Separation of 241Am3+ from 154Eu3+ Using 3,3’-Butyloxy-Bis-1,2,4-Triazinyl-2,6-Pyridine as a Potent Receptor

ABSTRACT The effective partitioning and transmutation of the minor An from the Ln in spent nuclear fuel in a non-proliferative manner is critical to lowering potential impact to the environment, moderating concern over the increase of nuclear power as a green alternative to fossil fuels, and solving a grand challenge in separation science. Although present in less than .1 wt.%, Am and Cm represent half-lives of over 400 years and significantly contribute to heat load through radioactive decay. Separation of the minor An from the neutron-poisoning Ln can advance the nuclear fuel cycle further toward closure and decrease the volume and radiotoxicity of daughter nuclides stored in a geologic repository. Similar physical properties and metal-complexant binding phenomena render this separation difficult. In this work, a comprehensive liquid-liquid separations study using the recently discovered 3,3’-butyloxy-bis-1,2,4-triazinyl-2,6-pyridine in the polar aprotic solvent trifluoromethylphenyl sulfone is presented. Unlike contemporary bis-1,2,4-triazinyl-2,6-pyridine complexants, the current system is stable and performs well in highly acidic systems. Separation of 241Am3+ from 154Eu3+, acid range tolerance, complexant concentration, and decomplexation studies are presented herein. Graphical abstract


Introduction
The management and disposal of spent nuclear fuel (SNF) which contains useable enriched uranium, transuranics, and fission products, continues to present challenges to closing the nuclear fuel cycle and increasing nuclear energy's viability as a carbon-neutral source.As the most dense [1] and efficient [2] source of clean energy, nuclear energy is predicted to have an increased contribution to the energy portfolio of the world toward meeting energy demands.Concerns with an open fuel cycle include long-lived radioactivity of plutonium and the minor An, specifically americium and curium, which pose an environmental risk for short and long-term storage.An additional issue lies within the need to increase nuclear capacity in the energy sector while discarding approximately 96% of usable uranium and plutonium that is present in SNF.Progressing toward a closed fuel cycle can address these challenges by implementing the well established PUREX process [3] for recycling plutonium and uranium; however, the highly acidic (3-5 M HNO 3 ) raffinate produced still contains long-lived minor An and fission Ln.
The partitioning and transmutation strategy [4] is proposed as an effective way to lessen the residual heat load from the minor An prior to the storage of residual SNF in a geologic repository. [5]Some solvent extraction processes for minor An 3+ partitioning include a combination of steps where the first step involves the co-extraction of Am 3+ , Cm 3+ , and the fission Ln 3+ from the PUREX raffinate (e.g., DIAMEX [6] ) followed by a second extraction of Am 3+ and Cm 3+ from the fission Ln 3+ (SANEX [7] ).Other processes, such as ALSEP, [8] 1c-SANEX, [9] and i-SANEX [10] extract Am 3+ and Cm 3+ directly from the PUREX raffinate.Chemoselective complexation and extraction of An 3+ over Ln 3+ , followed by An 3+ decomplexation must be realized before transmutation in fast neutron reactors can be implemented due to the high neutron capture cross-sections of the Ln fission products.
The selective separation of An over Ln in liquid-liquid separations presents difficulties due to physical and chemical similarities between these two rows of the periodic table including cationic radii, hardness, and oxidation number. [11]n 1999 Kolarik [12] disclosed a novel class of soft N-donor complexants which followed the CHON principle, [13] bis-1,2,4-triazinyl-2,6-pyridines (BTPs), which could address the challenge of chemoselectively facilitating speciation of the minor An 3+ over Ln 3+ (Figure 1).The effectiveness of BTPs was postulated by the complexant interacting with the more diffuse 5f orbitals of the An to exploit increased metal-complexant covalency. [14]Subsequent research has focused on exploration of the core of the Lewis-basic core with scaffolds including 2,2'-bipyridine-2,6-bis-carbonitrile (BTBP), [15] or 1,10phenanthroline-2,6-biscarbonitrile (BTPhen) [16] as a means to overcome inherent challenges relevant to SNF reprocessing, including limited solubility in process-relevant, nonpolar diluents, slow complexation kinetics, lack of robust stability to acidic and radiolytic degradation, in addition to challenges with decomplexation of the An 3+ post separation.Functional group interconversion on the periphery of the complexant toward improvement of complexant physical properties related to performance has also been reported. [17]etra-alkyl BTPs Kolarik (1999 SF Am / Eu up to 143  Frequently, studies on liquid-liquid separation of 241 Am 3+ over 152/154 Eu 3+ as simulated SNF with BTP complexants are limited to a single organic diluent due to solubility limitations, or must subsequently incorporate additives for phase modification to overcome solubility issues while simultaneously preventing third-phase formation.In order to gain fundamental insight into solvent effects on BTP ligand-mediated An separations, more experimental data in diluents with diverse dielectric constants is required. During development of a convergent method [18] to afford functionalized BTPs, an observation was made that 3,3'-dimethoxy-BTP (MOB-BTP) possessed interesting solubility properties in trifluoromethylphenyl sulfone (FS-13), in contrast to 4,4'-dimethoxy-BTP which was completely insoluble in FS-13.Furthermore, MOB-BTP displayed tolerance of HNO 3(aq) to 1 M with good distribution, D Am , and separation factors (SF) at 25 mM over bis-2,6-(5,6,7,8-tetrahydro-5,9,9-trimethyl-5,8-methano-1,2,4-benzotriazin-3-yl) pyridine (CA-BTP), at 50 mM complexant concentration. [19]n an effort to retain the chemoselectivity of the tricoordinate donor core of BTPs while modulating solubility toward improved separations performance, a series of alkoxy-BTP derivatives were developed. [20]Preliminary evaluation of 3,3'-dibutyloxy-BTP substantiated superior solubility in FS-13 (2,000 mM), Exxal-8 [21] (150 mM), and 1-octanol (12.5 mM) over previous BTP complexants, including MOB-BTP.Liquid-liquid separations assays of simulated SNF demonstrated excellent D Am and SF for 241 Am 3+ over 154 Eu 3+ from a scalable and economically viable complexant in Exxal-8 without phase modifiers. [22]ompletion of the separation cycle was also achieved without competitive ligand exchange.
With preliminary solubility success in both nonpolar and polar diluents, the 3,3'-dialkoxy-BTP class of complexants has great potential to achieve a deeper understanding of the fundamental behavior of metal-complex formation in support of these aforementioned areas.The present research reports distribution values of 241 Am 3+ and 154 Eu 3+ as a function of time, [HNO 3(aq) ], and 3,3'dibutyloxy-BTP concentration in liquid-liquid extraction experiments of simulated SNF.The polar aprotic diluent, FS-13, was chosen to compare and contrast previous separations results in Exxal-8. [23]Separation, acid tolerance, complexant concentration, and decomplexation studies of 3,3-butyloxy-BTP are reported herein.

Organic phase
An appropriate mass of 3,3'-butyloxy-BTP was transferred from a sample vial to a 5.0 mL Class A volumetric flask to make a concentrated stock solution.Weighing by difference was used on the original sample vial to calculate the actual mass of 3,3'-butyloxy-BTP dissolved.The solvent, FS-13, was added to the flask and manually agitated to facilitate dissolution.Generally, the complexant did not dissolve immediately and thus, was left to stand for approximately one hour.If the complexant was not fully dissolved after one hour, the solution was sonicated for five minutes.Various concentrations of 3,3'-butyloxy-BTP were prepared from the concentrated stock solution in 1.0 or 2.0 mL Class A volumetric glassware.

Aqueous phase
Eight concentrated stock solutions of 241 Am 3+ were prepared in various concentrations of HNO 3 (.25 − 5.0 M).These 241 Am 3+ stock solutions were used throughout the entirety of this study to spike the working aqueous phase of the matching [HNO 3 ].Initial counts of the aqueous phase were generally 3400 cpm/200 µL.This method was repeated for 154 Eu 3+ .

Extraction
Equal volumes (.6 mL) of the organic and active aqueous phase, in that order, were added to a 1.5 mL micro centrifuge tube.The centrifuge tubes were capped and wrapped in parafilm then placed on a pulsing vortex shaker at 1700 rpm for the appropriate time.After the allotted time on the vortex shaker, the samples were centrifuged for five minutes to achieve phase separation.The organic and aqueous phases were transferred to separate 1.5 mL centrifuge tubes with a disposable fine-tip transfer pipette.A 200 µL aliquot of each phase was transferred to a counting tube with sampling performed in duplicate.

Gamma spectroscopy
A 2480 Wizard2 automatic gamma counter with 3 inch NaI(Tl) was used for the determination of 241 Am 3+ and 154 Eu 3+ activity of samples.Samples were counted for 10 minutes, or until total counts reached 10,000, and the instrument software relayed the results as counts per minute (CPM).The distribution ratios of a given metal ion (D M ) were calculated according to Eq. ( 1) where [M] is the net CPM of either the organic phase (numerator) or aqueous phase (denominator).Separation factors (SF) were calculated using Eq. ( 2) Values are reported in figures as an average of duplicate samples with error bars representing uncertainties calculated at 2σ from the counting error.The supporting information for this work provides additional details, including raw data for separations experiments, numerical values for D M and SF calculations, and counting statistics.

Decomplexation
An initial liquid-liquid extraction was performed following the method described above in the solvent extraction section with the exception of volumes being increased from 600 to 800 µL.A 700 µL aliquot of the 241 Am 3+ loaded organic phase was transferred to a new micro centrifuge tube and an equal volume of HNO 3 was added.The concentration of HNO 3 was subsequently varied.The phases were contacted on a vortex mixer (1700 rpm) for 60 minutes followed by five minutes of centrifugation.The phases were separated and analyzed using the method described in for solvent extraction.

Solvent extraction experiments
Steadied by the preliminary success of complexant solubility in FS-13, a series of experiments were performed to explore separation efficacy of 241 Am 3+ -vs.-154Eu 3+ from simulated SNF.In preparation for subsequent separations studies, an equilibration experiment was performed to evaluate the optimal length of time of phase mixing to achieve extraction of 241 Am 3+ from 154 Eu 3 + at a constant concentration of 3,3'-butyloxy-BTP at 25 mM (Figure 2a).System equilibration was reached in 60 minutes with D Am = 9.86 and D Eu = 8.89 × 10 −2 .A 17-fold increase in the 241 Am 3+ distribution ratio occurred between 5 and 60 minutes of contact time with 65% of D Am occurring in the first 20 minutes of contact time.Changes after 60 minutes were within experimental uncertainty and negligible.Maximum D Eu was observed at 60 minutes, but the difference occurring between 20 and 120 minutes were within experimental error suggesting no significant increase in D Eu occurred after 20 minutes of contact time.The results of the equilibration study in FS-13 demonstrated that preferential 241 Am 3+ extraction from 154 Eu 3+ occurs rapidly with over half of the initial CPM of the 241 Am 3+ isotope extracted within the first 20 minutes of contact time as is, without incorporation of phase modifiers.Clean phase disengagement was observed after centrifugation of the biphasic mixture.Upon completion of the equilibration study that defined complete separation at 60 minutes of contact time, a study which varied the concentration of HNO 3(aq) while holding [3,3'-butyloxy-BTP] constant at 25 mM was performed to evaluate the impact of acid concentration on the separation efficacy of 241 Am 3+ extraction from 154 Eu 3+ (Figure 2b).
occurred.D Am at .25 M−1 M HNO 3(aq) were observed between 4.4-11.5 corresponding to 81-90% extraction of 241 Am 3+ to the organic phase.D Am increased significantly from 2 M−5 M HNO 3(aq) , 71.0-227; however, the percent extraction of 241 Am 3+ varied only slightly from 98.8 to 101% in these cases as benchmarked by the initial activity (CPM) of 241 Am 3+ in the aqueous phase indicating near to full extraction.It is important to note that the data point at 4 M HNO 3 is not included in Figure 2b due to the remaining CPM of the aqueous phase being below the critical limit and therefore, must be reported as not detected (see Table S4 for numerical values and counting statistics).This supports the percent extraction values indicating full extraction of 241 Am 3+ from the aqueous phase to the organic phase.Regarding 154 Eu 3+ , an equivalent trend was observed with D Eu increasing from .039 to .105,corresponding to 3.76-8.43%extraction, from .25 to 1.0 M HNO 3 .D Eu increased more rapidly from 2 to 4 M HNO 3 , .506 to 1.99, respectively, corresponding to a 64.3% extraction at 4 M HNO 3 .
The extraction of 241 Am 3+ to the organic phase was effective with increasing acid concentration in direct opposition to traditional alkyl-BTPs. [23]With an equilibration time of 60 minutes and an acid concentration of 1 M in FS-13, additional experimentation focused on varying the concentration of 3,3'-butyloxy-BTP while holding other parameters constant (Figure 2c).
Analysis of the data for the complexant concentration variance study of 3,3'-butyloxy-BTP in FS-13 at a constant [HNO 3 ] of 1 M for 60 minutes revealed a maximum SF of 264 at 6.25 mM 3,3'-butyloxy-BTP and a maximum D Am of 45 at 75 mM 3,3'-butyloxy-BTP.The increased SF at lower concentrations of 3,3'-butyloxy-BTP (6.25 and 12.5 mM) can be explained by the extremely small D Eu values which are an order of magnitude lower than at higher concentrations of 3,3'-butyloxy-BTP (37.5-75 mM).Overall, extraction of 241 Am 3+ and 154 Eu 3+ increased with increasing concentration of 3,3'-butyloxy-BTP.Distribution values for 154 Eu 3+ remained less than 1 until the concentration of 3,3'-butyloxy -BTP reached 50 mM.Interestingly, full extraction of 241 Am 3+ was not reached at 1 M HNO 3(aq) , as was observed at 3-5 M HNO 3(aq) , even when the concentration of complexant was increased threefold.The data from Figure 2c was transformed into the log-log plot of D M vs. [3,3'-butyloxy-BTP] for the purposes of slope analysis toward informing the approximate ratio of 3,3'-butyloxy-BTP relative to metal (Figure 2d).
Slope analysis of the log-log plot (m = 1.13 ± .01 for 241 Am 3+ ) of D M vs. [3,3'-butyloxy-BTP] likely suggests a mixture of 1:1 and 1:2 metal:ligand complexes, 1:1 predominating, which are formed in the extraction Eq. (3).These results are in direct contrast to the similar experiment performed in the polar protic solvent Exxal-8 where the complexant to metal ratio was 2:1 [22] , suggesting solvent effects impact organization and stability of the hydrated and solvated metal:ligand complex.
Complexants with high chemoselectivity for the minor An over Ln routinely necessitate the use of competitive ligand exchange to deposit the extracted minor An into the aqueous phase to complete the separation cycle.Decomplexation conditions were explored to ascertain the feasibility of completing the separations cycle with 3,3'-butyloxy-BTP, resulting in the data prsented in Figure 3a,b above.
An initial extraction of 241 Am 3+ in 5 M HNO 3 was performed with 25 mM of 3,3'-butyloxy-BTP in FS-13, comparable to the extraction in Figure 2b at 5 M HNO 3 , with the exception of an increased initial 241 Am 3+ CPM to load the organic phase with complexed 241 Am 3+ .The data observed in Figure 3a correlate 241 Am 3+ CPM of the loaded organic phase after the initial extraction of 241 Am 3+ in 5 M HNO 3 (Figure 3a, red bars) and 241 Am 3+ CPM of the decomplexed organic phase after 60 minute contact time with an inactive aqueous phase of either .1,.01,or .001M HNO 3(aq) (Figure 3a, purple bars).The decomplexation results were similar across all HNO 3(aq) concentrations with 22.7% of complexed 241 Am 3+ decomplexed with .1 M HNO 3 compared to 24.2% decomplexed with both .01 and .001M HNO 3 , thereby demonstrating proof of principle regarding the potential for An decomplexation, most notably without competitive ligand exchange to facilitate placement of the An back into the aqueous phase.A second decomplexation study was performed to determine if additional 241 Am 3+ could be decomplexed with subsequent additions of fresh inactive HNO 3(aq) and if 3,3'-butyloxy-BTP in FS-13 could be recycled for an additional extraction (Figure 3b).With no significant preference between [HNO 3(aq) ] from Figure 2a, .01M HNO 3 was chosen for comparison with decomplexation and recyclization results in Exxal-8 under identical conditions. [22]The results revealed that a total of 71.4% of the complexed 241 Am 3+ was deomplexed from the initial organic phase after three successive washes with .01M HNO 3(aq) .The percentages of 241 Am 3+ decomplexed in each sequence are presented in Figure 3b (blue bars) and reveal that an additional 42.4 ± 10% of the remaining 241 Am 3+ CPM in the organic phase after the first sequence (S1) was decomplexed during the second sequence (S2), followed by an additional 32.1 ± 4% in the third sequence (S3).After the third sequence wash, a second extraction of 241 Am 3+ in 5 M HNO 3 was performed with the remaining organic phase.The initial and second extractions revealed comparable 241 Am 3+ extraction efficiencies, with 97.4% of the initial 241 Am 3+ (CPM) extracted during the first extraction compared to 98.4% in the second extraction.
Separation of 241 Am 3+ with 3,3'-butyloxy-BTP from a biphasic system via liquid-liquid extraction was productive and the separation cycle for this system could be completed via decomplexation experiments in FS-13 with HNO 3(aq) .Recycling of the original complexant for subsequent extractions was also possible underscoring further potential for this complexant for minor An separations.

Conclusion
In summary, a comprehensive study of 241 Am 3+ from 154 Eu 3+ using 3,3'dibutyloxy-BTP as a complexant for liquid-liquid An separation in the polar aprotic solvent FS-13 has been reported.This system functions well as higher [HNO 3(aq) ] without decreased performance, third-phase formation, poor phase disengagement, or complexant degradation.The aforementioned is in contrast to most contemporary BTPs and reinforces the utility of 3,3'-alkoxy-BTP complexants for minor An separations.Solvent effects play a key role in complexant:metal stoichiometry and the separations outcome of this system with a majority ratio of metal:ligand at 1:1 for FS-13 as the diluent, as compared to a 1:2 ratio in Exxal-8.Higher D Am and improved decomplexation were able to be achieved in FS-13 than utilizing Exxal-8 as the diluent.Decomplexation and recycling of the initial organic phase were possible.Future work will focus on designing a BTP complexant soluble in a hydrocarbon diluent, such as kerosene, and radiolytic stability studies, in addition to evaluating the utility of this system for extraction of other relevant byproducts of nuclear fission including 244 Cm 3+ and will be reported in due course.

Figure 1 .
Figure 1.Summary of BTPs studied in liquid-liquid separations of minor An from Ln.