Side-reaction products identified for photo-nuclear production of⁹⁹Mo

Abstract

Production of ⁹⁹Mo by the ¹⁰⁰Mo (γ, n)⁹⁹Mo reaction through the bremsstrahlung process using an electron accelerator is one of the feasible options currently pursued by several countries. Here we report experimental results on identification of side-reaction products after the irradiation of natural and enriched ¹⁰⁰Mo targets. Side-reaction products identified include various Mo, Nb and Zr isotopes. Comparison of experimentally determined reaction production rates with those determined based on theoretical cross-sections will be presented. Moreover, activation products formed due to presence of impurities introduced during the manufacturing of the Mo targets will also be discussed.

Introduction

The most-used medical isotope for imaging, technetium-99m, the daughter of ⁹⁹Mo, is currently produced by fission of ²³⁵U in nuclear reactors because of the very high fission yield (6.1%) and the high thermal neutron fission cross section [1]. Most of the reactors used for the production of ⁹⁹Mo have been operating for decades. Recently, one of the major suppliers of ⁹⁹Mo, Nordion (Canada) ended processing of irradiated targets for production of ⁹⁹Mo, since the Chalk River reactor in Canada ceased its operation due to the age of the reactor. Many countries are looking for alternative production pathways for ⁹⁹Mo because of the regulatory approvals required for fission-made molybdenum and the complicated and expensive waste disposal options. Although neutron capture in a reactor using enriched ⁹⁸Mo targets by the ⁹⁸Mo(n,γ)⁹⁹Mo reaction can be used, the process still relies on the availability of a research or commercial reactor.

Accelerator technology provides a valuable alternative: The direct production of ^99mTc using the ¹⁰⁰Mo(p,2n)^99mTc reaction by cyclotrons [2,3,4,5,6] can be adopted by countries that have an infrastructure of strategically placed cyclotrons currently being used for the production of other medical isotopes, such as ¹⁸F. In addition, electron-beam accelerators can be used for production of ⁹⁹Mo through the bremsstrahlung process via the ¹⁰⁰Mo(γ,n)⁹⁹Mo reaction [7,8,9,10]. This is a very strong alternative because of the commercial availability of these accelerators, an easier waste disposal pathway, and the potential for producing kilocurie (kCi, 1kCi = 37 GBq) quantities of ⁹⁹Mo. For kCi quantities of ⁹⁹Mo, several-hundred-gram targets consisting of enriched ¹⁰⁰Mo are required, and, due to the high cost of enriched ¹⁰⁰Mo material, efficient recycling processes need to be in place. Recently, several studies of enriched Mo recycling [3, 11,12,13,14] reported recovery yields ranging from ~ 85 to ~ 98%.

Argonne National Laboratory, with support from the National Nuclear Security Administration’s Office of Material Management and Minimization, is providing technical development assistance to NorthStar Medical Technologies, LLC in its pursuit of ⁹⁹Mo production without the use of highly-enriched uranium. The U.S. Food and Drug Administration recently approved NorthStar’s RadioGenix® system for producing sodium pertechnetate from a low-specific-activity ⁹⁹Mo solution [15]. NorthStar is currently producing ⁹⁹Mo via a ⁹⁸Mo(n,γ)⁹⁹Mo reaction (neutron capture) pathway at the University of Missouri Research Reactor Center. Soon, a second production option will be available: the photonuclear reaction ¹⁰⁰Mo(γ,n)⁹⁹Mo from irradiation of enriched Mo targets by an electron accelerator. Although there are several reports discussing Mo cross sections [16, 17] and the formation of side-reaction products during the photonuclear reaction on ¹⁰⁰Mo targets [18, 19] most are based on theoretical cross sections or very short irradiation [20, 21]. Identification of the production rates for side-reaction products formed during the accelerator production of ⁹⁹Mo provides a better understanding of how the isotopic composition of the Mo target and the presence of other impurities can affect the final purity of the ⁹⁹Mo product. This is particularly important for tracking the level of certain impurities or when considering whether to use natural or isotopically ¹⁰⁰Mo enriched targets, due to long half-lives and a potential build-up of activity.

Here we report experimental data on side-reaction product yields collected after 0.5 and 4 h irradiations of natural ultra-high-purity (UHP) Mo and ¹⁰⁰Mo-enriched disks (both provided by NorthStar). Experimentally determined production rates are also compared to theoretically estimated production rates based on the JENDL photonuclear reaction cross section database and the particle and heavy ion transport-code system PHITS 3.02 simulation tool.

Experimental

Linac irradiations

Argonne’s low-energy and high-power electron linac was used as the source of the electron beam for the experiments. This electron linac operates with repetition rates up to 240 Hz with maximum beam power of up to 110 W per pulse. Effective beam energy is in the range of 20 to 50 MeV. The highest beam energy is about 53 MeV. The DC thermal gun produces electron beam pulses with amplitude up to 2.0 A and length up to 5.5 µs. RF power for the two accelerating structures is provided by two Thales TV2022A klystrons. After acceleration, the beam goes through an evacuated transport channel to the experimental hall and is delivered to the target face. Steering coils and quadrupole magnets keep the beam in the proper shape and position [22].

Mo irradiation runs were performed with a 40 MeV electron beam energy and beam parameters presented in Table 1. The typical beam energy spectrum at 37 MeV is shown in Fig. 1. The energy spectrum width was about ± 0.9 MeV or ± 2.3% at half of the maximum amplitude. The transverse beam shape at the converter was elliptical, with a width of 5 mm and a height of 3 mm full width at half maximum (FWHM). The accelerated beam pulse current for the 40 MeV beam energy was about 0.49 A.

Fig. 1

Typical beam energy spectrum produced by Argonne’s electron linac and the load lines

Table 1 Irradiation conditions and times for natural ultra-high purity (UHP) and ¹⁰⁰Mo-enriched Mo targets

The cooling water flow through the convertor was 11.3 L/min (3 gpm). To control the temperature of the window, two 1/16-inch thermocouples were installed at the tested window holder.

Monte Carlo calculations

In order to estimate production rates for all isotopes produced by the photonuclear reaction in the entire target system, simulations were done using the Monte Carlo simulation code PHITS 3.02. The major improvements to the photonuclear yield calculation in the latest version of PHITS include (1) replacement of the total reaction cross sections in the Japanese Evaluated Nuclear Data Library (JENDL/PD 2004), which contains measured data for 68 nuclei, (2) modification of the evaporation model for the giant resonance, and (3) implementation of the quasideuteron disintegration process. These improvements allow PHITS to examine a photonuclear reaction up to 140 MeV of incident photon energy and cover our bremsstrahlung photon energy range of 0–40 MeV. A new parameter was added to reflect the contribution of nucleus recoil from elastic scattering in the yield calculation. When comparing the production rates of simulations and experiments, simulations excluding elastic collision produce predictions that significantly underestimate weak channels such as (γ,x + n), (γ,p + n), and isomeric states. However, simulation predictions including elastic collision are quite close to the experiments, without any change in production rates for well-known strong channels such as (γ,n). Elastic collision may have been an important factor in the prediction of production rates for weak channels; therefore, we performed the simulations including elastic collision.

Geometry

For Monte Carlo simulations, we used an incident electron beam with a beam power of 1.5 kW at 40 MeV, assuming a FWHM beam size of 1 cm in a Gaussian shape. To produce the bremsstrahlung photons, the beam impinged on a water-cooled convertor consisting of six 0.5-mm-thick tantalum disks. The Mo targets were positioned on the right, and the beam struck from the left (Fig. 2). Dimensions of 1 mm thick and 12.7 mm in diameter were used for both the natural and enriched ¹⁰⁰Mo targets in the simulations. Enriched targets were split into two pieces: one portion was used for the 0.5 h irradiation and the second for the 4 h irradiation (Table 1). Natural and enriched ¹⁰⁰Mo targets were irradiated together.

Fig. 2

Geometry of clamshell setup for natural and enriched ¹⁰⁰Mo targets

X-ray flux and its energy distribution

The calculated (PHITS) bremsstrahlung photon fluxes generated by interactions with the convertors before the Mo targets are shown in Fig. 3. The photon beam flux vs. its energy as the target’s radius changes is shown in the left panel, and beam flux as a function of radius was plotted in the right panel. In this simulation, the radii of the targets were extended to 1.6 cm to determine the detailed beam flux dependence by radius. The beam flux at the edge of our target (6 mm in radius) decreased by about 50% compared to that at the center. These simulations can be used to increase the production yield by adjusting the diameter and thickness of the Mo target when the target mass is fixed for the production run.

Fig. 3

Left: flux vs. energy from 0.5 mm (black—on top) to 6 mm (yellow) in radius. Right: radial dependence of photon flux in the Mo target

Irradiated Mo targets

Natural isotopic composition Mo targets (⁹²Mo—14.84%, ⁹⁴Mo—9.25%, ⁹⁵Mo—15.92%, ⁹⁶Mo—16.68%, ⁹⁷Mo—9.55%, ⁹⁸Mo—24.13%, ¹⁰⁰Mo—9.63%) were made from ultra-high-purity (UHP) MoO₃ that was converted to Mo metal powder and pressed and sintered into 12 × 1 mm Mo disks. Elemental analysis of UHP Mo material for 77 elements showed only appreciable amounts of Ni (0.3 ppm), Cu (0.5 ppm), Zr (0.14 ppm), Sn (0.03 ppm), Sb (1 ppm), Cs (0.1 ppm), and U (0.06 ppm); other elements were below the detection limit. The disks were provided by NorthStar. In a similar way, NorthStar provided ¹⁰⁰Mo-enriched targets (97.39% ¹⁰⁰Mo content, 2.59% ⁹⁸Mo; the presence of other Mo isotopes was not reported). The major impurities identified in the enriched ¹⁰⁰Mo targets were as follow: Fe (540 ppm), Cr (64 ppm), W (75.1 ppm), Ni (39.4 ppm), Cu (14.9 ppm), Ge (11.4 ppm), and Mn (5.7 ppm). One UHP natural Mo and one ¹⁰⁰Mo-enriched target were placed together and wrapped in aluminum foil. Each target was then mounted in aluminum target holder for irradiation. Targets were mounted right on the surface of the tantalum water-cooled converter. Tantalum was chosen for its high Z value, high melting point, chemical stability, and good machinability. A schematic of the irradiation setup can be seen in Fig. 2.

Gamma counting of irradiated Mo targets

Short (0.5 h) irradiations were used to identify short-lived isotopes. The short irradiation time allowed for the retrieval of the irradiated targets right after irradiation, and for gamma counting within an hour of the end of bombardment (EOB). Longer-lived isotopes were counted after some delay to allow for the decay of short-lived isotopes and to decrease the dead time on the HPGe detector. Each irradiated target was counted more than three times at different times to confirm that the selected peak does not interfere with different gamma lines. This was confirmed by the decay correction to EOB. Major radionuclides identified together with their production pathway, major gamma peak, and half-life are listed in Table 2. Theoretical cross sections for photonuclear reactions listed in Table 2 could be found online at TENDL database. HPGe gamma detector was calibrated using multi-nuclide gamma standard (Eckert & Ziegler) and spectra were analyzed using GammaVission software (Ortec). The detector had appropriate energy and efficiency calibration obtained using multi-nuclide standard at each distance used for gamma counting of the samples. Decay correction during acquisition was used to account for decay of radionuclides during longer count times. Whole irradiated Mo disks and fractures (in case of ¹⁰⁰Mo enriched material) were counted facing the detector in flat orientation at various distances up to 2 m from the detector.

Table 2 Major radionuclides identified after irradiation of natural and ¹⁰⁰Mo enriched targets, their production pathway, main gamma energy peak and half-life (obtained from Nuclide Navigator, Ortec)

Due to ingrowth of ⁹⁵Nb from ^95mNb and ⁹⁵Zr, ⁹⁵Nb activity at EOB was calculated as follows:

$${A}_{Nb95}^{0}=\frac{{A}_{Nb95}-\frac{{\lambda }_{Nb95}{A}_{Nb95m}^{0}}{{\lambda }_{Nb95}-{\lambda }_{Nb95m}}\left({e}^{-{\lambda }_{Nb95m}t}-{e}^{-{\lambda }_{Nb95}t}\right)-\frac{{\lambda }_{Nb95}{A}_{Zr95}^{0}}{{\lambda }_{Nb95}-{\lambda }_{Zr95}}\left({e}^{-{\lambda }_{Zr95}t}-{e}^{-{\lambda }_{Nb95}t}\right)}{\left({e}^{-{\lambda }_{Nb95}t}\right)}$$

It was experimentally determined that due to the very small production of ⁹⁵Zr, ingrowth of ⁹⁵Nb from ⁹⁵Zr was minimal and could be ignored for simplicity. Ingrowth of ⁹⁵Nb is shown in Fig. 4. It should be mentioned that presence of activation products ⁵⁴Mn, ⁵¹Cr, and ⁵⁷Co was only observed in enriched Mo targets (after the decay of ⁹⁹Mo) and is due to the impurities mentioned earlier. Typical gamma spectra obtained after irradiation of natural UHP and ¹⁰⁰Mo enriched Mo targets are shown in Fig. 5.

Fig. 4

Ingrowth of ⁹⁵Nb from decay of ^95mNb

Fig. 5

Gamma spectra of irradiated natural UHP and ¹⁰⁰Mo-enriched Mo targets

Results and discussion

Average detected activities at EOB for natural UHP Mo and ¹⁰⁰Mo-enriched targets after 0.5 and 4 h irradiations are listed in Tables 3, 4, 5 and 6. Figure 5 compares the gamma spectra of irradiated natural UHP and ¹⁰⁰Mo enriched targets. The major peaks identified correspond very well to previous literature data [18, 20]. The data in Fig. 5 illustrate the minimal production of side-reaction products for enriched ¹⁰⁰Mo targets. The major side products identified after irradiation of enriched targets are ^95,96,97,98mNb isotopes and minor production of ⁹⁵Zr. No presence of ^95mNb was identified after irradiation of enriched ¹⁰⁰Mo targets; however, direct production of ⁹⁵Nb was confirmed. This indicates that there is no significant production pathway for ^95mNb from ⁹⁸Mo, as there is no production on enriched ¹⁰⁰Mo target containing ⁹⁸Mo. Due to the very short half-lives of ^97,98mNb (Table 2), they are not a concern; however, the production of ^95,96Nb (these are produced due to a small content of ⁹⁸Mo in enriched ¹⁰⁰Mo targets) with longer half-lives might be of concern if Nb is not removed from the Mo product, especially if natural Mo targets are used. Their content depends on the level of ⁹⁸Mo in the target, and for higher enrichments of ¹⁰⁰Mo, they should not be a concern. The removal of Zr and Nb isotopes, however, is not difficult, as coprecipitation with Fe(III) is fairly efficient [23]. It also should be noted that production of ^97mNb was not observed; only direct production of ⁹⁷Nb was detected.

Table 3 Activities, one-sigma uncertainties, and fraction of activities to that of ⁹⁹Mo for radioisotopes found in irradiated natural UHP Mo target at EOB. Irradiation: 0.5 h at 40 MeV and 1.5 kW, 0.9736 g Mo target

Table 4 Activities, one-sigma uncertainties, and fraction of activities to that of ⁹⁹Mo for radioisotopes found in irradiated natural UHP Mo target at EOB. Irradiation: 4 h at 40 MeV and 1.5 kW, 0.9719 g Mo target

Table 5 Activities, one-sigma uncertainties, and fraction of activities to that of ⁹⁹Mo for radioisotopes found in irradiated enriched ¹⁰⁰Mo (97.39%) target at EOB. Irradiation: 0.5 h at 40 MeV and 1.5 kW, 0.6254 g Mo target

Table 6 Activities, one-sigma uncertainties, and fraction of activities to that of ⁹⁹Mo for radioisotopes found in irradiated enriched ¹⁰⁰Mo (97.39%) target at EOB. Irradiation: 4 h at 40 MeV and 1.5 kW, 0.4294 g Mo target

The analysis of gamma spectra for irradiated enriched ¹⁰⁰Mo targets shows the importance of keeping the level of impurities low. As can be seen in Table 6, small levels of ⁵¹Cr, ⁵⁴Mn, and ⁵⁷Co were detected after irradiation of ¹⁰⁰Mo-enriched targets; these isotopes are due to the presence of impurities most likely coming from stainless steel. These impurities could have been introduced during target manufacturing or may have been originally present in the enriched Mo material. The source could be linked to equipment, tools, or storage vessels that might be used during enrichment, chemical processing, disk manufacturing, and handling processes. Although the production of these radionuclides is relatively low, if they are not removed, their accumulation in the target might be of concern for recycled targets due to their long half-lives.

For natural Mo targets, the high activity of ^91mNb (T_1/2=1536.1 h) at EOB is noticeable (Table 4) and represents ~ 2.6% of ⁹⁹Mo activity at EOB. Due to a long half-life, the activity of ^91mNb remains a significant dose contributor after the decay of ⁹⁹Mo for long irradiations as well. Sixty days after EOB, when ⁹⁹Mo activity reaches ~ 2.7E−7 of its original, the activity of ^91mNb is still at ~ 50% of its EOB activity. This should be considered a potential source for a high dose when handling waste streams after production of ⁹⁹Mo using natural targets. ^91mNb and ^95 m/95Nb are the major side-reaction product contributors to the radiation dose for freshly irradiated natural Mo targets. In contrast to enriched targets, in natural Mo targets production of ⁹⁶Nb is significant: ~16–17% of ⁹⁹Mo activity at EOB for a short irradiations. This, however, is not a concern for several-days-long irradiations due to the relatively short half-life of ⁹⁶Nb (T_1/2=23.35 h). From other Mo isotopes, only ⁹⁰Mo was detected in sufficient activities after the short 0.5 h irradiation. No detectable activity of ^93mMo in irradiated UHP natural Mo targets was detected by an analysis of gamma peak at 1477 keV, even when gamma counting was performed within an hour after EOB.

Production rates for radioisotopes containing natural UHP Mo and enriched ¹⁰⁰Mo identified after 0.5 and 4 h irradiation at 40 MeV are compared with calculated values obtained from the Monte Carlo simulation tool PHITS 3.02 in Tables 7, 8, 9 and 10, and are normalized per mass of Mo target irradiated. This comparison identifies major reaction pathways for certain side-reaction products occurring in ¹⁰⁰Mo-enriched targets, which in our case contained only ⁹⁸Mo and ¹⁰⁰Mo isotopes. In general, the most likely isotopic contaminant in highly enriched ¹⁰⁰Mo material is ⁹⁸Mo, due to its similar mass; however some presence of lighter Mo isotopes is also possible. Analysis of enriched ¹⁰⁰Mo material used in this study showed only the presence of ¹⁰⁰Mo (97.39%) and ⁹⁸Mo (2.59%).

Table 7 Comparison of experimental and theoretical production rates for natural UHP molybdenum target at EOB. The 0.9736 g target was irradiated for 0.5 h at 40 MeV and 1.5 kW

Table 8 Comparison of experimental and theoretical production rates for natural UHP molybdenum disks at EOB. The 0.9719 g target was irradiated for 4 h at 40 MeV and 1.5 kW

Table 9 Comparison of experimental and theoretical production rates for enriched ¹⁰⁰Mo (97.39%) target at EOB. The 0.6254 g target was irradiated for 0.5 h at 40 MeV and 1.5 kW

Table 10 Comparison of experimental and theoretical production rates for enriched ¹⁰⁰Mo (97.39%) target at EOB. The 0.4294 g target was irradiated for 4 h at 40 MeV and 1.5 kW

In general, experimentally obtained production rates were higher than those calculated using the Monte Carlo simulation tool PHITS 3.02. Average experimental production rates for ⁹⁹Mo from irradiating natural Mo targets (1.23E+09–9.63% ¹⁰⁰Mo) correlate very well to those obtained after irradiation of ¹⁰⁰Mo-enriched targets (1.21E+10–97.39% ¹⁰⁰Mo). When production rates for ⁹⁹Mo are normalized per mass of ¹⁰⁰Mo in the target, the average production rate is 1.26E+10 with a standard deviation of 5.7%. This suggest that experimentally obtained data for different lengths of irradiation and content of ¹⁰⁰Mo agrees very well with production rates of ⁹⁹Mo. For enriched targets, relatively good correlation of experimental and calculated production rates was obtained for ^98mNb and ⁹⁶Nb. For the rest of the radionuclides, computer models significantly under-predict production rates. It is important to consider this when predicting the purity of the final ⁹⁹Mo product for enriched ¹⁰⁰Mo in a production facility. Because enriched material will be recycled multiple times, those impurities can accumulate in the target material and might require additional chemical purification steps to meet purity requirements.

To determine the production rates of other radionuclides, we looked at irradiated enriched ¹⁰⁰Mo targets more closely (Table 11). Because only two Mo isotopes (⁹⁸Mo and ¹⁰⁰Mo) were present in significant quantities in enriched targets, the analysis is a bit easier than that for natural Mo targets.

Table 11 Experimental production rates at 40 MeV for some Nb isotopes normalized per gram of ⁹⁸Mo

For ⁹⁵Nb, there is no production pathway on ¹⁰⁰Mo, and therefore any production of ⁹⁵Nb on an enriched target is due to the presence of ⁹⁸Mo via ⁹⁸Mo(γ,2n + p)⁹⁵Nb. The experimental production rate for this reaction is 1.41E+07 atoms/s/kw/g ⁹⁸Mo. The contribution of reactions on ⁹⁶Mo and ⁹⁷Mo (⁹⁶Mo(γ,p)⁹⁵Nb, ⁹⁷Mo(γ,p + n)⁹⁵Nb) in natural targets is about 17 × higher than that on ⁹⁸Mo (2.41E+08 vs. 1.41E+07) when normalized per mass of ⁹⁸Mo.

Similarly, for ⁹⁶Nb there is no production pathway on ¹⁰⁰Mo, and, therefore, any production of ⁹⁶Nb on enriched targets is via ⁹⁸Mo(γ,p + n)⁹⁶Nb. For production of ⁹⁶Nb, it seems that the experimental production rate for the reaction ⁹⁸Mo(γ,p + n)⁹⁶Nb is 4.86E+07, and for the ⁹⁷Mo(γ,p)⁹⁶Nb reaction it is 2.58E+08 (3.07E+08 minus 4.86E+07). The major reaction pathway for the production of ⁹⁷Nb is from ⁹⁸Mo via ⁹⁸Mo(γ,p)⁹⁷Nb, as the theoretical production rate for reaction on ¹⁰⁰Mo is ~ 1000 × lower [17]. The experimental production rate of ⁹⁷Nb determined after irradiation of the natural Mo target is 5.4E+08 atoms/s/kW/g-⁹⁸Mo, and for irradiated enriched ¹⁰⁰Mo target is 6.34E+08, giving an average of 5.87E+08 with a standard deviation of 11%.

Experimental production rates for various Nb isotopes are further compared with calculated production rates and literature data in Table 12. It can be seen that for all these isotopes, experimental production rate ratios or normalized production rates are higher than the calculated and literature data [17].

Table 12 Comparison of normalized production rates for various Nb isotopes

Considerable production of ⁹⁵Zr was observed only after 4 h irradiation. If it is assumed that the major reaction pathway for ⁹⁵Zr in the enriched ¹⁰⁰Mo target is due to the presence of ⁹⁸Mo and the reaction ⁹⁸Mo(γ,n + 2p)⁹⁵Zr, it would be expected that the total activity of ⁹⁵Zr produced should be higher using a natural Mo target, due to its higher content of ⁹⁸Mo (~ 21 × more ⁹⁸Mo in natural Mo than in enriched ¹⁰⁰Mo). However, the opposite was observed, with a higher production of ⁹⁵Zr in the enriched ¹⁰⁰Mo target (Table 13). The content of ¹⁰⁰Mo in the irradiated enriched Mo target (0.4294 g Mo) is about 4.5 × higher than in the natural Mo target (0.9719 g of Mo), and the activity of ⁹⁵Zr produced is ~ 2.7 × higher in the enriched target. Although this doesn’t provide excellent correlation to the mass of ¹⁰⁰Mo, it does point to a conclusion that major production of ⁹⁵Zr is due to the reaction on ¹⁰⁰Mo via ¹⁰⁰Mo(γ,n + α)⁹⁵Zr. However, it should be noted that reactions on ^97,98Mo have comparable, but slightly lower, production rates. Table 13 summarizes this discussion. The presence of the Zr impurity in enriched ¹⁰⁰Mo material was 50 ppb, so activation of ⁹⁶Zr is very unlikely, due to a low content of Zr and low natural abundance of ⁹⁶Zr (2.8%).

Table 13 Comparison of ⁹⁵Zr production on natural and enriched ¹⁰⁰Mo targets for 4 h irradiation at 40 MeV

The cumulative experimental production rates for natural Mo targets for 0.5 and 4 h irradiations presented in Tables 7 and 8 agree reasonably well for most of the listed radionuclides. The experimental production rate for ⁹⁰Mo after 0.5 h irradiation is 3.14E+07 and 3.03E+07 after 4 h irradiation. Similarly good correlations can be seen for ⁹⁵Nb and ⁹⁶Nb. Differences in experimental production rates obtained for ^92mNb and ^95mNb after 0.5 and 4 h irradiation are slightly higher, but reasonable agreement was still observed. It should be noted that these production rates are for all combined reaction pathways listed in Table 2.

Comparison of short-term and long-term duration irradiations

Results of the long irradiation of Mo performed in 2015 were reported previously [24]. In the 2015 irradiation, enriched and natural molybdenum disks were irradiated for a period of 6.5 days. The target was composed of twenty-five Mo disks containing six ¹⁰⁰Mo enriched disks (positions 5–10) and nineteen natural Mo disks. A summary of gamma counting results for the natural Mo target from this irradiation run is presented in Table 14.

Table 14 Summary of gamma counting results from 6.5 d irradiation of one natural Mo target (position 2 out from 25 disks) irradiated at 42 MeV and 8 kW reported for long-lived radionuclides. Gamma counting of non ⁹⁹Mo radionuclides was performed after the decay of ⁹⁹Mo (4 weeks after EOB)

Total ⁹⁹Mo activity produced in 6 g of ¹⁰⁰Mo enriched target (95.08% ¹⁰⁰Mo) was ~ 12.4 Ci. The production of ⁹⁵Zr and Nb isotopes in the enriched ¹⁰⁰Mo target was not quantified, since most of the activity of Zr and Nb was removed after the co-precipitation with Fe followed by filtration [24]. However, the presence of ^92mNb in the irradiated enriched ¹⁰⁰Mo target confirms the presence of lighter Mo isotopes such as ⁹⁴Mo.

Table 15 compares experimental production rates for short (4 h irradiation at 40 MeV) and long-term (6.5 days at 42 MeV) irradiations normalized per mass of ¹⁰⁰Mo. It is important to point out the significant difference between short and long-term irradiation conditions. In short irradiation, we used a high-Z converter (Ta) to produce a hard x-ray spectrum, while in long-term irradiations, a high-Z converter was not used; instead, the conversion process was realized in the actual Mo target. In addition, the energy distribution of the bremsstrahlung photons was different due to the different thickness of the material in the converter. It should also be noted that the target for long-term irradiation was composed of twenty-five Mo disks, and for this comparison, gamma counting data obtained for the disk in position two, close to the converter, was used. Despite the differences in irradiation setup, normalized production rates for most isotopes (except ^92mNb) were quite similar and were within the same order of magnitude. The correlation of the ⁹⁹Mo and ⁹⁵Zr production rates for short (4 h) and long (6.5 d) irradiations is very good. Such a great agreement for ⁹⁹Mo production is probably coincidental given the dramatic difference in irradiation conditions; however, the production of ⁹⁵Zr is good example of the agreement of experimentally determined production rates.

Table 15 Comparison of experimental production rates calculated from 6.5 d irradiation of the natural target and results from short 4 h irradiations, normalized per mass of ¹⁰⁰Mo

Conclusions

Irradiations of a natural UHP Mo and an enriched ¹⁰⁰Mo target at 40 MeV were performed to obtain basic information about the identification of major side-reaction products. Experimental results were compared to computer simulations using Monte Carlo simulation code PHITS 3.02, and the experimental results obtained from 4 h irradiations at 40 MeV were compared to previously obtained results from 6.5 d irradiations at 42 MeV. A very good correlation for ⁹⁹Mo experimental production rates was observed for 0.5 h, 4 h, and 6.5 d irradiations. It was determined that production of ⁹⁵Nb and ⁹⁶Nb is associated with the presence of ⁹⁸Mo in enriched ¹⁰⁰Mo, and there is no appreciable production rate on natural ¹⁰⁰Mo. Moreover, the major production pathway for production of ⁹⁷Nb in enriched ¹⁰⁰Mo is due to the presence of ⁹⁸Mo.

Calculated production rates are largely in close agreement with experimentally obtained ⁹⁹Mo production rates, but, except for ⁹⁰Mo, the computer model significantly under-predicts the production rates of other radioisotopes. It is important to take this into account when predicting the purity of the final ⁹⁹Mo product for enriched ¹⁰⁰Mo. This was especially true of the calculated production rates for ⁵¹Cr, ⁵⁴Mn, and ⁵⁷Co (due to the presence of Cr, Mn, and Ni impurities in the enriched ¹⁰⁰Mo target), which were more than a 3 × lower than those determined experimentally. This is particularly important to tracking the impurity level in the original enriched material as well as to the introduction of elements during the recycling of enriched material, due to long half-lives and a potential build-up of activity. Because enriched material will be recycled multiple times, those impurities can accumulate in the target material and might require additional chemical purification steps to meet purity specifications.