Introduction

The Comprehensive Nuclear-Test-Ban Treaty (CTBT) was adopted by the United Nations General Assembly on 10 September 1996 [1]. It bans all nuclear explosions in any form from all signatory nations. While the CTBT has not yet entered into force, the International Monitoring System (IMS) is currently being built to monitor the world for nuclear explosions to look for possible treaty violations after its entry-into-force. The IMS consists of four main technologies: seismology, hydroacoustics, infrasound, and radionuclide. The radionuclide technology may be broken down into atmospheric particulate monitoring and atmospheric noble gas monitoring. The noble gas monitoring within the IMS is the focus point for this work.

The noble gas monitoring network is optimized for the measurement of 131mXe, 133mXe, 133Xe, and 135Xe [2]. These radioxenon isotopes are produced with high yield through nuclear fission. The noble gas properties of radioxenon facilitate the escape of these radionuclides from underground nuclear explosions. For example, 133Xe was the only fission product detected in the atmosphere resulting from the 9 October 2006 nuclear test conducted in the Democratic People’s Republic of Korea (DPRK) [3].

There are multiple sources of radioxenon in the environment. Anthropogenic radioxenon sources include commercial nuclear reactors, research nuclear reactors, and the radiopharmaceutical industry. Natural sources include spontaneous fission of 238U and 232Th in the Earth’s crust as well as cosmic ray activation of xenon in the atmosphere [4]. As a result, it is important to be able to differentiate between the sources of radioxenon through forensic signatures. Kalinowski and Tuma [5] demonstrated the differences in signature between commercial nuclear reactors and nuclear explosions. However, Saey et al. [6] and Biegalski et al. [7] illustrated the direct overlap in radioxenon signatures between nuclear weapon explosions and medical isotope production facilities. It is this overlap in forensic signatures that resulted in the initiation of the Workshop on Signatures of Medical and Industrial Isotope Production (WOSMIP) series [8].

It is the goal of this work to investigate possible tracers that may be utilized for identification of emissions from radiopharmaceutical facilities. It is recognized that the best solution would be to significantly reduce medical isotope emissions to the point where they would not be detected by IMS systems. However, implementation of strict emission controls at such facilities is not legally mandated and may have practical limitations. As a result, this work investigates an alternative solution. Tracers for radiopharmaceutical facilities may be in the form of radioactive or non-radioactive materials.

Radioactive tracers

Radioactive tracers have the advantage of being detectable with the IMS radionuclide monitoring stations without any modifications. For this work activation of Xe, Cs, Ba, and Ar were investigated. It was assumed that these elements would be irradiated in the same vessel as the U target. However, parallel irradiations and releases could also be conducted. If tracers are released from each target dissolution, then absence of that tracer in an IMS detection might exclude radiopharmaceutical production as a source of a particular detection. Conversely, if a tracer is detected in coincidence with other relevant radionuclides at a station, then the detection may be correlated to emissions from radiopharmaceutical production.

The radioxenon signatures of the medical isotope production was calculated via ORIGEN-ARP within the SCALE 6 software package [9]. ORIGEN-ARP models the complete reactor spectrum rather than just utilizing thermal neutron cross-sections. As a result, threshold reactions such as (n,α) and (n,p) are properly calculated. Uranium target specifications were taken as the average of parameters reported by Saey et al. [6]. Table 1 lists these values. The average values were utilized for modeling within this work. Figure 1 contains a multiple isotope ratio correlation (MIRC) plot illustrating the radioxenon signatures from a high enriched uranium (HEU) pulse reactor and medical isotope production. The HEU pulsed reactor is a bare sphere of 94 % 235U run for 1 μs to produce 1 kiloton of energy. It is assumed that the radioxenon signature from the HEU pulsed reactor is a close approximation of a nuclear weapon. The HEU pulsed reactor is broken into two lines that span the decay time from t = 0 to t = 10 days. The line on the right represents the radioxenon isotopic ratio signatures resulting from immediate fractionation from parent radionuclides within the fission product decay chain. The decay line on the left represents the radioxenon isotopic ratio signatures that results from the cumulative build-in of parent fission products. Figure 1 illustrates how the medical isotope production facility radioxenon signature overlaps the signature expected from a nuclear weapon.

Table 1 Parameters utilized for modeling medical isotope production
Full size table
Fig. 1
figure 1

Multiple isotope ratio comparison (MIRC) plot illustrating the radioxenon signature from medical isotope production

Full size image

An obvious radioactive tracer that we chose to calculate is the use of additional radioxenon in targets. The addition of different levels of radioxenon in the emissions would shift the signature in Fig. 1 so that it no longer overlaps with that of a nuclear explosion. Figure 2 illustrates the signature of naturally enriched radioxeon after undergoing the irradiation methodology identified in Table 2. It is shown that the signature of irradiated natural Xe is significantly different than that of a nuclear explosion. Figure 2 also displays the radioxenon signature resulting from the mix of 1 g of natural Xe (~0.2 L at STP) along with a U target for medical isotope production. The mixture causes the radioxenon signature to shift to the left due to additional 131mXe. Consequently such additions could be utilized to differentiate the source signature resulting from medical isotope production from that of interest to nuclear explosion monitoring.

Fig. 2
figure 2

MIRC plot for the activation of naturally enriched Xe

Full size image
Table 2 Perfluorocarbon tracer compounds [15]
Full size table

One additional benefit to the natural Xe irradiation is that it also produces 125Xe, 127Xe, and 129mXe which are not produced appreciably in nuclear explosions because their fission yields are low. The addition of these radioxenons could also be utilized as a tracer to the radiopharmaceutical facilities. It was also shown that the use of enriched Xe isotopes could be utilized. Isotopically enriched 130Xe produces pure 131mXe upon irradiation. This would have a similar effect to what is shown in Fig. 2. Isotopically enriched 124Xe, 126Xe, and 128Xe could also be irradiated to produce unique radioxenon signatures. It should be noted, however, that the inclusion of even 1 g of natXe to a target may prove to be operationally difficult.

The irradiation of Ba and Cs were also investigated, as they could potentially be included as solid material within the U targets. The activation of both of these elements results in radioxenon production and their source signatures are shown in Fig. 3. The radioxenon signature from Ba activation is significantly different from that of a nuclear weapon and the radioxenon signature of Cs represents that of a nuclear weapon explosion that has undergone significant decay. While interesting to study, calculations show that Ba and Cs activation are impractical for Xe production. In order to produce enough radioxenon to significantly shift the medical isotope production signature, one would have to irradiate 105 kg of Ba along with each U target.

Fig. 3
figure 3

MIRC plot for the activation of Ba and Cs

Full size image

The last radioactive tracers we investigated were Ar isotopes. Ar is another noble gas, but it is not actively monitored by the IMS. So, additional monitoring capabilities would have to be developed if radioargon was utilized as a tracer. Figure 4 illustrates the radioargon activities that would be produced for 1 g of Ar (~0.6 L at STP) activation with the methodology listed in Table 2. 41Ar has the highest activity after irradiation, but quickly decays after irradiation with a 1.83 h half-life. 37Ar is next in activity level and decays with a 35-day half-life. This puts 37Ar in a range where it may be practical as a tracer, but it decays only through β emission making detection challenging. 39Ar and 42Ar are produced at levels too low to be utilized as a practical tracer. As with the natXe, it may prove impractical to irradiate Ar gas during medical isotope production.

Fig. 4
figure 4

Activity of radioargon isotopes after irradiation of 1 g Ar

Full size image

Non-radioactive tracers

Non-radioactive tracers may also be utilized to identify emissions from medical isotope production facilities. Such tracers have been historically utilized for atmospheric transport studies. A good tracer is inert from an environmental chemistry perspective, is easily detected, and has low environmental background. Non-radioactive tracers identified for this work include SF6, 3He, perfluorocarbons, and trace elements.

Sulfur hexafluoride (SF6) is an anthropogenically produced compound that has been historically utilized for atmospheric tracer studies [10, 11]. It has a long atmospheric lifetime probability of 3,200 years and is easily detected through gas chromatography at levels below 1 ppt. The atmospheric background is largely from gas insulated switchgear systems [10]. The global SF6 background in the northern hemisphere has increased from 0.50 ppt in 1978 to 3.11 ppt in 1994 [10]. One negative aspect to SF6 is that it is a potent greenhouse gas [11].

3He has been employed as both an atmospheric tracer 12 as well as an underground water tracer [13]. The benefits of 3He include its low environmental background and ease of detection. Use of 3He as a super fluid and as a neutron detector makes it a highly desirable material. It has historically been produced as a byproduct of 3H decay. However, the production has now stopped and there are severe shortages of the material [14]. As a result, 3He will not be a practical tracer unless new production infrastructure is developed.

Perfluorocarbons (PFCs) are fluorocarbons made up by fluorine and carbon atoms only. They are chemically inert, thermally stable and have a low background. Table 2 shows a list of PFC compounds that may be possibly utilized for environmental tracers [15]. PFCs are detected via an electron capture detector or a negative ion mass spectrometer with detection limits as low as 1 part in 1015 by volume [16]. However, similar to SF6 PFCs are greenhouse gasses.

Trace metals are the last option identified in this work for non-radioactive tracers. Metal particulates have long been utilized as tracers for long-range atmospheric transport and wet deposition studies [17]. Elements of interest include Ag, Au, Eu, In, Ir, Os, Re, Ru, and Ta. These are largely rare earth elements that could be collected as atmospheric aerosols within the IMS system. However, analysis methods external to the IMS such as neutron activation analysis, inductively coupled plasma–mass spectrometry, or X-ray fluorescence would have to be utilized. In addition, metal particulates may react differently in the atmosphere than noble gases, and that would have to be taken into consideration.

Conclusions and recomendations

This study has identified possible tracers to be utilized by radiopharmaceutical facilities so that their emissions may be uniquely identified. Radioactive tracers including radioxenon and radioargon isotopes appeared feasible, though questions remain on the operational feasibility of irradiating these gasses. The benefit of radioxenon tracers is that they may be directly detected via IMS systems without additional efforts. Non-radioactive tracers including SF6, 3He, PFCs, and trace elements were discussed. The non-radioactive tracers have the benefit of being inert within the environment and easy to detect. However, additional detection systems would have to be implemented in the IMS for co-detection along with radioxenon.

If the use of tracers for identification of radiopharmaceutical facility emissions is to be seriously considered, work must be conducted to evaluate the practical implementation of such methods. A first step would be to evaluate the regulatory concerns of such emissions. Next one would need to conduct a cost analysis study for implementing these methods. Both infrastructure and operational costs need to be considered. Lastly, detection feasibility, data fusion, and system integration concerns should be assessed.