Separation of fission produced 106Ru from simulated high level nuclear wastes for production of brachytherapy sources

An effective and simple process for the isolation of 106Ru from high-level liquid wastes was developed. Radioactive ruthenium was oxidized by H5IO6 in HNO3 solution and was extracted to CCl4 phase in the form of RuO4. In order to obtain ruthenium in the suitable form for production of brachytherapy sources, RuO4 in organic phase was reduced and re-extracted to aqueous phase. The efficiency of extraction of 103Ru to organic phase was 86 %, re-extraction to aqueous solution was near 100 %, so the overall recovery of 103Ru is estimated at more than 80 %.


Introduction
Brachytherapy is the common method for treating various tumors, and currently the ruthenium-106 and iodine-125 applicators are the most frequently used. Considering that 106 Ru is a bemitter with maximum energy of 3.54 MeV, it is best indicated in the treatment of small melanomas, with up to 20 mm tissue range [1]. It also replaced 90 Sr/ 90 Y sources because of it higher energy of emitted bparticles [2] and possibility of simpler source preparation. 106 Ru is commercially obtained from neutron irradiated high enrichment 235 U target in process of production 99 Mo. After isolation of 99 Mo radioisotope and decaying of 103 Ru, ruthenium is separated from the wastes by multistep procedure. At present, there are only a handful of ageing reactors worldwide capable of producing the 99 Mo, therefore alternative strategies for production of this key medical isotope are explored. In our work, we propose to use liquid high-level radioactive waste as a source of high activity of 106 Ru.
The potential utilisation of fission-produced platinum metals (fission platinoids) as valuable products has attracted attention in the last few decades, as large amounts of spent nuclear fuel have accumulated worldwide [3]. Table 1 presents the isotopic composition of ruthenium isotopes after 5 years cooling of liquid nuclear waste [4].
Simple calculations indicate that 1 dm 3 of waste solution after reprocessing of nuclear fuel contains about 500 GBq of 106 Ru after 4 years of cooling. This amount of activity is enough for production of about few thousands of brachytherapy sources.
During reprocessing of the spent fuel, the metallic Ru is dissolved in concentrated nitric acid and forms stable Runitrosyl complexes [5]. In the high acidity the dominating ruthenium species are the [RuNO(NO 2 ) 2 (NO 3 )(H 2 O) 2 ] 0 and [RuNO(NO 2 ) 2 (H 2 O) 3 ] ? [6]. The concentration of different species depends mainly on the composition of the medium and also the time of ageing.
Ruthenium metal was efficiently separated from other fission products by oxidation and distillation of RuO 4 with absorption in NaOH solution. El-Absy et al. [7,8] separated Ru radionuclides from a 131 I-free fission product acidic solution containing KMnO4, by boiling for 40 min. In other work, ruthenium was electrochemically eliminated from a 3 M HNO 3 solution of high-level waste, as RuO 4 , in the presence of AgNO 3 at 60°C [9]. Gandon et al. [10] coprecipitated ruthenium with copper ferrocyanide neutral solution. D. Banerjee et al. [11] used conventional ion exchangers and chemical precipitation based processes for the effective removal of the 106 Ru activity from NH 4  Present communication reports results of our process development studies on the recovery of ruthenium radioisotopes from simulated solution of high level radioactive waste using oxidation-extraction method.

Radionuclide
For reasons of availability we used in experiments the 103 Ru nuclide instead of 106 Ru. The latter nuclide 106 Ru is separated in complicated procedure from fission products of 235 U, while 103 Ru is produced in a simple way by direct thermal neutron irradiation of natural ruthenium. 103 Ru was obtained by neutron irradiation of ruthenium salt (NH 4 ) 2 [RuCl 5 (H 2 O)] at a neutron flux 7 9 10 13 n cm -2 s -1 for 8 h in the nuclear reactor Maria at Świerk, Poland. The irradiated target was dissolved in 1 M HNO 3 .
Others radionuclides, 131 I in the form of Na 131 I solutions was obtained from NCBJ-Polatom Ś wierk and 99m Tc in the form of 99m TcO 4 was milked from 99 Mo/ 99m Tc generator.

Radioactivity measurements
The 103 Ru radioactivity was measured using an ORTEC system with a high resolution HPGe detector using photo peak at E c = 497.05 keV (88.7 %) and in NaI c-scintilation counter LG-1b, ICHTJ, Poland.

Reagents
The following commercial chemicals were used without additional purification: H 5

Solvent extraction studies
Experiments were carried out under ambient conditions by shaking equal volume (5 ml each) of organic and aqueous phase in a separatory funnel using wrist action shaker. Phase separation was done by centrifugation and suitable aliquots (1 ml) of each phase were assayed. The distribution ratio ''D'' of the metal was determined as the ratio of metal concentration in organic phase to that in aqueous phase. Percentage extraction of metal ion was calculated by equation:

Extraction of 103 Ru to CCl 4 phase
In oxidizing solutions ruthenium forms tetroxide, RuO 4 , which is easily extractable to organic phase. Formation of RuO 4 is indicated by color change from deep orange to golden yellow. The RuO 4 formed was extracted to an organic phase. Unfortunately, the RuO 4 is not stable in the CCl 4 phase and formation of black RuO 2 precipitate is observed after a few hours. To avoid reduction of RuO 4 to RuO 2 the organic phase was contacted with a solution generating Cl 2 molecules: 0.01 M HCl ? 0.05 M H 5 IO 6 . The Cl 2 molecules, formed in aqueous solution, are very soluble in CCl 4 and distributed among the two liquid phases keeping ruthenium in the form of RuO 4 in the organic phase for several month [12]. Influence of the various oxidants and acids on ruthenium oxidation-extraction process were studied to optimize the process. Table 2 presents results of 103 Ru extraction from solutions containing various oxidizing agents. Concentration of used oxidants was the same taking into account the number of electrons involved in the reaction.
As show in Table 2 the obtained results indicate that the best oxidant is orthoperiodic acid (86.0 % extraction), a  somewhat worse, but also possible to use is a potassium metaperiodate (75.1 %) and potassium permanganate (79.5 %). The obtained results well correlate with oxidation potential of reagent used. An important parameter was the selection of a suitable amount of oxidant to obtain complete oxidation of ruthenium to RuO 4 and thus its extraction into the organic phase. We have studied the 103 Ru extraction depending on the concentration of orthoperiodic acid. The results are presented in Fig. 1.
In concentration range from 5 to 40 g l -1 of H 5 IO 6 only insignificant increasing of 103 Ru extraction is observed. Therefore, it can be assumed that that solution containing only 10 g l -1 of H 5 IO 6 should be sufficient for effective extraction of 103 Ru to CCl 4 phase.
In the next step, influence of various acids and acid concentrations on 103 Ru extraction were studied. We examined the following acids: nitric acid, sulfuric acid, hydrochloric acid and perchloric acid. The results are presented in Tables 3 and 4.
In the solutions of HNO 3 , H 2 SO 4 , and HClO 4 extraction of 103 Ru was comparable. Only in HCl solution extraction was significantly lower. Additionally, the use of HCl solution is not desirable due to the formation of Cl 2 gas by reaction of orthoperiodic with hydrochloric acid. For further experiments HNO 3 solution was selected. This choice was dictated by the fact that the high-level radioactive waste are generally in the form of a HNO 3 solution. Table 3 presents dependence of the 103 Ru extraction on the HNO 3 concentration in range of 1-5 M. Table 4 only very small increasing of 103 Ru extraction was observed when HNO 3 concentration increased from 1 to 5 M. Summarizing our results on optimization of 103 Ru extraction process, we can conclude that 86 % of extraction could be obtained for using H 5 IO 6 -10 g l -1 as oxidant and 1 M HNO 3 solution. Using of higher H 5 IO 6 and HNO 3 concentrations gave only insignificant increasing of the process efficiency.

As shown in the
The PUREX raffinate contains also other long-lived fission products like 135,137 Cs, 90 Sr, 241 Am, 99 Tc, 129 I, 97 Zr, among which 99 Tc and 129 I could be potentially co-extract with 106 Ru. In oxidizing solution technetium could be extracted as HTcO 4 and iodine in I 2 or interhalogen form. The 135,137 Cs, 90 Sr, 241 Am and other metallic radionuclides in HNO 3 solution, not containing complexing agents, are present in either cationic form or nonextractable species.
For co-extraction studies of 99 Tc and 129 I we used shortlived isotopes 99m Tc and 131 I. The extraction of both radionuclides were performed in solution of concentration of 10 g l -1 H 5 IO 6 in 1 M HNO 3 . We did not observe extraction of radionuclide studied, radioactivity of the 99m Tc and 131 I in the organic phase was below the background level.
Since the 106 Ru sources for brachytherapy are usually obtained by electrochemical deposition from aqueous solutions [13], we investigated the possibility of ruthenium transfer from the organic to aqueous phase. Because RuO 4 is the only form of ruthenium, which is stable in CCl 4 phase, for re-extraction of 103 Ru we decided to reduce RuO 4 to Ru(III) and Ru(II) oxidation state. The following compounds were selected as reductants: sodium sulfite, hydroxylamine, hydrazine and sodium borohydride. Results of 103 Ru extraction from the organic into aqueous phase are shown in Table 5.   As shown in Table 5, the best results were obtained for 0.1 M aqueous solutions of hydrazine and for hydroxylamine hydrochloride. These reductants are most sufficient, because of their relatively high solubility in the organic phase, where reduction of RuO 4 to the Ru(III) and Ru(II) took place. Reduced forms of ruthenium are insoluble in CCl 4 phase and passed immediately to the aqueous phase.
Kinetic studies were carried out in the system 103 RuO 4 in CCl 4 (organic phase) and Na 2 SO 3 0.1 M HCl (aqueous phase). The results presented in Fig. 2 indicate that the process is relatively fast and after 40 min equilibrium state is achieved.

Conclusion
A highly effective and flexible process for the separation of 106 Ru from simulated high-level liquid waste was elaborated. It was found that the optimal way for extraction of 103 Ru to CCl 4 organic phase is oxidation of ruthenium nitrozyl complexes to RuO 4 by 10 g l -1 H 5 IO 6 in 1 M HNO 3 solution. It was found that in re-extraction process to aqueous phase the most efficient compounds for reduction of RuO 4 in CCl 4 phase are hydrazine and hydroxylamine hydrochloride. The overall recovery of 106 Ru is estimated at more than 80 %.
Production batches of hundreds GBq of 106 Ru radioisotope separated from 1 l of PUREX raffinate can be achieved using the above-mentioned separation technique. For verification of the obtained results further experiments with real wastes solutions is necessary.