The emerging technology of nanopore membranes holds promise for the future as one of the processes that could be used for separation of radio­isotopes from water or gas streams. Membrane processes are normally easy to operate and offer the additional advantages of immediate capture of the pollutant for cleaning and recycling. An additional advantage is that mem­brane reactors tend to be portable and can be built to occupy the smallest land footprint possible.

Strontium is an alkaline earth element, existing either in the elemental state or as the divalent ion, the latter predominant in consideration of the nuclear fuel cycle. It forms insoluble phosphates and carbonates, is only weakly hydrolyzed, and in generally forms rather weak coordination complexes in aqueous and most organic media. Bonding in coordination compounds of Sr is mostly ionic in nature. Generally, strontium would be expected to follow barium in spent fuel processing and radioactive waste management. The daughter product of 90Sr decay (90Y, t1/2 = 2.67 d) quickly comes to radioactive equilibrium with the 90Sr parent and usually behaves chemically like a heavy lanthanide. Yttrium’s chemistry will be discussed below with that of the lanthanides

1.1.1 99Tc

Technetium is a second row transition metal, d5 in the elemental state, and a chemical analog of manganese. It exists in a multiplicity of oxidation states from +1 through +7 in solution. In the context of nuclear fuel cycle chem­istry, the most important oxidations states are the tetra and heptavalent species. The most important Tc(IV) species is TcO2(s). Heptavalent techne­tium is present as the pertechnetate anion TcO4- in solution, chemically analogous to perrhenate (ReO4-) or perchlorate (ClO4-). In the solid state it exists as Tc2O7. Oxides of technetium can be volatilized during used fuel processing or waste glass preparation. Pertechnetate will co-extract with U(VI) in PUREX chemistry, thus potentially “contaminating” the primary product of PUREX extraction. As an anion, TcO4- is a species likely to be quite mobile in the environment, in essence able to travel with the solvent front if leaked from an underground storage facility. In the presence of chelating agents, a variety of Tc complexes are formed, as has been reported in selected underground waste tanks at the Hanford Site. [2,3]

1.1.2 129I

As a halide, iodine also can be found in a variety of oxidation states from -1 through +7. The most important species in used fuel processing and waste management are I-, I2 and IO3-. Volatility of I2 is a concern in used fuel processing. In fact, volatilization and capture of the gas during fuel dissolution is a preferred method of isolating iodine in used fuel processing. Though the specific activity of 129I is low, iodine will bioaccumulate in the thyroid (which regulates growth processes in mammals) hence represents a substantial hazard. Fortunately, this bioaccumulation can be reversed by isotopic dilution with natural iodide. Bioaccumulation and the mobility of anions and volatile I2 are each important drivers for isolation of iodine. In the context of ultimate waste disposal, iodine is also of concern because it is not compatible with most of the solid materials being contemplated for high level waste isolation.

The “Spintek” filtration unit is an intensified form of crossflow filtation. Instead of a bundle of porous tubes, the Spintek unit has a series of circular, hollow porous plates made from sintered stainless steel, arranged in a stack on a central hollow post (Fig. 3.8).

The stack is contained within an outer housing into which the slurry to be concentrated is pumped. The stack of plates is spun to increase shear forces across the surface of the plates, and the supernatant from the slurry

passes into the center of the hollow plates and exits via the central post. The slurry is thus concentrated and can be recycled through the system until the desired degree of concentration is achieved. Because of the increased shear forces, typical fluxes achieved are some ten times higher than those in conventional crossflow filtration. The stack of plates can be removed for maintenance or replacement by withdrawing it from the top of the unit, and this arrangement is well suited to installation in a type 3 PSC. At present there has been no industrial plant application of this type of filter system, though it remains under active development (Herman, 2006).

Referring to Table 6.2, current commercial reprocessing of LWR reactor fuels occurs in four countries: France (La Hague), the UK (Thorp), Japan (Tokai-mura, Rokkasho-mura), and Russia (Mayak). Additionally, India (Kalpakkam, Tarapu; PHWR fuels) and the UK (Sellafield B205; Magnox fuels) operate plants to reprocess other types of used nuclear fuels. It is apparent that the lion’s share of international reprocessing capacity is asso­ciated with LWR fuels from thermal NPPs and, as such, will be the focus of the ensuing description of the typical PUREX flowsheet. One of the most exciting developments of current times is the fact that the Rokkasho plant in Japan is in the midst of hot startup at the time of this writing. This addi­tional capacity of 800 tHM/year marks the first new PUREX plant that has been brought on line in well over a decade. Additional planned capacity throughout the world speaks well for the renewed interest and commitment to nuclear power and the prominence of the nuclear fuel cycle to the global economy.

a tHM = metric tonnes heavy metal. b As of the end of 2006.

c An additional 100 tHM FBR and 150 tHM MOX fuels have also been reprocessed.

d An additional 18 tHM MOX fuel have also been reprocessed. e Includes the MOX and FBR quantities processed at La Hague & MOX at Tokai (see above).

Sources: OECD/NEA 2006 Nuclear Energy Data, Nuclear Eng. International Handbook 2007.

Spent Fuel Reprocessing Options, IAEA-TECDOC-1587, 2008.

Equally important is the fact that the “next generation” of used fuel reprocessing will likely be based on technologies that tweak standard PUREX process chemistry to achieve very similar results. Consequently, these advanced flowsheets will appear very similar to the traditional PUREX flowsheets in operation today. It is therefore of practical utility to review the PUREX process with greater scrutiny. The typical PUREX solvent extraction flowsheet used in the modern commercial plants of La Hague and Rokkasho-mura is set forth schematically in Figs 6.2 and 6.3. The primary difference between this flowsheet and that used at the other commercial plants indicated in Table 6.2 is inclusion of the Tc scrub (scrub 2) and complementary extraction, which is specific to the La Hague and Rokkasho-mura facilities (Baron 1993, 2003). The individual units opera­tions and process chemistries are reviewed with greater detail in the ensuing description.

6.2 First cycle (codecontamination and partitioning) of the PUREX flowsheet.

6.3 U and Pu purification cycles.

Two variations of the UREX process flowsheets were developed as part of the UREX+ demonstrations (Pereira et al., 2005). The original flowsheet was developed to allow recovery of technetium by solvent extraction as shown in Fig. 7.2. Solid lines are aqueous phase streams and the dashed lines are organic phase streams. In the first process segment, uranium and technetium are extracted from the bulk of the dissolved fuel. Co-extraction of Pu and Np is prevented by introduction of a complexant/reductant in the scrub section. Technetium is stripped into concentrated acid, and uranium is subsequently recovered via a low acid strip. Technetium recovery was lower than desired, and so in subsequent tests, the loaded solvent is stripped of U and Tc by dilute nitric acid, as shown in Fig. 7.3. Technetium is removed, as pertechnetate, from the U/Tc-strip product by an anion exchange process.

The co-extraction process designed has three parts and is shown in Fig. 7.4 (Pereira, 2005). In the first process segment plutonium, neptunium, uranium and technetium are extracted from the bulk of the dissolved fuel. Plutonium and neptunium are then stripped by a complexant/reductant in dilute acid in the Np/Pu-strip segment. The Np/Pu product stream is then scrubbed of uranium in the U/Tc-Re-extraction section. The combined solvent is scrubbed of excess nitric acid with a feed of dilute nitric acid

before entering the U/Tc-strip section, where a dilute nitric acid feed removes uranium and technetium from the solvent.

In connection with the treatment of salted HLW to produce a radionuclide concentrate to be solidified, the task was set to combine extraction and

stripping of Cs, Sr, An and REE. The developed flowsheet involved the following operations:

• combined extraction of Cs, Sr, An and REE by the UNEX-extractant (8 stages);

• scrubbing of the extract with a solution of 0.3 M citric acid in 0.05 M HNO3 (2 stages);

• combined stripping of Cs, Sr, An and REE by a solution containing 1 M guanidine carbonate and 10 g/l DTPA (8 stages);

• regeneration of the extractant by a 1 M HNO3 solution (6 stages).

This variant of the UNEX-process (see Fig. 9.6) was also tested on actual acidic HLW at RI and Idaho National Laboratory facilities.

Improved UNEX process flowsheet with combined stripping of Cs, Sr, An and REE

The objective of optimizing the UNEX-process flowsheet with a combined stripping of Cs, Sr, An and REE was to achieve the largest possible decrease in the number of operations, and to achieve a reduction in the volume and number of categories of secondary wastes. As a result of investigations conducted at pilot plants operated by RI and Idaho National Laboratory, the flowsheet presented in Fig. 9.7 was developed and tested.

The electroreduction cell is exposed to oxygen gas and LiCl-Li2O salt at 650 °C. According to the small-scale electroreduction tests, stainless steel is expected to be a durable material as both cathode and cell crucible. However, the severe corrosion of Fe-Cr-Ni alloys and Ni alloys such as Inconel-600 and Hastelloy C-276 was reported in material tests carried out by KAERI (Cho, 1999). This difference can be ascribed to the extreme experimental conditions of the latter experiments, such as the higher temperature (up to 1000 °C) and higher Li2O concentration (up to 25 wt%). When MgO is used for the cathode material, corrosion of the cell has to be considered because initial Li2O concentration must be high enough to compensate the con­sumption of Li2O during electrolysis. In cases when meshed or percolated material is used for the cathode, the Li2O concentration can be kept low at approximately 1 wt% during electrolysis, resulting in moderate corrosion of the cell material.

On the other hand, the anode material undergoes severe corrosion because the liberation of oxygen occurs at the interface between the anode surface and the LiCl-Li2O electrolyte. Although the integrity of the Pt anode is expected from many experiments, different materials have been tested to find cheaper alternatives. The requirements for the anode material are that it is nonconsumable and has a high current density. In addition to metals, different types of materials such as ceramics and crystals are cur­rently being tested; however, an alternative material to meet the require­ments has not yet been reported.

A number of extractants for partitioning An and/or Ln elements from dis­solved nuclear fuel solutions or accompanying waste streams have been developed or improved upon during the past twenty years. The extractants are generally classified on the basis of structure, extraction and stripping chemistry, and the type of metal complex formed (Sudderth et al., 1986). The four primary groups are: solvating or neutral extractants, chelating extractants, organic acid extractants and ion pairing extractants. They can be further divided into those that co-extract the An(III, IV, VI) and Ln ele­ments and those that are more selective for the trivalent An(III) and Ln(III) species. Examples of the former are the bi-dentate Octyl(phenyl)-N, N-diis — obutylcarbamoylmethylphosphine oxide (CMPO, Fig. 13.4) and N, N’- dimethyl-N, N’-dibutyl-2-2-tetradecylmalonamide (DMDBTDMA, Fig. 13.5) extractants, which are the basis of the well-known TRUEX (TRansUranium Extraction) and DIAMEX (DIAMide Extraction) pro­cesses, respectively.

The development of CMPO was the result of numerous investigations aiming to combine phosphine oxide or neutral phosphonate functional groups with carbamates (Horwitz et al., 1985, Schulz et al., 1988, Schulz et al., 1982, Gatrone et al., 1987a, Gatrone et al., 1987b). The CMPO ligand exhibits excellent extraction properties in nitric acid and diluents that are compatible with a conventional PUREX process. It has been the subject of much study over the last two decades and represents one of the most char­acterized reagents for total actinide separations from nuclear waste streams (Mathur et al., 2001). A primary disadvantage of the TRUEX process is that

13.3

Chemical structure of CMPO ligand.

13.4 Chemical structure of DMDBTDMA ligand.

the polar solvent modifier tributyl phosphate (TBP) must be added to reduce hydrolytic and radiolytic degradation of CMPO, improve acid dependences and to prevent third-phase formation. This results in a large phosphorus content which translates into a significant phosphorus residue requiring disposal if incineration is used for solvent destruction. It should be noted that, although the use of reagents that leave no solid residue when incinerated is a reasonable objective, the favorable aspects of organophos- phorus extractants may justify a phosphorus ash compatible waste form such as phosphate glass (Van Hecke et al., 2006). In an effort to eliminate the phosphorus ash, French scientists have championed the CHON prin­ciple for designing extractants, which means they consist only of C, H, O and N atoms. They may therefore be incinerated to gaseous products and avoid the contaminated solid residue that results from combustion of phos­phorus extractants. The malonamide-based extractants (e. g., DMDBTDMA) meet the CHON criteria and are known to be some of the best bidentate diamide ligands (Sasaki et al., 2002).

Variations of extractants with soft donor atoms (N, S) have been pro­posed for partitioning the group III An elements (Am, Cm) from Ln ele­ments. Examples are the dithiophosphinic acid extractants such as the commercially available Cyanex 301 (Sole et al., 1993, Modolo et al., 1998a, Modolo et al., 1999, Zhu et al., 1996b, Zhu et al., 1996a, Hill et al., 1998). The above mentioned extractants, and numerous others at various stages of maturity, have been documented quite well and the reader is referred to the literature for the applications, advantages and disadvantages of each. Recent reviews are given by Mathur et al., (2001) and by Van Hecke and Goethals (2006).

Microorganisms in nature and in reactor systems rarely grow as separate cells. The microorganisms form complex communities either in the form of agglomerations called flocs or as biofilm on the surfaces of inanimate objects and other organisms. The performance of a microbial culture is not only a function of its capability to degrade or transform a pollutant but also the configuration of the community in which it resides. There are complex interrelationships that occur within the microstructure that affects the avail­ability of substrates, symbiotic existence through toxicity shielding of more susceptible species, and transfer of metabolites to organisms that could otherwise not grow on the only primary substrate in the bulk liquid. The biofilm itself is often a complex structure constructed by the bacteria. The formation of the biofilm especially in the initial stages is believed to be an active process coupled to the cell’s central metabolism (Kjelleberg and Hermanson, 1984; Paul, 1984). Within the biofilm, complex processes take place such as nutrient cycling, mass transport resistance, cell and substrate diffusion, and biofilm loss at the surface that make prediction of the per­formance of the culture in the biofilm mathematically challenging.

A. PAULENOVA, Oregon State University, USA

Abstract: This chapter provides a short insight into multifaceted chemistry of actinides. The chapter first reviews the unique features of the f-block elements and compares the lanthanide and actinide transition series. The chapter then discusses the coordination chemistry of actinides with hard — and soft-donor ligands, hydrolysis, redox reactions, and radiation effects. Kinetic and thermodynamic aspects of solution chemistry of actinides and their effect on their separation behaviour are discussed.

Key words: reactivity, speciation, complexation, disproportionation, radiolysis.

2.1 Introduction

There are more than 400 operating nuclear reactors around the world. They produce about 15% of the world’s electricity, almost 24% of electricity in OECD countries, and 34% in the European Union (WNA, 2010). Nuclear power is the most environmentally benign way of producing electricity on a large scale. Without it most of the world would have to rely almost entirely on fossil fuels for a continuous, reliable supply of electricity. The use of electricity is continuously increasing; it is expected that global electricity demand will double from 2004 to 2030. It is also likely that the fraction of nuclear energy in the total energy produced will increase; therefore, it is essential that the expected need for increased supplies of nuclear fuel be addressed.

One of the unique characteristics of nuclear energy is that used fuel may be reprocessed to recover fissile and fertile materials to provide fresh fuel for existing and future nuclear power plants. Several European countries, Russia, and Japan have a policy of reprocessing used nuclear fuel, although government policies in many other countries have not yet addressed the various aspects of reprocessing (WNA, 2010). As early as the Manhattan Project (when the idea of reprocessing and closing the fuel cycle was first expressed), the principal motivation was the recovery of unfissioned uranium and plutonium from the used fuel elements. Such recycling results in some 25% more energy from the original amount of uranium in the

process, thus contributing to energy security. A second reason for reprocess­ing of used fuel is to reduce the volume of material to be disposed of as high-level waste to about one fifth of the volume of waste at discharge from the reactor. At the same time its radioactivity level is decreased such that after about 100 years the activity decreases much more rapidly than in used fuel itself (WNA, 2010). At the moment, only a few countries are reprocessing LWR used fuel (France, UK, Russia, and Japan) and other reactor fuel (India), with the largest reprocessing capacity existing in France at La Hague (WNA, 2010).

This multistep technology, spanning from mining, conversion, enrich­ment, fuel fabrication, burning during reactor operation, storage of irradi­ated fuel, reprocessing (separation, recycling), to conversion into a stable waste form and final disposal involves a highly sophisticated array of many chemical transformation processes. The chemistry of the nuclear fuel cycle is an example of interdisciplinary applied science; it demands a combination of inorganic, coordination, and environmental chemistry of actinides (and fission products). The chemistry of reprocessing is quite complex and includes many aspects of fundamental inorganic and physical chemistry: Complexation (coordination), redox reaction (thermodynamics and kinet­ics), acid-base equilibria, hydrolysis, and so on. The coordination chemistry of actinides with both hard — and soft-donor ligands plays a crucial role through the entire nuclear fuel cycle.

This chapter will provide a review of selected chemical and physico­chemical properties of the actinide elements, their typical compounds and their ions in aqueous solutions. The f-block elements have many unique features, and comparison of similar species of the lanthanide and actinide transition series provides valuable insights into the chemistry of both series.

Since Mendeleyev’s development of the modern concept of periodicity of elements, the table of elements has significantly changed. Numerous experiments confirmed the position of actinides in the periodic table as a 5f series of elements, first proposed by G. T. Seaborg (Seaborg et al., 1949, pp. 1492-1524, 1978). Chemical behavior and electronic structural evidence established the actinides (Ac-Lr; atomic numbers Z = 89-103) as an inner transition series with actinium as the first member, analogous to the lantha­nide transition series (La-Lu; Z = 57-71). Electronic configuration of ele­ments may be significantly different in the gaseous atoms, in ions in solutions or solids, and in the metallic state. Comparing the electronic structures of lanthanides and actinides (Table 2.1), it can be seen that in the Ln3+ ions, 14 4f electrons are added in the sequence beginning with cerium (Z = 58). In the An-series, the addition of 14 5f electrons in the sequence from thorium (Z = 90) to lawrencium (Z = 103) is not as regular as in 4f series. While protactinium metal displays 5f electron character, as is expected for

the third member of an actinide series (Zachariasen, 1973; Fournier, 1976; Haire et al., 2003), no compelling evidence exists to show that thorium metal, or thorium ions in solution or in any of its well-defined compounds, contain 5f electrons (Edelstein, 2006).

Both the similarities in and the differences between the actinide and lanthanide series have had great heuristic value in actinide element research (Edelstein, 2006). The metallic (zero valent) and 3+ ionic forms with half — filled f-electron shells, are of special interest because of the enhanced stabil­ity of this particular electron configuration. For example, curium (Z = 96) has seven 5f electrons in the elemental form and the 3+ oxidation state and magnetic, optical, and chemical properties that are remarkably similar to those of gadolinium (Z = 64) with seven 4f electrons.

The principal differences between the two transition series arise largely from the lower binding energies and less effective shielding by outer elec­trons of 5f as compared to 4f electrons. The 4f orbitals of lanthanides are deeply buried and completely screened by 5s and 5p electrons that causes 4f electrons to have limited importance in chemical bonding. The trivalent oxidation state is the most stable because it is formed by ionization of two 5s and one 5p electrons. On the other hand, the 5f orbitals have greater spatial extension, and penetrate the core. The energy differences between the 5f, 6d, 7s and 7p orbitals are relatively small; hence, multiple oxidation state are possible, and covalent bonding interaction with other atoms is possible.