‘Electroreduction’ is a relatively new technique to convert oxide fuel into a metal form, generating less radioactive waste than conventional reduction steps using a Ca reductant. In an electroreduction cell, schematically shown in Fig. 10.9, cathode baskets loaded with decladded oxide fuels and inert anode rods are immersed in a LiCl-Li2O electrolyte, and electrolysis is carried out at approximately 650 °C to remove oxygen from the oxide fuel at the cathode and to generate oxygen gas at the anode. The main reactions at both electrodes are described as:

Anode: yO2-in salt ^ (y/2)O2 + 2ye — 10.12

Cathode: MxOy + 2ye — ^ xM + yO2-in salt 10.13

where M denotes actinides such as uranium and plutonium. The oxygen is electrochemically ionized, and the actinide metal remains at the cathode. The ionized O2- is transported through the salt and is discharged at the anode to form O2 gas. When a graphite anode is used, CO2 or CO is evolved instead of O2. During the reaction, some LFP elements such as

10.9

Schematic figure of electroreduction process.

cesium, strontium, barium and europium are dissolved into the salt phase. As the redox potentials for the reduction reaction have not yet been assessed, the formation energies of the oxides provide useful indexes for determining the reaction. Table 10.3 gives the standard Gibbs free energy of formation of oxide obtained from the thermodynamic database MALT-II (Japan Calorimetry Society, 1992). As this table shows, the pos­sible reductions are of UO2 and Pu2O3 because they are less stable than Li2O. However, reductions of americium were observed in experiments on unirradiated mixed oxide fuel (Iizuka, 2006) and irradiated UO2 fuel (Herrmann, 2005), suggesting the decrease of the chemical activity of americium due to the formation of a plutonium-americium solid solution in addition to the stabilization of O2- in LiCl. The metal product obtained in the electroreduction step is then charged in the anode basket of the electrorefiner, and consequently treated in the same way as chopped spent metal fuels.

New tridentate diamide extractants — diglycolamides, whose amidic func­tions are linked by an oxypentyl bridge (Fig. 11.8) — were designed by Tachimori et al. at the end of the 1990s to compete with the bidentate malonamides studied in Europe (Tachimori et al., 2002, Sasaki and Tachimori, 2002, Sasaki et al., 2001, 2007a, b, Sugo et al., 2002). As the

AAA’jA’-tetraoctyl-S-oxapentanediamide (TODGA) presented a high extraction efficiency toward An(III), it was thoroughly investigated in the frame of the successive collaborative projects PARTNEW (Madic et al., 2004) and EUROPART, as a potential substitute for DMDOHEMA in the DIAMEX process. The solvent formulation (i. e., TODGA and TBP, respec­tively dissolved at 0.2 and 0.5 mol. L-1 in HTP) was optimized to compensate for the low loading capacity of TODGA (Modolo et al., 2007b). Several countercurrent spiked and hot tests were carried out in centrifugal contac­tors at the FZJ (Julich, Germany, Modolo et al., 2008) and the ITU (Karlsruhe, Germany, Magnusson et al., 2009a). They confirmed the poten­tiality of the TODGA solvent to reprocess genuine PUREX raffinates by enabling the recovery of more than 99.99% of the An(III) and the Ln(III), with very high feed decontamination factors (DF ~ 40 000 for Am and Cm), provided strontium extraction is hindered by oxalic acid scrubbings (Magnusson et al., 2009b). Unfortunately, these scrubbing streams increase the volume of liquid waste generated (Fig. 11.15).

Development of highly selective compounds 345

The prospect of biological treatment is relatively new. Biological treatment is based on the principle of emulating the natural occurring processes to treat waste. During three billion years of existence, microorganisms have evolved mechanisms to survive in hostile environments and to adapt to changes in the environment (Bush, 2003). Environmental engineers around the world have undertaken to find ways to tap into the mysteries of nature by diligently studying the action of microorganisms as they adapt to extreme conditions.

One of the most conserved mechanisms in the living cell is the biochemi­cal pathway for electron-transport through the cytoplasmic membrane to conserve energy through the oxidation of an electron donor and reduction of an electron acceptor such as oxygen. This process has been conserved over billions of years, such that, to this day, all life on earth depends on variants of this pathway (Bush, 2003; Thomas et al., 1985; Nealson, 1999; Kalckar, 1974). Most biochemical processes for degradation and/or detoxi­fication of compounds are linked to this process.

Lately, microorganisms have been isolated that are capable of reducing the toxic forms of heavy metal and transitional metal elements in transura­nic waste (TRU) to less mobile precipitable forms (Lloyd, 2003). Other researchers have found microbial cultures with the capability to resist high radiation doses (Battista, 1997; White et al., 1999). However, most of the resistant microbial species have not yet demonstrated the capability to degrade priority compounds such as polycyclic phenolics and polychlori­nated compounds. But recent research shows promise that this scenario is soon to change. For example, recent studies have shown that microorgan­isms may not only resist radiation, but may to a certain degree utilize the radiation for metabolic advantage. One example was illustrated in a recent study in which cultures of melanizing fungi from the cold regions utilized ionizing radiation to derive metabolic energy (Dadachova et al., 2007).

One aspect that has puzzled scientists in nuclear microbiology is the apparent capability of some species of bacteria to biologically separate dif­ferent isotopes of the same element (Whiticar, 1999). This process has been observed for the lighter elements such as H and D and for some heavy metals such as selenium (Se(IV)/Se(II), sulfur S(-II), and C-14 (Habitch and Canfield, 1997; Habitch et al., 1998; Herbel et al., 2000). This process, if engineered correctly, could be used to separate different isotopes of a com­pound from a mixture for further processing. For example, the capability to separate the radioactive carbon-14 isotope from carbon-12 in irradiated graphite could result in cleaner graphite suitable for recycling (Molokwane and Chirwa, 2009).

Cesium is the heaviest stable alkali metal, existing under most circumstances relevant to the nuclear fuel cycle as the weakly hydrated cation, Cs+. As the largest of the un-reactive alkali metal cations, it is weakly hydrated and resistant to the formation of coordination complexes. Size-recognition phenomena have proven the most viable means of manipulating the specia — tion of Cs+ for successful separations. Separation procedures have been developed around the use of crown ether or calixarene compounds, typically

large polyether species with molecular cavities designed to accept large cations of low charge. They can be isolated from solution using ion exchange materials, either organic or inorganic in nature. Cs+ should be relatively mobile in the environment, though this ion is readily taken up by clays and other layered minerals with appropriate interlayer spacing to accept a large cation. Like 90Sr/90Y, 137Cs and 137mBa (t1/2 = 2.55 min) rapidly establish secular radioequilibrium, hence 137Cs is always accompanied by 137mBa, which decays with a characteristic 661 kev gamma ray to produce stable 137Ba.

Ion exchange uses solid porous sorbents that have either anions or cations within the pores that are relatively loosely attached. The cations and anions can be exchanged for anions or cations in a solution that is passed through the ion exchange material. Contact between the solution and the ion exchange material is usually achieved within a column, with the solution flowing either upwards or downwards though the ion exchange bed. Ion exchange is typically employed for separating radionuclides at low concen­trations from aqueous effluent streams from radioactive processing facili­ties, though it has also been used occasionally to carry out bulk separations. Ion exchange materials are categorized into three main types:

• Inorganic. These are typically naturally occurring zeolite and other alu — mino-silicate minerals with a well-defined crystal lattice structure. They can be selected so that only contaminants with a certain molecular size, that fit within the lattice, are sorbed.

• Organic. These are carbon-based polymeric materials, often known as “resins”. They can be porous and sorb species themselves, or they can form a passive substrate that is coated with other materials which are chemically active.

• Specialist. These are inorganic or organic materials which contain “tai­lored” molecules, such as crown ethers. These molecules have a molecu­lar structure which contains rings enclosing a space that is right-sized for the atoms or molecules of specific contaminating species. Such mate­rials are thus highly specific for such species.

Each type of ion exchange material can also be anionic, capable of sorbing negative ions, or cationic, capable of sorbing positive ions from solution.

Ion exchange process operations fall into two categories.

The aqueous solution resulting from the fuel dissolution/clarification/chem- ical adjustment operation provides the liquid feed to the first extraction cycle in which codecontamination and U/Pu partitioning steps are lumped together. Regardless of the semantics, there is some value in describing the codecontamination and partitioning steps separately. The objective of the codecontamination cycle is to extract U and Pu together and provide the major decontamination from (or scrubbing of) the fission products. Since codecontamination includes several strongly coupled unit operations; the initial “extraction” ensures adequate extraction efficiency, while the subsequent “scrubbing” operations are aimed at removing those problem­atic fission products mentioned previously. To satisfactorily explain and appreciate the careful balances associated with the initial extraction/scrub portion of the flowsheet mandates a brief review of the conflicting chemis­tries prevalent in the process.

The raffinate from the UREX separations module becomes the feed to the CCD-PEG process segment, shown in Fig. 7.5 (Pereira, 2007a), or the alter­native FPEX process shown in Fig. 7.6. Because the supply of CCD-PEG solvent was limited, a wash was incorporated into the flowsheet to enable its recycle. In the extraction section, Cs and Sr (with Rb and Ba) are

extracted into the solvent. In the scrub section, a solution of nitric acid at moderate concentration, scrubs other species, primarily transuranic and lanthanide elements (TRU), from the solvent. In the strip section, the alkali and alkaline-earth cations are stripped by a combination of an organic ammonium carbonate salt and a complexing agent.

Because of uncertainty in the supply of both CCD and FS-13, an alterna­tive to the CCD-PEG process was developed at Oak Ridge and Idaho National Laboratories. The FPEX process, fig. 7.6 (Pereira, 2007b), was based on the CSSX process developed to treat the alkaline tank wastes at the Savannah River Site and at Oak Ridge National Lab. The FPEX solvent was designed to be stable at low acid concentrations. The solvent uses a solution containing a calixarene to extract Cs, a crown ether to extract Sr, a modifier, and trioctylamine in a kerosene diluent, Isopar L. The use of a kerosene diluent is significant as it is highly compatible with the hydrocar­bon diluents used in the other separations modules.

The flowsheets developed based on the UNEX process were proven in the course of a series of tests at RI and Idaho National Laboratory plants. At RI, the tests were performed using simulated HLW solutions with actual,

long-lived radionuclides, whilst in Idaho actual HLW from this nuclear center were used.

Three types of centrifugal contactors served as the extraction equipment:

• 12-staged miniature centrifugal contactor (MC 12-30) with a stage working volume of 30 cm3, manufactured by NIIAR (Dimitrovgrad);

• centrifugal contactors with rotors of 3.3 cm in diameter and with stirring and settling chamber capacities of 22 cm3 and 32 cm3, respec­tively; the contactors were designed and manufactured at NIKIMT (Moscow);

• centrifugal contactors with rotors of 2 cm in diameter, designed and manufactured at the Argonne National Laboratory (USA).

The typical composition of the HLW solutions used in the experiments performed is shown in Table 9.13.

Results of the UNEX process tests with selective recovery of cesium (flowsheet in Fig. 9.4)

In tests with selective recovery of cesium, the following indices for recovery of long-lived radionuclides (%) were obtained: Cs137 — 99.94, Sr90 — 99.81, Eu154 — 99.76, U238 — 99.98, Np237 — 99.95, Pu239’240 — 99.98, Am241 — 99.18.

The degree of concentration of Sr, An and REE in the strip product of these radionuclides was 2.5. During the experiments, no precipitates and suspensions in the process products were detected.

An important requirement of pyrochemical processing is being able to limit the interaction between pure metals and the processing equipment. Graphite crucibles with a zirconia coating have been utilized as heating crucible for the cathode processor. After each run, the coating has to be removed and a new coating applied. This action is labour-intensive and limits the through­put of the equipment. The coating also reacts with uranium to form dross, which is primarily uranium oxide. This uranium oxide must be reduced to a metal form and recycled. To minimize the formation of dross and to increase actinide metal throughput, alternative materials for the crucible of the cathode processor are under study. INL has developed a niobium cru­cible with a hafnium nitride coating and a zirconia-lined graphite crucible (Benedict, 2007). They have reported that very little dross was formed in the former and that five runs could be carried out before the coating had to be reapplied, while in the latter, 0 to 1% dross was formed, as shown in Fig. 10.26.

Researchers in the field of solid-phase extraction have sought to exploit the extraction and partitioning characteristics of the extractants often utilized in liquid-liquid extraction by fixing them onto various substrates as dis­cussed in the previous section. Although the field is not new to separation scientists, the use of solid-phase extraction for radiochemical separations is, relatively speaking, a more recent application of the technology. As such, there are very few examples of its use at technical or even pilot scales. The greatest application of solid-phase extraction pertaining to An and Ln par­titioning has been in the performance of very pure separations for analytical and radio-analytical chemistry. A very robust suite of these resins has been developed by E. P. Horwitz and co-workers and is sold commercially by EIchrom Technologies Inc., Darien, Illinois. These resins have come to be the primary separation tool in many radiochemistry laboratories over the last two decades and are also gaining substantial acceptance in the field of radio-pharmaceutical chemistry. Many technologists have recognized the potential benefits of employing the technology for treatment of primary and secondary streams in nuclear fuel reprocessing. However, the use of solid-phase extraction technology at the industrial scales necessary for nuclear fuel and/or waste processing involves a number of issues not encountered in very small-scale processes. Potential advantages and disad­vantages associated with solid-phase extraction technology for use at nuclear industrial scale are addressed in a later section; however, predomi­nate questions include performance, robustness, and economics of the tech­nology as compared to conventional (e. g., liquid-liquid extraction) or other mature processes. Several investigations have been performed with a small number of established solid-phase extraction sorbents using feed solutions from actual process streams in an effort to obtain data for comparison to conventional process methods. European scientists (Eschrich et al., 1980) evaluated a variant of the PUROCHROMEX (Plutonium Uranium Recovery On CHROMatographic Extraction columns) process designed primarily for the recovery of uranium and plutonium (Eschrich et al., 1968). The chromatographic column process consists of a silica gel column fol­lowed by three TBP solid-phase extraction columns in series. The solid — phase extraction columns used in their investigations were the commercial Levextrel-TBP resins (Lewatit CA 9221, OC1023 — Bayer AG) consisting of 60 wt.% TBP on inert polystyrene-divinylbenzene copolymer beads. A process flow diagram of the PUROCHROMEX process is shown in Fig. 13.6.

A feed solution of dissolved light water reactor (LWR) fuel, adjusted to 5-6 M HNO3, was introduced via the silica gel columns which retain highly

active undissolved fuel particles, in addition to 50-90% of the zirconium/ niobium and more than 60% of the antimony (Van Ooyen et al., 1964, Eschrich et al., 1970). This step was continued until uranium breakthrough was detected by on-line monitors. The train was then washed with 5.7 M HNO3 to remove unextracted fission and activation products from the system. The authors indicate that, in most cases, uranium is retained by the first TBP column and plutonium by the second, with the third column acting as a polisher. Uranium was eluted from the first column with a few bed volumes of distilled water. Plutonium was removed from the second TBP column via standard reductive stripping, i. e., by eluting with a reducing agent (e. g., dilute solutions of hydrazine or ferrous sulfamate) and convert­ing extractable Pu(IV) to the non-extractable Pu(III) species. Tests were performed with this process and various modifications using solutions of dissolved LWR fuel and different column sizes ranging from 10 to 1000 mL. The authors report single-cycle beta-gamma decontamination factors on the order of 107 for the column I uranium product and 104-105 for the plu­tonium product stripped from column II. Plutonium contamination of the uranium fraction was insignificant, but uranium carryover in the plutonium product did occur in most cases necessitating further purification of the plutonium fraction. However, some carryover of uranium in the plutonium fraction may actually be beneficial in light of contemporary concerns with

producing pure plutonium and the potential proliferation issues. A mixed plutonium-uranium product may also be compatible with a mixed oxide (MOx) fuel fabrication scheme if the proper ratio can be maintained.

The data from the small-scale PUROCHEOMEX tests are certainly promising although information regarding column performance vs. total throughput was not provided. Problems due to the radiolytic and hydrolytic degradation of the sorbent material were, however, noted and the authors indicate that the availability of a more resistant support would increase the attractiveness of this process.

Lumetta and co-researchers (Lumetta, et al., 1993) investigated the utility of commercially available TRU resin (ElChrom Industries Inc., Darien, Illinois) for use in separating americium and plutonium from actual neutral­ized cladding removal waste (NCRW) taken from a radioactive waste tank at the US Department of Energy’s (DOE) Hanford Site. The TRU chro­matographic resin consists of 13% CMPO and 27% TBP adsorbed onto Amberchrom™ CG-71 (Barney et al., 1992) and can be used for the separa­tion of transuranic elements. The NCRW sludge was washed and dissolved in HNO3-HF to provide a feed solution of composition shown in Table 13.1.

A small aliquot of feed solution (11 mL) was passed through a precon­ditioned column containing 500 mg of TRU resin. The bed was washed with 2 M HNO3 and then eluted with 5 mL of 0.01 M HNO3 followed by 5 mL of 0.01 M 1-hydroxyethane-1,1-diphosphonic acid (HEDPA). The func­tional ligands in TRU resin are the same extractants used in the TRUEX liquid extraction process, which consists of 0.2 M CMPO and 1.4 M TBP in

a normal paraffin hydrocarbon diluent (Horwitz et al., 1985). The authors noted that the separation properties of the TRU resin should be similar to those seen with the TRUEX process. The average decontamination factors for Am and Pu were however substantially lower than those achieved with the TRUEX process and dissolved NCRW feed solution (Swanson, 1991). The authors concluded that this was a result of a higher resin affinity for the U as well as some competition from the uncomplexed Zr present in the feed solution. The small mass of solid-phase extraction material used was not sufficient to accommodate loading by U and Zr and did not correlate with the amounts of CMPO available for TRU metal complexation in the TRUEX process. It was noted that the competition from Zr could be reduced by adding a complexing agent to the feed (e. g., fluoride ion). Uranium loading is problematic since increasing the amount of TRU resin to provide enough CMPO capacity to accommodate U loading would be quite expensive at large scale. However, the TRU resin may be a viable option for separating Pu and Am provided the U is first removed from the NCRW solution by some other means (Lumetta et al., 1993).