry and material composition threshold. This arms control use focuses on are well known, calculations can also adjust the Trilateral Initiative, the Plutonium detector parameters for these changes. Production Reactor Agreement, and the A large counter was demonstrated to a Mayak Transparency Initiative. In these Russian delegation during a Trilateral arms control regimes, classified weapons Workshop at Lawrence Livermore National components or plutonium materials with Laboratory in 1997. This detector measured classified isotopic compositions must be bulk plutonium in sample sizes up to and examined. By making separate measure- including a 30-gallon drum. The twin of ments of single neutron events and coinci- this instrument at the Rocky Flats dence events that involve two and three Environmental Technology Site has been neutrons, the large measurement inaccura- used since 1995 by the IAEA to measure cies associated with plutonium mass deter- excess U.S. weapons plutonium. This mination from singles neutron counting are counter has also been used in a series of avoided (see page 36). experimental tests on U.S. weapons compo- The major advantages of NMC for these nents. Both of these instruments were cali- applications include its accuracy, robustness, brated entirely with unclassified materials. and calibration using unclassified reference The components measurements yielded the materials. Representative materials that would mass of the plutonium accurately to about be classified for these initiatives are not 6% in 30 minutes. required for this measurement technique. Typically, an NMC instrument is character- Diana G. Langner 252 ized using a series of Cf sources. Then, the Los Alamos National Laboratory

Experimental test measurements of plutonium-bearing weapons components using neutron multiplicity counting at the Rocky Flats Environmental Technology Site.

ACNT • Technology R&D for Arms Control • Spring 2001 37 Plutonium

Minimum-Mass Estimates of Plutonium

o confirm that inspected items are high-resolution spectra are unique to pluto- authentic, some arms control agree- nium, the spectral characteristics cannot be Tments may require measurements of reproduced by substituting other materials. nuclear weapons and their components. The exact computation of the intensity Inspected items should exhibit two chosen distribution of radiation emitted by a weapon attributes: that the mass of plutoni- nuclear weapon is a complex problem that um exceeds a declared threshold and that requires defining the source geometry. This the isotopic composition is consistent with creates an impasse because the plutonium weapons-grade material. A Minimum–Mass mass cannot be computed without knowing Estimate (MME) method confirms plutoni- classified information. Therefore, rather um mass threshold and isotopic composi- than attempting to replicate the actual tion using only a high-purity germanium source, the MME method describes a config- detector. The masses of all gamma ray-emit- uration that could produce the spectrum ting isotopes are estimated by fitting fea- using the minimum quantity of plutonium. tures in the spectrum associated with inten- In the hypothetical minimum-mass model, sities of the gamma-ray peaks and low-angle plutonium self-attenuation is ignored and scattering. The MME method provides high intervening materials are assumed to be confidence that the amount of plutonium is uniform. Given these assumptions, comput- at least as great as the MME, but the actual ing the intensities of gamma rays and low- amount of material may be substantially angle scattered photons is greatly simplified. greater if the plutonium is thick or if the The process of fitting the spectrum using shielding is nonuniform. Because features in nonlinear regression estimates the masses of

5 Based on a minimum mass esti- 10 mate model, the gamma-ray 239Pu 232U spectrum recorded for a 241 238 400–gram plutonium plate is Am U 240 226 compared to a computed spec- 5 Ra 10 Pu trum. The ﬁlled regions repre- 238Pu scatter sent the contributions from each of the isotopes included in the model. The gray region represents an empirical continuum. 105 The yellow region represents ounts / keV

photons that scatter at low C angles relative to the incident photon trajectories. Note that the region between 470 keV 105 and 590 keV is not represented, producing a discontinuity at about 470 keV.

105 320 360 400 440 620 660 700 740 780 Energy (keV)

38 ACNT • Technology R&D for Arms Control • Spring 2001 Plutonium

all the isotopes that emit gamma rays in produce good fits to spectra recorded for a two energy regions: 315 to 470 keV and 590 large range of nuclear weapons and their to 780 keV. These isotopes include 238Pu, components. 239Pu, 240Pu, and 241Am. A limitation of the MME method is that Among the sources evaluated using the the spectrum must exhibit well-defined MME approach was a 400-gram, 2.3-mil- peaks for the isotopes in plutonium. If the limeter-thick plate of weapons-grade pluto- gamma rays from plutonium are highly nium. The plate is surrounded by aluminum attenuated and if there is a strong continu- cladding, a steel drum, and various thick- um such as that produced by Bremsstrah- nesses of lead. Intervening materials have lung radiation, it may not be possible to little impact on the mass estimates. The plot determine either peak intensities or the con- compares one measurement to the comput- tinuum associated with low-angle scattered spectrum based on the MME model. The ing. Masses are also underestimated signifi- MME for 239Pu is about 90% of the actual cantly when the plutonium thickness mass for the measurements of the plutoni- exceeds several millimeters. um plate. The ratios of masses of the other isotopes to 239Pu are approximately correct, 238 Dean J. Mitchell though the relative uncertainty for Pu is Sandia National Laboratories large due to statistical uncertainties because only 0.05 grams of this isotope were present. Despite the simplifying assumptions, the MME method is sufficiently robust to

ACNT • Technology R&D for Arms Control • Spring 2001 39 Plutonium

Symmetry Measurements of Plutonium Objects

nder some circumstances, it may be symmetry measurements simultaneously by important to know if the object in a tapping the neutron counts from the eight, Ustorage container is cylindrically equally spaced neutron detector banks in symmetric. We test for cylindrical symmetry the multiplicity counter. by looking for an isotropic neutron radia- Another requirement for the FMTTD was tion field emitted by a plutonium object in that the measurements needed to be made a sealed storage container. Vitaliy Dubinin behind an information barrier (see page 6). of the All-Russian Scientific Research As a consequence, data acquisition and Institute of Experimental Physics (VNIIEF) analysis needed to be done by an unattend- suggested this method during the Mutual ed computer. Data from the eight scalers Reciprocal Inspections discussions in were automatically adjusted to reflect small Moscow in 1996 (see sidebar, page 9). efficiency differences in the eight detectors. Shortly after, the method was tested in joint The symmetry attribute was determined by U.S.–Russia experiments at Lawrence a simple analysis of the adjusted net (back- Livermore National Laboratory. ground subtracted) detector counts to find For these experiments, neutron counts the one detector that deviated the most from unclassified plutonium objects in sealed from the average value of the adjusted net storage containers were recorded following counts from all of the detectors. The incremental rotations of 15°. Ideally, if the absolute fractional deviation, s, about the item is cylindrically symmetrical, the neu- average was computed. tron count will be equal at each rotational The values of s were compared to an arbi- position. The 1996 experiments employed trary threshold value chosen specifically for free-field measurements conducted on a low- the FMTTD. To be declared asymmetric, the scatter platform in the Radiation Measure- value of s had to exceed 0.15. When this ment Facility (see page 26). occurred, a red light indicated a failure of For the recent Fissile Material the symmetry attribute; otherwise, a green Transparency Technology Demonstration light indicated adequate symmetry. (FMTTD), held at Los Alamos National Laboratory, a number of different plutoni- Thomas B. Gosnell um attributes needed to be measured simul- Lawrence Livermore National Laboratory taneously (see page 14). One of these attributes was the mass of the object in a storage Diana C. Langner container, measured by a neutron multiplic- Los Alamos National Laboratory ity counter. It was possible to conduct the

During joint experiments in 1996, U.S. and Russian detectors measured unclassiﬁed plutonium objects in storage containers. Total neutron measurements were made with a Russian SRPS7 detector (white box) and a U.S. SNAP-2 detector (blue box).

40 ACNT • Technology R&D for Arms Control • Spring 2001 Plutonium

0º 0º

270º 90º 270º 90º 0.6 0.81 1.2 0.6 0.8 1 1.2

180º 180º

(left) Polar plot of the neutron counts recorded from a plutonium sphere in an AT400R container during the 1996 joint experiments. The cir- cular pattern indicates the presence of a cylindrically symmetric object in the container. (right) This polar plot, produced from counts recorded from a thick plutonium disk, shows that the neutron ﬁeld is anisotropic, indicating asymmetry.

Pulse processing electronics

Multi- PC/104 NMC input Bus SBC scaler

Symmetry analyzer

Block diagram of the neutron symmetry measurement made during the FMTTD. The eight outputs from the neutron multiplicity counter are recorded on an eight-input scaler and then analyzed by the PC-104 single-board computer.

ACNT • Technology R&D for Arms Control • Spring 2001 41 Plutonium

Gamma-Ray Imaging To Determine Plutonium Shape

he design of a nuclear weapon Classification concerns are met in two requires careful choice of the shape of ways. The image resolution of such an Tthe nuclear materials used to generate instrument (number of centimeters per the explosion. As such, this shape is an pixel) can be adjusted in an easily verifiable important attribute to verify that material fashion, so that details of the plutonium in a disassembly stream originates from piece below a certain size are not visible. Or, weapons. The shape is so much a part of the inspection can be automated so that no the design that the details of the shape, i.e., image is ever displayed. Here, a computer exact dimensions, are classified. routine analyzes the data behind an infor- Nevertheless, if we could verify that the fis- mation barrier, responding with a “red sile material is of approximately the correct light” or “green light.” dimensions, this would strongly indicate that the component originates from a dis- Klaus-Peter Ziock assembled weapon. Such a measurement Lawrence Livermore National Laboratory can be made remotely on plutonium even if it is in a sealed, opaque container. To measure shape, we take advantage of the fact that plutonium, like other fissile materials, emits gamma radiation characteristic to that material. Gamma radiation is nothing more than high-energy light that can penetrate through a significant amount of matter. Plutonium glows with this radiation, much like a light bulb, with the difference being that the gamma rays will continue on through the walls of a sealed container as if the plutonium were in a clear vessel. With a suitable instrument, we can capture this “light” and make an image of the plutonium (see image). Although this image is at somewhat worse resolution than we might produce with a camera, it is sufficient to verify that the radioactive material in the container has the correct shape to meet transparency requirements, and—based on Using unclassified sources, a simple detector measured the gamma- the spectral information—that it is the ray flux at a number of positions outside the container. The different material claimed. readings generated an image. More advanced instruments can perform the same inspection at a distance of several feet in much shorter time intervals (well under an hour.)

42 ACNT • Technology R&D for Arms Control • Spring 2001 Plutonium

Autoradiography Adaptation Using Optically Stimulated Luminescence

unique, radiation-sensitive film com- generated by the OSL film cannot be seen bined with the property of optically with the human eye. It takes the OSL reader Astimulated luminescence (OSL) can to make the image visible. The OSL film is generate a low-resolution, two-dimensional reusable, can be erased, and does not gener- image of a nuclear weapon component in a ate chemical waste, as does ordinary radi- storage container. It can determine the ographic film. presence of nuclear materials and define To identify the presence of nuclear the attribute of a distributed versus point weapons components within containers, the radiation source. The film is read using a OSL film can discern the distributed-versus- reader that limits the resolution of the point source attribute. Computer calcula- images to no finer than 1 centimeter-by- tions of the OSL film placed behind a tin 1 centimeter pixels. collimator have demonstrated the ability to After exposure to a radiation field, the distinguish a point source from a distributed internal crystal structure of OSL materials is source. The tin collimator, if spaced every displaced, making the OSL materials sensi- one centimeter to collimate the intrinsic tive to light stimulation at specific frequen- radiation, measures the extent of the source. cies. When stimulated by green light, the Note the clearly different patterns generated material emits blue light directly propor- by the two types of radiation sources. tional to the radiation-field exposure (see Monitoring warhead materials and com- image). A specially constructed reader scans ponents during transport and warhead dis- the OSL film with green light, producing mantlement are envisioned. OSL film blue emitted light. accomplishes this without the intrusiveness The magnitude of the blue light is pro- inherent to conventional x-ray film radiog- portional to the radiation intensity. Light- raphy, without generating an intrusive, emitting diodes excite the OSL film, illumi- high-resolution image, and without generat- nating a 1 centimeter-by-1 centimeter area. ing a toxic waste stream. This large illumination area limits the resolution of the detection system to no better Steve D. Miller than 1 centimeter-by-1 centimeter. Unlike Pacific Northwest National Laboratory ordinary x-ray film, the radiation patterns

Computer calculation of the dif- 25 ferentiation of point and spherical Point Source sources. The collimator is 40 mm 20 Spherical Source long and 3.175 mm wide.

0 0 5 10 15 20

Interval Count Relative to Mimimum Collimator Interval Number

ACNT • Technology R&D for Arms Control • Spring 2001 43 Plutonium

Attributes and Templates from Active Measurements with NMIS

he Nuclear Materials Identification estimation. Template-matching compares System (NMIS) was developed jointly radiation signatures with known reference Tby Oak Ridge National Laboratory and signatures. For treaty applications, authenti- the Y-12 Plant for nuclear-material control cating the reference signatures along with and accountability. NMIS has been used storing and retrieving templates may be dif- with active neutron interrogation for both ficult. Attribute estimation, on the other highly enriched uranium (HEU) and pluto- hand, determines the fissile mass and other nium, and with passive neutron interroga- properties from various features of the radi- tion (no external source) for plutonium. ation signatures and does not store radia- In active techniques, fissile material— tion signatures. It does require calibration, stimulated by an external neutron source— which can be repeated as necessary. fissions, producing neutrons and gamma NMIS is now used to verify weapons rays. The time distribution of particles leav- components received and stored at Y-12 by ing the fissile material can be measured template-matching. NMIS also estimates with respect to the source emission in a two attributes, fissile mass and enrichment, variety of ways and with a variety of accel- for HEU metal. NMIS employs a 252Cf erator and radioactive sources. source of low intensity (<5 × 106 The data from interrogating nuclear neutrons/second) such that the dose at one weapons and components can be used in meter is approximately twice that of a com- two ways: template-matching and attribute mercial airliner at cruising altitude. Such a

Detectors Source

A typical active measurement of a fully assembled weapon at the Pantex Plant. The radioactive source, 252Cf in an ionization chamber, is on the right and four (at least one is required) detectors are on the left. Radiation emanating from the 252Cf source is transmitted through the weapon, scattered by the weapon, and induces ﬁssion in the weapon with the resulting radiation from all three processes reaching the detectors.

44 ACNT • Technology R&D for Arms Control • Spring 2001 Plutonium

source presents no significant safety con- now used at the All-Russian Scientific cerns, either for personnel or nuclear- Research Institute of Experimental Physics explosive safety, and has been approved for (VNIIEF) in Sarov, Russia, and another has use at the Pantex Plant on fully assembled been shipped to the All-Russian Scientific weapons systems. The same NMIS tech- Research Institute of Technical Physics nique has also been used for HEU metal at (VNIITF) in Snezhinsk, Russia. Recent pas- the Y-12 Plant, where both the fissile mass sive measurements were completed with and the enrichment were measured to ±2%. plutonium metal (1.77 wt%) at VNIIEF, Presently, three systems are routinely used showing that NMIS could estimate mass at Y-12 to confirm or verify the weapons and thickness of the metal. components in containers. This method was non-intrusively demonstrated to Russian vis- John T. Mihalczo and J. K. Mattingly itors in 1997 and 1998. One such system is Oak Ridge National Laboratory

The equipment setup for verifying HEU metal at the Y-12 Plant in Oak Ridge, Russian visitors watch a demonstration of the active interrogation technique. The 252Cf Tennessee. The 252Cf source is located near the part. The uranium mass and enrich- source and detectors are located around a container with a weapon component inside. ment of the HEU metal in birdcages were determined to within ±2%.

ACNT • Technology R&D for Arms Control • Spring 2001 45 Uranium

Determining the Presence of HEU with Passive Detection Methods

ighly enriched uranium (HEU) is one Even if 235U is detected, its presence alone of two fissile materials incorporated does not mean the uranium is HEU. To Hin nuclear weapons. During the Cold determine this, we must have some indica- War, hundreds of metric tons of HEU were tion of the enrichment. The other dominant produced by the U.S., Soviet Union, and isotope of HEU is 238U. Knowing the concen- other nuclear-weapons states. The ability to tration of both isotopes provides an estimate detect HEU in nuclear weapons and their of uranium enrichment. However, even if dismantled components is viewed by the 235U is detected, the dominant gamma rays U.S. policy community as a potentially from both isotopes are sufficiently well sepa- important transparency measure. rated in energy (186 keV for 235U and 1,001 Passive detection methods exploit signa- keV for 238U) that unknown relative attenua- tures intrinsic to the undisturbed material. tion of these gamma rays within the urani- To protect classified information, measure- um items and through their storage contain- ments are made outside the weapon or stor- ers precludes knowing their true relative age container. Nuclear weapons and compo- emission intensities. An exception is the nents in their storage containers are large, “enrichment meter” method that examines dense, and nonhomogeneous. Signatures the 186-keV peak and the adjacent continu- must be sufficiently penetrating to escape um to determine uranium enrichment (see from the interior of the weapon or container page 52). The method requires calibration and reach a detector. Signatures must also be against known standards, a condition unlike- High-resolution gamma-ray sufficiently intense to complete an inspec- ly to occur in arms control scenarios of spectrum of a 2.2-kilogram tion measurement in a reasonable period of increasing interest. These scenarios include spherical source of uranium time. For HEU, the only signature that heavily shielded HEU so that methods enriched to 93% in 235U. The approaches these criteria is from gamma rays detecting the 186-keV gamma ray are not 232U content is 100 parts per emitted by the radioactive decay of uranium applicable. trillion. The red, orange, and isotopes. Unfortunately, the signature of Because determining the presence of yellow areas indicate where the 235U is so weakly penetrating that simply shielded HEU by measuring its key isotopes most prominent peaks are locat- detecting shielded HEU—let alone quantify- is intractable in some arms control settings, ed for 235U, 238U, and 232U, ing it—is a task that ranges from straightfor- an indirect alternative signature is being respectively. ward to nearly impossible. explored—detecting the impurity isotope 232U. 232U does not occur in nature but is 106 235U introduced into HEU as a result of reprocessing uranium reactor fuel. In the 1960s, 5 10 uranium separated from irradiated plutonium production reactor fuel was introduced 104 into U.S. gaseous-diffusion plants to be enriched. As a result, trace quantities of 232 103 U were entrained in the gaseous-diffu- 238U sion cascades where they remain today, con- 2 taminating new feed stock as it is being

et counts 10 232 232 N U introduced into the cascade. U is found typically at the 100–200 parts per trillion 101 (ppt) level in U.S. HEU. We believe that a

0 similar circumstance occurred in the Soviet 10 Union. Evidence indicates that, during the enrichment process, 232U is preferentially 10–1 swept into the light isotope fraction that 0 500 1000 1500 2000 2500 3000 becomes HEU and amounts too small to Energy (keV) measure go into the heavy isotope fraction

46 ACNT • Technology R&D for Arms Control • Spring 2001 Uranium

that becomes depleted uranium. Therefore, is produced which decays to 232U and the presence of 232U in uranium is consis- remains in the plutonium. Arms control tent with that uranium being HEU. regimes must consider this possibility, but it 232U is part of the collateral radioactive may be of small consequence, because in decay series associated with the thorium this case, the presence of the 236Pu signa- decay series. The thorium series is one of ture is evidence of the presence of a fissile three sources of natural gamma-ray back- material. The final concern is that because ground radiation. This series begins with both natural thorium and depleted uranium 232Th and decays through a series of are plentiful and relatively inexpensive, radioactive daughters including 208Tl. A they could be placed in the sealed container number of emissions are associated with to spoof a measurement. The naturally this decay series, but the most distinctive is occurring thorium chain begins with the a highly penetrating 2,614-keV gamma ray very long-lived 232Th, which decays to emitted following the beta decay of 208Tl. 228Ra then to 228Ac before reaching 228Th. Because 232U enters the thorium series at A telltale clue that natural thorium is pre- 228Th, it too exhibits the 2,614-keV gamma sent is a cluster of gamma rays emitted by ray as a signature. Because of the relatively 228Ac in the neighborhood of 900 keV with short half-life of 232U and the short half- a 911-keV line being the most intense. lives of its daughters, it is readily observable A simple measurement technique would The naturally occurring in a gamma-ray spectrum. At the 100-ppt be to use a high-resolution gamma-ray radioactive-decay series of level, the 2,614-keV peak is of comparable detector to detect the penetrating 1,001-keV thorium is indicated in blue. height to the 1,001-keV peak from 238U in line from 238U to confirm the presence of The prominent gamma ray at 93% enriched uranium. uranium (but not HEU), the 2,614-keV line 2,614-keV (see figure on oppo- Two difficulties must be overcome in (consistent with the presence of HEU or site page) is emitted due to the 232 determining the presence of HEU by detect- possibly weapons-grade plutonium), and presence of U from a collat- ing 232U. First, gamma-ray emissions from the absence of a line at 911 keV to ensure eral decay series that enters the 228 232U are relatively weak, requiring large that the 2,614-keV line is associated with thorium series at Th. This 236 detectors and possibly longer measurement fissile material. Using the same measure- series also includes Pu. times than normally desirable. In the unlike- ment, determining the absence of a gamma ly event that a more satisfactory passive ray at 414 keV builds additional confidence 236Pu 2.9 y means of detecting shielded HEU is found, that the 2,614-keV line is associated with arms control regimes requiring the detection HEU rather than weapons-grade plutonium, of HEU must account for this difficulty. The 232U which might be desirable in some arms 72 y second difficulty is that the salient features control regimes. of the 232U signature, notably its association 208 228Th 232Th with the decay of Tl and its 2,614-keV 1.9 y 1.4×1010 y 232 Thomas B. Gosnell gamma ray, are not unique to U. Lawrence Livermore National Laboratory 228Ac The signature associated with the decay 6.1 h 228 224Ra 228Ra of Th and all of its daughters, including 3.6 d 6.7 y 208Tl, is found in natural background radiation due to traces of thorium ubiquitous in 220Rn the earth’s crust. Discrimination against 55 s background 2,614-keV gamma rays can be largely accomplished by massive shields 212 216 Po Po – –7 β around the detector and behind the object 3.0×10 s 66% 0.16 s α 212Bi under inspection. Another source of the 61 m 2,614-keV signature is weapons-grade pluto- 208Pb 212Pb nium. During the creation of plutonium, a stable 34% 11 h 236 208Tl trace quantity of the impurity isotope Pu 3.1 m

ACNT • Technology R&D for Arms Control • Spring 2001 47 Uranium

Active Interrogation for Monitoring HEU

issile materials in shielded configura- erators producing less than 6-MeV electrons tions are difficult to detect and charac- by relying on the fissioning properties of Fterize through passive measurements. In HEU. some cases, actively interrogating these Another approach involves interrogation materials with neutrons or energetic pho- with a pulse of deuterium-deuterium (d-d) tons determines some properties of such or deuterium-tritium (d-t) reaction-generat- materials. Active techniques may provide ed neutrons and measurement of the result- reliable ways to determine the presence of ing neutron intensity as a function of time highly enriched uranium (HEU) and other (the differential die-away curve) inside a attributes for arms control initiatives. As in moderating cavity containing the material. passive methods, an information barrier will This technique was originally developed at likely be required to protect sensitive design Los Alamos for waste assay and more information (see page 6). A principal R&D recently has been applied to monitoring goal is identifying signatures that can serve packages for the presence of special nuclear as attributes capable of being verified with materials. The detection and analysis tech- unclassified sources rather than using tem- nology is well developed for these applica- plates which require a “trusted” classified tions. Current work focuses on determining object to verify the measurement. signatures from weapon components and In an approach being examined at Los developing high-flux, long-lived neutron Alamos National Laboratory, an object is generators. The d-d reaction has a lower interrogated with pulses of high-energy cross section and thus an inherent lower photons (up to ~10 MeV) of sufficient ener- flux for a given ion current. On the other gy and intensity to produce neutrons from hand, long-life d-d generators can be devel- photofission or (γ,n) reactions in the fissile oped because the target deuterium is easily material. The resulting neutrons then replenished and no tritium handling is induce fission chains in the material. needed. Efforts to develop a high-current Neutron events detected in a high-efficiency ion source that can lead to a high-intensity detector are time-tagged and analyzed to d-d generator are underway at Lawrence determine a correlation signature. This sig- Berkeley National Laboratory. nature is a measure of the multiplication in Another technique being studied at the assembly and thus could discriminate Lawrence Livermore National Laboratory HEU from other materials. Electrons from employs an accelerator-based neutron gen- linac or betatron accelerators striking a erator to look above and below the 238U fis- high-density (high Z) target produce the sion threshold at ~1 MeV to determine the energetic photons. presence of HEU. To detect the neutrons A related technique has been developed from the induced fission of uranium, we are by Idaho National Engineering & looking at conventional scintillation detec- Environmental Laboratory in which the tors as well as thorium fission chambers. electron energy is varied from 8–12 MeV to Thorium fission chambers have the advan- produce a variety of interrogating photon tage of being insensitive to gamma rays and spectra. Characteristics of the resulting sensitive only to neutrons above 1 MeV. gamma rays and neutrons are measured as These advantages make their use a type of a function of the interrogation energy. The “information barrier,” which would be help- resulting data provides a signature unique ful for transparency measurements. New to a particular material. Results indicate a forms of thorium fission chambers are being possible complementary photonuclear developed to greatly enhance their intrinsic inspection method that uses electron accel- efficiency.

48 ACNT • Technology R&D for Arms Control • Spring 2001 Uranium

Major considerations in active approaches are personnel safety from intense interrogating radiation sources, nuclear-explosive safety if weapons are being interrogated, and operational effects that might include separate facilities and measurement schemes which are signiﬁcantly more complex than passive methods currently being considered. In summary, active approaches may offer the only high-conﬁdence means of meeting potential HEU monitoring requirements, but they will be more expensive and complex and have a greater operational impact than passive measures currently being considered for verifying plutonium.

Robert Scarlett Los Alamos National Laboratory The Active Interrogation Package Monitor can detect the presence of special nuclear materials in a few James Jones seconds even when the material is heavily shielded. Idaho National Engineering and Environmental Laboratory

Arden Dougan Lawrence Livermore National Laboratory

An inspection or veriﬁcation technology for highly enriched uranium (HEU) stored in containers at a storage facility. The system uses a transportable, electron accelerator and a neutron detection system. The technology can inspect stored material and containers entering or leaving the facility.

ACNT • Technology R&D for Arms Control • Spring 2001 49 Uranium

Blend-Down Monitoring System for HEU

he Highly Enriched Uranium (HEU) unit length of pipe). Fissile flow is traced Purchase Agreement between the U.S. through a blending tee by detecting time- Tand Russia provides for monitoring modulated fission fragments in the LEU the blending of HEU (500 metric tons) with stream at a detector downstream of the low-enriched uranium (LEU) to produce blending point produced by the neutron- commercial reactor-grade material for the source modulation in the HEU stream. U.S. (see page 10). The Blend-Down The enrichment monitor, installed down- Monitoring System (BDMS) was developed stream from the mass flow monitor, uses by DOE for the Russian facilities. It incorpo- gamma-ray spectrometry to determine the rates the fissile mass flow monitor devel- enrichment. The BMDS confirms mass flow oped by Oak Ridge National Laboratory and and traces fissile material. It is self-con- the 235U isotopic enrichment method devel- tained and fully automated, designed for oped by Los Alamos National Laboratory unattended operation. (see next page). The BDMS has been installed and was The system measures the flow rate of fis- successfully demonstrated in the Paducah sile material in process pipes, tracing the fis- Gaseous Diffusion Plant to a team of sile material from the HEU leg to the LEU Russian experts in 1998. In these tests, the leg. The mass flow monitor induces fissions enrichment varied from 1.5% down to in the fissile material and detects the 1.1%. It has successfully operated in the delayed gamma rays emitted by fission frag- Ural Electrochemical Integrated Plant (UEIP) ments at a detector downstream. The at Novouralsk in Russia, which has one unit induced fissions are modulated using a neu- operating. Another monitoring system has tron-absorbing shutter to create a time- been shipped to the Electrochemical Plant dependent signature detected by the down- at Zelegnogorsk for future installation. stream detectors. The mass flow monitor determines the flow rate by measuring two John Mihalczo, James Mullens, Jose March-Leuba, things: (1) the time required for the fission and Taner Uckan fragments to travel along a given length of Oak Ridge National Laboratory pipe (inversely proportional to the fissile- material flow velocity), and (2) an amplitude measurement proportional to the fissile concentration (in grams of 235U per

One leg of blending tee: A LEU Blending point 57Co HEU gamma-ray spectrometer using a 252Cf source measures the 186-keV gamma ray from BDMS 252Cf 235U to determine the 235U content of the HEU gas. It also measures the 122-keV gamma BDMS Enrichment Flow Flow source rays from a 57Co source to detector detectors modulator obtain the total uranium in LEU Blend-Down Monitoring System (BDMS) the gas. Both measurements product determine the enrichment of the product.

50 ACNT • Technology R&D for Arms Control • Spring 2001 Uranium

Blend-Down Enrichment Monitor for HEU

n February 1993, the United States and To measure the total amount of uranium the Russian Federation signed a bilateral in the gas, we use a technique called IAgreement for the U.S. to purchase low- gamma-ray transmission. A 57Co gamma- enriched uranium (LEU) derived from highly ray source is mounted on the pipe opposite enriched uranium (HEU) removed from dis- the detector. This source has a prominent mantled Russian nuclear weapons. The gamma ray with an energy of 122 keV. The Agreement calls for the establishment of NaI detector measures simultaneously the transparency measures that provide both par- 186-keV count rate and the 122-keV count ties with confidence that the nuclear non- rate. These two quantities are first measured proliferation objectives of the Agreement are with the pipe empty of process gas, then being achieved. The U.S. has the right to with UF6 gas present. The empty pipe mea- install non-intrusive, nondestructive assay surement determines the 186-keV count instruments to independently and continu- rate for the room background from nearby ously monitor the 235U enrichment at the pipes and uranium deposits on the pipe blend point (see previous page). interior. The difference between the 186- Los Alamos National Laboratory devel- keV count rates in these two measurements oped a detector to watch the enrichment of determines the 235U signal originating in uranium–hexafluoride (UF6) gas. The UF6 in the process gas. the HEU leg, at approximately 90% enrich- The change in the 122-keV count rate in ment, mixes with 1.5% LEU blend stock to these two measurements determines the produce LEU product in the 3–5% range. total amount of uranium present in the gas. The enrichment monitor measures the Combining the 186-keV measurement with ratio of 235U to the total uranium. Sodium the 122-keV measurement gives a value of iodide (NaI) is the photon detector. The iso- the enrichment of the UF6 process gas. tope 235U emits a prominent gamma ray with an energy of 186 keV. The count rate Phil Kerr of the 186-keV gamma ray depends on the Los Alamos National Laboratory number of 235U atoms in the gas, which in turn depends on the uranium enrichment and the gas pressure.

The Ural Electrochemical Integrated Plant is located in Novouralsk, Russia, where a per- manent ofﬁce supports DOE’s transparency activity.

ACNT • Technology R&D for Arms Control • Spring 2001 51 Uranium

Portable Equipment for Uranium Enrichment Measurements

ince January 1997, the HEU The system measures the 235U enrich- Transparency Implementation Program ment by using an enrichment meter. In has used nondestructive assay (NDA) this method, the enrichment of the sample S 235 equipment to determine the U enrich- is proportional to the rate of the 186-keV ment of highly enriched uranium (HEU) in gamma rays emitted by 235U. This method containers. Ten portable NDA systems are is valid for measuring the enrichment of currently deployed at the four Russian sites bulk uranium samples that are homoge- that process material covered by the trans- neous and large enough to fill the view of parency agreements (see page 10). The the gamma-ray detector with an “infinite equipment measures several forms of urani- thickness” of uranium. The gamma-ray um, including metal components, metal detector views the sample through a colli- shavings, uranium oxide, and uranium mator made of dense material, restricting hexaﬂuoride. the detector’s field of view so that reason- ably sized samples can fulfill the requirements of the enrichment meter. “Infinite thickness” is defined as that amount of material that reduces the intensity of the 186-keV gamma ray by a factor of one hundred. For uranium metal, this thickness is about 3 millimeters; for solid uranium hexafluoride, 1.4 centimeters. The uranium samples measured in the HEU Transparency Implementation Project are always large enough to satisfy this “infinite thickness” condition. The portable NDA system is composed of commercially available products: a collimat- ed NaI gamma-ray detector, a Canberra Inspector, and a laptop computer. The gamma-ray detector is a 1 inch-by-1 inch NaI crystal coupled to a photo-multiplier tube. The signals from the detector are processed by electronic components in the Canberra Inspector unit. A laptop computer controls the system and allows the operator to input the necessary sample parameters, e.g., container-wall thickness and material type. The computer also analyzes the

235 gamma-ray spectrum created by the The portable NDA equipment used for measuring the U enrichment of uranium in containers. Inspector and then calculates and displays the measured enrichment.

Daniel Decman Lawrence Livermore National Laboratory

52 ACNT • Technology R&D for Arms Control • Spring 2001 Uranium

Determining Uranium Enrichment

e have developed a method to calibration has little to do with the result. determine the uranium enrich- The squares in this figure are the result Wment of a sample in a container. using a low-enriched uranium source (NIST The sample can be in any chemical form, standard) for calibration. The triangles use either an oxide or a hexafluoride. In addi- declared values of highly enriched samples tion, the sample can be in any kind of for calibration. Both results are good and container. The system uses a high-purity tell us that the selection of calibration germanium gamma-ray detector, a sources does not affect the result. Canberra InSpector data-acquisition system, and a laptop computer. S. John Luke and David A. Knapp The collimator for the gamma-ray detec- Lawrence Livermore National Laboratory tor is designed so that the field of view is filled for all measurements. The software corrects for the attenuation of the container 100 A comparison of measured ura- and the self-attenuation of the sample. The nium enrichment with declared software requires a rough energy calibration enrichment for uranium sam- 80 and an enrichment calibration using NIST ples. The red squares are low- uranium-oxide standards. The system does enriched uranium sources to cal- not require that the enrichment calibration 60 ibrate the system. The blue tri- be performed with attenuators, which is a angles are declared values of great leap forward from previous methods. highly enriched uranium. We have tested this system on a wide vari- 40 ety of samples in different forms and in different types of containers and found it to be Measured Enrichment very robust. 20 This method works very well, as the figure shows. A perfect measurement would lie on a line with a slope of one. This is very 0 0 20406080100 much the case. The data are interesting Declared Enrichment because they show that the enrichment

ACNT • Technology R&D for Arms Control • Spring 2001 53 Chain-of-Custody

Chain-of-Custody Monitoring of Warhead Dismantlement

o meet potential needs of the next minimizing the impact of treaty monitoring stage of U.S.–Russian nuclear arms on stockpile stewardship. For example, during Treductions, Los Alamos National normal operations, inspectors can remotely Laboratory has developed two prototype monitor IFMS data at a central station located systems for monitoring the dismantlement outside the dismantlement facility. and storage of nuclear weapons. The All monitoring data collected by the IFMS Integrated Facility Monitoring System are displayed in real time at the central sta- (IFMS) was tested at the Device Assembly tion, and past events can be retrieved from Facility (DAF), Nevada Test Site in early the archives. Expert system software inte- 1999. The Magazine Transparency System grates sensor information with known disas- (MTS) added the ability to monitor nuclear sembly protocols to ensure treaty compliance weapons and components in storage prior and maintain the inventory of treaty-limited to and after dismantlement. Both proto- items. Examples of protocol violations that types were demonstrated at the Pantex the expert system can detect include excess Plant in October and November 1999. time of movement between process stages, These monitoring systems rely heavily on inventory discrepancies, and the verification the chain-of-custody concept. Chain-of-cus- of tamper-indicating seals and tags. tody offers inspectors confidence in the data A key subcomponent of the IFMS, called that confirm the status and location of all the Integrated Tamper-Indicating Device treaty-limited items (in this case, nuclear (ITID), provides the uninterrupted surveillance warheads and components). Inspectors are of sealed containers as opposed to periodic also confident that no unauthorized tam- checks of seal integrity. The ITID, mounted on pering with the items has occurred. a weapon or component container, consists of a seal, camera, infrared tag, and UHF transmit- Integrated Facility ter. This subcomponent tracks the location of Monitoring System (IFMS) items between steps in the dismantlement Using a combination of sensors, tamper- process and ensures that no unauthorized indicating seals and tags, and a computer- tampering occurs. based expert system, IFMS monitors the Other sub-components, called integrated nuclear warhead dismantlement process monitoring stations, are placed outside point to point. It is designed to protect entryways to disassembly bays and cells. sensitive and classified information while These stations track treaty-limited items as

Device Assembly Facility at the Nevada Test Site. Integrated Tamper-Indicating Device (ITID).

54 ACNT • Technology R&D for Arms Control • Spring 2001 Chain-of-Custody

they enter and exit disassembly bays and and the RFID system interrogates the RF monitor the removal and re-attachment of tag on the MAGTAG blanket for additional ITIDs on weapon and component contain- motion detection. ers. These subcomponents relay their data All modules in the MTS run on a single to the central station where the data can be computer. The barcode is read on the arms- authenticated by inspectors. control seal for containers when they enter or leave the magazine. Data from the barcode Magazine Transparency System (MTS) reader—time stamp, reader number, and bar- The MTS detects unauthorized movement code number—are transmitted from the MTS of weapon containers from storage maga- computer to the IFMS central computer. zines and maintains their inventory. The A single software system monitors each system uses only passive tags and seals to of the MTS sensors (gaussmeter, video, and reduce host-country safety and security con- RFID) for movements within the magazine. cerns. It maintains and transfers data on the If no movement is detected, the software inventory of stored treaty-limited items to sends an “ALL OK.” signal to the IFMS cen- the IFMS central station. The system tral system. An alarm is triggered if one or demonstrated at Pantex included the fol- more of the sensors detects movement. lowing elements: Monitoring systems with the capabilities • Gaussmeter of the IFMS and MTS can help meet the • Low-light video camera objectives of the 1997 U.S.–Russian Helsinki • Radio-frequency identiﬁcation (RFID) tag Summit statement. This statement commit- • MAGTAG blanket ted the U.S. and Russia to increase the • Barcode reader transparency of nuclear weapons stockpiles • One-time keypad. and warhead dismantlement. These proto- The gaussmeter measures the magnetic types are a ﬁrst step in meeting these long- field within the magazine and detects term objectives. changes in that field caused by the movement of magnets in the MAGTAG James E. Doyle blankets covering the containers. The Los Alamos National Laboratory video camera detects any scene changes,

Integrated Monitoring Station.

ACNT • Technology R&D for Arms Control • Spring 2001 55 Chain-of-Custody

Fission-Product Tagging

persistent challenge in transparency SNM. Following processing within the facili- regimes is that of reconciling special ty, the resulting pieces of SNM are examined Anuclear materials (SNM) entering a with a high-resolution, gamma-ray detector closed facility (i.e., one to which inspectors to determine whether they bear inventories do not have access) with the materials leav- of fission products consistent with the irradi- ing the facility following the dismantlement ation as it was performed. Most of the activi- of a weapon or the processing of the SNM, ty of the fission products—the “pink paint” without revealing anything classified about of the analogy—decays away within a few either the materials themselves or the days, so that the intrinsic radiation of the processes they undergo within the facility. SNM is not permanently increased (an The search for a means to ensure that the important radiation-safety consideration for material observed going in is the same operations at a real processing facility); how- material coming out was succinctly summa- ever, key activities persist for the several tens rized by a policymaker: “Why can’t you of hours necessary for most forms of SNM paint the SNM pink?” This pithy question is processing to be completed. at the heart of a transparency tool called fis- Proof-of-concept experiments centered sion-product tagging. around the ARIES process (see sidebar, next The SNM-bearing item is irradiated at the page) to convert a classified weapon compo- entrance of the facility with a neutron nent into an unclassified form. The source that produces fission products in the GODIVA fast-burst reactor was used to induce fissions in a component passing through ARIES. Measurements following the passage of the SNM through ARIES revealed that the fission products were transported effectively through the ARIES process; that is, the physics “worked.” However, the neutron source required to induce a detectable number of fissions proved to be rather large. This creates health-physics implications for the facility where fission-product tagging is implemented. The research program, accordingly, has not yet proceeded to the development of a prototype tagging tool, but the satisfactory transport of the fission products in the ARIES-oriented experiments gives reasonable confidence that a tagging tool could be built should a dismantlement or conversion regime require one.

M. William Johnson Los Alamos National Laboratory

56 ACNT • Technology R&D for Arms Control • Spring 2001 Chain-of-Custody

Nondestructive Assay of Special Nuclear Materials

Nondestructive Assay (NDA) methods quantify the pluto- Materials and residues from over 50 years of weapons pro- nium and uranium contents of sealed containers without open- duction, now in storage, are being repackaged in long-term ing them by measuring the naturally occurring radioactive storage containers at DOE sites around the country. For ulti- emissions from the items inside the containers. NDA instru- mate disposition, plutonium will be either mixed with ura- ments include a calorimeter (measuring the heat output from nium to form mixed-oxide (MOX) fuel for nuclear reactors, or the radioactive decay of plutonium), a neutron multiplicity it will be immobilized in glass logs to prevent its escape into counter (measuring the neutrons produced by the sponta- the environment or its theft by terrorists. neous radioactive decay of plutonium or the induced ﬁssion of Los Alamos National Laboratory is also demonstrating uranium), and gamma-ray detectors (identifying unique signa- technologies to dismantle nuclear weapons, convert the plu- tures from the isotopes to determine their relative fractions). tonium to an unclassified shape, and place it in DOE- NDA measurements can be integrated with robotics and approved containers. Products and residues from the disman- automated under the control of a central computer. Robotic tlement process are quantified through integrated and control improves consistency, increases throughput, lowers robotically operated NDA methods. operating costs, decreases radiation exposure to operating personnel, and improves nuclear-material safeguards.

Thomas E. Sampson Los Alamos National Laboratory

ACNT • Technology R&D for Arms Control • Spring 2001 57 Alternative Signatures

Identifying Weapon Components in Sealed Containers by Electromagnetic Induction

cceptable verification methods must the container and the contents inside. furnish just enough information to Parameters that significantly influence the Auniquely identify items of concern, electromagnetic signature include electrical without providing unnecessary information. conductivity, magnetic permeability, total Additional features that may be required of mass, mass distribution, and the orientation a verification method include high speed, of each object interacting with the coil. By portability, and simple operation. themselves, two-dimensional coil imped- Pacific Northwest National Laboratory is ance measurements are insufficient to developing an electromagnetic induction derive (invert) explicit information because method (EM induction coil) as a non- the dimensionality of the complete parame- nuclear approach to verification methods. ter set influencing the signature is more The EM induction coil shows promise for extensive. obtaining unique, non-intrusive signatures The EM induction coil method has been of weapon components. evaluated at the Pantex Plant and Pacific A low-frequency, voltage-driven coil Northwest. Test items in these evaluations induces a magnetic field (of a magnitude included weapon components, models, and similar to the earth’s magnetic field) inside a wide assortment of metal objects selected the container, causing eddy currents to flow to characterize and optimize the capability throughout conductive components. The of this method. To date, our results show complex-valued coil impedance, measured that this method can— in seconds, varies in response to a multi- • Uniquely identify metal containers, and tude of independent parameters describing their contents

6.4 Al (hollow) Bronze 6.4 Brass Al (solid) Copper 6.4 j

RC 6.4 Tall BB cylinder

304 Stainless Steel

e(t) L 6.4 C LS RS Vertical Voltage

6.4 Titanium Wide BB cylinder

Steel Primary Coil Circuit Secondary Induction Circuit 6.4 -1.0 -0.5 0.0 0.5 1.0 1.52.0 2.5 Horizontal Voltage

Eddyscope output for different metal objects placed in a steel drum surrounded by a coil.

58 ACNT • Technology R&D for Arms Control • Spring 2001 Alternative Signatures

• Sort weapon components according to The EM coil method requires less than a type minute to obtain the signature and analyze • Sort objects made from different metals it. Positioning the coil over the container having the same geometry requires more time than the one-button • Sort objects made from the same metal measurement. The sealed container remains having a different geometry untouched during coil placement and removal. These features make this method • Distinguish between a metal and an appropriate for rapid evaluation of large oxide. stockpiles. The necessary electronics and A failure modes and effects analysis per- apparatus are approved for export to all formed at Pacific Northwest and approved countries. For warhead and fissile material by Pantex shows minimal risk, even in a transparency programs, this issue may worst-case scenario. This results from the become a limitation for more complex mea- fact that even with the maximum possible surement methods. current in the coil, the magnetic fields are similar to the earth’s intrinsic magnetic field in amplitude. The EM coil system was oper- Ronald Hockey ated in the laboratory for several hours on Pacific Northwest National Laboratory battery power, demonstrating remote operation where AC power is unavailable.

ACNT • Technology R&D for Arms Control • Spring 2001 59 Alternative Signatures

Thermal Infrared Signatures of Containers

ffective confirmation of nuclear- the resistance of the path to the heat flow, weapon dismantlement and subse- causing higher temperatures at the surface Equent transportation and storage of at locations of higher heat. the nuclear components poses two seeming- Three containers were imaged using a ly incompatible requirements: detecting suf- commercial infrared camera. A 20-watt heat ficient information about the enclosed com- source was placed inside a single-wall AL-R8 ponents to ensure they contain the pre- container and a double-wall AT400-R con- sumed nuclear components, while preclud- tainer. A second, empty AL-R8 container ing access to sensitive information. Infrared was placed nearby for reference. The ambi- cameras can monitor containers for thermal ent temperature was about 20°C—a typical (heat) sources that continue to generate room temperature—and the temperature of heat and at rates and over periods of time the warmest areas of the containers was 2°C inconsistent with anything other than to 4°C higher. nuclear materials. Some features are thermal reflections. For A double-walled, heavily insulated stor- example, the two similarly shaped, colored age container encloses a heat-generating circles on the faces of the two rear contain- source. The heat source results in a higher ers are thermal reflections of the container temperature inside the container than out- in front. However, the circle pattern on the side. This temperature differential across lid of the front container clearly maps the the walls drives the thermal energy out- interface between the lid and its seat. The ward. Heat flows at a rate dependent on

Representation of a heat source inside a sealed ﬁssile-material storage Infrared image of two containers with heat sources and one empty container showing typical heat ﬂow patterns. container.

60 ACNT • Technology R&D for Arms Control • Spring 2001 Alternative Signatures

pattern of the red area on the rear right con- inch squares and arranged randomly in a 25 tainer is consistent with the location of the x 25 “chip” array on the container lids (see source and the geometry of the container. figure below). A low-cost alternative to infrared-camera Applications include any monitoring of thermography uses liquid-crystal appliqués the dismantlement operation and long-term that convert the infrared to visible light, storage, verification of dismantled compo- allowing visual monitoring and document- nents being transferred, and confirmatory ing to be done with low-cost consumer-grade measurements. Infrared-based imagery’s video or digital cameras. Experiments used salient feature is that it exploits the noncommercial, self-adhesive liquid crystal nuclear, nonvisual, intrinsic heat-generating sheets having temperature sensitivities rang- characteristic of the items of interest, with- ing from 20°C to 40°C in four 5°C ranges. To out compromising engineering data. cover a wider range with sufficient sensitivity to detect the expected few-degree tempera- Charles R. Batishko ture difference, the sheets were cut into half- Pacific Northwest National Laboratory

A low-cost alternative to infrared cameras is a random array of liquid- crystal “chips” on the lid of an AL-R8 storage container with an enclosed 20-watt source. Red represents the cooler portions of the lid while blue shows the warmer portions.

ACNT • Technology R&D for Arms Control • Spring 2001 61 Tags and Seals

Tags and Seals for a Potential Strategic Arms Control Monitoring Regime

amper-indicating devices (TIDs) that plex reflective pattern, etc. that cannot be support security technologies and removed without being detected. TIDs can Toperations will play a critical role in be expected to contribute to treaty-compli- transparency and nonproliferation. A TID is ance transparency for controlling nuclear- a tag or seal that detects tampering, or weapon inventories under the Mayak Fissile attempts at tampering, of a controlled item, Material Storage Facility (FMSF) such as a canister containing a weapon Transparency and the Trilateral Initiative. component. A tag is a unique identifier The Tags and Seals Working Group based on either applied or intrinsic features. (TSWG) was formed by the Joint DOE–DoD A seal is a tag applied across an interface to Integrated Technology Steering Committee indicate a breach. A TID also provides a in 1998. TSWG was chartered to assess past means for identifying and monitoring an and current TID technology supporting ver- The first-generation Smart Bolt item to be secured, such as radiation detec- ification, compliance, and transparency and Reader, intended to seal tion equipment stored at a point of entry. needs and to make recommendations of AT-400R fissile-material storage TIDs should have a unique signature that candidate TID technologies for potential containers, was developed by cannot be counterfeited to identify the con- monitoring regimes such as Mayak FMSF the Russian Institute of trolled item. The signature may contain ser- Transparency and the Trilateral Initiative. Experimental Physics, VNIIEF. ial numbers, digital identification, a com- TSWG surveyed existing TID technologies, both commercial and noncommercial, to provide a baseline for future evaluations, recommendations, and demonstrations. Many of the commercial technologies are used in domestic safeguards. Some of the very secure TID technologies were developed under the START I program but never implemented. TSWG surveyed adhesive seals, reflective particle tags, radio-frequency tags, loop TIDs, mechanical TIDs, and intrinsic TIDs. TIDs serve a very important role in the transparency regimes being developed. As a central feature of any chain-of-custody architecture for nuclear systems and components during transportation and storage, they create a documentation record that

62 ACNT • Technology R&D for Arms Control • Spring 2001 Tags and Seals

can be audited. TIDs can also monitor com- tion-and-selection process, it is likely that ponent or material inventories in intermedi- the host will insist on using devices devel- ate and final storage, and they can indicate oped within its own country. tampering of data and equipment during Minimizing verification, thus reducing and between inspections. Finally, they are inspection effort, depends on using TIDs also key components in containment and acceptable to all parties. Efforts to adopt surveillance systems composed of video common certification criteria, already cameras, recording stations, and access con- underway within the international safe- trols (such as motion detectors). guards community, would be useful in Technical specialists from the participat- improving the efficiency of the selection ing parties must prove to their respective process and the acceptability of the chosen policymakers that it is possible to maintain devices. continuity of knowledge over nuclear mate- Reaching an international consensus on rials and the corresponding data and equip- acceptable TIDs and developing testing and ment used to verify and monitor them. certification protocols are not easy. The Without technical measures to seal and tag level of dependence placed on these nuclear items and materials, it is difficult to devices, and the consequences of their fail- ensure that the materials are as declared ure make evaluation and selection a critical without requiring an inordinate amount of step on the path to effective transparency in effort verifying them. The importance of international disarmament activities. successfully evaluating and deploying effective TIDs is great. Cooperative participation Jennifer Tanner in selecting the devices to be used will help Pacific Northwest National Laboratory reduce threats to the U.S., Russia, and international security. Arden Dougan Selecting appropriate TIDs, establishing Lawrence Livermore National Laboratory procedures for their use, and actually imple- menting those procedures are important to any inspection system for warhead-dismantlement transparency and weapons-material bilateral or trilateral verification. By engag- ing in a fully cooperative effort, confidence in the devices and their assurances can be maximized. Without a cooperative evalua-

ACNT • Technology R&D for Arms Control • Spring 2001 63 Tags and Seals

Seals for Transparency and Treaty Monitoring

amper-indicating seals are designed to Transparency Laboratory at Los Alamos detect unauthorized access to a contain- National Laboratory, has undertaken the fol- T er or door. Unlike locks, seals do not lowing tasks: necessarily bar an unauthorized entry—they 1. Optimize the security of existing seals by simply record that it took place. This type of developing improved (but still cost-effec- “tamper detection” has long been considered tive) protocols. This work has been con- an important component of overall security ducted for DoD, DOE, IAEA, and private and safeguards for nuclear weapons and mate- companies. rials. Seals will continue to play an important 2. Develop new seals that have better securi- role in monitoring trilateral agreements, and ty and more of the attributes needed for U.S.–Russian joint nuclear safeguarding and transparency and treaty monitoring. So processing programs. Two serious problems with existing seals far, this work has led to the development for transparency and treaty monitoring are of four novel seal designs and four U.S. vulnerabilities and conflicting goals. patent applications. These seals provide Regarding vulnerabilities, the unhappy fact excellent security and transparency at a is that many types of seals potentially avail- modest cost. They are envisioned for use able for nuclear applications can be defeat- at Mayak and other venues. ed, at least when they are used in a conven- 3. Demonstrate new transparency and mon- tional manner. “Defeating” a seal means itoring protocols that overcome many of gaining access to what it is protecting (by the problems associated with seals and manipulating the seal or replacing it with a with conventional security and safe- counterfeit) such that the unauthorized guards approaches. These protocols access goes undetected. emphasize certain attributes that we The second serious problem with existing seals is that virtually none were designed believe are essential in any effective (and with transparency and treaty monitoring in negotiable) international transparency mind. Conventional tamper detection pro- and monitoring program: vided by existing seals differs significantly • Classified information is not released. from what is needed for effective trans- • Monitoring hardware is provided by parency and treaty monitoring in terms of the host, eliminating the use of foreign goals, likely adversaries, environment, per- hardware and thus avoiding the safety, sonnel, economics, visibility, seal handling, security, and espionage concerns that and consequences of a security failure. inevitably arise. To address both problems, the Vulnerability Assessment Team, in conjunction with the new Applied Monitoring &

64 ACNT • Technology R&D for Arms Control • Spring 2001 Tags and Seals

• The physical presence of foreign ed by the host facility are installed and inspectors inside each side’s nuclear which are kept by the inspectors for facilities during dismantlement opera- analysis and reverse engineering to tions is minimized, again reducing safe- check for signs of tampering; and ty, security, and espionage concerns. “keep the used parts” protocols that • The use of conventional seals, encryp- permit the inspectors to keep and ana- tion, or information barriers is limited. lyze monitoring hardware after it has Confidence in the integrity of host- fulfilled its monitoring functions. supplied monitoring equipment is pro- • Tags, seals, and use protocols are specifi- vided via more trustworthy, low-tech cally designed for treaty monitoring and methods. These include live, dynami- transparency that neither replace nor cally tested continuous sensor feeds; compromise those required for domestic “choose or keep” protocols that let the nuclear security and safeguards. inspectors randomly choose at the last minute which of several monitoring Roger G. Johnston modules, components, or seals provid- Los Alamos National Laboratory

A few of the estimated 5,000 commercial seals are shown. No signiﬁcance is attached to which seals have been chosen for this photograph other than that a number of them have been used, are being used, or are under consideration around the world for various nuclear-related applications.

ACNT • Technology R&D for Arms Control • Spring 2001 65 Detection of Nonnuclear Components

Chemical-Explosive Detection by Neutron Interrogation

ll types of nuclear weapons use chem- identifies range-recovered munitions that ical explosives to rapidly assemble have lost their identifying markings due to Afissionable materials to criticality. The corrosion and exposure to the elements. presence of chemical explosives is another PINS was designed to nondestructively iden- nuclear-weapons attribute, as is the pres- tify munitions and containers filled with ence of the special nuclear material itself. chemical warfare agents, such as sarin nerve Unlike uranium and plutonium, explosives gas. Chemical munitions are its most fre- do not naturally emit radiation—they are quent application, but PINS can also identi- composed of stable, non-radioactive chemi- fy projectiles filled with smoke-generating cals. Because explosives cannot be identified chemicals, such as white phosphorus, and by passive radiation counting, detecting munitions filled with high explosives. explosives is more challenging than identi- PINS, like most neutron-interrogation sys- fying most radioactive materials. tems, shines neutrons from a radioisotopic The presence of explosives inside a closed source or a neutron generator into the object container can be detected by active radia- being tested. Neutrons easily penetrate the tion methods, for example, neutron interro- object’s housing, even the half-inch thick gation. For the past 10 years, Idaho steel wall of a 155-millimeter artillery shell. National Engineering & Environmental Inside the object, the neutrons excite the Laboratory has developed and helped the atomic nuclei of the fill chemicals, and these U.S. military field a neutron-interrogation nuclei in turn de-excite by emitting charac- system called Portable Isotopic Neutron teristic gamma rays. The gamma radiation Spectroscopy (PINS). The PINS system escapes the item and is measured by a

The Portable Isotopic Neutron Spectroscopy (PINS) setup in TA- Graphite reflector 18 at Los Alamos National Laboratory. The shapes simulate Polyethylene high explosives. moderator Steel shipping container Explosive parts

HPGe detector

252Cf source Tungsten shadow shield Liquid nitrogen dewar Polyethylene moderator

66 ACNT • Technology R&D for Arms Control • Spring 2001 Detection of Non-nuclear Components

spectrometer. The gamma-ray energy spec- A second series of measurements used trum identiﬁes those chemical elements polyethylene to enclose the simulated high inside the item and their relative abundance. explosive. The polyethylene moderated and Neutron-interrogation measurements on redirected the neutrons that passed through simulated high explosives were recently the simulated explosive without interaction. carried out at Los Alamos National The polyethylene boosted the intensity of Laboratory’s TA-18 (see page 24). An off-the- the nitrogen gamma rays by a factor of six. shelf PINS system used hollow hemispheri- Adding the polyethylene showed that PINS cal shells made from simulated Composi- could identify the simulated explosive in as tion B high explosive. The mass of the sim- little as 500 seconds, or about 9 minutes. ulated explosive ranged from 1 to Because active methods for detecting explo- 13 kilograms. Three gamma-ray peaks pro- sives are well-understood, we see this neu- duced by the 10.829-MeV nitrogen capture tron-interrogation method as a promising gamma are shown in the graph (red line). technique for identifying nuclear weapons The blue line shows the low background in through their high-explosive components. this energy region when no simulated explosive is present. A.J. Caffrey and B.D. Harlow Idaho National Engineering & Environmental Laboratory

20 A single gamma-ray energy at 10.829 MeV indicates the pres- Simulant ence of nitrogen. Three peaks Nitrogen Background arise from this energy because of single escape 15 energy losses due to pair produc- Nitrogen full tion in the high-purity germani- Nitrogen energy peak um spectrometer crystal. The double escape middle energy peak, or “single 10 escape” peak is the most intense. Counts The “double escape” peak is also visible. The three peaks are separated by exactly 0.511 MeV, the 5 rest mass energy of the electron.

0 9.6 9.8 10.0 10.2 10.4 10.6 10.8 11.0 Gamma-ray energy (MeV)

ACNT • Technology R&D for Arms Control • Spring 2001 67 Points of Contact

Points of Contact

ACNT lists the following names for the beneﬁt of readers who may wish to contact programmatic representatives directly.

David Spears Jessica Kehl Program Manager Office of the Secretary of Defense DOE/NNSA, NN-20 Cooperative Threat Reduction Policy Office 202.586.1313 703.614.3501 [email protected] [email protected] Al Boni Ed Mastal Savannah River Technology Center Department of Energy, Germantown 803.725.2628 301.903.3197 [email protected] [email protected] Leigh Bratcher Dean Mitchell Pantex Plant Sandia National Laboratories 806.477.5366 505.844.8868 [email protected] [email protected] Gordon Dudder Ken Sale Pacific Northwest National Laboratory Lawrence Livermore National Laboratory 509.376.9530 925.423.0686 [email protected] [email protected] M. William Johnson Michele Smith Los Alamos National Laboratory Department of Energy 505.667.2047 202.586.8909 [email protected] [email protected] James Jones Idaho National Engineering & Environmental Laboratory 208.526.1730 [email protected]

68 ACNT • Technology R&D for Arms Control • Spring 2001 NN-50’s Materials Protection, Control, and Accounting (MPC&A) program focuses on enhancing the physical protection and material-accountancy infrastructure at former Soviet Union sites containing weapons-usable nuclear material. This work is augmented by the program’s involve- Accelerator Applications and ment with enhancements to the regulatory framework within Russia, national-level material Radiation Science accounting, nuclear-material transportation, as well as other related activities. The program’s goal is to reduce the risk to U.S. national security of an undetected theft of weapons-usable material from the former Soviet sites by enhancing their MPC&A capabilities. The placement of a yellow box around a given type of facility on the diagram signifies that the program is involved, or is planning to be involved, in infrastructure upgrades at these sites.

EDUCATION Offering undergraduate and graduate degrees in physics, health physics, engineering, and applied science

PARTNERSHIPS AND COLLABORATIONS Teaming with university government and commercial partners on applied research, testing and technology deployment.

IAC offers technical expertise and state-of-the-art facilities for collaborations with universities, government agencies, and commercial and industrial organi- zations. These partnerships promote practical advances in nuclear and radiation science and facilitate the transfer of technology to the private sector. IAC collaborations range from agricultural applications of accelerator-produced radiation to nondestructive examination through gamma-ray spectroscopy, radiography, and tomography. IAC partners conduct research and testing on applications in environmental remediation, waste management, chemical-weapons verification, contraband detection, and radiation effects. • • • • • • • • • • • • • • • • SERVICES • • FACILITIES and INSTRUMENTATION using x-ray generators from 30 Fixed facility and field digital radiography computed tomography systems data acquisition system-Calibrated PIN diodes-Internet 11 connection Supporting Instrumentation-Calibrated 20-GHz sampling scope-Multi-parameter D/T Neutron Generator-Sodern Genie 16 Tandetron-High-Voltage Engineering, Inc. 1.5 MeV/amu Two positive ion Van de Graaffs-High-Voltage Engineering, Inc.-2-MV potential – – 18-MeV Linac (Varian Associates, Inc.) 3000 R/min (@ 1 m) widths: 500 ns to 4 Varitron Accelerator (Varian Associates Inc.)-2 Two 4-MeV Linacs-Field radiography/neutron source capability experimental needs and applicable resources. Single point-of-contact: simplified coordination between researchers and modifications Customized accelerator performance Experimental verification of predictions and objectives Customized radiation detection system development Photon and neutron dosimetry Photon, electron, and neutron transport calculations for system applications Customized user support Wide range of accelerator types available to researchers 0.5 space 21,000 sq. feet of laboratory energy spread-Long-pulse mode: pulse width: 4 Pulse widths: 15 ns to 2 Beam energy and current analysis Short-pulse mode: pulse widths: 30 Repetition rate: single shot to 240 Hz-Three beam ports – 25 MeV electron Linac µ s Beam energy and current analysis-Capable of up to µ s – 450 kV – 0p,1 nC/pulse, 0.5% beam 50 ps, 10 – 12-MeV energy range-Pulse µ s, 2,000 nC/pulse- www.iac.isu.edu Pocatello, Idaho 83209 1500 Alvin Ricken Drive Idaho Accelerator Center Email: [email protected] Fax: 208.282.4538 Phone: 208.282.2902 Pocatello, Idaho 83209 Campus Box 8060 Idaho State University Associate Director Dr. Jay Kunze Email: [email protected] Fax: 208.526.5208 Phone: 208.526.1730 Idaho Falls, 83415-2802 PO Box 1625 (INEEL) Environmental Laboratory Idaho National Engineering and Associate Director Dr. James L. Jones Email: [email protected] Fax: 208.282.5878 Phone: 208.282.5877 Pocatello, Idaho 83209 Campus Box 8263 Idaho State University Director Dr. Frank Harmon CONTACTS See reverse side for an explanation of this chart Uranium mining Plutonium production reactors Civilian reactors uranium Enrichment Natural LEU fabrication Fuel processing Uranium Production Reprocessing Materials

P Spent U

l Strategic

Reprocessing Reserve u storage

Naval HEU r

t fuel o a

n n

i i u u

m m or LEU Natural

Military use Military

Strategic Strategic Reserve

fabrication Pu Post-1997, naval fuel, commercial Fuel

Pre-1997, weapons-grade Pre-1997 fabrication fabrication Enriched Uranium Pu parts Highly Parts Military use Military Assembly Weapons HEU weapons component Pu weapons component Assembly

l U

u r t a

o n

i i u Warhead

u storage/

l u

r t a

o n n i

u i

Storage

excess P

u U

t r

o a

n n

i i

u u m m fuel production Pu conversion and blending Mixed-oxide fabrication Fuel fuel rods

Uranium

Material Storage immobilization Pu waste Civilian reactors Civilian reactors reactor Naval rods High-level storage waste Spent fuel

NNSA/NN/ACNT-SP01

National Nuclear Security Administration