2

Radioactive Source Uses, Risks, and Control

This chapter provides a broad overview of current uses of radioactive sources and discusses factors that affect safety and security risks associated with use of these sources, the roles and responsibilities within the government and other organizations to reduce these risks, and efforts to track and dispose of radioactive sources at the end of life.

2.1 RADIOACTIVE SOURCE APPLICATIONS

Radioactive sources have many beneficial uses in a range of medical, research, and commercial applications both within the United States and elsewhere. In medicine, radioactive sources are used for blood irradiation to prevent transfusion-associated graft-versus-host disease and in cancer therapy to treat tumors by irradiating them from outside the body (in external beam therapy and stereotactic radiosurgery) or inside the body (in high-dose-rate brachytherapy). In research, radioactive sources are used to treat cells, tissues, or small experimental animals to advance knowledge in radiobiology, hematology, and other branches of medicine. Commercial applications include sterilization to eliminate microorganisms on medical devices and health care products, to extend the shelf life of food and agriculture products, to eliminate harmful bacteria and a variety of microorganisms, and to manage reproduction in insects or pests. Other commercial applications include industrial radiography to visualize structures and detect defects, well logging to explore the structure and composition of the rocks and fluids in the subsurface and to measure fundamental petrophysical properties, industrial gauges for measuring the thickness, density, or fill level of a product, and radioisotope thermoelectric generators (RTGs) to power systems that cannot be easily accessed. These applications are discussed in some detail in Chapters 4 through 6.

The radioisotopes most commonly used in medical, research, and commercial applications discussed in this report are cobalt-60, cesium-137, iridium-192, and americium-241. About 90 percent of the activity from these radioisotopes (in particular, cobalt-60 and cesium-137) is used in Category 1 and Category 2 sources for radiation therapy, blood irradiation, research, sterilization, and other industrial applications. Most of the remaining activity from these radioisotopes is used in Category 3 sources for well logging, high-dose-rate brachytherapy, industrial gauges, and other applications.

Since the previous National Academies review of the applications of radioactive sources (NRC, 2008), no new applications of Category 1 and Category 2 radioactive sources have emerged. Although Category 3 sources were not examined in the previous National Academies report, no new applications of these sources have emerged either. However, one application of Category 1 sources, the RTGs for land-based power, has been phased out. RTGs continue to be used in space applications (see Section 6.5).

2.2 RADIOACTIVE SOURCE CHARACTERISTICS AND SAFETY AND SECURITY RISKS

Radioactive sources pose both safety and security risks. The safety risks involve failure, damage, human error, and other inadvertent acts that could result in accidental radiation exposures. The security risks involve theft, sabotage, and other malicious acts, including their incorporation in a radiological dispersal device (RDD) that could result in intentional radiation exposures. Radiological events involving either inadvertent or intentional acts, depending on the scenario and magnitude, can lead to a range of adverse health and socioeconomic effects. The level of exposure of persons would depend on many factors such as the physical and chemical form of the radioactive material used in the RDD and proximity of persons to the event. For example, people in the immediate vicinity to an RDD could be killed or injured by the explosion, and likely only a few people would suffer from deterministic health effects such as acute radiation sickness or radiation burns. Additional injuries and deaths due to chaos during evacuations of the affected areas could occur. Radiological events also have the potential for causing stochastic effects such as development of future cancers due to radiation exposure and long-lasting mental health effects due to perceived or actual radiation exposure. In addition, RDDs would have severe socioeconomic impacts resulting from contamination of structures and land, and the subsequent area denial.

In principle, safety and security risks can be assessed using a risk triplet: What can go wrong? How likely is it to go wrong? and What are the consequences if it does go wrong? (Kaplan and Garrick, 1981). Answering these questions for security events generally poses greater challenges than answering them for safety events, because determining the likelihood and consequences of malevolent acts is difficult. Specifically, for the characterization and quantification of the likelihood of terrorist acts, the universe of malevolent actors is unknown and the motivations, intentions, and capabilities of these actors are therefore difficult to assess. In addition, in contrast to adverse safety events, the probabilities of various security events involve strategic uncertainties in the sense that malevolent actors can modify their actions in response to defensive measures.¹

The International Atomic Energy Agency (IAEA) Code of Conduct for the Safety and Security of Radioactive Sources helps national authorities create and strengthen the regulatory infrastructures so that radioactive sources are used within an appropriate framework of radiation safety and security (IAEA, 2004). In addition, the IAEA has identified the need for Member States to perform threat assessment for radioactive material, associated facilities, and associated activities based on available intelligence, law enforcement, and open-source information (IAEA, 2019d). The IAEA does not discuss likelihood of any threats but encourages Member States to establish representative threats for analysis or to identify design-basis threats. The IAEA has produced a series of documents on security of devices and facilities that use radioisotopes.²

As noted in Chapter 1, in the United States the Interagency Task Force on Radiation Source Protection and Security was established to assess and provide recommendations to the President and Congress related to the security of radioactive sources from potential terrorist threats. In addition, emergency preparedness and response guidelines related to RDDs have been developed by several federal agencies including the U.S. Nuclear Regulatory Commission (U.S. NRC, 2020a), the Department of Homeland Security (DHS, 2017), the Environmental Protection Agency (EPA, 2017), and the Centers for Disease Control and Prevention (CDC, 2014).

The committee identified eight characteristics of radioactive sources that affect the likelihood of a source being involved in a radiological event and the likely severity of the consequences of that event. Generally, characteristics that affect safety risks related to radioactive sources also affect the security risks associated with these same sources. These characteristics follow:

1. Total activity of the radioactive source. This characteristic is the primary factor for defining Category 1 and Category 2 sources as high risk because it affects their potential to cause deterministic harm. It also affects the potential of a source to cause stochastic harm and area contamination and denial. High-activity sources are the primary focus of regulatory agencies to reduce safety and security risks. However, Category

___________________

¹ Risk analysis of and decision making on security issues have been under consideration over the past decade. The U.S. NRC has conducted public meetings on this topic, and a summary of a recent presentation is available (U.S. NRC, 2019b). Approaches to quantifying terrorist risk events are given by Garrick (2008, Chapters 2 and 5).

² A complete listing is provided at the IAEA website, https://www.iaea.org/publications/search/type/implementing-guides.

1. 1 sources, such as panoramic irradiators containing the highest activity sources available in commercial applications, are installed in large stationary devices within specially designed facilities and are therefore harder to access and remove. In addition, the above-mentioned sources could cause immediate or near-term harm to the person who handles them, making them an unlikely target for theft unless the source is adequately shielded as it is being removed and transported.
2. Half-life of the radioisotope contained in the source. The half-life dictates the timescale of the risk. Radioactive sources with half-lives on the order of hours to minutes or less can pose severe risks to the person who handles them or to anyone in close proximity to them, but, because they decay quickly, they do not contaminate areas for long time periods. Radioactive sources with very long-lived radioisotopes (e.g., thousands of years) are less active. The radioisotopes that are the focus of this report and are the most frequently used Category 1, 2, and 3 sources (cobalt-60, cesium-137, iridium-192, americium-241, and selenium-75) have half-lives that range from 74 days to 432.7 years and make attractive candidates for malicious use due to their potential to contaminate areas for long time periods and to cause area denial. Long-term consequences of an event that involves loss of control of cesium-137 with a half-life of 30.17 years would cause area denial for much longer than iridium-192 with a half-life of 74 days.
3. Physical and chemical forms and dispersibility of radioactive sources. Powders, a typical form of cesium-137, are easier to aerosolize and disperse than solid pellets, a common form of cobalt-60. More dispersible sources can potentially lead to more external contamination of people and structures and denial of larger areas. Consequently, remediation costs, especially if the radioisotopes involved diffuse into building surfaces, will be higher. Additionally, more dispersible sources are more likely to cause internal contamination to exposed individuals through respiration or ingestion. For these reasons, governments around the world have identified the cesium chloride salt form of cesium-137 as a major threat if used in an RDD, and many are taking measures to reduce or eliminate its use in radioactive sources used in medical, research, and commercial applications.
4. Aggregation of sources to Category 1 or Category 2. Although some individual sources may have low activity, if they are aggregated in close proximity in a single storage or use location, their total activity can be higher and therefore pose higher safety and security risks. Several sources in different settings are vulnerable to aggregation. These sources are typically portable and can include high-dose-rate brachytherapy sources, as well as sources used in radiography and well logging. The IAEA, the U.S. NRC, and other agencies and organizations recognize the risks related to aggregation of sources. For example, the U.S. NRC provides the requirements for the physical protection program for any licensee that possesses an aggregated Category 1 or Category 2 quantity of radioactive material in 10 CFR Part 37 (see Section 2.4.1).
5. Prevalence of sources. The widespread use of sources creates more potential for accidents or diversion. In the scope of this study, the most common radioactive sources contain cobalt-60. In 2020, the U.S. NRC reported that 93 percent of Category 1 sources and 85 percent of Category 2 sources contain cobalt-60. The greater the number of facilities using a source, the greater are the safety and security risks associated with that type of source.
6. Portability of sources. Radioactive sources that are installed in large stationary devices that are difficult to remove are more secure than sources that are installed in mobile devices that are more easily removed or in devices whose use requires frequent and long-distance transportation. Cobalt-60 sources, although prevalent, are almost all installed in stationary devices. However, iridium-192 sources used in industrial radiography are installed in mobile devices and are transported frequently and require frequent replacement due to the shorter half-life of the radioisotope. According to event tracking databases (see Section 2.5.3), iridium-192 sources used in industrial applications contribute to the most accidents that involve deterministic health effects because of their portability and vulnerability to misplacement or loss. Iridium-192 is also the radioisotope mostly involved in radiological source transactions in the United States due to its shorter half-life (74 days), which necessitates source replacement every 3 to 4 months. According to the U.S. NRC, about 97 percent of transactions of Category 1 and Category 2 sources involve this radioisotope.³

___________________

³ Margaret Cervera, U.S. NRC, presentation to the committee on June 11, 2020.

7. Accessibility of sources. Sources in facilities with restricted access may be less vulnerable than sources in facilities that have less restricted access. For example, industrial sterilization facilities limit access to authorized personnel only, whereas medical and research facilities typically can be accessed by visitors.
8. Safety and security protocols. Availability of safety and security protocols, their quality and effectiveness, and the level of adherence to them affect the likelihood of radiological events. Effectiveness of the protocols typically depends on the physical security of the location in which the radioactive source is used and stored as well as the level of training of the operating personnel. Once a radiological event has occurred, the emergency response capability locally available as well as the mitigation capacity that can be mobilized affect the consequences of the event.

Safety- and security-related radiological events may occur throughout the life cycle of a radioactive source’s use. Figure 2.1 presents possible safety- and security-related radiological events in relation to the source characteristics described above and provides key inputs to the assessment of safety and security risks from a radiation source. For each event type, a range of off-normal initiating occurrences would be hypothesized (from the risk paradigm: What can go wrong?) and then failure modes would be delineated to define the outcomes. Quantification of sequences of events that lead to these outcomes would be performed (from the risk paradigm: How likely is it?).

2.3 RADIOLOGICAL SAFETY AND SECURITY EVENTS

This section summarizes safety- and security-related radiological events that have occurred over the past 10–15 years, that is, since the 2008 National Academies report was released. Major events that occurred prior to 2008, including the events in Juarez, Mexico (1983); Chernobyl, Ukraine (1986); Goiânia, Brazil (1987); Acerinox, Spain (1988); and Nigerian oil fields (2002 and 2004), and resulted in relevant lessons learned are described elsewhere (see, e.g., IAEA, 1988, 1998; NRC, 2008). Brief descriptions of the events at Mayapuri, India (2010); Fukushima, Japan (2011); Tepojaco, Mexico (2013); and the University of Washington, United States (2019) follow.

2.3.1 Mayapuri, India, 2010

In February 2010, a research irradiator at Delhi University containing about 3,600 Curies (Ci) or 1.33 × 10¹⁴ becquerel (Bq) of cobalt-60 was auctioned to the Mayapuri scrap market in New Delhi. The irradiator had been purchased in 1968 but had not been used since 1985 and was in storage at the university’s chemistry department. The auction violated national regulations for radiation protection and safety of radioactive sources.

At the scrap market, the source was dismantled by workers who were unaware of the hazardous nature of the device. Seven people were exposed to radiation doses ranging from 0.6 to 6.8 Gray (Gy), and one person died. India’s Atomic Energy Regulatory Board and national emergency response agencies were involved at the source recovery operations (IAEA, 2013a). All 16 cobalt-60 pencils were recovered, of which 4 were recovered intact and the rest scattered in pieces. The event highlights the risks associated with inappropriate disposal of radioactive sources.

2.3.2 Fukushima Daiichi Nuclear Power Plant Accident, Japan, 2011

On March 11, 2011, the 9.0 magnitude Great East Japan earthquake followed by an unexpected 15-m tsunami damaged the power supply and cooling system of the three operational reactors at the Fukushima Daiichi nuclear power plant. The reactors proved robust seismically but could not withstand the tsunami. The Fukushima Daiichi nuclear power plant accident is considered the worst accident at a nuclear power plant since the Chernobyl disaster (IAEA, 2015c).

Assessments of the total radioactivity released have been performed by many organizations using different models. Most recently, the United Nations Scientific Committee on the Effects of Atomic Radiation issued an updated report (UNSCEAR, 2020) on the levels and effects of radiation exposure due to the Fukushima Daiichi nuclear power plant accident, which broadly confirms the major findings and conclusions of its previously published report (UNSCEAR, 2014). However, some of the doses to the public described in the previous report were overestimated. In addition, there was a considerable amount of new information becoming available on the levels of radionuclides in the environment, and in particular on concentrations of released radionuclides in the air as a function of time and on their physicochemical forms.

In general, a mixture of fission and activation products were released and now estimated at 120 petabecquerels (PBq) for iodine-131 and at 10 PBq for cesium-137 (UNSCEAR, 2020). These estimates are about 7 and 12 percent, respectively, of the corresponding releases estimated for the 1986 Chernobyl accident. Much (approximately 80 percent) of the released material was dispersed over the Pacific Ocean, but a significant fraction was dispersed over eastern mainland Japan. In addition, there were direct liquid releases and discharges from the Fukushima Daiichi nuclear power plant into the sea at the site.

There were no immediate radiation-induced health effects among plant workers or members of the public attributed to the accident. Doses received by workers involved in response and cleanup over the first 12 months after the accident had been about 13 millisievert (mSv), but a small percentage of the workforce (0.8 percent corresponding to fewer than 200 individuals) had received effective doses higher than 100 mSv with a maximum effective dose of about 680 mSv (UNSCEAR, 2020). Annual effective doses to the workers involved in cleanup have been declining since April 2012. Effective doses to members of the public were low: doses to adult evacuees were less than about 6 mSv, and average absorbed doses to the thyroid were less than about 15 mGy, compared with around 30 mSv and 500 mGy for the Chernobyl accident.

Thyroid exposure to radiation among children was also low on average because intake of iodine-131 was limited due to early evacuations, administration of stable iodine to residents, and restrictions placed on drinking water, fresh milk, and food (Kim et al., 2016). The thyroid dose of a few children ranged from 1 to 15 mGy. For comparison, the average thyroid dose among young children exposed during the Chernobyl accident was 1,500 mGy (Samet et al., 2018).

Following the Fukushima Daiichi nuclear power plant accident, thyroid ultrasound examination was offered to children age 18 or younger as a means to reassure members of the public that the radiation doses received were low and therefore the health risks were also low. However, the examination revealed an unexpectedly large number of thyroid cancers among children screened (200 cases up to 2019 among about 300,000 children screened) and raised concerns among residents and the public that it might be due to exposure to radiation from the accident. The examination findings also raised concerns within the scientific and medical community about overdiagnosis following thyroid screening caused by the use of highly sensitive ultrasound technology.

The Fukushima Prefecture government implemented the Fukushima Health Management Survey to monitor the health of the affected populations. By far the most prominent health effect reported in the survey is psychological effects among the evacuees and residents of Fukushima due to loss of family and friends, homes, employment, and sense of community, drastic relocations, and the perceived health risk due to radiation exposure (Suzuki et al., 2015, 2018). Because a portion of the population had suffered the combined impacts of the triple disaster (a major earthquake, a devastating tsunami, and a nuclear accident), it is difficult to assess to what extent the psychological effects observed could be attributed to each separately (IAEA, 2015c). To help address these effects, the prefectural government established the Fukushima Mental Health Care Center.

The economic consequences of the natural disasters and the Fukushima Daiichi nuclear power plant accident were enormous for the Fukushima prefecture because of severe losses in agriculture, manufacturing, and tourism. The economic consequences were also severe for Japan overall, particularly in the manufacturing and energy sectors. Over the past 10 years there has been gradual progress in economic reconstruction of Fukushima including cleanup of the nuclear site, but it has been a difficult and costly process. The Japanese government continues to face difficult decisions regarding decontamination of the plant, such as the fate of contaminated water from remediation activities and fuel debris removal and disposition from the reactors. In April 2021 the Japanese government approved the release of more than 1 million tonnes of contaminated water from the plant into the sea. The cost for cleanup of the nuclear site is estimated to reach 35 trillion to 80 trillion yen (approximately $350 billion to $800 billion) over 40 years (JCER, 2019). In addition, as of November 2020, more than 40,000 residents who lived near the nuclear power plant and were evacuated at the directive of the Japanese government due to this triple disaster are still under evacuation orders (FPG, 2020).

Although the Fukushima Daiichi nuclear power plant accident did not involve a radioactive source of the types addressed in this report, it demonstrated that an event that involves radiation can have large socioeconomic consequences even if it does not cause any immediate deaths due to radiation.

2.3.3 Tepojaco, Mexico, 2013

In December 2013, a truck transporting a cobalt-60 teletherapy source from a hospital in the northwestern city of Tijuana to a radioactive waste storage center was stolen in Tepojaco near Mexico City. The hijackers were unaware that the truck was carrying a high-activity (about 1,800 Ci or 70 TBq) source; their goal was to steal the truck (IAEA, 2013a). Law enforcement authorities recovered the truck and the teletherapy machine 2 days after the event. The radioactive cobalt-60 source had been removed from its protective shielding, but it remained intact (IAEA, 2013a). The six men who were arrested by the authorities on suspicion of stealing the truck did not appear to be exposed to high levels of radiation. This event highlights the risks associated with transportation of radioactive sources.

2.3.4 University of Washington Irradiator Event, 2019

On May 2, 2019, Department of Energy (DOE) subcontractor International Isotopes (INIS) was tasked with recovering a 2,900-Ci (approximately 107-TBq) sealed cesium-137 source from the Harborview Research and Training Facility in Washington State as part of the National Nuclear Security Administration’s (NNSA’s) work to remove and dispose of disused radioactive sources. The subcontractor’s attempts to mitigate unanticipated difficulties in dismantlement and disposition resulted in the release of a small amount of cesium, estimated to be about 1 Ci (37 GBq), resulting in internal and external contamination of 13 workers and observers who received effective doses no greater than 0.55 mSv. The facility was also contaminated, and more than 200 researchers and laboratory staff were forced to relocate into other spaces during recovery operations, with direct effects on more than 80 funded research programs with budgets estimated at tens of millions of dollars.⁴

A Joint Investigation Team co-led by NNSA and Triad National Security, LLC, completed a 9-month review of the event to identify the root and contributing causes (DOE, 2020). Lessons learned from the event include the importance of proper training of operations staff and improved understanding of roles and responsibilities of the several organizations involved in the immediate and longer-term recovery operations. Following the event, NNSA reevaluated the risk associated with all radioactive source recoveries and is changing how it recovers sources. Regardless of complexity, a hazard analysis is now required for any source recovery work. As of February 2021, remediation activities are almost complete, and reoccupation of the facilities is expected in fall 2021. The projected costs for response, recovery, remediation, and reconstruction, paid by NNSA, will exceed $100 million.⁵

The University of Washington irradiator event demonstrated that even small releases of radioactivity, in this case 1 Ci (37 GBq) of cesium-137 (a less than Category 3 amount), can incur a significant economic cost resulting from disruption of normal operations of the facilities involved.

2.4 RADIOACTIVE SOURCE CONTROL RESPONSIBILITIES IN THE UNITED STATES

The regulatory structures for licensing radioactive sources and ensuring their safe and secure use differ from country to country and a comprehensive review of these structures was outside the committee’s scope. This section focuses on the regulatory structure within the United States, which shares some similarities with other high-income countries. When possible, the committee makes comparisons with other countries.

In the United States, several government agencies and entities have regulatory or other authoritative responsibilities to ensure safe and secure use of radioactive sources in medical, research, and commercial applications, and to protect members of the public and the environment from possible adverse effects in case of safety or security events involving these sources. The role of agencies and entities that are relevant to the committee’s task are described in the following sections.

2.4.1 The U.S. NRC and Agreement States

The U.S. NRC licenses and regulates civilian use of radioactive materials and provides safety and security requirements associated with their use.⁶ Regulations in 10 CFR Part 20, “Standards for Protection Against Radiation,” Subpart I, “Storage and Control of Licensed Material” include security requirements for all radioactive materials except those that are specifically exempted by the regulations.⁷ The U.S. NRC issued Increased Control Orders in November 2005, which require licensees that possess Category 1 and Category 2 materials to provide for additional security for these materials.⁸ These Increased Control Orders were eventually replaced by the regulations in 10 CFR Part 37 titled “Physical Protection of Category 1 and 2 Quantities of Radioactive Material,” that

___________________

⁴ Lance Garrison, NNSA, presentation to the committee on June 12, 2020.

⁵ Lance Garrison, NNSA, presentation to the committee on February 25, 2021.

⁶ The complete U.S. NRC mission statement is as follows: “The NRC licenses and regulates the Nation’s civilian use of radioactive materials to provide reasonable assurance of adequate protection of public health and safety and to promote the common defense and security and to protect the environment.” See https://www.nrc.gov/about-nrc.html.

⁷ See https://www.nrc.gov/reading-rm/doc-collections/cfr/part020/part020-1801.html.

⁸ See Previously Issued Orders: Increased Controls at https://www.nrc.gov/security/byproduct/orders.html.

became effective in May 2013.⁹ Part 37 outlines the requirements for physical security, source monitoring, staff background checks, facility security plan, local law enforcement protection, training, and documentation. The security requirements contained in Part 37 use the IAEA source categorization system as the basis and focus on the potential of that material to cause prompt fatalities and the deterministic effects of radiation. Consequently, Part 37 only applies to Category 1 and Category 2 sources because of their potential (according to the IAEA system) to cause severe damage. Part 37 does not apply to Category 3, Category 4, and Category 5 sources unless they are aggregated at or above the Category 2 threshold.

In addition to the U.S. NRC, the 39 Agreement States also regulate radioactive materials use in accordance with the compatibility requirements of Section 274b of the Atomic Energy Act. Agreement States must, at a minimum, meet U.S. NRC regulations (be compatible with U.S. NRC regulations), and, in some cases, they fulfil their responsibility by promulgating regulations that are more stringent than those of the U.S. NRC. License conditions may also be imposed by the U.S. NRC or the Agreement States.

Radioactive material licenses fall under two broad categories: general and specific licenses. There are 19,300 specific licenses and 31,000 general license device holders in the U.S. civilian sector for use of radioactive material.¹⁰ The vast majority (80 percent) of these licenses are regulated by Agreement States.

The U.S. NRC defines a general licensee as “a person or organization that acquires, uses, or possesses a generally licensed device, and has received the device through an authorized transfer by the device manufacturer/distributor, or by change of company ownership where the device remains in use at a particular location.” Items and materials that may be possessed or used by operators pursuant to the general authorization are specified in regulations. In contrast, possession of a specific license requires the user to submit a license application and receive a license prior to obtaining radioactive material. Additional requirements for specifically licensed sources include adherence to license conditions, periodic renewals, and periodic inspections by the U.S. NRC or an Agreement State. Category 1 and Category 2 materials can only be obtained under a specific license. Category 3, Category 4, and Category 5 sources can be issued under either general or specific license.

A specific license is required in order to allow for the distribution of a generally licensed device. The specific licensee may distribute a generally licensed device to those who do not have a radioactive materials license, but the general licensee must comply with the regulatory requirements. The vast majority, approximately 80 percent of general license device holders, do not possess radioactive sources that are the topic of this report but instead possess sources such as road self-luminous exit signs (containing tritium), gas chromatographs (containing nickel-63), or static elimination devices (containing polonium-210).

In 2009, the Organization of Agreement States petitioned the U.S. NRC to increase the regulatory control over certain generally licensed sources. The petition resulted in a nondecision because of a tie vote within the Commission (U.S. NRC, 2009). However, the U.S. NRC did authorize Agreement States to increase controls on generally licensed sources at their own discretion, and some States enacted such increased controls for some of their general licensees (LLRW Forum, 2014).

National Source Tracking

The IAEA Code of Conduct (IAEA, 2004) calls for Member States to establish a national register of radioactive sources that tracks, at a minimum, all Category 1 and Category 2 sources. The Code of Conduct also notes that Member States should aim to harmonize the format of the registries to facilitate exchange of information regarding tracked sources. The U.S. NRC’s National Source Tracking System (NSTS) serves as the national registry of Category 1 and Category 2 radioactive sources.

The NSTS is part of a set of information technology tools known as the Integrated Source Management Portfolio (ISMP) developed to support the security and control of radioactive material. Other systems include the Web-Based Licensing (WBL) and the License Verification System (LVS). The individual systems that comprise the ISMP perform the following functions:

___________________

⁹ See https://www.nrc.gov/reading-rm/doc-collections/cfr/part037/index.html.

¹⁰ Margaret Cervera, U.S. NRC, presentation to the committee on January 30, 2020.

• The NSTS is an information technology system that tracks Category 1 and Category 2 sources from the time they are manufactured or imported through the time they are disposed of or exported or until they decay below Category 2 thresholds. All licensees, whether of the U.S. NRC or of an Agreement State, must report their Category 1 and Category 2 sources to the NSTS.
• The WBL is a Web-based repository of U.S. NRC and Agreement State licenses in possession of Category 1 and Category 2 quantities of radioactive material. All U.S. NRC licenses are reported in this database. Agreement States may choose for themselves whether to utilize WBL as their licensing database. However, if Agreement States do not utilize WBL as their database, they still must provide current copies of Category 1 and Category 2 licenses to the U.S. NRC for inclusion in the database.
• The LVS is a national verification system for Category 1 and Category 2 licenses aiming to ensure that only authorized licensees obtain radioactive materials in authorized amounts. LVS makes use of information stored in the WBL and the NSTS.

The data stored at the NSTS are organized by discrete sources, not by device or use, and the data in WBL are organized by total possession, not by device. According to the U.S. NRC, the deployment of the NSTS has enhanced accountability for risk-significant sources. Specifically, it enhanced the ability of the U.S. NRC and Agreement States to conduct inspections and investigations, communicate information to other government agencies, and verify legitimate possession and use of the tracked sources.¹¹ The U.S. NRC requires that licensees possessing Category 1 and Category 2 sources update the NSTS when they transfer or receive a source from another licensee. The U.S. NRC does not require licensees to declare whether a source is currently in use or in long-term storage. The NSTS tracks approximately 80,000 Category 1 and Category 2 radioactive sources. Of those sources, about 52 percent are Category 1 sources.¹²

The U.S. NRC and Agreement State regulations require licensees that possess Category 1 and Category 2 sources to annually reconcile physical source inventories with the source inventory in the NSTS.¹³ The reconciliation process involves confirming that the data in the NSTS are correct and resolving any discrepancies between the NSTS and the physical source inventory.

Information stored in the NSTS, the WBL, and the LVS is not publicly available, and the U.S. NRC does not publish aggregated data in public reports. The U.S. NRC routinely shares data from the NSTS with other federal partners, including NNSA, the Department of Homeland Security (DHS), and the Federal Bureau of Investigation, to support situational awareness.¹⁴

In 2020, NTI released its first Radioactive Source Security Assessment, which, without ranking or scoring countries, evaluated national policies and commitments in 175 countries and Taiwan to prevent the theft of high-risk radioactive materials. The NTI report raised concerns about the security of high-risk radioactive sources by noting, among other things, the absence of an active registry of radioactive sources in almost two-thirds of the countries examined. The authors of the report noted that the methodology used in the analysis did not involve in-depth country research but instead, the assessment relied on existing databases and other sources of consolidated information (NTI, 2020). For reasons unclear to the committee, the NSTS did not qualify as an “active registry” in this analysis.¹⁵

Category 3 Sources

Similar to all specifically licensed radioactive and nuclear material, Category 3 sources are subject to the security requirements of 10 CFR Part 20 and must be secured from unauthorized access or removal when in storage and be under constant surveillance and control when not in storage. Additionally, all Category 3 radioactive sources are subject to controls that are specific to their application. For example, radioactive sources used in well

___________________

¹¹ See https://www.nrc.gov/security/byproduct/ismp/nsts/overview.html.

¹² Letter from George Smith, U.S. NRC, to Ourania Kosti, National Academies, on February 5, 2021.

¹³ Reconciliation is required by 10 CFR § 20.2207(g), “Reports of transactions involving nationally tracked sources,” and by equivalent Agreement State regulations.

¹⁴ Letter from George Smith, U.S. NRC, to Ourania Kosti, National Academies, on February 5, 2021.

¹⁵ Sammantha Neakrase, NTI, presentation to the committee on January 6, 2021.

logging (see Section 6.3) are also subject to the regulations in 10 CFR Part 39 that include requirements for inventory maintenance, surveillance of operations, and secure transport. Other licensees who possess radioactive sources have regulations specific to their intended uses. However, as noted in the previous sections, security regulations and reporting requirements in 10 CFR Part 37 do not apply to Category 3 sources unless they are aggregated and meet or exceed the Category 2 thresholds. On several occasions, the U.S. NRC has considered changing its regulations related to Category 3 sources to more stringent ones.

In 2008 the U.S. NRC considered the inclusion of Category 3 sources and a subset of Category 4 sources in the just-established NSTS. These sources included fixed industrial gauges, well logging devices, high- and low-dose-rate brachytherapy, and certain radiography devices. At that time, it was estimated that the inclusion of Category 3 sources would add NSTS reporting requirements to approximately 1,000 licensees, or almost double the number of licensees required to report to the NSTS. Most of the comment letters on the proposed rule opposed this expansion of the NSTS for the following reasons:

1. The rule was premature and needed to be delayed to allow time to analyze the regulatory burden of adding these sources at the NSTS using actual experience from the operational NSTS for Category 1 and Category 2 sources.
2. The NSTS needed to be fully operational and successfully tracing Category 1 and Category 2 sources before the U.S. NRC adds a number of Category 3 and Category 4 sources to the system.
3. There needed to be additional justification regarding security risks posed by these sources before incurring additional burden of accounting for them at the NSTS (U.S. NRC, 2009).

On June 30, 2009, the U.S. NRC announced that the Commission “was unable to reach a decision on the staff’s recommendation to issue a final rule expanding the number and type of radioactive sources” covered under the NSTS (U.S. NRC, 2009).

The appropriateness of more stringent Category 3 source regulations was revisited again in 2014, when the U.S. NRC was required, via legislative mandate, to evaluate the effectiveness of the requirements of 10 CFR Part 37 and determine whether the requirements provided adequate protection. The agency concluded that the scope of 10 CFR Part 37 that is limited to Category 1 and Category 2 sources was appropriate based on threat, vulnerability, and consequence assessments at the time (U.S. NRC, 2014b). Most recently, the U.S. NRC Commissioners requested that staff evaluate whether there is a need to revise regulations or processes related to radioactive source protection and accountability. Impetus for that request was the Government Accountability Office (GAO) report on controls of dangerous materials (GAO, 2016), which recommended that the U.S. NRC take steps to

1. Include Category 3 sources in the U.S. NRC’s electronic system for managing information on sources.
2. Require that licensees who transfer Category 3 quantities of radioactive material to a potential recipient licensee confirm with the regulator or with the U.S. NRC’s electronic information management system the validity of the recipient’s license.
3. Consider requiring an onsite security review for all applicants for Category 3 licenses that the regulator does not know.

In response to the Commissioners’ request, the U.S. NRC and Agreement States formed a joint working group, the Category 3 Source Security and Accountability Working Group, to assess whether the agency should require additional security measures for Category 3 material. The working group determined that such material did not meet the threshold of prompt fatalities and deterministic health effects set by the U.S. NRC; that there is no evidence of adversarial interest in acquiring Category 3 quantities of material by theft; that security weaknesses at facilities that contain Category 3 quantities of radioactive material had not increased since first evaluated by the U.S. NRC; and the consequences of an RDD using Category 3 material are not significant enough to require additional security measures.¹⁶

___________________

¹⁶ The analysis and recommendations from these efforts are documented in SECY-17-0083 (U.S. NRC, 2017).

Based on the working group’s threat, vulnerability, and consequence analysis, U.S. NRC staff determined that data do not justify the cost associated with regulatory changes and recommended that the Commission not amend regulations to require license verification of Category 3 radioactive material or impose security requirements to prevent the aggregating of Category 3 material to a Category 2 quantity. The report recommended that the Commission approve the pursuit of rulemaking to require safety and security equipment be in place before granting a license for an unknown entity and clarify license verification methods for transfers involving quantities of radioactive material below the Category 2 threshold. At the time of this writing, the U.S. NRC had not yet issued direction on the proposed staff reevaluation of Category 3 source accountability.

Trends in Radioactive Source Inventory

The committee made several inquiries to the U.S. NRC to understand the current inventory of sources and inventory trends over the past 10–15 years. The committee learned that determining the number of devices licensed within periods of time (e.g., number of cesium-137 irradiators licensed in the past 10 years) would require access to both the NSTS and the WBL to match up discrete sources with possession limits and authorized uses. Doing so would be resource intensive, and the U.S. NRC could not provide the information to the committee. The U.S. NRC also told the committee that the number of material licensees has declined slightly over the past few years, and the U.S. NRC and Agreement States expect that general trend to continue. Some of this decline can be explained by merging of facilities (e.g., medical facilities) and consolidation of licensees.

Prior to the NSTS (from 2004 to 2008), the U.S. NRC maintained an interim database that was designed to collect a one-time inventory of devices and sources. Reporting to the interim database was voluntary. About 40,000 Category 1 and Category 2 sources were recorded in the interim database (NRC, 2008). Data from the interim database were transferred to the NSTS in 2008. In its first year of implementation (2009), the NSTS tracked approximately 60,000 sources, implying that the inventory of Category 1 and Category 2 sources was previously underestimated due to the voluntary nature of the interim database. An additional factor that explains the higher number of sources tracked at the NSTS was the inclusion of DOE sources that were not part of the interim database. As of February 2021, approximately 3 percent of sources in the NSTS were under DOE authority.¹⁷ As noted previously, as of February 2021, approximately 80,000 Category 1 and Category 2 sources are tracked by the NSTS. That is, the inventory of Category 1 and Category 2 sources has increased by about 30 percent from 2009 to 2021.

The U.S. NRC is in the best position to provide accurate information on what has contributed to the increases in the inventory of Category 1 and Category 2 sources in the United States. This would require an analysis of data stored at the different databases that comprise the ISMP. The committee was unable to receive information from the U.S. NRC on what contributes to the increase in the inventory. However, based on its own analysis of use of radioactive sources in the different applications and evaluation of trends for the past 10–15 years, the committee concludes that the biggest contributor to the increase in the radioactive source inventory is likely the use of cobalt-60 in industrial sterilization and specifically in medical device sterilization.

As discussed in Chapter 5, the medical device market in the United States is growing 5 to 7 percent annually¹⁸ because of increasing demand for existing medical devices and because of the availability of new products. The U.S. market for medical device sterilization grows at about the same rate as the medical device market. Despite the growing use of alternative technologies for medical device sterilization, use of cobalt-60 remains the most common radiation-based modality for this application in the United States and internationally.

Because of the lack of reporting requirements for Category 3 sources in the NSTS, the U.S. NRC does not have information on the number of Category 3 sources currently licensed in the United States. Prior to 2008, the U.S. NRC performed a one-time data collection and estimated that the number of Category 3 sources then was approximately 5,200. This number is likely an underestimate. If similar factors apply to Category 3 sources compared to Category 1 and Category 2 sources (i.e., the number of sources was underestimated by 50 percent when the U.S. NRC performed the one-time data collection in 2008, which was also voluntary, and the inven-

___________________

¹⁷ Email communication between Margaret Cervera, U.S. NRC, and Ourania Kosti, National Academies, on February 24, 2021.

¹⁸ Kathleen Hoffman, Sotera Health Services, LLC, presentation to the committee on October 13, 2020.

tory increases at the same rate as that of Category 1 and Category 2 sources), there would likely be more than 10,000 Category 3 sources in the United States today. Anecdotally, some experts have assumed that the number of Category 3 sources surpasses that of Category 1 or Category 2 sources. Tracking of Category 3 sources would elucidate this issue.

The committee did not carry out a comprehensive examination of international trends in radioactive source inventory. However, it requested and received information from the Canadian Nuclear Safety Commission (CNSC) on the country’s Sealed Source Tracking System (SSTS).¹⁹ Similar in concept to the NSTS, the SSTS tracks the creation and movement of all Category 1 and Category 2 sealed sources in Canada, as well as those imported and exported. Also similar to the United States, the inventory of Category 1 and Category 2 sources in Canada has increased over the past decade, but the increase has been greater. Specifically, in 2010, there were about 2,600 Category 1 sources and 22,500 Category 2 sources tracked by the SSTS, and in 2019 there were almost 7,000 Category 1 sources and 65,000 Category 2 sources tracked; that is, the number of sources tracked more than doubled. According to the CNSC, the increase in the source inventory can be attributed to disused or depleted sources being returned to manufacturers for recycling or long-term storage, to sources changing category due to decay, and to an increase in the number of sealed sources manufactured in Canada. Trends in source tracking in Canada are publicly available in annual reports and posted online on the CNSC website.

Category 3 sources are not tracked by the SSTS, and therefore the number of Category 3 sources in Canada has not been accurately determined. However, the CNSC requests licensee inventories of all sources annually that are stored in a searchable file. On the basis of this process, the CNSC reported more than 57,000 Category 3 sources in 2019. The CNSC does not plan to include Category 3 sources in its source tracking system, noting that the decision is “based on a risk-informed approach that takes into account the current mechanisms for reporting inventories.”²⁰ In making the tracking of Category 3 sources mandatory, the CNSC identified challenges related to three prerequisite tasks:

1. Upgrading the existing database to support the increase in the number of transactions and online users.
2. Amending a large number of licenses to mandate tracking of Category 3 sources.
3. Requesting and obtaining the required information from licensees within a short time frame.

2.4.2 National Nuclear Security Administration

The Office of Radiological Security (ORS) within the DOE NNSA engages in a three-pronged strategy for the security of high-activity radioactive sources:

1. Protection of radioactive sources used for medical, research, and commercial purposes;
2. Removal and disposition of disused radioactive sources; and
3. Reduction of the global reliance on radioactive sources by promoting the adoption and development of nonradioisotopic alternative technologies.

For the first effort, NNSA works with the U.S. NRC, the materials licensees, state, local, and tribal governments, and other federal agencies to build on the existing regulatory requirements by providing voluntary security enhancements. Examples of these enhancements include voluntary security upgrades such as cesium chloride irradiator hardening and facility-specific security upgrades; specialized training of local law enforcement to better respond to alarms at facilities with nuclear and radioactive materials; and security, including a testbed and voluntary pilot demonstration of cargo hardening, alarm assessment, and shipment tracking. To date, nearly 575 licensees (representing almost 950 buildings containing risk-significant sources) have partnered with NNSA to upgrade their physical security measures. These upgrades provide additional protection beyond that required for

___________________

¹⁹ Letter from Eric Lemoine, CNSC, to Ourania Kosti, National Academies, on February 25, 2021.

²⁰ Letter from Eric Lemoine, CNSC, to Ourania Kosti, National Academies, on February 25, 2021.

compliance with the requirements in 10 CFR Part 37. NNSA also offers additional security enhancements and tracking technology for mobile sources used in well logging and industrial radiography.

For the second effort, ORS implements source removals through the Los Alamos National Laboratory and the Idaho National Laboratory. ORS removes excess, unwanted, or disused radioactive sealed sources that pose a potential risk to national security, health, and safety (see Section 2.8.5).

For the third effort, ORS implements reduction of reliance on radioactive sources through the DOE and NNSA national laboratories. In 2014, ORS launched the Cesium Irradiator Replacement Project (CIRP) to reduce the number of cesium-137 and cobalt-60 sources in use in the United States by offering incentives to replace cesium-137 and cobalt-60 blood and research irradiators with nonradioisotopic alternatives. This project is described in some detail in Sidebar 1.2. ORS also funds research and development in alternative technologies (see Section 3.6) as well as comparative studies to assess equivalency of alternative technologies with radioactive source applications (see Section 5.2.3). Finally, ORS works with international partners to develop alternative technology consideration via political engagement, outreach, implementation, or technical exchanges. Part of this effort is supporting networking and information exchange through the Ad Hoc Meeting of Stakeholder States Involved with Technological Alternatives to High Activity Radioactive Sources. NNSA has supported 5 working group meetings to date with roughly 60 participants from 26 countries. In addition, NNSA co-sponsors workshops and resulting publications on alternative technology–related topics.

2.4.3 Other Agencies

Several other agencies within the United States have roles and responsibilities related to the use of radioactive sources for specific applications.

Food and Drug Administration (FDA)

FDA regulates the manufacturers of radiation generating devices (i.e., electronic products and medical devices) and irradiated products including blood and food. Different offices within FDA are responsible for these regulations. For example, the Center for Devices and Radiological Health (CDRH) and the Center for Biologics Evaluation and Research (CBER) regulate blood irradiators and procedures for irradiation of blood through a joined memorandum of understanding (MOU). CDRH consults with CBER for market clearance of blood irradiators via the 510(k) process,²¹ which involves a comparison to a legally marketed predicate. The Center for Food Safety and Applied Nutrition regulates irradiation of food products.

Department of Transportation (DOT)

DOT, together with DHS, and the U.S. NRC are co-regulators for the safe and secure transport of radioactive material. The roles of DOT and the U.S. NRC in the regulation of the transportation of radioactive materials are described in an MOU (U.S. NRC, 2015a). The MOU includes development of safety standards and regulations; package review, inspection, and enforcement; reporting of accidents and events; and information sharing.

All shipments of radioactive materials must be packaged and transported according to DOT and U.S. NRC regulations. The radioactive shipping container and packaging required are determined by the nature and form of material to be shipped and its level of radioactivity. Shipping containers and packaging for radioactive materials can be classified as excepted packaging, industrial packaging, Type A packages, and Type B packages. Only Type A and Type B packages are relevant to transfer of the types of radioactive sources (Category 1, 2, or 3 sources) examined in this report. All Category 2 and Category 3 sources are required to be shipped in either Type A or Type B packages, depending on the activity of the source being transported. All Category 1 sources are required

___________________

²¹ A 510(k) is a premarket submission made to FDA to demonstrate that the device to be marketed is as safe and effective as a legally marketed device. A 510(k) is required when introducing a device into commercial distribution for the first time, when there is a change or modification to a marketed device that could significantly affect its safety or effectiveness, and some other situations.

to be shipped in Type B packages. Type A and B packages must pass certain tests to demonstrate the capacity to maintain their integrity without the release of their contents.²²

U.S. Department of Agriculture (USDA)

USDA’s Animal and Plant Health Inspection Service regulates measures (phytosanitary treatments, see Section 5.3) to prevent the introduction or spread of pests by killing or sterilizing with high efficacy.

National Institute of Standards and Technology (NIST)

The Radiation Physics division at NIST owns several radioactive sources, both cesium-137 and cobalt-60, to maintain the national standard for air kerma (kinetic energy released per unit mass) and absorbed dose. One of the primary missions of NIST is to disseminate the air kerma and absorbed dose standard data to secondary calibration facilities and end users (see Section 6.4) for use in calibration of radiation detection instruments.

2.5 RADIOACTIVE SOURCE INCIDENT TRACKING

The committee is aware of three databases that track incidents of misplaced or stolen nuclear material and radioactive sources. The goal of these databases is to improve nuclear and radiological security by identifying common characteristics of the incidents and trends. Incidents reported to or by these databases demonstrate that losses and unauthorized activities involving radioactive material such as theft and illicit trafficking occur often. The reported information also underscores the need to enhance the regulations governing their use, storage, transport, and disposal. By examining the data reported, the committee was able to draw the following observations:

1. Incidents involving high-risk Category 1 and Category 2 sources are relatively rare;
2. Incidents involving Category 4 and Category 5 medical and industrial radiography sources are the most common;
3. Transportation of radioactive sources creates vulnerabilities;
4. More stringent security regulations reduce the risks of sources being stolen or going missing;
5. Not all sources are retrieved after they are reported missing or stolen; and
6. The databases likely underrepresent the actual number of stolen or missing sources.

2.5.1 The IAEA’s Incident and Tracking Database

The IAEA maintains the Incident and Trafficking Database (ITDB) on incidents of illicit trafficking and other unauthorized activities and events involving nuclear and other radioactive material out of regulatory control, which includes lost or stolen radioactive sources. Member States voluntarily participate in its reporting system and set their own standards for what to disclose to the agency. Aggregated ITDB data are made publicly available through IAEA reports, in particular, the ITDB Factsheet. The latest ITDB Factsheet was issued in 2020 and contains data up to 2019 (IAEA, 2020a).

Between 1992 and 2019, the ITDB reported 3,689 entries of which 8 percent involved incidents with a confirmed or likely act of trafficking or malicious use and 64 percent were not related to trafficking or malicious use. The remaining entries (28 percent) were reported to be undetermined in nature; that is, there was insufficient information to determine whether the incidents were related to trafficking or malicious use (see Figures 2.2a and b). Over the 27-year reporting period of the ITDB, nearly two-thirds (60 percent) of the incidents have involved radioactive sources. Overall, fewer than 200 radioactive sources were reported to the ITDB as stolen or obtained with the intention of using them for trafficking or malicious use—roughly 8 per year—compared to approximately 80 each year that appear to simply have been losses or other incidents unrelated to trafficking or malicious use.

___________________

²² See https://www.nrc.gov/reading-rm/doc-collections/cfr/part071/full-text.html.

The majority of thefts and losses reported to the ITDB were Category 4 and Category 5 sources used in industrial or medical applications. Most industrial sources that are reported stolen or lost are those used for nondestructive testing and for applications in construction and well logging and contain isotopes such as iridium-192, cesium-137, and americium-241.

The recovery rate for Category 1, 2, and 3 radioactive sources is high. Most incidents relating to Category 4 and Category 5 radioactive sources do not have a follow-up report confirming their recovery (IAEA, 2020a). In 2019, 189 incidents were reported to the ITDB, and they involved Category 2 to 5 sources.

2.5.2 Nuclear Material Events Database

The U.S. NRC, with support from the Idaho National Laboratory, has maintained the Nuclear Material Events Database since 1990. The database contains records of events involving licensed radioactive materials including lost, abandoned, or stolen radioactive sources or other material reported to the U.S. NRC by licensees or Agreement States. Since inception, the database has accumulated approximately 25,000 records of events involving radioactive material. Annual reports from 2007 to the present are available publicly; the latest report reviewed by the committee includes data up to 2019 (U.S. NRC, 2020c). Most of the radioactive materials reported as lost, stolen, or misplaced are Category 4 or Category 5 and involve mobile or portable sources and devices. When these materials are lost, stolen, or misplaced, they most often do not result in injury to either workers or the public and most are recovered.

An analysis of events demonstrated that the number of thefts reported in the Nuclear Material Events Database has decreased since issuance of 10 CFR Part 37. Specifically, since issuance of the regulations, there have been no thefts of Category 1 sources and six thefts of Category 2 sources of which five were recovered. The source that was not recovered decayed below the Category 2 threshold (U.S. NRC, 2018).

In 2017, approximately 249 sources were lost, abandoned, or stolen. One-third of these have not been recovered. Of the 249 lost sources, none were Category 1, seven were Category 2 iridium-192 radiography sources, and one was a Category 3 iridium-192 brachytherapy source. Except for one Category 2 source, all were recovered (U.S. NRC, 2018).

2.5.3 Global Incidents and Trafficking Database

The James Martin Center for Nonproliferation Studies (CNS) has maintained the Global Incidents and Trafficking Database with funding from NTI since 2013 (CNS, 2019). Since the database was created, researchers at CNS have identified more than 1,000 incidents worldwide using open-source reports and other publicly available information. The majority of the incidents were reported from North America (see Figure 2.3), but this is probably related to more transparent reporting systems in the United States and Canada. About half of these incidents involved radioactive material, with cesium-137 sources most often reported as lost or stolen (see Table 2.1).

In 2018, five cases of intentional trafficking of nuclear and other radioactive materials were recorded:

• Ukrainian security services arrested six individuals believed to be part of an international radioactive materials smuggling ring. The individuals were arrested after attempting to sell police an unspecified quantity of radium-226 in a sting operation. It is unclear how the individuals acquired the material.
• Ukrainian security services seized a device containing radioactive material from an individual who planned to sell and mail the device to an unnamed European country.
• Four scrap metal dealers in the Netherlands were arrested after authorities determined they were illegally selling radioactive scrap metal used in ballast blocks on ships.
• Sheremetyevo airport customs in Russia found a “yellow, radioactive mineral” in a package arriving from Italy. Presumably, the material was confiscated.
• Customs officials in Orenburg, Russia, confiscated 292 “medical medallions” from a truck driven by a Kazakhstani citizen. The medallions were reportedly being smuggled into the country, and gamma radiation was registered at 20 times above background level.

TABLE 2.1 Reported Incidents by Material Type in CNS Global Incidents and Trafficking Database

SOURCE: CNS, 2019. Produced by the James Martin Center for Nonproliferation Studies (CNS) for the Nuclear Threat Initiative.

2.6 PHYSICAL RADIOACTIVE SOURCE TRACKING

As noted in previous sections, radioactive sources that are portable and often in transit are vulnerable to theft or diversion. Portable sources include radiography cameras and oil well logging devices, which are often in transit due to their applications in shipyards, power plants, and oil and gas fields.

Incidents that involve theft of portable radioactive sources do not necessarily imply that perpetrators are trying to steal radioactive sources to commit mayhem. Instead, they often want to steal the vehicles themselves or devices inside the vehicles because of the perceived value of the stolen objects. Having a means to track the physical movement of the devices containing radioactive sources while in transit can help timely recovery.

In 2012, the World Institute for Nuclear Security and the World Nuclear Transport Institute published a guide to help users and transporters of radioactive sources make informed decisions about source tracking (WINS and WNTI, 2012). That guide describes the general features of a tracking system that consists of an electronic device that would be attached to a transport vehicle carrying a radioactive source or affixed to the equipment containing the source or both approaches. The signal from the electronic device would be detected by Global Positioning System (GPS) satellites (providing location accuracy within 3 meters) and by cellular general radio packet service (GRPS) (giving redundancy in ensuring position determination). The guide points out that there are some remote parts of the world where GPS and GRPS coverage is not available or is unreliable. Additional considerations are ensuring that the signaling device has reliable power such as long-lived batteries, providing for continuous staffing of an operations and control center, and coordinating and communicating rapidly with response forces.

Ensuring the readiness of response forces can be particularly challenging when transportation crosses international borders or even borders within countries such as the individual states in the United States. Different response forces could have different requirements for notifying them and for handing off from one jurisdiction’s response force to another’s. The major requirement for a tracking and response system is to alert the response force and have the force arrive on the scene before the attackers can accomplish the task of misusing the source.

Tracking technologies have focused on having secure and reliable means to know the whereabouts of the types of portable sources mentioned above. For example, the Republic of Korea has developed the Radiation Source Location Tracking (RADLOT) system, which uses GPS and Code Division Multiple Access (CDMA)

networks²³ and is designed for real-time monitoring of industrial radiography sources. The RADLOT system has many mobile terminals, a central control center, and a secure communications network. In addition to real-time tracking of radiography sources in transit, it features a radiation monitor that detects radiation from the source to let operators know whether the detector has been detached from the radiography device (Jang, 2019).

During the 2012 Nuclear Security Summit in Seoul, the Republic of Korea announced its intention to work with the IAEA to deploy the RADLOT system as a pilot project in Vietnam. Vietnam has about 700 mobile sources in use, 600 sources in storage, and about 40 companies or groups with licenses for possession of these sources (Phi et al., 2018). In 2015, the Korea Institute of Nuclear Safety began transfer of the RADLOT technology to Vietnam, and in 2017, 30 RADLOT detectors were deployed to licensees and companies that use radiography cameras for testing (Phi et al., 2018).

In the United States, ORS has sponsored development by the Pacific Northwest National Laboratory (PNNL) of the Mobile Source Transit Security (MSTS) for tracking of industrial radiography and well logging sources. During a presentation to the committee, PNNL project leaders explained that the required characteristics of the tracking system are enhanced security of the device, cost-effectiveness, reliability, robustness, and effective alert and alarm communication.²⁴ Like RADLOT, MSTS uses secure satellite and cellular communication networks and has radiation sensors to detect the proximity of the sources in transit to the signaling devices. The next steps for MSTS are to form commercial partnerships, implement a manufacturing and distribution plan, and establish responsibilities among the manufacturers, distributors, and users.

Another physical radioactive source tracking device is the NucTrack Solution tracking system deployed in France. This system has features similar to those of the RADLOT and MSTS systems. Nuc21, the company that developed the NucTrack Solution, aims to deploy the system to other European countries (Moreau, 2019).

In the coming years, the challenge will be to provide for deployment of tracking systems across dozens of countries around the world where more than 10,000 portable sources are in use. PNNL representatives who briefed the committee noted that the incentive for industry to adopt these systems is the benefit to their business in protecting against the misuse of radioactive sources in transit. Countries that lack resources may need financial and technological assistance and may benefit from technical cooperation facilitated by the IAEA, as demonstrated by the example of Vietnam.

2.7 RADIOACTIVE SOURCE CONSEQUENCE ANALYSES

A number of potential scenarios for RDDs in terms of the source used and location have been formulated for establishing priority of protective actions and response guidelines and for assessing possible immediate and long-term exposures to the populations. Given that the likely goal of the terrorist group that uses an RDD is to cause massive panic and economic devastation, many of these scenarios are based on an RDD detonation in metropolitan or agricultural areas.

Studies conducted by Sandia have estimated the economic impacts of three scenarios involving RDD events to inform radiological material security programs:

1. Detonation of an RDD using a Category 1 radioactive source in lower Manhattan;
2. Detonation of an RDD using a Category 3 radioactive source in lower Manhattan; and
3. Detonation of an RDD using a Category 1 radioactive source in an agricultural area in California.

The analyses and conclusions of the RDD scenario in California are not publicly available. The analyses of the RDD scenarios in Manhattan are also not publicly available, but the general conclusions have been reported, including in a presentation to the committee.²⁵ These analyses employed modeling to assess physical impacts, emergency responses, and economic impacts of the event phases. Economic impacts were estimated as loss of gross domestic

___________________

²³ A CDMA network allows several transmitters to send information simultaneously across a single communication channel.

²⁴ Brian Higgins and Fredrick Mauss, Pacific Northwest National Laboratory, presentation to the committee on September 9, 2020.

²⁵ Larry Trost and Vanessa Vargas, Sandia, presentation to the committee on April 29, 2020.

product (GDP) during the entirety of the event phases and considered both physical and psychosocial impacts (see Table 2.2). The Sandia analyses reported that events involving an RDD that uses either a Category 1 or Category 3 source can have large negative economic impacts and that those impacts go beyond the actual contaminated area. The analyses also reported that RDDs using less regulated (both in terms of security and in terms of source tracking in national databases) Category 3 radiation sources can have impacts similar to those of RDDs using Category 1 sources. RDDs involving either a Category 1 or a Category 3 source are expected to draw resources from outside the region, potentially affect supply chains, and influence consumer demand for goods from the affected region. Therefore, the economic consequences ripple within the larger economy. Specifically, an RDD using a Category 1 quantity of radioactive material reduced GDP by approximately $30 billion. An RDD using a Category 3 quantity of radioactive material reduced GDP by approximately $24 billion. That is, a reduction of greater than 90 percent in the activity of the released material only produced a reduction in economic effect of 20 percent.

The Sandia analyses estimated that detonation of an RDD using a Category 1 or Category 3 source would have a comparable order-of-magnitude level of casualties due to evacuations (GAO, 2019). Few fatalities were estimated from the initial explosion and no fatalities due to direct radiation exposure in either scenario. The Sandia analyses did not provide projections of stochastic health effects such as development of cancer in the future following the two RDD scenarios.

GAO used the conclusions from the Sandia analyses to recommend to the U.S. NRC that it

• Consider socioeconomic consequences and fatalities from evacuations when determining requirements for security measures for radioactive materials that could be used in an RDD.
• Implement additional security requirements for smaller quantities of high-risk material.
• Require all licensees to implement additional security measures, if they own certain Category 3 sources at a single facility that in aggregate can reach Category 1 or Category 2 levels (GAO, 2019).

The U.S. NRC staff generally disagreed with the recommendations and responded that “the basis for GAO’s recommendations for further regulatory changes was not well-founded and did not consider all aspects of risk (i.e., threat, vulnerability, and consequence)” (U.S. NRC, 2019a). The U.S. NRC also noted that GAO’s recommendations lacked context by ignoring the federal, state, and local efforts to guard against possible malicious use of radioactive sources and the response and mitigation capabilities (U.S. NRC, 2019b).

The conclusions of the Sandia analyses make a strong point that Category 3 sources, if used in an RDD, can have significant economic consequences comparable to those of a Category 1 source. However, the accuracy of the economic estimates derived from the analyses cannot be fully assessed because the inputs and assumptions are not publicly available. The committee’s review indicates that there are several factors that appear to be unaccounted for in these analyses. Specifically, the economic loss estimates prepared by Sandia do not include loss-of-life estimates, and therefore they almost certainly underestimate the economic costs as they would be estimated in a Regulatory Impact Analysis required to support major rules issued by federal administrative agencies. For example, DOT uses a value of statistical life (VSL) of $9.6 million to monetize fatalities (DOT, 2016).

Sandia’s economic loss estimates also do not appear to account for substantial costs due to loss of time at work for those in the affected area and those subject to evacuation. For instance, assuming that those who are evacuated lose 4 days of work, there would be an additional productivity loss on the order of $124 million and $65

TABLE 2.2 Socioeconomic Consequences of RDDs Using Category 1 and Category 3 Sources

SOURCE: David Trimble, Edwin Woodward, and Jeff Barron, GAO, presentation to the committee on January 30, 2020.

million, respectively, for the two RDD scenarios.²⁶ Furthermore, there would be both immediate and longer-term psychological impacts that could potentially be monetized through increased need for mental health surveillance programs. Taking account of these costs on affected individuals would substantially increase estimates of the total social cost of the RDD events as estimated by Sandia.

An alternative approach to the focus on loss of GDP pursued by Sandia is the cost-benefit analysis concepts employed in Regulatory Impact Analyses of major federal rulemakings. Rather than estimating changes in economic activity as measured by changes in GDP, this alternative approach sums the various costs borne by those affected by the RDD. In addition to monetizing the loss-of-life and time costs of those evacuated, it would monetize the cost of area denial directly, possibly as the present value of the rent that would have been paid had the use of buildings not been denied.

Several federal agencies and others (Rosoff and von Winterfeldt, 2007) have conducted economic modeling analyses similar to Sandia’s to inform specific priorities and regulatory frameworks. The details of these other analyses are also not publicly available. Likely, the inputs, assumptions, and modeling programs used by the various analysts differ and in some cases they could over- or underestimate the impact assessments (Dombroski and Fischbeck, 2006). Comparing conclusions on the potential severity of an RDD across different analyses is difficult unless the inputs and assumptions are described in some detail.

To gain an informed appreciation of the consequence analyses performed by Sandia and other government agencies, a formal peer review by knowledgeable and independent experts would be needed. Such peer review could compare the inputs and assumptions used in the different analyses and independently verify and validate the modeling programs used. In addition, the peer review could result in a set of best practices for conducing such economic analyses following hypothetical RDD scenarios that involve radioactive sources.

2.8 END-OF-LIFE MANAGEMENT OF DISUSED RADIOACTIVE SOURCES

Disused sources are either excess, unwanted, or spent sources that can pose safety and security problems if not properly dispositioned. Spent sources, even when they can no longer be used in the practices for which they were authorized because of radioactive decay, may still be significantly radioactive and potentially hazardous to human health and the environment. Because there are approximately 80,000 Category 1 and Category 2 sources in the United States and approximately 2 million sealed sources in total, tens of thousands of sources are estimated to be disused (DSWG, 2021). While the exact number of disused sources in the United States is not known, the Off-Site Source Recovery Program gives a sense of the scale: from 1997 to January 29, 2021, it secured 41,070 sources that were no longer wanted by the users.²⁷ Internationally, the IAEA estimated that of the millions of sources known to have been produced, approximately 20 percent are disused, and almost all countries have existing inventories of disused sources in storage (IAEA, 2005). More recent IAEA data provide further insight into the scale of the numbers of disused sources. Specifically, from 2014 to 2019, the IAEA provided assistance services to about 20 Member States in recovering and conditioning more than 4,200 disused sources. Also, during that period, the IAEA helped with source removal of 155 disused high-activity teletherapy sources in 12 countries. The IAEA has ongoing projects for disused sources in 15 additional Member States.²⁸

Licensees in the United States and most countries are not required to declare whether or when radioactive sources in their possession are disused, nor are they required to provide for prompt disposition. Disposition can involve different options depending on the level of activity in the disused source and pathways available for safely and securely managing the disused sources. If source suppliers offer recycling options, certain disused sources can be returned to suppliers and manufacturers for recycling. Also, a disused source may have enough radioactivity in it for it to be repackaged for reuse in another application. Reuse can involve transferring the source to another

___________________

²⁶ The Census Bureau provides values for this back-of-the-envelope calculation in its fact sheet for New York City. See https://www.census.gov/quickfacts/fact/table/newyorkcitynewyork/PST040219. Putting per capita New York City income into 2020 dollars yields $39,580. A rough estimate of productivity loss per evacuee is the 4 lost days divided by 250 workdays per year. This $633 times the number of evacuees gives the estimate of total productivity loss.

²⁷ See https://osrp.lanl.gov.

²⁸ Ian Gordon, IAEA, presentation to the committee on September 9, 2020.

user or returning it to a manufacturer who can perform repackaging (WINS, 2020a). Recycling and reuse options are discussed in more detail in Section 2.8.1.

Disposal is the ultimate disposition option for disused sources that cannot be recycled or reused. Ideally, at the end of life, licensees safely and securely dispose of their spent sources in a commercial or government-operated or -approved depository. However, the high disposal costs, lack of adequate depositories, and insufficient guidance for disposal create an environment in which users have little or no incentive to dispose of disused sealed sources.

In the worst scenarios, some dispose of high-risk sources outside of regulated channels by abandoning or orphaning them. As has been demonstrated too often, orphaned radioactive sources can end up mixed with scrap metal. Scrap metal dealers and processors in the United States report hundreds of alarms on suspicious material each year. The problem is international, and the IAEA has reported that the agency is aware of hundreds of events each year involving sources mixed with scrap metal. Events include Ciudad Juarez, Mexico, in 1983; Goiânia, Brazil, in 1987; Samut Prakarn, Thailand, in 2000; and most recently, Mayapuri, India, in 2010 (see Section 2.3.1). These events resulted in fatalities and exposure to members of the public (Gasdia-Cochrane, 2018; IAEA, 1988, 2013b).

If a missing or orphaned source is not detected before the recycled device material is sent off site, exposure to members of the public may occur. In the past 40 years, recalls of table legs, elevator buttons, tissue box holders, rebar, stainless steel mesh, and fashion belt buckles have resulted from radioactive sources containing cobalt-60 being melted into other metals. In addition to the dangers of personnel exposure to the radioactive sources, the actual melting of a radioactive source can result in significant economic losses. In the United States, typical costs for cleanup range from $10 million to $12 million and can be as high as $30 million (Gasdia-Cochrane, 2018).

The IAEA Code of Conduct on the Safety and Security of Radioactive Sources and its implementing guidance documents—Guidance on the Import and Export of Radioactive Sources (IAEA, 2012b) and Guidance on the Management of Disused Radioactive Sources (IAEA, 2018a,b)—provide advice on end-of-life management options for disused sources. The IAEA recommends that countries have a policy and strategy for radioactive waste management including for disused sources. Options for managing disused sources include reuse and recycle, return to a supplier, storage, or disposal. The focus of this section is on disposal; other options are discussed only briefly.

2.8.1 Reuse and Recycle

Recycling is an effective way to delay the actual disposal of a source until another option becomes available. According to the International Source Suppliers and Producers Association, which represents about 95 percent of the globally produced and distributed sources, recycling is industry’s preferred option because it reduces the amount of radioactive material that needs to be produced (Fasten, 2012). Recycling involves disassembling the source and recovering the radioactive material as a single element, for example, cobalt-60, or mixed or chemically combined in more than one element, for example, americium-241/beryllium. Qualified technicians are trained to perform the procedure safely. The recovered material can often be reused as is after packaging it with the same type of material from other disused sources to reach the level of activity needed for a particular application. A recycled source needs to be reencapsulated either by overencapsulating it into a new tertiary capsule or removing the old outer capsule and replacing this with a new outer capsule of the same or different design. A different option is to destroy the source completely and reprocess the radioactive material and any other valuable components (Fasten, 2012).

Reuse means redeploying a source in the same or different application. There is no physical change made to the source, and its original identity is retained. This also defers the need for waste disposal and puts the source to better use, where it will have more controls than if placed into storage.

Both reuse and recycle have been implemented effectively, but they address only a small percentage of the large number of sources that require disposal. There are commercial entities that are licensed to process some sources for recycling and reuse. Organizations such as the Conference of Radiation Control Program Directors (CRCPD) and commercial entities can facilitate the transfer and reuse of some sources from one licensee to another.

2.8.2 Return to Supplier

Many source manufacturers and suppliers have programs to collect disused sources for reuse or recycling, or for transfer to another licensee. Return of a source to the manufacturer or supplier typically requires a prior arrangement with the user, and in some cases a “one-for-one” exchange where the user returns a disused source and concurrently purchases a replacement source. This practice is common with some source applications such as industrial radiography, panoramic irradiators and other irradiators that use cobalt-60, teletherapy, and brachytherapy. When the user replaces a spent source for a new source, there is typically no cost for taking the old source back.

It can still be challenging to return a disused source to a manufacturer or supplier even when there is a return program in place. One challenge is for the user to provide documentation of the country of origin of the source and where manufacturing occurred. A second challenge arises when source manufacturers have production facilities in several countries and different components of a source are manufactured in facilities in different countries, thus making it difficult to determine where to return the disused source. Additional challenges are the limited availability of certified transportation containers and the requirement for sources to have a special-form²⁹ certification to show which containers can transport them (Fasten, 2012).

2.8.3 Disposal

At the end of useful life, many radioactive sources will be classified as Class A, B, C, or Greater-Than-Class-C (GTCC) low-level radioactive waste according to the classification criteria in 10 CFR § 61.55, with Class A being the least hazardous and Class C being relatively more hazardous (see Table 2.3 for summary). Class A, B, and C waste are appropriate for “near surface disposal”; for these classes, there are four licensed low-level waste disposal sites in the United States (U.S. NRC, 2020b):

• EnergySolutions Barnwell Operations in Barnwell, South Carolina. This site currently accepts waste from Connecticut, New Jersey, and South Carolina. Barnwell is licensed by the State of South Carolina to dispose of Class A, B, and C waste.
• U.S. Ecology, in Richland, Washington. This site accepts waste from the Northwest and Rocky Mountain compacts. Richland is licensed by the State of Washington to dispose of Class A, B, and C waste.
• EnergySolutions Clive Operations in Clive, Utah. This site accepts waste from all regions of the United States. Clive is licensed by the State of Utah for Class A waste only.
• Waste Control Specialists (WCS), LLC, near Andrews, Texas. This site accepts waste from the Texas Compact generators and outside generators with permission from the Compact. WCS is licensed by the State of Texas to dispose of Class A, B, and C waste.

DOE has statutory responsibility for development of disposal capacity for GTCC waste, which cannot be disposed of in currently licensed commercial low-level waste disposal facilities. DOE has made significant progress toward establishing a disposal pathway for GTCC waste, including the highest activity cesium-137 sources used in blood and research irradiators. In October 2018, DOE issued the environmental impact assessment of GTCC waste disposal at the WCS Federal Waste Facility located in Andrews County, Texas, and stated its preference to dispose of the entire inventory of GTCC and GTCC-like wastes at this facility (DOE, 2018). The WCS site near Andrews, Texas, was permitted to accept Class A, B, and C wastes from 34 states without a commercial disposal facility.³⁰ As of this writing, no decision has been made on disposal of GTCC waste.

Updated U.S. NRC disposal guidance (U.S. NRC, 2015b) enables radioactive material licensees to dispose of many Category 2 cesium-137 sources at currently operational commercial radioactive waste disposal facilities.

___________________

²⁹ Special-form radioactive sources are sources in which the radioactivity is sealed inside a capsule so that it is much less likely to cause contamination. Sealed sources are examples of special-form material. Because they are more rugged and less likely to leak, the amount of activity that can be transported in a Type A or Type B package is much higher for special-form sources than for normal forms such as radioactive liquids and radioactive waste.

³⁰ See http://www.wcstexas.com/about-wcs/overview.

TABLE 2.3 Most Common Risk-Significant Sources and Waste Categorization

NOTE: See Table 1.1 for threshold activities for Category 1 and Category 2 sources that contain these radioisotopes.

SOURCE: Sarah Norris and John Zarling, NNSA, presentation to the committee on September 9, 2020.

Notably, in this guidance, for longer-lived radioisotopes such as cesium-137, the average concentration in Curies per cubic meter, not the total Curie content of the disused source, determines whether it can be disposed of at a commercial facility. The guidance also enables commercial disposal of most Category 1 and Category 2 cobalt-60 sources as Class A or B waste due to their short half-life. The U.S. NRC has rulemaking in process examining the potential for disposing of GTCC waste streams that are high-activity cesium and transuranic plutonium or americium in other than a geological waste repository. This rulemaking, which is in the early stages, could open up the potential for disposal at or deeper than 30 m, which is near the surface but not so deep as to be considered geological disposal.

Disposal of disused sources containing foreign-origin americium-241 is currently not permitted in the United States. The permit for the Waste Isolation Pilot Plant (WIPP) only allows for disposal of transuranic material such as americium that can be traced to the U.S. defense production programs such as those run by NNSA or its predecessor agencies. In 2004, NNSA’s stockpiles of americium-241 were exhausted, and as a result, the U.S. NRC permitted importation of americium-241 from Russia. Although it is not known exactly how much Russian-origin americium-241 has been imported to the United States, NNSA ORS estimates that about 39,000 americium-241 sources cannot be disposed of in the United States because of current legal restrictions, that about 7,500 of these are at the end of their working life, and that the number of these disused sources will likely increase to about 20,000 by 2025. If WIPP restrictions for disposal of these disused sources are lifted, NNSA estimates that disposal would only require a small fraction (about 0.003 percent) of the available disposal space at WIPP.³¹

2.8.4 Disposal Challenges

There are many challenges to radioactive waste disposal around the world. Users who are faced with a decision to adopt an alternative technology need to determine what to do with the radioactive source that they will no longer use. As noted in the previous sections, some organizations may have considered the full life cycle of the radioactive sources they possess and have made arrangements for their disposal, but many have not. Disposal costs can be a reason for an organization’s reluctance to adopt an alternative technology.

Access to a Waste Facility

The IAEA (2018a) reports that there are few countries that have full access to waste disposal facilities for disused radioactive sources. For radioactive sources with half-lives less than 30 years, most can be disposed of in near-surface facilities, and these are primarily located in countries with a nuclear power industry. Countries without nuclear power industries tend to store disused sources on the user premises or at a designated centralized facility until a waste facility might be built in the future. For transuranic sources such as americium-241 and plutonium-238, the only licensed facility in the United States is WIPP, but as discussed in Section 2.8.3, WIPP is only licensed for U.S.-origin defense-related transuranic waste and not foreign-origin material. As also discussed in Section 2.8.3, for Classes A, B, and C low-level wastes, the United States has four licensed sites. Unlike in most countries, the U.S. waste classification system does not have the category “intermediate-level” waste.

___________________

³¹ Sarah Norris and John Zarling, NNSA, presentation to the committee on September 9, 2020.

The Republic of Korea and Sweden have both licensed and operated geological disposal facilities for short-lived low- and intermediate-level waste. Some countries (Germany and Switzerland) expect that all low-level and intermediate-level waste will ultimately be disposed of in one multipurpose, deep geological facility as yet to be built.

Financial Guarantees

Paragraph 22(b) of the IAEA’s Code of Conduct states that Member States’ regulatory bodies should “ensure that arrangements are made for the safe management and secure protection of radioactive sources, including financial provisions where appropriate, once they have become disused” (IAEA, 2004). It is broadly recognized that those who benefit from using a source should pay for its disposition. In many cases the user has not considered the full life-cycle costs of owning and using radioactive sources, which should include the disposal costs. A financial guarantee is the licensee’s acknowledgment and commitment that there will be sufficient resources at the end of life of a source to manage its safe disposition. The financial guarantee also aims to mitigate the risk that users could go out of business before honoring their obligations to manage disposal.

In the United States, the cost for disposal of sources is based primarily on volume and activity; the higher the activity of the source, the higher the price. Although disposal costs of most sources under 1 Ci (37 GBq) range from $500 to $5,000, costs for disposal of larger sources can range from tens to hundreds of thousands of dollars. In addition, there are costs for interim storage, packaging and conditioning, and transportation associated with disposal of these sources (DSWG, 2012). According to ORS, domestic removal operations for high-activity disused sources have ranged from $100,000 to $175,000.³² These costs are well above what some users, such as a small hospital, can afford, and they were not considered when the source was purchased or received as a donation. Although most countries now require “take back” agreements between purchasers and suppliers, these are usually only a commitment by the supplier to take back the source, rather than a commitment to do so at a specified cost. Consequently, many users are not prepared or able to cover these unexpected disposal costs.

Current U.S. NRC financial assurance requirements in 10 CFR § 30.35, “Financial assurance and recordkeeping for decommissioning,” are not intended to address disposal of sealed sources but instead are aimed at licensees who possess certain byproduct material with half-life greater than 120 days and at activity levels above certain thresholds for decommissioning of facilities that may require decontamination prior to release. Specifically, regarding sealed sources or plated foils, 10 CFR § 30.35 requires a fixed dollar amount ($113,000) of financial assurance or a decommissioning funding plan for licensees who possess byproduct material with a half-life greater than 120 days and at activity levels above certain thresholds. The sealed byproduct material thresholds in 10 CFR § 30.35 for which financial assurance is required only apply to a subset of Category 1 and Category 2 sources. This fixed dollar amount for financial assurance is inadequate for the transportation and disposal costs of many Category 1 and Category 2 sources.

In 2016 the U.S. NRC staff submitted to the Commission a scoping study to identify the main factors that may affect decisions regarding development of new or modified requirements and guidance for byproduct material financial planning. The scoping study showed that financial planning for end-of-life radioactive source management can ensure that the full costs of purchasing and using these sources are appropriately considered. However, the implementation of new requirements would lead to increased regulatory costs and, according to the U.S. NRC, has the potential to adversely affect beneficial uses of these sources (U.S. NRC, 2016).

Some Agreement States have already implemented financial assurance requirements and storage time limits. For example,

• Texas has implemented a 2-year time limit for storage of disused sealed sources and collects fees from licensees to cover the cost of orphaned and abandoned source recovery;³³
• Illinois has implemented financial assurance requirement for most sources;³⁴ and

___________________

³² Sarah Norris and John Zarling, NNSA, presentation to the committee on September 9, 2020.

³³ Texas’ financial provisions for orphan sources (Health & Safety Code, Subtitle D, Nuclear and Radioactive Materials, Chapter 401, Radioactive Materials and Other Sources of Radiation, Subchapter H, Financial Provisions).

³⁴ Illinois has strict financial assurance requirements for sources (Title 32, Energy Chapter ii: Emergency Management Agency, Subchapter B: Radiation Protection, Part 326).

• Florida has a radiation protection trust fund covering all costs associated with licensee bankruptcy and orphaned sources.³⁵

Some countries, such as Canada, France, Germany, Switzerland, and the United Kingdom, have followed direction from the IAEA Code of Conduct and require the use of a financial assurance plan (Volders and Sauer, 2016). For example, licensees in Canada must make a financial guarantee, which is “tangible commitment” to having sufficient financial resources to safely terminate radiation source use (CNSC, 2020). Licensees can meet their financial guarantee obligations by participating in an insurance program administered by the CNSC. The licensee’s total liability for sealed source use is calculated according to a formula that yields a total liability proportional to the costs for the safe disposal of radioactive sources at the end of their useful life. Currently, annual premiums range from $25 to approximately $4,500 (CNSC, 2020).

Shipping Containers

A historical challenge to waste disposal efforts for high-activity sources is the limited availability of authorized Type B containers (referred to as packages by the IAEA) needed to transport the sources from the licensee to the secure disposal facility. Over the past decade, NNSA has completed the development, testing, and certification of two new transportation packages: the 435-B Type B package and the 380-B Type B package. The 435-B Type B package is lightweight, easy to transport, and capable of transporting a broader variety of radioactive devices compared to other packages. The 380-B package is able to transport devices that are challenging for other packages currently available. The first source removal using the 435-B Type B package was completed in March 2018 at a hospital that replaced a cesium-137 irradiator through CIRP (see Figure 2.4). Use of the 380-B Type B package is

___________________

³⁵ Florida has a Radiation Protection Trust Fund of 5 percent of the annual licensing and inspection fee to cover the cost for abandonment of radioactive materials, default on lawful obligations, and insolvency (64E-5.206 Section 404.122 and 404.131(2)).

expected to begin in spring 2021. The cost of these packages is very high; NNSA notes that the price of the 380-B Type B package is $1.5 million (NNSA, 2019). The leasing cost is tens of thousands of dollars.

NNSA has made available a 435-B Type B package for domestic use and provided one to the IAEA to assist in source removals worldwide. This package, because of its smaller and lighter design, makes it easier to move in countries with fewer resources. While this helps reduce the cost of transportation, that cost remains high and adds substantially to the overall disposal costs. There is a need to identify several other internationally certified Type B shipping packages that would have widespread applicability to disused sources.

2.8.5 Disposal Programs

In the United States, a few government-sponsored programs, such as the Off-Site Source Recovery Program (OSRP), help defray the disposal cost of sources, making it possible for licensees to register sources for disposal at a lower, subsidized cost. The OSRP is sponsored by NNSA and is implemented through the Los Alamos National Laboratory and the Idaho National Laboratory. The initial scope of the program included sealed sources comprising GTCC radioactive waste and was later expanded to include recovery of beta- and gamma-emitting sources. Once registered in the program, the sources are prioritized for waste disposal according to the radionuclide, activity, and other factors such as whether the source is intact or leaking. Some sources have a higher priority for removal based on safety or security concerns (e.g., cesium-137 and plutonium-238). In total, as of January 2021, the OSRP has recovered more than 41,000 sources located in the United States and more than 3,400 internationally from more than 1,250 sites.³⁶ More than 3,500 of those are Category 1 and Category 2 sources (Itamura et al., 2018). ORS also facilitates international removals by sharing expertise, for example, via advisory groups on removal planning and training to service providers on packaging.

NNSA also provides funding for the Source Collection and Threat Reduction (SCATR) program, which is administered by the CRCPD. The SCATR program aims to reduce the footprint of disused or unwanted radioactive material stored at the licensees’ facilities by providing assistance regarding their disposition. The SCATR program also provides financial incentives for disposal of Class A, B, and C sources with access to a commercial disposal facility in the form of cost-shared support for the packaging, transport, and disposal. Common SCATR-eligible sources are used in medical and industrial applications, such as calibration, brachytherapy, radiography, and density gauges. SCATR has collected and disposed of more than 30,000 sources since its inception in 2007.

Despite the obvious contributions by the OSRP and the SCATR programs, there are concerns that these programs also create unintended disincentives for users to take responsibility for the costs associated with disposal of their disused sources and instead rely on the government to contribute toward costs. As commercial disposal options for Category 1 and Category 2 disused sources become more widely available, access to subsidized disposal may become limited.

Internationally, several other programs exist to identify viable waste disposal options for disused sources and to ensure that all countries have access to dispose of disused radioactive sources. Some countries currently dispose of disused sources along with other radioactive waste. Countries with nuclear power programs have developed near-surface radioactive waste disposal facilities for low- and intermediate-level waste. However, the specific activity of many sources exceeds the waste acceptance criteria for such facilities. Countries without nuclear power programs are investigating borehole disposal (typically 100 m deep) as a potential management option for long-lived and high-activity disused sources (see Figure 2.5). The borehole disposal concept involves the placement of solid or solidified radioactive waste in an engineered facility of relatively narrow diameter bored and operated directly from the surface. Generic post-closure safety assessments using different scenarios and radionuclides have demonstrated that borehole disposal provides an appropriate degree of long-term safety. Safety and other considerations in the disposal of disused sealed radioactive sources in borehole facilities have been summarized by the IAEA (2003b). There is an ongoing initiative with two countries, Ghana and Malaysia, to develop and implement this waste disposal methodology. This is currently a pilot program, and it is envisioned that the concept will be

___________________

³⁶ See https://osrp.lanl.gov/images/Maps/Recoveries_to_Date.pdf.

further developed and standardized for use in countries where they do not have nuclear power and will not have a large-scale waste facility (van Marcke, 2019).

Many low- and middle-income countries are not able to prioritize the management of higher-activity disused sources, including their disposal. The IAEA has provided technical and financial assistance to remove high-activity disused sources in many countries through repatriation or recycling initiatives. These activities are typically funded by higher-income country donations into the Nuclear Security Fund and through the Technical Cooperation programs. For example, in 2018, the IAEA helped remove 27 disused highly radioactive sources from five countries in South America. These sources were mainly used for medical purposes and sterilization; some were disused and stored at hospitals for more than 40 years (IAEA, 2018a). As mentioned earlier, NNSA’s OSRP is also assisting countries with disposal of disused sources. For example, the program will assist with the decommissioning and removal of three cobalt-60 sources currently stored at a hospital in Guatemala. Other countries, including Canada, France, and Germany, also provide direct assistance to low- and middle-income countries for removal of radioactive sources.

Under a multiyear IAEA Technical Cooperation interregional project, Cradle-to-Grave Management of Radioactive Sources, the IAEA provides training and assistance for management of disused radioactive sources to participating countries. The IAEA strongly encourages countries to establish end-of-life management options prior to purchasing a new source and has developed programs to help countries understand the available waste disposal options and determine what option is best for their inventory of disused sources (Yusuf, 2020).

2.9 CHAPTER 2 FINDINGS AND RECOMMENDATIONS

Finding 1: Radioactive sources continue to be used broadly, both nationally and internationally, for medical, research, sterilization, and other commercial applications. No new applications of high-risk (Category 1 and Category 2) and moderate-risk (Category 3) radioactive sources have emerged during the past 10–15 years. One application of Category 1 sources, the use of radioisotope thermoelectric generators for land-based power, has been phased out.

The radioisotopes mostly used in medical, research, and commercial applications examined in this report are cobalt-60, cesium-137, iridium-192, and americium-241. About 90 percent of the activity from these radioisotopes (in particular, cobalt-60 and cesium-137) is used in Category 1 and Category 2 sources for blood irradiation, research, radiation therapy, sterilization, and other commercial applications. Most of the remaining activity from

these radioisotopes is used in Category 3 sources for high-dose-rate brachytherapy, industrial gauges, well logging, and other applications. Use of radioactive sources for RTGs for land-based power has been phased out. However, RTGs continue to be used in space exploration.

Finding 2: The U.S. government and the international community have taken actions to strengthen the security and accountability of radioactive sources. These actions focus primarily on high-risk (Category 1 and Category 2) sources because of their higher potential to cause deterministic effects in persons handling or coming in contact with them. Security and accountability for Category 3 sources has a lower priority because of their lower potential to cause deterministic effects.

The U.S. NRC increased its security regulations for Category 1 and Category 2 sources in 2013 by enacting 10 CFR Part 37, which outlines the requirements for physical security, source monitoring, staff background checks, facility security plan, local law enforcement protection, training, and documentation. Part 37 does not apply to Category 3 sources, because they are considered less risk-significant according to the IAEA’s source categorization system and U.S. NRC’s regulatory system, which are based on the potential of sources to cause deterministic health effects in a person handling or coming in contact with them, if these sources are not safely managed or securely protected. Despite revisiting the appropriateness of more stringent Category 3 source regulations at least three times in the past 10–15 years, the U.S. NRC maintains the position that additional security measures for Category 3 sources are not needed.

Finding 3: In the United States, Category 1 and Category 2 sources are tracked by the National Source Tracking System, a nonpublic centralized database maintained by the U.S. Nuclear Regulatory Commission since 2008. The number of Category 1 and Category 2 sources has increased over the past 12 years by about 30 percent.

Tracking of radioactive sources increases accountability for these sources both for the licensees and for the regulator. Today, the U.S. NRC tracks approximately 80,000 Category 1 and Category 2 sources from the time they are manufactured or imported through the time they are disposed of or exported or until they decay below Category 2. Of those sources, about 52 percent are Category 1 sources. The number of Category 1 and Category 2 sources combined in 2009 was 60,000. Because of the lack of reporting requirements for Category 3 sources at the NSTS, the U.S. NRC does not have information on the number of these sources licensed in the United States. Based on a one-time voluntary data collection prior to 2008 and assuming similar trends in production and use as with Category 1 and Category 2 sources, today there are likely more than 10,000 Category 3 sources in the United States.

Finding 4: The less stringent security measures and lack of national and international tracking of Category 3 sources make them vulnerable to unauthorized transactions and theft.

The U.S. NRC has reported that since implementation of 10 CFR Part 37 the number of thefts of Category 1 and Category 2 sources has decreased. The agency has also reported that implementation of the NSTS has enhanced the ability of the regulators to conduct inspections and investigations and verify legitimate possession and use of the tracked sources. In the absence of these security measures and a national tracking system for Category 3 sources, both in the United States and internationally, Category 3 sources, many of which are portable, are more vulnerable to unauthorized transactions and theft compared to Category 1 and Category 2 sources.

Finding 5: Recent modeling analyses of radiological events concluded that small radiation releases and small radiation exposures of populations below the levels that can cause deterministic effects may have serious and long-term economic consequences. Various real life radiological events are supportive of this conclusion. A safety system that is based solely on deterministic effects of radioactive sources may provide an inadequate level of protection to society.

Analyses by Sandia reported that Category 3 sources, if used in an RDD, can have significant economic consequences, comparable to those of a Category 1 source. The 2011 Fukushima Daiichi nuclear power plant accident in Japan and the 2019 incident during the sealed source recovery at the University of Washington in Seattle are two very different-size events in terms of the quantities of radioactive material released and land area contaminated, but both demonstrated the negative socioeconomic impacts in the absence of immediate fatalities from radiation exposures. The radioactive source recovery at the University of Washington showed that response, cleanup, and remediation of just 1 Ci (37 GBq) of cesium-137 (a less than Category 3 amount) can cost more than $100 million.

Recommendation A: The International Atomic Energy Agency, the U.S. Nuclear Regulatory Commission, and other organizations should consider reframing their source categorization schemes to account for both (a) probabilistic health impacts such as development of cancer later in life and (b) economic and social impacts. This reframing would lead to a more holistic description of overall risk, including potential consequences if the sources are not safely managed or securely protected.

Significant progress has been made since the IAEA’s categorization system was implemented in 2003 in understanding and quantifying the probabilistic health impacts and socioeconomic impacts of events that involve radioactive sources. In fact, a number of countries including the United States have developed methodologies to quantify these impacts as part of their radiological emergency response planning. Although the projected impacts can vary significantly depending on the scenarios considered and the inputs and assumptions of the modeling, this does not make them less important. The IAEA, the U.S. NRC, and other organizations should incorporate probabilistic health impacts and economic and social impacts in their source risk analyses and, subsequently, in their categorization schemes to fully address the consequences to the society if a radioactive source is not safely managed or securely protected. These more holistic analyses could demonstrate that the current numeric source-activity thresholds defining the lower limits for radioactive sources need to be adjusted.

Recommendation B: The International Atomic Energy Agency, the U.S. Nuclear Regulatory Commission, and other organizations should make changes to their security and source tracking guidance and regulations based on the outcome of the reframing in Recommendation A.

Following the reframing of the radioactive source categorization system as indicated in Recommendation A, the IAEA, the U.S. NRC, and other organizations will likely need to make adjustments to their security and source tracking guidance and regulations so that sources that pose high risks in terms of deterministic, stochastic, and economic and social impacts are strictly regulated.

Recommendation C: In parallel, the U.S. Nuclear Regulatory Commission should phase in tracking of Category 3 sources in the existing National Source Tracking System. Such tracking would provide a more accurate accounting in the national inventory of Category 3 sources and would increase accountability for owning these sources and regulating their use. The U.S. government should make informed decisions about potential security enhancements for Category 3 sources at the facilities where these sources are located.

Adoption of Recommendations A and B by the IAEA, the U.S. NRC, and other organizations, if successful, would take a long time, at least 5 years, because the organizations will need to receive buy-in from various stakeholders and go through their processes of preparing guidance or rulemaking. However, the committee judges that some action to improve accountability and security of Category 3 sources is warranted immediately. Tracking of Category 3 sources through the NSTS is a step toward that goal.

With more than a decade-long operating history, the NSTS has demonstrated that it can successfully trace Category 1 and Category 2 sources. Expanding it to include Category 3 sources would add some administrative burden to the U.S. NRC and the licensees but the benefits outweigh the challenges.

Finding 6: The U.S. government’s risk reduction goal of replacement of radioactive sources with nonradioisotopic alternatives will not be realized until disused sources are properly removed and disposed of. The high costs of disposal and the limited options, resources, and guidance for disposal domestically and internationally, may be prohibitive both for adoption of alternatives and for appropriate end-of-life disposal of radioactive sources.

There are many challenges to radioactive waste disposal around the world. Users who are faced with a decision to adopt an alternative technology will need to determine what to do with the radioactive source that they will no longer use. Some organizations may have considered the full life cycle of the radioactive sources they possess and have made arrangements for their disposal, but many have not. The high disposal costs, lack of adequate depositories, and insufficient guidance for disposal create an environment where users have little or no incentive to dispose of disused sealed sources. These challenges can also be a reason for an organization’s reluctance to adopt an alternative technology.

Anecdotal evidence suggests that only a fraction of sources that are replaced are appropriately disposed of. The remaining sources are typically stored in the licensee’s facilities, because there is no available disposition pathway or disposition costs are high and cannot be borne by the licensee. In those cases, introduction of an alternative technology may increase security risks instead of reducing them.

Recommendation D: The U.S. Nuclear Regulatory Commission should expand its current requirements for financial guarantees to ensure that they adequately cover the end-of-life management for newly licensed radioactive sources. The U.S. government should also develop and implement a national strategy for end-of-life management of currently owned and orphan Category 1 and Category 2 radioactive sources and should consider it for Category 3 sources.

It is broadly recognized that those who benefit from the use of the radioactive source should be responsible for its disposition. While existing U.S. NRC financial assurance requirements address sealed sources, they are inadequate to cover current disposal costs. The existing U.S. NRC financial assurance requirements are largely aimed at licensees who possess certain (unsealed) byproduct material with a half-life greater than 120 days and at activity levels above certain thresholds for decommissioning of facilities that may require decontamination prior to release. The U.S. NRC should expand its existing financial assurance requirements to ensure that licensees who consider purchasing new sources commit sufficient financial resources to dispose of radioactive sources at the end of their useful life.

Although applying financial assurance requirements for new sources is feasible to implement, applying these requirements retroactively is less feasible. Few users might have anticipated the high costs of disposal. To safely remove these sources, government assistance is needed in the form of technical support, subsidies, and other means. The Off-Site Source Recovery Program and the Source Collection and Threat Reduction program are examples of successful end-of-life management programs for radioactive sources. As part of the recommended national strategy to safely and securely dispose of radioactive sources, the U.S. government would need to identify solutions to repatriate and dispose of sources that currently cannot be repatriated or disposed of because of existing regulations.