A worldwide resurgence in interest in nuclear power plant construction has recently been underway. According to the International Atomic Energy Agency (IAEA) in 2010:
In recent years, in every region of the globe, many countries have expressed a new or renewed interest in nuclear power. In the context of growing energy demands to fuel economic growth and development, climate change concerns, and volatile fossil fuel prices, as well as improved safety and performance records, some 65 countries are expressing interest in, considering, or actively planning for nuclear power (IAEA, 2010a).
If a resurgence of nuclear power does occur,1 fuel will be required for new power plants. In this situation, many countries—particularly new entrants—may need assurance that fuel for these power plants will be cost effective and available. Thus, unless an alternative solution to producing or recycling nuclear fuel domestically is proposed and universally accepted, some nations may be interested in constructing nuclear fuel cycle facilities, including enrichment or reprocessing facilities.
Nuclear power plants themselves are often considered a lesser proliferation2 risk than other nuclear fuel cycle facilities, particularly those that
__________________
1 The global economic crisis and the recent events at the Fukushima nuclear power plants in Japan may change or delay these plans.
2 Defining the point at which a state is considered to have proliferated is complicated. For example, is a nation a proliferator if it produces highly enriched uranium or plutonium, or is it necessary to begin other efforts to construct a weapon? There is no standard that the workshop briefings summarized in this report are necessarily adhering to in the discussions of the subject.
are used to produce or recycle nuclear fuel. These facilities are considered to have a significant potential to increase the risk of host state proliferation3 if sited in states that do not currently possess nuclear weapons technology (NRC, 2009).4 This is because these facilities in many cases are or can be easily converted to dual use; potentially, they can be used to produce weapons-usable material and to spread nuclear weapons technology as well as fuel for nuclear reactors.
There exist both technical and policy efforts aimed at managing and reducing the proliferation risks associated with nuclear fuel cycle facilities. In particular, a range of technical methodologies are currently being developed to assess some of the proliferation risks associated with these facilities. For the most part (as discussed in more detail later in this report) these methodologies remain immature (TAMU, 2010).
The current report is a summary of a workshop on improving the assessment of proliferation risk associated with nuclear fuel cycles. The workshop was held on August 1-2, 2011 by the National Research Council (NRC) of the U.S. National Academies at the National Academies’ Keck Center in Washington, DC.
The workshop was organized as part of a larger project undertaken by the NRC, the next phase of which (following the workshop) will be a consensus study on improving the assessment of proliferation risks associated with nuclear fuel cycles.5 This study will culminate in a report prepared by a committee of experts with expertise in risk assessment and communication, proliferation metrics and research, nuclear fuel cycle facility design and engineering, international nuclear nonproliferation and national security policy, and nuclear weapons design. This report is planned for completion in the spring of 2013.
The overall project was originated in response to a 2011 joint request from the U.S. Department of Energy (DOE) National Nuclear Security Administration’s Office of Nonproliferation and International Security and the DOE Office of Nuclear Energy. DOE asked that the workshop feature discussions about key nonproliferation policy questions capable
__________________
3 Throughout this workshop, participants discussed “host-state” proliferation, or actions taken by the government of the state in which the facility is located to use the facility to illicitly produce nuclear materials for non-peaceful uses.
4 Throughout this report, the risk associated with the physical security of the facility or materials against attack, theft, or diversion of nuclear materials (security risk) is considered to be separate from host state proliferation risk. Security risk is explicitly not intended to be considered by the current study.
5 The task statement for the full study can be found in Appendix A
of being answered by a technical assessment of the host state proliferation risk and the utility of these questions for informing nonproliferation policy decisions. The statement of task for the workshop is included as Appendix B.
DOE and NRC agreed that the workshop would not result in consensus findings or conclusions among the participants, but would instead focus on encouraging discussion on the topics in the statement of task. For this reason, the current report does not feature findings, conclusions, or recommendations, but instead serves as a summary record of the briefings and discussions that occurred during the symposium. Many of these themes will be expanded upon in the consensus study.
The workshop was organized by a four-member committee with extensive expertise in nonproliferation policy, proliferation resistance assessment, nuclear weapons policy and technology, and nuclear fuel cycle technology. Biographical sketches of the committee members and staff are provided in Appendix C. The committee met twice over the course of the project: first, in May 2011, to plan the workshop; and second, in August 2011, to hold the workshop. The workshop agenda is provided as Appendix D, and the list of workshop participants is provided as Appendix E. Appendix F contains biographical sketches of the workshop speakers.
The workshop featured a range of expert briefings on proliferation and the nuclear fuel cycle as well as extensive audience participation in the form of breakout and discussion sessions. The workshop was organized into three sessions, reflected in chapters 2, 3, and 4 of this report: (1) key policy issues associated with proliferation and the nuclear fuel cycle; (2) methods and methodologies to assess the proliferation resistance of fuel cycle facilities; and (3) summary and future directions for the assessment of proliferation resistance.
The remainder of the chapter is based on a white paper distributed to the workshop participants (Case and Ferriss, 2011) and is intended to provide the reader with background information regarding proliferation and nuclear energy, the nuclear fuel cycle, and some basic concepts associated with assessing the proliferation resistance of nuclear fuel cycle facilities.
NUCLEAR FUEL CYCLES
Uranium nuclear fuel cycles consist of three stages: (1) the front end, in which uranium is mined, milled, enriched,6 and fabricated into fuel; (2) power generation; and (3) the back end, in which spent fuel is either
__________________
6 Enrichment is not a part of all fuel cycles; for example, the CANada Deuterium Uranium (CANDU) reactor design can operate using natural (i.e., unenriched) uranium.
disposed of (open fuel cycles) or processed to produce new fuel (closed or partially closed fuel cycles). 7
The power generation (second) stage is often considered to pose a lesser risk than facilities associated with the front end and the back end of the fuel cycle.8 As stated in the 2009 National Research Council report on America’s Energy Future, “Nuclear power plants themselves are not a proliferation risk, but nuclear fuel cycle technologies such as enrichment and reprocessing introduce the risk that weapons-usable material could be produced” (NRC, 2009, p. 491). This risk is discussed in the following two sections: (1) the front end, or enrichment; and (2) the back end, or recycling (reprocessing) or disposal.
Front-End Facilities: Enrichment
Nuclear fission reactions—in which a neutron is used to split an atom, releasing energy that becomes the power plant’s heat—are sustained in materials that are “fissile,” such as certain isotopes of uranium (e.g., uranium-235) and plutonium (e.g., plutonium-239).
More than 99 percent of natural uranium is uranium-238, rather than uranium-235, the uranium isotope whose fission is used to power most nuclear reactors. Typically, in uranium fuel produced for power generation (or material for use in a nuclear weapon), the concentration of uranium-235 is increased from natural concentration, i.e., enriched. The first working technologies to enrich uranium—thermal diffusion and electromagnetic isotope separation—were developed for the Manhattan Project, the United States’ massive effort in the 1940s to develop a nuclear weapon. However, these technologies are both expensive and time-intensive, and were eventually dropped from the U.S. program in favor of a technology known as gaseous diffusion enrichment. However, gaseous diffusion enrichment is very energy-intensive, and in the decades since, more efficient gas centrifuge enrichment technology has been developed and deployed around the world. Either technology can be used to produce power reactor fuel or nuclear weapon-usable material, by using a different number of enrichment stages.
Both gaseous diffusion and gas centrifuge enrichment processes rely on the slight difference in mass between uranium-238 and uranium-235. In gaseous diffusion enrichment, uranium hexafluoride gas (UF6) is forced through a series of semi-permeable membranes. Because lighter molecules
__________________
7 This is a simplification; DOE, for example, has research underway on modified open fuel cycles.
8 While reactors alone may present little risk, a clandestine reprocessing capability (or a possible breakout scenario) can significantly increase this risk.
pass through the membrane more easily than heavier molecules, at each membrane stage slightly more uranium-235 passes through the membrane than uranium-238. After many repetitions, the UF6 gains a higher proportion of uranium-235 until it is enriched to the desired level.
Gas centrifuge enrichment uses a cascade of centrifuges (rapidly rotating containers) to gain a slightly higher percentage of uranium-235 at each stage. Uranium-238, being slightly heavier, is driven farther in the radial direction from the center of the centrifuge. The lighter molecules of uranium-235 remain closer to the center of the centrifuge, and are drawn out and then input into the next stage of the centrifuge.
Other technologies that have been proposed to produce nuclear material include laser isotope separation, in which the uranium isotopes are separated using laser light to excite the molecules. There are currently no operating commercial-scale laser separation facilities in the world; however, General Electric-Hitachi plans to build a laser separation facility in Wilmington, North Carolina.9
As of 2010, three new commercial enrichment facilities, all using centrifuges, were planned: George Besse II in France, as well as the American Centrifuge Plant and the National Enrichment Facility in the United States. The U.S. Nuclear Regulatory Commission is currently reviewing plans for an additional centrifuge facility in Idaho and the previously mentioned laser enrichment facility in North Carolina (IAEA, 2010b).
Back End Facilities: Reprocessing
After highly radioactive spent nuclear fuel (SNF) has been removed from a power reactor it must either be stored for eventual disposal or reprocessed to recycle the remaining fissionable material. These options are referred to as a once-through or open fuel cycle and a closed or partially-closed fuel cycle, respectively. The United States is currently using a once-through fuel cycle; however, the Obama administration’s recent decision to cancel the planned repository program at Yucca Mountain in Nevada has resulted in no clear path for the disposal of U.S. spent fuel, most of which is currently being stored on-site at the power plants (BRC, 2011; NRC, 2008). Over the past few years, the complicated U.S. situation with respect to spent fuel and high-level radioactive waste disposal has led to a renewed interest in closed fuel cycles for waste management. This interest is due in part to their potential to reduce the long-lived radioactive waste burden (USDOE, 2007; USDOE, 2010). Russia, India, Japan, France, and the U.K. maintain (at least in part) a closed fuel cycle for civilian SNF (IAEA, 2011).
__________________
9 See: http://www.starnewsonline.com/article/20100415/ARTICLES/100419797
Reprocessing (also called recycling) spent fuel requires several steps, notably including a chemical or electrochemical process to separate and recover fissile components of the fuel. Most proliferation and security discussions focus on this step, because it is at this point that a stream of material that could be used in weapons (or easily treated to be usable in weapons) is produced.
The most common and best understood method used for separating fissile components is a variant of the PUREX (Plutonium and URanium EXtraction) process, which was initially used in the 1950s. This process results in a separated, pure stream of plutonium as well as uranium. Alternatives to PUREX currently under investigation (e.g., advanced Uranium Reduction and EXtraction [UREX+]; CO-Extraction of uranium and plutonium [COEX]; and electrochemical separations) include both evolutionary modifications of PUREX and entirely different separations technologies.
In many cases, alternatives to PUREX are intended to avoid the generation of the pure stream of plutonium that PUREX produces, and so reduce proliferation and theft risk. However, the reduction in risk is not always robust if minor changes are made to the process. For example, a modified version of PUREX has been suggested to provide increased proliferation resistance, but in 2008, the NRC reported that small adjustments could convert this process to PUREX (NRC, 2008).
ASSESSING PROLIFERATION RISK AND PROLIFERATION RESISTANCE OF NUCLEAR FUEL CYCLE FACILITIES
Whether or not proliferation occurs is based on individual and group decisions. These decisions can be reversed, modified, paused, or substantially altered as the potential proliferator’s political and economic environment changes and new technical options and opportunities emerge. However, there are some technical tools that have been proposed to evaluate and manage the risk of proliferation as well as the resistance of nuclear facilities to proliferation attempts.
Managing any risk—whether proliferation risk, safety risk, or security risk—involves three interacting elements: risk assessment (understanding the risk), risk communication (informing decision makers about the risks), and risk control (arriving at and implementing decisions to manage the risk) (NRC, 2011). In the following section, some background is provided for the reader on proliferation risk and resistance as connected to the nuclear fuel cycle. When discussing the probability of a nuclear facility being used to produce material for weapons, two terms are frequently used: proliferation resistance and proliferation risk. Throughout the report, a distinction has been made between these two concepts.
Proliferation resistance assessment focuses on how difficult it is to divert a particular dual use technology for non-peaceful uses.10 On the other hand, proliferation risk assessment recognizes and attempts to quantify the likelihood of threats, the barriers that must be overcome (proliferation resistance) and the consequences resulting from a potential threat (IAEA, 2002; Charlton, 2011). Proliferation risk and resistance are discussed in more detail in Chapter 3 of this report.
The proliferation resistance associated with a fuel cycle system or facility is frequently described in terms of extrinsic and intrinsic barriers to proliferation. A 2001 report from Lawrence Livermore National Laboratory (Hassberger et al., 2001) defines intrinsic barriers as “those features fundamental to the nuclear fuel cycle that deter or inhibit the use of materials, technologies or facilities for potential weapons purposes.” The same report states that extrinsic barriers “depend on implementation details and compensate for weaknesses in the intrinsic barriers. Safeguards, material control and accountability are examples of these extrinsic barriers, often referred to as the institutional barriers.”
The two most commonly used approaches to assessing the proliferation resistance of a nuclear fuel cycle are attribute-based and scenario-based analyses. Attribute-based analyses, such as the approach developed by the Texas A&M University’s Multi Attribute Utility Analysis (MAUA), attempt to identify and quantify both intrinsic and extrinsic barriers and then aggregate them. Scenario-based analyses model specific proliferation scenarios11 in their entirety rather than individual attributes of a particular system12 (TAMU, 2010). The Proliferation Resistance and Physical Protection (PR&PP) Working Group within the Generation IV Initiative Forum (GIF) has developed a scenario-based approach (TAMU, 2010).
Extrinsic barriers have been established by a number of diplomatic efforts, such as international treaties. They include the Nuclear Non-Proliferation Treaty (NPT)—which opened for signature in 1968 and entered into force in 1970—and the IAEA, founded in 1957. As of November 2010, the IAEA, which facilitates the peaceful use of nuclear material and facilities,
__________________
10 This definition of proliferation resistance draws on the International Atomic Energy Agency definition used by William Charlton in his Chapter 3 discussion of proliferation resistance. However, other definitions of proliferation resistance are used, for example, the Generation IV International Forum’s Proliferation Resistance and Physical Protection methodology uses a definition based on six measures of proliferation resistance (http://www.gen-4.org/Technology/horizontal/proliferation.htm).
11 Scenarios describe the sequence of events of an attempt to proliferate, beginning with the initial efforts and continuing through each stage (and associated barriers) to the final objective.
12 In this context, a “system” refers to the technological system being modeled, i.e., the fuel cycle or the facility.
had 151 member states willing to comply with various inspections and safeguards.13 One hundred ninety nations are parties to the NPT,14 with the notable exceptions of Israel, Pakistan, and India. North Korea signed the treaty in 1985 but withdrew in 2003 (UN, 2011). Another example of an extrinsic barrier might be the adoption by the Nuclear Suppliers Group of criteria for sales of sensitive fuel cycle technologies.
Other, related, extrinsic barriers—specifically, safeguards measures15—can be highly technical, and require technical assessments to understand, improve, and maintain their efficacy. Nuclear safeguards measures are applied primarily to non-weapons states and require nuclear facility operators to maintain and declare detailed accounting records of all movements and transactions involving nuclear material. These records are verified through official in-person inspections by the IAEA as well as surveillance cameras and other instrumentation.16
The remainder of the report discusses the key questions for non-proliferation policy and the potential for technical assessments to provide valuable input to these questions.
__________________
13 See: http://www.iaea.org/About/Policy/MemberStates/
14 See: http://disarmament.un.org/TreatyStatus.nsf/
15 The IAEA defines “safeguards measures” as “activities by which the IAEA can verify that a State is living up to its international commitments not to use nuclear programmes for nuclear-weapons purposes.” (http://www.iaea.org/Publications/Factsheets/English/sg_overview.html)
