Study of new routes for purification of fission 99 M

Tc is the most used medical radioisotope in the world, especially for diagnosis procedures. It is originated in the radioactive decay of Mo, which in turn is one of the fission products of the uranium irradiation that occurs in nuclear reactors. The chemical processing of Mo might be lined up in several steps according to the characteristics of the targets or the local requirements in order to separate it from other fission products. In this work, two routes of Mo purification, MR1 and MR2, were purposed as an alternative method to be set up in the Brazilian Multipurpose Reactor project. The MR1 route consisted by three consecutives chromatographic columns packed with strong anionic resin, Chelating resin, and aluminum oxide, respectively. The MR2 route was arranged in two chromatographic columns and a sublimation oven. The final yield for the MR1 was 84.4% and the overall time process was about 7 hours, performing the highest final Mo recovery efficiency, high decontamination degree and a shorter process time compared to the MR2.


INTRODUCTION
Tc is the main radioisotope used in nuclear medicine worldwide due to its pure gamma decay of low energy (140 keV) and short half-life (6.02 hours). It is commonly combined to other molecules known as markers, which in turn have an affinity for some specific organ or tissue [1,2]. In the world, more than 22 million medical procedures are performed annually for diagnostic imaging tests applying 99m Tc [3]. 99m Tc can be provided for medical centers throughout the country using 99 Mo/ 99m Tc generators. Therefore, 99 Mo plays an important role as a precursor for the global supply of 99m Tc. The production of fission 99 Mo initiates with the irradiation of 235 U targets (UAlx/Al) into a nuclear reactor, followed by a chemical processing for purification, where the 99 Mo is separated from the other undesirable radionuclides that are produced in the fission process as well. Then, the bulk 99 Mo is divided and shipped to generators manufacturing units [4,5].
After the Mo shortages in 2008-2009, when some producing reactors were disabled, several countries were led to planning new strategies for 99 Mo production. Particularly in Brazil, it was created the BMR project (Brazilian Multipurpose Reactor) as an action to meet the local demand for fission radioisotopes. The BMR project presumes the construction of a nuclear reactor, chemical processing facilities, and research laboratories [6,7].
The radiochemical purification is a relevant step on the 99 Mo production process due to the complexity of the separation of 99 Mo from the other fission radionuclides generated in the uranium decay, and the purification method depends on the production process adopted in the nuclear plant.
There are many processes for 99 Mo production currently in use worldwide, and some of them are derived from the well-known KSA process (Karlsruhe-Sameh Aluminide) [8]. This process was developed initially in Germany and applied on an industrial scale in the Netherlands in 1996, where it became a reliable, safe, and high efficient industrial-scale process ran by the Mallinckrodt Company at Petten (Netherland). Later, it was adopted by other countries such as Argentina and South Africa [9][10][11][12].
The ROMOL-99 (Rossendorf Molybdenum-99) also is a high efficient German process that was implemented in Pinstech (Pakistan). It also can be implemented on a smaller scale of production and presents an interesting sublimation step at the end of the purification process [13].
According to KSA and ROMOL-99 methodology, the uranium target is dissolved in a strong alkaline media, resulting in a solution containing 99 Mo and other radionuclides as contaminants.
This solution is treated by packed-columns chromatography and sublimation (ROMOL-99) to achieve the required radiochemical purity.
In this work, two routes (MR1 and MR2) of 99 Mo purification were studied in order to contribute to setup of the production process on BMR project facilities. These routes were based on two established processes, KSA and ROMOL-99. A flow chart of the reference processes routes and the studied is shown in Fig. 1.

MATERIALS AND METHODS
The MR routes were composed mainly of chromatographic columns as in the reference processes. In MR1 99 Mo purification was arranged in three followed chromatographic columns. The first column was packed with strong anionic resin Dowex 1x8, the second one with chelating resin Chelex-100, and lastly, a column with alumina an inorganic ion-exchanger. In MR2 also was performed in three steps, where the first and second ones were carried out by chromatography using a Dowex 1x8 and alumina columns, respectively, and the third step was a sublimation process. Both routes' steps are shown in Figure 1.
The routes were tested using loading solutions previously prepared, which were added various elements and radionuclides in order to simulate yielded a sample produced by an alkaline

MR1 route
For the first step of the MR1 route were used glass column 13x2.5 cm packed with 35 g of Dowex 1x8 anionic resin with no pre-conditioning procedure. Firstly, the column was percolated with 1.5 L of the loading solution at a flow rate of 40 mL•min -1 . After that, the packed-bed was washed twice: first, with 500 mL of 1 mol•L -1 NaOH at 20 mL•min -1 , and then, with 500 ml of distilled water at 20 mL•min -1 . Finally, the 99 Mo was collected by elution with 1.0 mol•L -1 Na2SO4 at 3 mL•min -1 . In the third step, it was used a glass column (10x1 cm) filled with 10 g of acid alumina, with no pre-conditioning treatment. The eluted solution from the Chelex column was acidified adding 1 mol•L -1 HNO3 until completing the volume of 80 mL, approximately. This solution was percolated through the alumina bed at 3 mL•min -1 flow rate After that the column was washed with 50 mL of distilled water, and then with 50 mL of 0.01 mol•L -1 NaOH. The elution was done with 1 mol•L -1 NaOH solution at 1 mL•min -1 flow rate.

MR2 route
The MR2 route also used ion-exchange chromatography, but a thermal separation by sublimation was included. The purification was started using a chromatographic column packed with Dowex 1x8 resin, and a second column filled with the alumina was adopted in the followed step.
Both columns (with Dowex and alumina, respectively) were carried out based on the same method described previously for the MR1 route, however, instead of 1.0 mol•L -1 NaOH as the eluting solution, it was used 1.0 mol•L -1 KHCO3 in the first column (Dowex) and 0.1 mol•L -1 NH4OH in the second one (alumina).
Before the sublimation step, the sample was pre-treated by evaporation. The eluted solution from chromatographic steps was heated at 80 ºC in a heating plate, for removing of excessed water and reducing of volume until about 5 mL. Then, this volume was transferred to a small platinum crucible, and heated until dryness.
After cooling, the crucible containing 99 Mo solid residue was placed inside a quartz tube, which was accommodated in a horizontal tubular oven under a controlled temperature. The sample was heated to sublimation temperature at a heating rate of 15 ºC•min -1 , an isotherm of 10 minutes at 850 ºC, and an airflow rate of 5 L•h -1 inside the tube.

Measures of radioactive activities
All experiments were carried out in duplicate at 25 ºC and 10 mL of solutions (from loading, eluting, and washing) were collected at the end of each step for the measure of activities and step efficiency calculations.
The measurements of the radioactive activities of the samples studied were performed using a gamma spectrometer with coaxial hyperpure Germanium detector (HPGe) model GX 1518, coupled to a Canberra Inc. multi-channel acquisition system.
Where R is the elution, distillation or sublimation yield of the given radioisotope, Af and Ai are two respective final and initial areas of the principal peaks of radioactive activities at a given acquisition time.
The result adopted for each study of the purification steps corresponds to the arithmetic averages of representative experiments among the others.

7
In the complete radioisotope purification route tests, total yields were obtained by successively multiplying the yield means of the steps involved by equation 2: Where R t is the total recovery yield of the radioisotope and R (1, 2,…,η) corresponds to the arithmetic means of the yields in the purification steps of the tested routes.

RESULTS AND DISCUSSION
As mentioned previously, on the first step of both routes were used ion-exchange columns filled

MR1 route
The final yields of elution, effluent, and wash solutions obtained after the first column packed with Dowex resin and Na2SO4 eluting solution are shown in Table 1. The results indicated a low concentration of the 99 Mo in the effluent samples and high retention of it in the resin. The 99 Mo recovery was greater than 90% after the final elution, confirmed by the low loss observed after the percolation of the loading solutions and the washes. For the other contaminants, the total of elimination (effluent + washes) was about 79.3% for 95 Zr, 82.9% for 121 Te, and 84.5% for 103 Ru. Due to the low concentrations of the contaminants in some samples, their radioactivity measurements presented the results lower than the limit of detection (< DL), in the conditions applied. The Dowex resin was significant for the 99 Mo separation from the other fission elements present in the initial process solution.
In addition, the adoption of Na2SO4 solution for the 99 Mo elution was important because it does not lead to a chemical interference in the 99 Mo complexation reactions, which would affect the next separation by chelation chromatography [14]. In this step, the complexation of the 99 Mo with thiocyanate ions was performed using the eluted solution from the Dowex column. The obtained results in this process are presented in Table 2. The molybdenum recovery was also elevated in this step, presenting elution average higher than 97% and 99 Mo loss about 2%, during the washes. The measure of the contaminants' activities indicated values lower than the limit of detection (DL).
The alumina column was used to reduce the concentrations of the remaining contaminants, as well as the elimination of possible residues from the hoses, connections, and resins. Besides, the capacity to concentrate the 99 Mo in the final eluted solution also was observed. The recovery of the 99 Mo was approximately 88%, revealing the lower results than the previous columns, however, the purity of the sample improved, once the activities of the contaminants were under the limit of detection. The results of the samples obtained from alumina column are shown in Table 3.  The MR1 arrangement presented advantages in comparison with the so-called Sameh process, in which two redundant chelating columns are used [12].

MR2 route
For the 99 Mo recovery from the Dowex column in the MR2 route, the KHCO3 solution was applied instead of Na2SO4. The potassium bicarbonate solution was chosen considering the next purification step, once the sulfate as eluent affects the 99 Mo retention in the alumina, increasing the loss in the effluent [14]. The results for these experiments are shown in Table 5.   Table 6.  The ammonium hydroxide solution was adopted as the eluent for 99 Mo in the alumina column considering the next step. NaOH residues in the sample during the sublimation process produce sodium oxide along with molybdenum oxide by thermal decomposition, making difficult the 99 Mo recovery [13]. The sublimation yields are shown in Table 7. Results obtained in each step and the final yield of the MR2 route, considering the entire process is presented in Table 8. These results could be improved by developing of specific glassware to integrate the evaporation and sublimation phases inside the oven reducing losses.
The total time demanded in the process steps, and material transfers were approximately 9 hours. This route had a final yield of 75.2% and all the contaminants presented activities below the limit of detection, thus evincing a high purification degree of 99 Mo.
This route is similar to the ROMOL-99 process [15], but the position change of the resin (first step) and alumina (second step) columns in the process enabled to avert a labored step of the reference process [14], which involves acidifying the solution to a narrow pH range (0.3 to 0.5) to perform on the severe working conditions due to high activities of the solutions.
In the MR2, the sublimation step took a longer time than a chelating column, so the processing time of the MR1 route was lower than the MR2. So operational improvements in the sublimation step are required, such as higher heating and cooling rates on the oven to reduce the operating time.
Also, an inversion of the elution direction on the alumina column and the development of a more suitable concentrator system could contribute to the improvement of efficiency and reducing the overall process time.
Among the routes tested, the MR1 route presented the highest final purification yield of 99 Mo (84.4 %) and with a high decontamination degree. The results indicated that after the initial 99 Mo separation on anionic resin column, only two more chromatographic columns were sufficient for satisfactory purification of this radioisotope. Two comparative HPGe gamma spectra of radionuclides and their main energy peaks utilized on MR1 route are shown in fig. 3. This result resembles the process used in a South African producer center that also employs three chromatographic columns in its process and has a yield of approximately 85% [16].
In the next studies, the measuring difficulties of low activities of the contaminants on the purified solution can are avoided by increasing the irradiation time of the preceding salts or making tests in the up-scale.

CONCLUSIONS
The ion exchange technique using an anionic resin column was applied in MR1and MR2 as the initial process step to purify the alkaline solution as adopted in the process adopted in Argentina and South.
The use of anionic resin in the first column of the process offered effective elimination capacity of most of the contaminants present in the loading samples as well as the possibility of a selective elution of the 99 Mo radioisotope in both routes.
Besides, this process configuration avoids the pre-acidification of the loading solution, as needed in ROMOL-99 process, which represents an operation challenge due to the control of aluminum precipitation in the loading solution under critical conditions.
Between the routes tested, MR1 route presented the highest final 99 Mo recovery efficiency, high decontamination degree and a shorter process time compared to the MR2 route.
The results indicate the possibility of up scaling tests of MR1 route as a short and efficient production process using only three chromatographic columns, like in South Africa´s process.
The results also provide subsidies for further purification investigations on pilot cells, targeting the production at BMR facilities.