INTRODUCTION

One of the promising options for the processing of MNUP SNF from the BREST-OD-300 reactor at the reprocessing module of the pilot demonstration power complex (RM ODEC) is a combined technological scheme that includes pyrochemical processing at the first stage to remove the main fission products and reduce the specific SNF activity, at the second stage hydrometallurgical reprocessing of SNF is designed [1–4].

Volume oxidation (voloxidation) of spent nuclear fuel [5, 6] is considered as one of the main operations for reprocessing MNUP SNF at RM ODEK. The operation is designed to separate the fuel composition from the cladding and remove part of the volatile fission products from the fuel.

The oxidation of uranium and plutonium nitrides brings about the arrangement of the crystal lattice of the main SNF components, which leads to the fragmentation of fuel pellets and the separation of the fuel composition from the fuel cladding. Additionally, voloxidation will make it possible to separate tritium and radiocarbon from the fuel composition for their localization at the initial stage of SNF reprocessing [7–9].

The process of fuel separation from the cladding as a result of fuel composition oxidation is known and tested on model systems and on real oxide SNF samples [7, 10–18]. Data on the patterns of oxidation and dissolution of the voloxidized model MNUP fuel were reported in [9, 19].

This work is aimed at evaluating the efficiency of voloxidation of the MNUP SNF to separate the fuel composition from the cladding and remove volatile fission products.

EXPERIMENTAL

The studies were carried out with MNUP fuel spent in the BN-600 reactor as part of combined experimental fuel assemblies KETVS-1,7 and experimental fuel assembly ETVS-10.

To conduct experimental studies, the fuel rods were fragmented mechanically, from which 8 batches were formed and weighed. When cutting fragments of fuel rods, no local ignition of MNUP SNF or sparking occurred.

The main characteristics of the spent fuel and the formed batches of fuel rod fragments are listed in Table 1.

Table 1. Characteristics of the formed batches of fuel rod fragments
Full size table

The voloxidation of prepared SNF batches was carried out using an experimental reactor, the scheme of which is shown in Fig. 1.

Fig. 1.
figure 1

Experimental apparatus for MNUP SNF voloxidation.

Full size image

The device is designed on the ground of a resistance furnace with a range of operating temperatures up to 900°C. The furnace contains a reactor made of heat-resistant stainless steel, equipped with inlet and outlet nozzles for supplying gases and removing volatile fission products (VFPs). To ensure the oxidation of SNF in a stream of moist air, the inlet pipe of the reactor is connected to a stainless steel container filled with distilled water heated to a temperature of 40–50°C. The air supplied to the SNF oxidation zone first passes through a layer of heated water, then enters the SNF oxidation zone.

To monitor the process temperature, a chromel–alumel thermocouple is placed in the reactor vessel. An alumina ceramic crucible is suspended inside the reactor vessel for placing fuel rod fragments therein. In the lower third of the ceramic crucible, there is a heat-resistant stainless steel grid with a mesh size of 1 mm, designed to separate the voloxidized SNF from the fuel element cladding.

In the reactor vessel there is a gas outlet pipe connected to the inlet branch pipe of the voloxidation reactor. The lower end of the gas supply tube is located below the crucible level. The air supplied to the reactor first enters the bottom of the reactor and then rises. This solution makes it possible to avoid stagnant zones in the reaction region and increase the efficiency of SNF oxidation.

The top of the reactor is hermetically sealed with a lid made of stainless heat-resistant steel. The outlet of the voloxidation reactor is connected to bubblers to trap VFPs.

MNUP SNF voloxidation was performed according to the following algorithm. A pre-weighed batch of fuel rod fragments with MNUP SNF in a ceramic crucible was placed in the reaction chamber of the voloxidation reactor. All parts of the experimental setup were hermetically connected. The voloxidation reactor was heated at a rate of 10°C/min with a continuous argon supply at 150 mL/min. After reaching the preset temperature mode, the inert gas supply was turned off and air was supplied at a preset flow rate. The isothermal regime was maintained for a specified time, after which the heating was turned off, the air supply was stopped, and argon was fed to the internal volume of the voloxidation reactor at a rate of 150 mL/min for 1 h. Afterwards, the bubblers with nitric acid and sodium hydroxide solutions were disconnected, the argon purge was turned off, and the voloxidation reactor was cooled at ambient temperature. The main experimental conditions for voloxidation are presented in Table 2.

Table 2. Experimental conditions for voloxidation
Full size table

After the process completion, the powder of the voloxidized SNF and fuel rod fragments were weighed. Powders of oxidized fuel spent as part of KETVS-1 were fractionated by shaking on control laboratory sieves made of stainless steel with a woven wire mesh (mesh size of 400, 200, 100, and 50 µm). The fraction weight was determined by the difference in the weights of laboratory sieves before and after fractionation. Weighing was carried out on a laboratory balance with an accuracy of ±0.1 g.

A portion of the averaged powder of voloxidized SNF was transferred for radiochemical analysis. The obtained data were compared with the content of VFPs in the MNUP SNF sample before voloxidation. The initial content of VFPs in MNUP SNF was found from the results of a destructive radiochemical analysis of the fuel spent in the composition of KETVS-1 and KETVS-7. The procedure and results of destructive radiochemical analysis are reported in detail in [20, 24–26].

The completeness of SNF separation from claddings was found by calculation and experimental methods. If, after the voloxidation completion, SNF residues were detected in fuel rod fragments by the lack of a gap in the fuel rod tubes, the degree of fuel separation from the cladding was estimated by the calculation method, determining the following values:

– the weight of fuel rod fragments with spent nuclear fuel residues after the voloxidation completion, g (m1);

– the linear weight of an unirradiated fuel rod tube made of the ChS 68-ID alloy, the same as in the fuel rods with MNUP fuel spent in KETVS-1 and KETVS-7, g/mm (m2);

– the length of fuel rod fragments taken for research, mm (L);

– the weight of the separated SNF powder, g (m3).

The weight of SNF remaining on the claddings of fuel rod fragments (m4) was calculated in accordance with the expression (1):

$$ {{m}_{4}}={{m}_{1}}-{{m}_{2}}L. $$
(1)

The completeness of SNF separation from the cladding was calculated as the weight fraction of the separated SNF powder in accordance with Eq. (2):

$$ \omega =\frac{{{m}_{3}}}{{{m}_{3}}+{{m}_{4}}}\times 100. $$
(2)

The application of the calculation method is associated with a number of assumptions on the change in the linear weight of the fuel rod tube during irradiation and oxidation of MNUP SNF. The values obtained by the calculation method should be considered as an estimate.

In the case when fuel rod fragments did not visually contain SNF residues, the determination of nuclear materials was carried out by an experimental method. The fuel rod fragments were subjected to two-stage washing: at the first stage, with 10 M nitric acid; at the second stage, with a solution of 10 M HNO3 containing 0.02 M NaF. The solvent temperature was 95 ± 5°С, and the dissolution time was 6 h at each stage of the process.

To confirm the completeness of the transferring the voloxidized fuel into the solution, after completion of the two-stage dissolution, the claddings were washed in a 10 M nitric acid solution containing 0.1 M HF at a temperature of 95 ± 5°C for 6 h.

The solutions obtained at each stage of dissolution were analyzed for the content of nuclear materials and fission products. The amount of SNF found in solutions after the dissolution completion was equated to the amount of fuel not separated during voloxidation.

RESULTS AND DISCUSSION

At the first stage of the research, a series of six experiments was carried out to determine the optimal conditions for the voloxidation of MNUP SNF. Data on the separation efficiency of the fuel composition from the fuel rod cladding vs. the experimental conditions are given in Table 3.

Table 3. Completeness of separation of the MNUP fuel spent as part of KETVS-1 from the fuel cladding depending on the experimental conditions
Full size table

The weight of SNF on fuel claddings in runs nos. 1–3 was estimated by the calculation method, in runs nos. 4–6 it was found by the experimental method.

It can be seen that the completeness of separation of the fuel composition from the fuel element cladding increases with the process temperature rise up to 450°C. With a further increase in temperature, the fraction of the separated SNF powder remains practically unchanged. In runs nos. 3 and 4, as the temperature increased from 350 to 450°C, the completeness of SNF separation from cladding increased from 76.8 to 98.6%. In runs nos. 5 and 6, with an increase in temperature from 450 to 550°C, the SNF separation efficiency from the cladding remains unchanged.

The low completeness of separation at a temperature of 350°C confirms the observations made in [9], where it was found that uranium nitride begins to rapidly oxidize at temperatures above 300°C.

Increasing the rate of air blowing through the reactor volume also leads to a rise in the fraction of SNF separated from the claddings only in a certain temperature range. In runs nos. 2 and 3, carried out at a temperature of 350°C, with an increase in the air supply rate to the reaction zone, the completeness of SNF separation from the claddings increased from 58.4 to 76.8%. In runs nos. 4 and 5, carried out at a temperature of 450°C, a twofold change in the rate of air supply into the voloxidation reactor has practically no effect on the fraction of the separated SNF powder. The observed patterns in the MNUP SNF oxidation are probably due to the change of the limiting stage of the heterogeneous reaction from the kinetic region in the temperature range of 350–450°C to the diffusion one at temperatures of 450–550°C. There is no a significant effect of the air supply rate into the MNUP SNF oxidation zone which indicates the intradiffusion limitation of the process. At the same time, the experimental results obtained in this work are insufficient for an unambiguous formulation of the kinetic patterns of MNUP SNF voloxidation.

The established experimental facts are consistent with the data in [27], the authors of which showed that the oxidation rate of briquettes of mononitride uranium–plutonium fuel increases with a temperature growth and reaches a maximum value of 160 mg/min at a temperature of 400°C.

The obtained experimental results indicate that the optimal conditions for the MNUP SNF separation from fuel cladding are the conditions of runs nos. 4–6, in which the completeness of fuel separation is at the level of 99%.

The results of determining the fractional composition of the voloxidized MNUP SNF are listed in Table 4.

Table 4. Fractional composition of oxidized MNUP SNF
Full size table

In all the studied oxidation modes, more than 40% of the particles have a size not exceeding 100 µm, and the fraction of particles larger than 400 µm does not exceed 10%. It was found that in the range of experimental conditions studied, the fractional composition of oxidized SNF does not depend on the temperature and air supply rate to the oxidation zone.

The presented results indirectly evidence that there occurs no significant fuel sintering during voloxidation. The possibility of local sintering of the fuel composition in the voloxidation process was established by the authors of [14] when studying the uranium oxide fuel oxidation patterns. In particular, it was shown that with a decrease in the voloxidation temperature below 480°C, the degree of refinement of the fuel composition increases. In the temperature range of MNUP SNF voloxidation, we studied, this pattern was not revealed.

Data on the residual content of some volatile SNF components in voloxidized fuel samples are presented in Table 5.

Table 5. Content of 3H, 14C, 106Ru in samples of voloxidized MNUP SNF
Full size table

Data in Table 5 indicate that, over the entire range of experimental conditions studied, a high efficiency of 3H removal from the fuel composition is detected. In runs nos. 3–6, the tritium content in the voloxidized SNF sample is below the detection limits of the applied analysis technique and is less than 0.2% of its content in the SNF sample before voloxidation. The analogous indicator for 14C during voloxidation in the temperature range of 450–550°C is 2%. The amount of ruthenium in the voloxidized SNF samples diminishes with increasing voloxidation temperature. This experimental fact is probably due to the volatility of ruthenium(VIII) oxide.

We emphasize that in runs nos. 1–3, after the voloxidation completion, the specific activities of 3H, 14C, and 106Ru were determined only in the separated MNUP SNF powder. The content of 3H, 14C, and 106Ru in the fuel remaining after voloxidation on the fuel cladding was not evaluated. Therefore, the specific activity of the listed radionuclides in the unseparated part of the fuel composition may exceed the obtained values.

The reported data on the completeness of tritium removal are consistent with the results obtained during the voloxidation of uranium and uranium–plutonium oxide fuel in [28–31]. The authors note that when voloxidation is carried out in the temperature range of 420–500°C for 6–8 h, the degree of tritium separation from the fuel composition ranges from 99.4 to 99.9%. Most authors report that at a process temperature exceeding 650–700°C, the yield of VFPs decreases due to sintering of the fuel composition.

Based on the data on the separation of the fuel composition from the claddings and the removal of volatile products, the conditions of run no. 5 are optimal for carrying out the MNUP SNF voloxidation, ensuring the efficiency of the process at the minimum required temperature and air supply rate.

At the second stage of the research, experiments were carried out with larger batches of MNUP nuclear fuel spent in KETVS-7 and ETVS-10 for testing the voloxidation efficiency under the selected optimal conditions. Voloxidation was conducted under the conditions of run no. 5. The results on the completeness of fuel composition separation from the fuel element cladding are given in Table 6. Data on the efficiency of removing 3H, 14C, 106Ru from SNF are given in Table 7.

Table 6. Completeness of MNUP SNF separation from fuel cladding
Full size table
Table 7. Content of 3H, 14C, 106Ru in samples of voloxidized nuclear fuel spent in KETVS-7, run no. 7
Full size table

The presented results generally agree with the data obtained in a series of experiments to determine the optimal conditions for the MNUP SNF voloxidation.

The smaller fraction of separated SNF in run no. 7 is probably due to the smaller diameter of the KETVS-7 fuel rods in comparison with those of ETVS-10 and the higher burnup of the spent fuel irradiated in the KETVS-7 composition. The data obtained in the work on the completeness of the fuel composition separation from the fuel claddings disagree with the results of experiments obtained earlier during the MNUP SNF voloxidation, in which it was found that the residual content of nuclear materials on the fuel cladding after the voloxidation completion does not exceed 0.1% [32]. This discrepancy is probably caused by a decrease in the efficiency of the heterogeneous oxidation reaction due to a rise in the weight of oxidized SNF, which leads to an increase in the thickness of the layer of fuel fragments in the crucible of the voloxidation reactor and a deterioration in the air supply to the SNF. To eliminate the influence of this factor in voloxidation reactors, it is necessary to provide for the possibility of agitating the prepared fuel rod fragments during their oxidation.

The residual content of tritium in the voloxidized MNUP SNF is less than 0.2% of its initial content in the fuel composition. The fraction of 14C removed during voloxidation exceeds 98%, and the residual content of 106Ru in SNF is 80.4%.

Thus, experiments with enlarged batches of MNUP SNF proved the voloxidation efficiency under the chosen optimal conditions for separating SNF from fuel cladding and removing tritium and radiocarbon.

CONCLUSIONS

The optimal conditions for MNUP SNF voloxidation are a temperature of 450°C and an air flow rate of 150 mL/min. The completeness of fuel composition separation from the fuel element cladding during MNUP SNF voloxidation under optimal conditions is 98–99%. The fuel cladding residues after voloxidation contain a significant amount of SNF on their surface. To reduce the loss of nuclear materials and their involvement in the fuel cycle, it is required to include the cladding washing procedure in the MNUP SNF processing chain.

MNUP SNF voloxidation makes it possible to remove tritium from the fuel composition by more than 99.8%, reduce the residual content of 14C in SNF to 2% of its initial amount, and prevent the spread of these radionuclides throughout the water-extraction cascade during SNF reprocessing.