Introduction

Organohalides are often considered as diluents for tributyl phosphate (TBP) in the fractionation of highly radioactive waste. Among the organohalides that attract attention are carbon tetrachloride, hexachlorobutadiene (HCBD), freons, hexachlorotetrafluorobutane, trifluorodichloroethylpentachloroethyl ether and the like [1,2,3,4,5]. The TBP–HCBD–HNO3 extraction system is used to extract, separate and purify uranium and plutonium in the chemical-metallurgical processing of decommissioned standard uranium blocks. However, HCBD and other chlorine-containing compounds are highly toxic, and their radiolysis is accompanied by the formation of chlorine and/or HCl, which at elevated temperatures is fraught with severe corrosion of process equipment. Thus, at a dose of 1.18 MGy in water drops released from a 30% TBP–HCBD–HNO3 extract, the concentration of HCl can reach 6.6 M [6].

Unlike chlorine-containing compounds, organofluorine compounds attract close attention due to the higher stability of C–F bonds and the low probability of radiolytic formation of F2 [7,8,9,10,11,12,13,14]. However, the joint irradiation of TBP and organofluorine diluents in the presence of HNO3 has not been practically studied, which hinders the development of recommendations on the composition and degree of fluorination of a potentially acceptable diluent.

This work focuses on the radiolytic transformations of formal-n2 (diluent) in its mixture with TBP (40%; extractant) in the presence of dissolved nitric acid, water, and plutonium nitrate (organic phase of the extraction system). The radiolytic transformations of the aqueous phase are not discussed, since formal-n2 is concentrated in the organic phase and, accordingly, the properties of the organic phase are most dependent on the radiolytic transformations of formal-n2. Formal-n2 is a fluorinated diester (acetal) with a density of 1.67 g cm–3 and a viscosity of 13.4 mPa s (at 20 °С). The number of fluorine atoms is twice the number of H atoms (the electron fraction of fluorine is 61.5%) [15]. The applicability of formal-n2 as a diluent in the extraction of rare-earth and transplutonium elements (Eu and Am) with phenyloctyl-N,N-diisobutylcarbamoylphosphine oxide (CMPO), including in the presence of 30% TBP, was investigated by V.G. Khlopin Radium Institute [16, 17]. Extraction processes using metanitrobenzotrifluoride (F3) and phenyltrifluoromethylsulfone (FS-13) as diluents were compared under the same conditions. At the same concentrations of CMPO, the most favorable results in the extraction of HNO3, Am, and Eu were observed in the cases of formal-n2 and F3 (especially when 30% TBP was added to both systems), despite the lower density and viscosity of F3 (1.436 g cm–3 and 3.05 mPa s, respectively) [16]. A noticeable susceptibility of formal-n2 to ionizing radiation was previously shown [18, 19], but the effect of TBP and HNO3 on its radiolytic transformations has not been studied.

Experimental

Extraction system

The extraction of plutonium with a solution of TBP in octafluoropentanol formal (formal-n2, CH2[OCH2(CF2)3CF2H]2) from an aqueous solution of 0.21 M 239Pu(IV) nitrate and 3 M HNO3 was studied. TBP was preliminarily purified by successive alkaline-permanganate, oxalate, and alkaline treatments, followed by washing with distilled water until neutral. Octafluoropentanol formal (99.6%) supplied by Mayak was not subjected to additional purification.

The nitric acid solution of 239Pu(IV) was preliminarily purified (from 241Аm accumulated over time) on a column with VP1–AP anion exchanger. Partially oxidized to the hexavalent state, plutonium (~ 15%) was reduced with hydrogen peroxide to Pu(IV), after which it was stabilized by purging with nitrogen oxides for 6 h. Nitric acid solutions were prepared using extra pure HNO3. Plutonium was extracted in a glass separating funnel with a glass (rotation speed ≈ 1200 rpm) or steel (rotation speed 600–800 rpm) screw stirrer. The stirring time was 30 min, and the ratio of the volumes of the aqueous and organic phases was 1/1.

The role of radiolysis was studied under two irradiation modes: under the action of plutonium’s intrinsic alpha radiation and under the action of accelerated electrons.

Irradiation with alpha particles

Initially, the organic phase (40% TBP in formal-n2; 200 ml) was mixed with an equal volume of an aqueous solution of 0.21 M 239Pu(IV) nitrate and 3 M HNO3 (30 min; steel stirrer). Stirring ensured partial redistribution of nitric acid, water, and plutonium nitrate from the aqueous phase to the organic phase. The resulting two-phase system (without further stirring) was kept in a glass cuvette with a water seal (filled with 3 M HNO3 solution) for 61 days (under normal room lighting and temperature). Thus, natural mass transfer occurred between the phases, and 239Pu α-particles caused radiolysis in both phases. Periodically, equal aliquots were taken from both phases. These aliquots were then mixed with each other (30 min; steel stirrer) and, after phase separation, aqueous and organic samples were used for analysis.

The dose rate was calculated according to the Eq. 

$$I = \frac{{\uplambda } \times C \times {N}_{A} \times {E}_{\alpha } \times {K}_{E}}{A}$$
(1)

where λ = 9.11 s− 1 atom− 1, 239Pu decay constant; C = 50 g dm− 3, plutonium concentration (metal); NA = 6.02 × 1023 mol− 1, Avogadro’s number; Еα = 5.14 MeV, average energy of 239Рu α-particles; KE = 1.602 × 10− 13 J MeV− 1, conversion factor; A = 239, atomic mass. The average energy of γ-quanta (0.05 MeV, about 1% of Eα) was not taken into account due to the insignificance of the value. The dose rate calculated in this way was 0.094 W dm− 3, or 0.11–0.15 Gy s− 1, depending on the density of the sample. Dose Do absorbed in the organic phase is: Do = DKd (1 + Kd)−1, where D is the total absorbed dose in the extraction system; Kd is the distribution coefficient of plutonium, defined as the ratio of the alpha activities of the organic and aqueous phases.

Electron beam treatment

The object of electron beam radiolysis was an organic solution − 40% TBP in formal-n2 (initial volume 200 ml). The experiment consisted of simulating 50 extraction cycles, each of which included the following stages: (1) saturation of the organic solution with nitric acid by contact with 3 M HNO3 aqueous solution (3 stirring periods of 20 min each); (2) electron beam irradiation of the resulting organic phase at a dose of 2.8 kGy; (3) extraction of the radionuclide from a 3 M HNO3 aqueous solution into the organic phase (stirring for 30 min); (4) re-extraction of the radionuclide with a 0.2 M HNO3 aqueous solution from the organic phase (3 stirring periods of 10 min each); (5) regeneration of the organic phase with a 5% Na2CO3 aqueous solution (2 stirring periods of 10 min each) to remove acidic radiolysis products. Thus, only the periodically regenerated organic phase containing nitric acid but free of plutonium nitrate was subjected to irradiation (a stepwise increase in the absorbed dose up to 150 kGy). In turn, for each operation of acidification, extraction, re-extraction and regeneration (stages 1, 3, 4 and 5), fresh aqueous solutions were used in a volume equal to the volume of the organic solution.

All operations were carried out at room temperature. At the irradiation stage, a UELV-10-10TM linear electron accelerator with a vertically scanned 7.5 MeV electron beam (6 µs pulses, 300 Hz pulse repetition rate, 800 µA average beam current, 245 mm scan width, 1 Hz beam scanning frequency) was used. The average dose rate was 1.2 kGy s− 1. To avoid radiation heating, irradiation was carried out for 2.3 s (2.8 kGy) in each cycle. For dosimetry, films of copolymer with phenazine dye CO PD(F)R-5/50 (GSO 7865 − 2000) were used.

Analysis

Activity in aqueous and organic solutions was measured on Beckman Instruments, Inc. scintillation spectrometer using Optiphase Hi Safe 3 universal fluid as a scintillator. The effect of the component composition of the analyzed organic and aqueous samples was taken into account using the corresponding calibration coefficients.

The concentration and valence state of plutonium in nitric acid solutions were determined spectrophotometrically (SF-2000). The rate of phase separation of the emulsion, as well as the density and viscosity of the organic phase, were periodically monitored. The viscosity of the organic phase was measured using a glass capillary viscometer calibrated against n-tetradecane. Dynamic viscosity (mPa s) at t = 20 °C was calculated using the equation:

$${\eta }_{20} = 0.0674 \times {\rho }_{t} \times {\tau }_{t} \times {1.0352}^{(t-20)}$$
(2)

where ρt is the sample density (g cm− 3), τt is the liquid outflow time (s) at temperature t.

Results and discussion

Autoradiolysis with 239Pu alpha particles

During all two months of exposure to 239Pu alpha radiation, there is a change in the properties of the organic phase, 40% solution of TBP in formal-n2. A gradual decrease in the plutonium distribution coefficient Kd between the organic and aqueous phases is observed; it decreases from 3.6 to 1 (Fig. 1). At the same time, some other properties of the organic phase change less. For example, the density remains unchanged, and the increase in viscosity does not exceed 10–12% (from 12.1 to 13.5 mPa s).

Fig. 1
figure 1

Influence of the absorbed dose D of 239Pu’s intrinsic alpha radiation on the distribution coefficient of plutonium, Kd. Do is the dose absorbed in the organic phase

Full size image

At doses up to 100 kGy, a linear decrease in Kd with dose is observed, which may indicate the dominant effect of primary radiolytic intermediates on the transformations of plutonium nitrate complexes. The primary radiolytic processes include the formation of anions and fluorine radicals (from formal-n2), as well as the decomposition of nitrate ions (dissolved in the organic phase) in reduction and oxidation reactions. Unlike CHF3 and CH2F2, which have negative electron affinities, larger organofluorine molecules can participate in the dissociative capture of thermal and hot electrons with the elimination of fluoride ions [7, 18].

C11F16H8O2 + e− → ⋅C11F16H8O2− → ⋅C11F15H8O2 + F− (3).

The stability constants of \(\text{P}\text{u}{\text{F}}_{3}^{+}\) complexes are 3–4 orders of magnitude higher than the corresponding values for nitrate complexes [20,21,22]. Substituting nitrate ions in the coordination sphere of plutonium, F− anions hinder its dissolution in organic media and, thereby, reduce Kd.

HF precursors can also be fluorine radicals, which arise, in particular, upon homolytic cleavage of the C–F bond in excited formal molecules and ions. Reacting with alkyl groups of TBP or formal (RH), the ⋅F radical is able to abstract hydrogen to form HF.

⋅F + RH → ⋅R + HF. (4)

Thus, the radiation-induced partial defluorination of the formal-n2 can serve as one of the most significant reasons for the decrease in Kd. Another reason may be the reduction of nitrate to \(\dot{\text{N}}{\text{O}}_{3}^{2-}\), \(\text{H}\dot{\text{N}}{\text{O}}_{3}^{-}\) or \(\dot{\text{N}}{\text{O}}_{2}\) in fast reactions with free ⋅Н radicals or electrons [23, 24]. Nitrate can also decay in the reaction with the radical cation (\({\dot{\text{R}}}^{+}\)) of TBP or formal-n2

$$\text{N}{\text{O}}_{3}^{-} + {\dot{\text{R}}}^{+} \to \dot{\text{O}\text{N}{\text{O}}_{2} + \text{R}}$$
(5)

.

The short range (≤38 μm [7]) and the high ionization density (local) characteristic of α-particles contribute to a high probability of nitrate degradation directly in the Pu coordination sphere, as well as the formation of HF and F− in the immediate vicinity of this coordination sphere.

In parallel with the radiation-induced decrease in Kd, a slight acceleration (from 0.68 to 0.77 mm s–1) of the separation of the aqueous and organic phases is observed in the dose range up to ~ 300 kGy (Fig. 2). With a further increase in the dose, the phase separation rate decreases down to 0.53 mm s–1.

Fig. 2
figure 2

Effect of the absorbed dose of 239Pu alpha radiation on the rate of phase separation

Full size image

Several competing processes affect viscosity and the rate of phase separation. The presence of an acid (HF and HNO3) negatively affects the formal-n2 stability due to acid catalysis [15, 25]:

$$\text{R}\text{O}\text{C}{\text{H}}_{2}\text{O}\text{R} \underleftrightarrow{{\text{H}}^{+}} \text{R}\text{O}\text{H} + \text{R}\text{O}\text{C}{\text{H}}_{2}^{+} \underleftrightarrow{{\text{H}}^{+}, { \text{H}}_{2}\text{O}} 2\text{R}\text{O}\text{H} + \text{C}{\text{H}}_{2}\text{O}$$
(6)

.

Radicals ⋅R arising from TBP and formal-n2 as a result of direct absorption of radiation energy, as well as in elimination reactions of type (2), participate in the formation of larger molecules, in particular, dimers [18, 23]. At the same time, the radiolytic cleavage of C–O bonds typical of simple ethers occurs, followed by the formation of more mobile octafluoropentanol [26]. The probability of cleavage of C–F and C–C bonds in fluoroalkyl groups is comparable to each other, but lower than the probability of cleavage of C–O skeletal bonds [7].

Also, one of the important processes is the formation of butanol and dibutyl phosphoric acid from TBP. Debutylation of TBP occurs during both hydrolysis and radiolysis [23]. Dibutyl phosphate, in particular, is able to slow down the decrease in Kd. However, in order to significantly increase the extraction of plutonium, the concentration of dibutylphosphoric acid should be about 2 M [27]. But at the doses used in this work, such a concentration is unattainable (≤10–2 M [23]). At the same time, even at a relatively small fraction in the irradiated organic phase, strong complexes of plutonium with dibutyl phosphate can be poorly removed during re-extraction and even during regeneration of the extractant [27].

Despite the variety of the above chemical transformations, the valence of plutonium in the aqueous phase remains practically unchanged (Fig. 3): the fraction of Pu(VI) at 376 kGy does not exceed 1% of the total Pu in the aqueous phase, which can contribute to a decrease in Kd by no more than 2–3%. Plutonium in aqueous solutions is characterized by the simultaneous presence of ions with different oxidation states [22, 28]. The autoradiolysis of the aqueous phase contributes to a certain variety of plutonium valence states due to the simultaneous occurrence of both oxidation (mainly with the participation of ⋅OH and ⋅HO2 radicals) and reduction (with the participation of hydrated electron and ⋅H) [7, 28]. In the absence of other metals [28], the radiolytic oxidation of plutonium seems more likely, since the oxidizing radicals are rather inert to nitrate, while the reducing radicals are effectively scavenged by nitrate. This is the reason for the appearance of Pu(VI), shown in Fig. 3b.

In the organic phase, the radiation-induced redox transformations of Pu(IV) weaken, since both TBP and formal-n2 are present in high concentrations and serve as strong scavengers of both oxidizing (alkoxy) and reducing (electron and ⋅H) radicals [7, 18].

Fig. 3
figure 3

Optical absorption spectra (a) of plutonium solution (diluted to 1.7 g dm− 3 with 3 M HNO3 solution): (1) after evaporation at 35–40 °C and 20–25 mm Hg; (2) after additional treatment with hydrogen peroxide and nitrogen oxides; (b) of aqueous phase (diluted 10 times with 3 M HNO3 solution) after 46 days of irradiation (376 ± 15 kGy dose)

Full size image

The effect of HF noted above can be reduced by the cyclic use of the organic phase, when HF can be washed out in each cycle during the regeneration of the organic phase with an alkaline solution. The removal of HF can help reduce the dependence of Kd on the dose, as well as reduce the corrosion of equipment that comes into contact with the organic phase during irradiation.

Radiolysis with accelerated electrons in a cyclic mode

The observed value of Kd during extraction from 3 M HNO3 solution fluctuates within 10–15%, slightly increasing with the dose (Fig. 4). At the same time, the concentration of nitric acid in the organic phase is practically constant over the entire range of absorbed doses and is ≈ 0.13 mol dm− 3.

Fig. 4
figure 4

Change in the plutonium distribution coefficient in a cyclic mode with electron beam treatment

Full size image

At the same time, radiolysis negatively affects the stage of plutonium re-extraction (Fig. 5). The fraction of plutonium remaining in the organic phase during re-extraction increases at doses of 110–125 kGy by a factor of 3–5 compared to the initial value. Almost the same type of correlation between absorbed dose and plutonium content is observed for the regeneration stage. Increasing the dose to 110 kGy leads to an almost tenfold increase in the residual activity of plutonium in the organic phase after regeneration. The retention of plutonium in the organic phase can be facilitated by the products of radiolysis, in particular, the products of nitration and nitroxylation [23] of tributyl phosphate and octafluoropentanol. Moreover, radiolysis of both TBP and formal-n2 leads to the formation of high molecular weight products, in particular, via combination of radicals [18, 23]. For example, the TBP dimer and radical adducts of TBP retain the functional groups that ensure extraction, but their mobility in the organic phase and permeability with respect to water are undoubtedly reduced compared to TBP. Large molecules of this type can contribute to the retention of plutonium in the organic phase. The high dose rate provided by the electron beam creates a sufficiently high concentration of radicals and thus increases the probability of radical recombination.

The butyl groups in the TBP molecule can serve as the main source of hydrogen in the formation of HF by reaction (4). H-abstraction is characteristic not only for ⋅F and ⋅H radicals, but also for alkoxy radicals arising from the homolytic cleavage of the C–O bond in the formal. In a viscous condensed medium, the combination of radicals dominates over their disproportionation [7]; accordingly, large TBP and formal radicals can serve as the main precursors of bulkier molecules that promote the retention of plutonium in the organic phase.

Fig. 5
figure 5

Influence of the absorbed dose on the content of residual plutonium in the organic phase (1), the retention of plutonium in the organic phase during re-extraction (2) and the residual content of plutonium in the organic phase after regeneration (3)

Full size image

The effect of radiolysis and hydrolysis of the organic phase is observed visually when monitoring foaming and emulsion stability. Previously, it was shown that even in the absence of other components, the transparency of the irradiated formal-n2 deteriorates, and the viscosity increases (up to 6 times at a dose of 1.76 MGy) [18]. The increase in viscosity is non-linear, accelerating with increasing dose. Moreover, the irradiated formal forms a gelatinous liquid mass with water, the separation of which is significantly slowed down.

In a funnel with a steel stirrer having a sharp edge, phase separation slows down by 6–8 times when absorbed doses reach 110–140 kGy (Fig. 6, a). This effect is not observed when using a glass stirrer with rounded edges - the rate of phase separation practically does not change with the dose. The sharp edge contributes to a strong dispersion of the solution, which makes the rate of separation of the emulsion more dependent on the composition, size, and surface properties of the droplets. Even with a slight increase in the speed of rotation of the steel stirrer, the rate of phase separation can decrease by an order of magnitude. At doses above about 40 kGy, colorless transparent micro-emulsions with a more stable phase boundary are observed at the phase separation stage. At the stages of extraction and re-extraction of plutonium, the decrease in the rate of phase separation with dose occurs rather monotonically, with the exception of the dose range of 70–100 kGy (Fig. 6, b).

Fig. 6
figure 6

Influence of the absorbed dose on the rate of phase separation when the organic phase is saturated with 3 M HNO3 aqueous solution (at the 1st and 2nd contacts), during the extraction of plutonium from 3 M HNO3 and during re-extraction (1st contact)

Full size image

With all types of intercycle treatment, the change in the viscosity of the organic solution with the dose occurs in a similar way (Fig. 7). The initial viscosity is about 12 mPa s. With increasing dose, it almost does not change until the 15th cycle. Then there is a noticeable increase, and by the 31st cycle (dose of 86.8 kGy), the viscosity reaches its maximum values, which are 1.4–1.5 times higher than in the original sample. With a further increase in the dose, the viscosity gradually decreases almost to the initial level. It is noteworthy that at a dose of 80–90 kGy, not only the maximum values of viscosity are observed, but also noticeable deviations from monotonicity in the dynamics of the change in the rate of phase separation during the extraction and re-extraction of plutonium (Fig. 2). The observed effect can be explained by the competition of differently directed processes: the accumulation of heavier products of the radiolytic combination of radicals [7, 18, 23], which contribute to an increase in viscosity, and the accumulation of products of radiolytic and hydrolytic degradation of the solution components, which contribute to a decrease in viscosity and some slowdown in phase separation.

Fig. 7
figure 7

Effect of cyclic treatment with electron beam irradiation on the viscosity of the organic phase at the end of the stages of irradiation, extraction, re-extraction and regeneration

Full size image

Conclusion

The extraction stability of a system containing 40% TBP in formal-n2 in the presence of 3 M HNO3 and 0.21 M Pu(IV) was studied under continuous irradiation with intrinsic α-particles and under cyclic irradiation with accelerated electrons. There is no doubt that the radiolytic transformations of formal-n2 make a significant contribution to the formation of products that affect the properties of the organic phase, including the retention of plutonium. Moreover, formal-n2 is able to scavenge radiolytic intermediates originating from other components of the organic phase and thus affect both ion and radical exchange. The negative effect of radiolysis on the extractability of plutonium is associated primarily with changes in the Pu(IV) coordination sphere due to the partial replacement of nitrate by the fluoride ion, as well as the radiolytic degradation of nitrate. This changes the solubility of plutonium salts and complexes in the organic phase.

The cyclic mode of using the organic phase in the range of absorbed doses of 100–140 kGy contributes to a decrease in the equilibrium concentration of radiolysis products responsible for the decrease in the interfacial distribution coefficient of plutonium. At the same time, the residual concentration of plutonium in the organic phase after re-extraction for 45–50 cycles (120–150 kGy) increases by 3–5 times, and after regeneration - by 4–6 times.

The rate of phase separation for 45–50 cycles is reduced by 5 times during extraction, by 4 times during re-extraction and by 2–3 times during regeneration. In the latter case, already after 10–15 cycles (30–40 kGy), phase separation becomes difficult due to sufficiently long-lived microemulsions with a more stable phase boundary (bubbly structures).

The increase in viscosity and parallel deceleration of phase separation is facilitated by the radiolytic formation of high molecular weight products. The mode of cyclic regeneration of the organic phase at doses up to 50–60 kGy minimizes the undesirable accumulation of radiolysis products that impair the extraction process.

This work dealt with special cases of irradiation with alpha particles and accelerated electrons. At the same time, the radiolytic effects observed for formal-n2 can also be projected onto other types of ionizing radiation due to the generality of the initial process, ionization, i.e. the transformation of a molecule into a radical cation due to the expulsion of an electron. Moreover, such emitted electrons (secondary) make a key contribution to the subsequent set of ionizations and chemical transformations [7]. At the same time, the effect of radiolysis also depends on the composition of the irradiated object, and on the properties of ionizing particles (mass, charge, energy, etc.), which determine the linear transfer of energy to the substance subjected to irradiation. Therefore, further studies of radiolytic transformations of formals, depending on the type of ionizing radiation and the composition of the extraction system, seem to be relevant from both fundamental and applied points of view.