Process for separating at least one first chemical element E1 from at least one second chemical element E2, involving the use of a medium comprising a specific molten salt

Abstract

The invention pertains to a process for separating at least one first chemical element E.sub.1 from at least one second chemical element E.sub.2 coexisting in a mixture in the form of oxides, comprising the following steps: a) a step to solubilise a powder of one or more oxides of the said at least one first chemical element E.sub.1 and a powder of one or more oxides of the said at least one second chemical element E.sub.2 in a medium comprising at least one molten salt of formula MFAlF.sub.3 wherein M is an alkaline element, after which there results after this step a mixture comprising the said molten salt, a fluoride of the said at least one first chemical elements E.sub.1 and a fluoride of the said at least one second chemical element E.sub.2; b) a step to contact the mixture resulting from step a) with a medium comprising a metal in the liquid state, the said metal being a reducing agent capable of predominantly reducing the said at least one first chemical element E.sub.1 relative to the said at least one second chemical element E.sub.2, after which there results after this step a two-phase medium comprising a first phase called metal phase comprising the said at least one first chemical element E.sub.1 in oxidation state 0, and a second phase called saline phase comprising the molten salt of above-mentioned formula MFAlF.sub.3 and a fluoride of the said at least one second chemical element E.sub.2.

Claims

1. A method for separating at least one first chemical element E.sub.1from at least one second chemical element E.sub.2coexisting in a mixture in the form of oxides, comprising the following steps: a) a step to solubilise a powder of one or more oxides of said at least one first chemical element E.sub.1and a powder of one or more oxides of said at least one second chemical element E.sub.2in a medium comprising at least one molten salt of formula MFAlF.sub.3, where M is an alkaline element, resulting after this step in a mixture comprising said at least one molten salt, a fluoride of said at least one first chemical element E.sub.1, and a fluoride of said at least one second chemical element E.sub.2; and b) a step to contact the mixture resulting from step a) with a medium comprising a metal in the liquid state, the said metal being a reducing agent capable of predominantly reducing said at least one first chemical element E.sub.1 relative to said at least one second chemical element E.sub.2, resulting after this step in a two-phase medium comprising a first phase which is a metal phase comprising said at least one first chemical element E.sub.1 in oxidation state 0, and a second phase which is a saline phase comprising the least one molten salt of above-mentioned formula MFAlF.sub.3, and a fluoride of the said at least one second chemical element E.sub.2.

2. The process according to claim 1, wherein the element(s) E.sub.1 are selected from the group formed by the actinides, transition elements, and the element(s) E.sub.2 are selected from the group not comprising any actinides.

3. The process according to claim 2, wherein the element(s) E.sub.2 are selected from the group formed by the lanthanides, transition elements other than those of E.sub.1, alkaline or alkaline-earth elements, and/or pnictogenic elements.

4. The process according to claim 1, further comprising reprocessing spent nuclear fuel, transmutation targets used for nuclear physics experimentation, or refractory matrixes included in the composition of nuclear reactors, using said steps a) and b).

5. The process according to claim 1, wherein the molten salt is a salt of formula LiFAlF.sub.3.

6. The process according to claim 1, wherein AlF.sub.3 is contained in the molten salt up to a molar content of 10 to 40 mole %.

7. The process according to claim 1, wherein the metal in the liquid state at step b) is selected from among aluminium and the alloys thereof.

8. The process according to claim 7, wherein the alloy is an alloy of aluminium and copper.

9. The process according to claim 1, further comprising, before step a), a step to prepare the mixture of powders intended to be used at step a).

10. The process according to claim 9, wherein, when the process relates to the reprocessing of uranium oxide spent nuclear fuel, said step to prepare the mixture of powders further comprises: an operation for mechanical treatment of the spent fuel to form a powder of oxide(s); and a heat treatment operation by voloreduction to remove volatile fission products.

11. The process according to claim 9, wherein, when the process relates to the reprocessing of uranium oxide spent fuel, said step to prepare the mixture of powders comprises a voloxidation operation after which uranium oxide UO.sub.2 is converted to uranium oxide U.sub.3O.sub.8.

12. The process according to claim 1, wherein above-mentioned step a) and step b) are performed successively.

13. The process according to claim 12 which, wherein, step a) and step b) are performed successively, further comprises a digestion step of elements selected from among the platinum-group elements and/or molybdenum contained in the mixture resulting from step a), said digestion step being performed after step a) and before step b).

14. The process according to claim 13, wherein the digestion step consists of contacting the mixture resulting from step a) with a medium comprising a metal in the liquid state, said metal being capable of selectively absorbing the platinum-group elements and/or molybdenum relative to the elements E.sub.1 and E.sub.2 contained in the at least one molten salt, the following being obtained after this step: the mixture of step a) being free of said platinum-group element(s) and/or molybdenum; and a metal phase comprising the above-mentioned metal in the liquid state and the said platinum-group element(s) and/or molybdenum.

15. The process according to claim 14, further comprising, after the digestion step, a step to separate the mixture of step a) and the metal phase.

16. The process according to claim 1, further comprising, after step b), a step c) to separate the metal phase from the saline phase.

17. The process according to claim 16, wherein when the process relates to the reprocessing of spent fuel, the metal phase thus separated is subjected to the following successive treatments: a back-extraction step of the actinide(s) by contacting the metal phase with a molten chloride medium in the presence of an oxidizing agent belonging to the chloride family to convert the actinides in the metal state to actinide chloride(s), after which there subsists a metal phase free of actinide(s) and a chloride saline phase; and a step to convert the actinide chloride(s) to actinide oxide(s).

18. The process according to claim 16, wherein the saline phase derived from separation step c) is subjected to the following successive treatments: a distilling step, to regenerate the medium comprising at least one molten salt of MXAlF.sub.3 type; and a vitrifying step of elements E.sub.2 removed from the saline phase after the distillation step.

19. The process according to claim 1, wherein above-mentioned step a) and step b) are performed simultaneously.

20. The process according to claim 13, wherein the platinum-group elements comprise Ru, Rh, or Pd.

Description

DETAILED DESCRIPTION OF PARTICULAR EMBODIMENTS

(1) The following particular embodiments relate to study campaigns conducted on the behaviour of oxides of strategic interest (more specifically the oxides PuO.sub.2, UO.sub.2, U.sub.3O.sub.8, Nd.sub.2O.sub.3, Eu.sub.2O.sub.3, CeO.sub.2, ZrO.sub.2, MoO.sub.3, Y.sub.2O.sub.3, SrO, Sb.sub.2O.sub.3, PdO, RuO.sub.2 and Rh.sub.2O.sub.3 in specific proportions) in media of LiFAlF.sub.3 molten salt type having different molar compositions (different molar compositions for examples 1 to 3 explained below).

(2) The above-mentioned oxides in terms of concentration added to the medium of molten salt type meet the specificities given in the following Table.

(3) TABLE-US-00001 Concentration of the element added to medium of molten Type of oxide salt type (in mg/g of salt) PuO.sub.2 101.1 UO.sub.2 33.3 U.sub.3O.sub.8 33.3 Nd.sub.2O.sub.3 5.2 Eu.sub.2O.sub.3 4.5 CeO.sub.2 5.2 ZrO.sub.2 5.0 MoO.sub.3 5.7 Y.sub.2O.sub.3 4.7 SrO 4.2 Sb.sub.2O.sub.3 4.1 PdO 5.0 RuO.sub.2 5.8 Rh.sub.2O.sub.3 3.2

(4) The study campaigns were all conducted following the same protocol which comprised the following steps: a step for intimate mixing of the LiFAlF.sub.3 salt (15 to 20 g depending on experiments) with one or more oxides in the above Table; a step to place the mixture resulting from the preceding step in a reaction crucible containing an ingot of AlCu alloy (78-22 mole %) of same weight as the LiFAlF.sub.3 salt; a step to place the crucible in a controlled atmosphere under a constant stream of argon followed by a heating step to about 300 C. to dehydrate the whole; a step to heat the crucible to about 835 C. to obtain melting of the crucible content; a step to leave the molten mixture under agitation at constant speed for 4 hours; a step to take a sample of the two phases (saline phase and metal phase respectively) followed by return to ambient temperature; a step for hot dissolution of the samples in 1M Al(NO.sub.3).sub.3, 3M HNO.sub.3 for the saline phases (100 mg salt dissolved in 10 mL solution), and in a hot solution of 3M HNO.sub.3 and 150 L of concentrated HF for the metal phases (100 mg of metal dissolved in 10 mL of solution); a filtration step to retain the insolubles; an optional dilution step of the filtrate in 0.5 M HNO.sub.3.

(5) The concentration of the elements was determined in the two above-mentioned phases: via count and spectrometry for (.sup.239Pu+.sup.240Pu); via liquid X fluorescence for uranium; and via ICP-AES elementary analysis for the other elements (namely Nd, Eu, Ce, Zr, Mo, Y, Sr, Sb, Pd, Ru and Rh).

(6) On completion of the above analyses a material balance was determined for each element (designated <<m>> below), and compared with the material initially added to the crucible.

(7) The percentage of non-solubilised oxide can be estimated rom the following equation:
xm.sub.insoluble=100(xm.sub.met+xm.sub.salt)
where: xm.sub.met is the percentage of element m contained in the metal phase; and xm.sub.salt is the percentage of element m contained in the saline phase.

(8) The quantification of each element m in the saline phase and metal phase allows calculation of the distribution coefficient D.sub.m, using the following equation:

(9) $D_m=\frac{{\mathrm{Xm}}_{\mathrm{met}}}{{\mathrm{Xm}}_{\mathrm{salt}}}$
where: Xm.sub.met is the molar fraction of the element m contained in the metal phase; and Xm.sub.salt is the molar fraction of the element m contained in the saline phase. The calculation of the distribution coefficient D.sub.m does not include the fraction of non-solubilised oxide.

(10) The values xm.sub.met et Xm.sub.salt can be calculated using the following equations:
Xm.sub.met=xm.sub.met/(xm.sub.met+xm.sub.salt) and Xm.sub.salt=xm.sub.salt/(xm.sub.met+xm.sub.salt)

(11) A global distribution coefficient E.sub.m, integrating the fraction of element contained in oxide form after equilibrium (non-dissolved fraction) was evaluated for each element m by determining the ratio:
E.sub.m=xm.sub.met/(xm.sub.salt+Xm.sub.insoluble)

(12) This distribution coefficient (although not determined on thermodynamic equilibrium) indicates the amount of element m present in the metal at the end of the experiment in relation to the initial amount added in oxide form to the crucible.

(13) It is considered that the implementing of the process of the invention is efficient for selective extraction of actinides, if the following conditions are combined: the actinides are largely in majority in the aluminium phase; and a minimum amount of other elements is contained in the aluminium;

(14) which, in other words, means that the value of the global distribution coefficient of the actinides (symbolised E.sub.AN) must be high (i.e. Log E.sub.AN>0 and ideally Log E.sub.AN>1) and that the value of the global distribution coefficient of the fission products (symbolised E.sub.FP) must be low (i.e. Log E.sub.FP<0 and ideally Log E.sub.FP<1).

EXAMPLE 1

(15) This example illustrates a campaign of studies conducted on the behaviour of oxides of strategic interest such as defined above conforming to the above-mentioned operating mode in a specific molten salt medium of LiFAlF.sub.3 type (containing 35 mole % of AlF.sub.3).

(16) This campaign included several tests: a first test with UO.sub.2 alone entailing the use of 15.7 g of LiFAlF.sub.3; a second test with U.sub.3O.sub.8 alone entailing the use of 15.2 g of LiFAlF.sub.3; a third test with a mixture comprising PuO.sub.2 and Nd.sub.2O.sub.3 entailing the use of 17 g of LiFAlF.sub.3; a fourth test with a mixture comprising ZrO.sub.2, MoO.sub.3, RuO.sub.2, Rh.sub.2O.sub.3, PdO and Nd.sub.2O.sub.3 entailing the use of 20.1 g of LiFAlF.sub.3; a fifth test with a mixture comprising SrO, Y.sub.2O.sub.3, Sb.sub.2O.sub.3, CeO.sub.2 and Eu.sub.2O.sub.3 entailing the use of 15.1 g of LiFAlF.sub.3.

(17) For each of these tests the global distribution coefficient E.sub.m and the distribution coefficient D.sub.m were determined, the methods of determination being explained below for the element(s) of the oxide(s) involved.

(18) The logarithmic values of these coefficients are grouped together in the Table below.

(19) TABLE-US-00002 Element xm.sub.salt xm.sub.met xm.sub.insoluble Log D.sub.m Log E.sub.m Pu 4.5 90 5.5 1.30 0.95 U (derived 4.1 97.9 0.0 1.38 1.38 from UO.sub.2) U (derived 3.3 93.6 3.1 1.45 1.16 from U.sub.3O.sub.8) Nd (derived 47.1 8.9 44.0 0.72 1.01 from fourth test) Eu 92 1.6 6.5 1.76 1.80 Ce 80.1 12.3 7.6 0.81 0.85 Zr 28.2 11.5 60.3 0.39 0.89 Mo 50.6 10.8 38.6 0.67 0.91 Y 96.7 2.1 1.2 1.66 1.67 Sr 98.8 1.2 0.0 1.90 1.90 Sb 1.8 4.9 93.3 0.44 1.29 Pd 18.5 27.0 54.5 0.16 0.43 Ru 2.1 9.4 88.5 0.65 0.98 Rh 1.3 24.8 73.8 1.27 0.48

(20) Several important points emerge from this Table.

(21) The solubilisation of the actinide oxides in LiFAlF.sub.3 (35 mole % of AlF.sub.3) (namely PuO.sub.2, UO.sub.2 and U.sub.3O.sub.8) is total or near-total (>94.5%). As a result, the coefficients D.sub.m and E.sub.m show similar values. After dissolution of the oxides and extraction equilibrium reached, nearly all the actinides are present in the metal phase which translates as logarithmic values of the distribution coefficients D.sub.m and E.sub.m close to 1 or higher than 1. It is probable that the high extraction of the actinides by the aluminium leads to a shift in equilibrium further promoting solubilisation of the oxide (solubility saturation in the salt never being reached).

(22) The results obtained with UO.sub.2 and U.sub.3O.sub.8 show very similar behaviour both regarding the solubilisation of the oxides and the uranium extraction yield by the metal. This is an important result since it allows validation of the two choices of heat treatment upstream of the solubilisation/extraction step, namely: either recourse to conventional heat treatment (in H.sub.2 or Ar atmosphere, after grinding of the fuel) or heat treatment of the fuel via voloxidation process.

(23) The other elements studied all show a very low Log E.sub.m value (<0, even <1 in most cases) which translates their reduced presence, after equilibrium, in the metal phase.

(24) Two causes could explain the Log E.sub.m values obtained for these elements: either low or very low solubility of the oxides in LiFAlF.sub.3; or low extraction yield in the metal phase.

(25) The first of these two causes is fully illustrated for the platinum-group elements (Ru and Rh). They display positive Log D.sub.m values meaning that these elements once solubilised in LiFAlF.sub.3, are mostly extracted by the metal phase. As shown in the Table above, their content is very small in the metal phase due to the very low solubility of these elements in LiFAlF.sub.3 (it is more or less zero if the reducing metal is not present).

(26) The second of these causes is fully illustrated for the lanthanide elements and for yttrium and strontium. These elements show very close Log D.sub.m and Log E.sub.m values, demonstrating that their respective oxides are well solubilised in LiFAlF.sub.3. These elements are scarcely extracted from the metal phase during reducing extraction.

(27) Finally, some elements (Nd, Zr, Mo and Pd) are penalised by the accumulation of these two causes, which translates as low solubility and low extraction yield, resulting in Log E.sub.m<0 values (even <1 for some thereof).

(28) Example 1 shows 90% recovery of the plutonium initially placed in the crucible and near-quantitative recovery of uranium, all in a single step. It demonstrates that the initial form of the uranium oxide is compatible with the different envisaged heat treatments for the fuel. Example 1 is a perfect illustration of the feasibility of the separation of actinides/fission products within a DOS process.

EXAMPLE 2

(29) This example illustrates a campaign of studies conducted on the behaviour of oxides of strategic interest (more specifically PuO.sub.2, UO.sub.2, Nd.sub.2O.sub.3, ZrO.sub.2, MoO.sub.3, PdO, RuO.sub.2 and Rh.sub.2O.sub.3) conforming to the above-mentioned operating mode in a specific molten salt medium of LiFAlF.sub.3 type (comprising 15 mole % AlF.sub.3).

(30) This campaign comprised several tests: a first test with UO.sub.2 alone entailing the use of 15.6 g of LiFAlF.sub.3; a second test with a mixture comprising PuO.sub.2 and Nd.sub.2O.sub.3 entailing the use of 17 g of LiFAlF.sub.3; a third test with a mixture comprising ZrO.sub.2, MoO.sub.3, RuO.sub.2, Rh.sub.2O.sub.3, PdO and Nd.sub.2O.sub.3, entailing the use of 21 g of LiFAlF.sub.3.

(31) The global distribution coefficient E.sub.m and distribution coefficient D.sub.m were determined, the methods of determination being explained below for the element(s) of the oxide(s) concerned.

(32) The logarithmic values of these coefficients are grouped together in the Table below.

(33) TABLE-US-00003 Element xm.sub.salt xm.sub.met xm.sub.insoluble Log D.sub.m Log E.sub.m Pu 2 70 28 1.54 0.37 U (derived 3.7 87.7 8.6 1.37 0.85 from UO.sub.2) Nd (derived 33.9 28.2 37.9 0.08 0.40 from third test) Zr 17.5 8.8 73.7 0.30 1.02 Mo 35.5 6.6 57.9 0.73 1.15 Pd 19.8 14.0 66.2 0.15 0.79 Ru 2.2 4.7 93.1 0.32 1.31 Rh 5.4 24.2 70.4 0.65 0.49

(34) From this Table the following important points emerge.

(35) As previously, extensive solubilisation of the actinide oxides was observed.

(36) As in Example 1, the distribution coefficients D.sub.An and E.sub.An obtained after the experiments show values (Log E.sub.An>0) fully compatible with the implementing of the process of the invention to separate actinides/fission products for the reprocessing of spent fuel.

(37) The other elements studied all show a very low Log E.sub.m value (<0) translating their slight presence, after equilibrium, in the metal phase. As in the preceding example this can be accounted for by the low solubility of their respective oxides in LiFAlF.sub.3, or by a low extraction yield in the metal phase.

EXAMPLE 3

(38) This example illustrates a campaign of studies conducted on the behaviour of oxides of strategic interest (more specifically PuO.sub.2, Nd.sub.2O.sub.3, ZrO.sub.2, MoO.sub.3, PdO, RuO.sub.2 and Rh.sub.2O.sub.3) conforming to the above-mentioned operating mode in a specific molten salt medium of LiFAlF.sub.3 type (comprising 25 mole % of AlF.sub.3).

(39) This campaign comprised several tests: a first test with a mixture comprising PuO.sub.2 and Nd.sub.2O.sub.3 entailing the use of 17 g of LiFAlF.sub.3; a second test with a mixture comprising ZrO.sub.2, MoO.sub.3, RuO.sub.2, Rh.sub.2O.sub.3, PdO and Nd.sub.2O.sub.3 entailing the use of 15.7 g of LiFAlF.sub.3.

(40) The global distribution coefficient E.sub.m and distribution coefficient D.sub.m were determined, the determination methods being explained below, for the element(s) of the oxide(s) involved.

(41) The logarithmic values of these coefficients are grouped together in the Table below.

(42) TABLE-US-00004 Element xm.sub.salt xm.sub.met xm.sub.insoluble Log D.sub.m Log E.sub.m Pu 2.2 94 3.8 1.63 1.19 Nd (derived 44.9 19.5 35.6 0.36 0.62 from second test) Zr 23.6 10.0 66.4 0.37 0.96 Mo 50.1 9.2 40.8 0.74 0.99 Pd 27.6 12.1 60.3 0.36 0.86 Ru 1.4 6.9 91.6 0.68 1.13 Rh 1.0 20.5 78.5 1.31 0.59

(43) Several important points emerge from this Table.

(44) As previously, extensive solubilisation of plutonium oxide was observed.

(45) As in the preceding examples, the distribution coefficients D.sub.An and E.sub.An obtained after the experiments show values (Log E.sub.An>0 or .sup.1) that are fully compatible with the implementing of the process of invention to separate actinides/fission products for the reprocessing of spent fuel.

(46) The other elements studied all show a very low Log E.sub.m value (<0, even <1 in most cases) translating their slight presence, after equilibrium, in the metal phase. As in the preceding example, this can be accounted for by the low solubility of their respective oxides in LiFAlF.sub.3, or by a low extraction yield in the metal phase.

* * *

(47) It follows from these examples that the campaigns of experiments allowed results to be obtained in terms of recovery of actinides (by the metal phase) that are fully satisfactory. The separation factors between the actinides and the other elements show that the process of the invention can be fully applied to different molar compositions of the salt LiFAlF.sub.3.

(48) As previously indicated, the following conditions must advantageously be met: a high global distribution coefficient E.sub.An, i.e. Log E.sub.An>0 (ideally, Log E.sub.An>1) for the actinides and low global distribution coefficient E.sub.FP (FP designating the fission products) i.e. Log E.sub.FP<0 (ideally, Log E.sub.FP<1) for all the other elements. The above examples allowed successful fulfilling of these conditions using LiFAlF.sub.3 salts having a composition varying between 15 and 35 mole % of AlF.sub.3.

(49) The salt of composition LiFAlF.sub.3 (comprising 35 mole % AlF.sub.3), preferred to the others for facilitated solubilisation of the oxides, is well suited. The yields E.sub.m obtained after experimental validation show that it would be possible to recover more than 99% of the actinides when setting up two stages of extraction. After extraction, the separation factors between actinides and fission products are sufficient to envisage efficient fuel reprocessing. The addition of a washing stage before oxidative back-extraction should further increase the fission product decontamination rates of the actinides.

(50) The integration of the process of the invention in a scheme for the reprocessing of nuclear fuel of oxide or carbide type via reducing extraction in molten fluoride medium (LiFAlF.sub.3) leads to a very good actinide recovery rate (typically more than 99% with fewer than three extraction stages) and allows high selectivity between actinides and fission products.

(51) In the developed process, the actinides contained in the irradiated fuel (U, Np, Pu, Am and Cm) remain grouped within one same flow which imparts good proliferation resistance to the process and meets the objectives of fourth generation reactors to reduce the noxiousness of waste with long lifetime. This process scheme can be applied to oxide fuels but also to carbide fuels (through application of suitable heat treatment). The field of application of this process can be extended to the reprocessing of all fuels (such as nitrides) or irradiated targets provided it is possible for them to be converted to an oxide at the head-end of the process thereby providing the process with large flexibility.