Minor Actinides: Partitioning, Transmutation and Incineration

Abstract

Plutonium isotopes, but also the isotopes of minor actinides: mainly neptunium, americium and curium can be fissioned by neutrons in the core of nuclear reactors. They also can be transformed as non-fissile isotopes by neutron capture into fissile nuclides (transmutation). Incineration of 99% of the plutonium, neptunium, americium and curium would decrease the long term radiotoxicity of the high active waste (HLW) such that the radiotoxicity level of natural uranium would be underrun already after about $3\times 10^{4}$ years. This requires chemical separation of plutonium, neptunium, americium and curium and the fabrication of fuel elements with these actinides. These chemical separation methods and the fuel refabrication methods were already developed by research and development programs and demonstrated in pilot plants or at laboratory scale. The possible incineration rates for the different actinides in different reactor types (light water reactors, liquid metal cooled fast breeder reactors and accelerator driven systems) have been thoroughly investigated. Reactor strategies with light water reactors operating in symbiosis with liquid metal cooled fast breeders or accelerator driven systems are feasible. The different reactor and fuel cycle strategies have different radioactivity loads and different radiotoxicity levels within the different parts of their fuel cycle. Whereas the radiotoxicity can be drastically decreased in the back end of the fuel cycle, the masses of plutonium and minor actinides and their radioactivity and radiotoxicity can be higher during reprocessing and refabrication. Transmutation and destruction of long-lived fission products is only feasible with reasonable efficiency for Iodine-129 and Technetium-99.

9.1 Introduction

Safety and health impact analyses of the emplacement in deep geological repositories of HLW described in Sect. 7.6.3 have shown the C-14, Cl-36, and Tc-99 as well as I-129 nuclides to be first in reaching groundwater close to the surface because of their solubility in water. The time when these nuclides will turn up in lakes, rivers or drinking water wells is a function of the technical barriers enclosing the HLW container. All safety studies for deep geological repositories so far have indicated that point in time to be on the order of a few thousand years (Sect. 7.6.3). The radioactive burden possible at that time is below the strict regulations of USEPA and of European licensing authorities. After that period of time, Cs-135, Se-79, Nb-93m, and Zr-93 could reach the aquifers close to the surface in 10$^{4}$–10$^{6}$ years without, however, causing a radioactive burden higher than that of C-14, Cl-36, Tc-99 and I-129. Later, the neptunium, uranium, plutonium and other actinide nuclides would reach the biological environment. That point in time depends on whether these longlived nuclides will form so-called colloidal species after an intrusion of water, and on the efficiency with which colloidal species of the longlived actinides can be retained by an engineered barrier (near-field) during their transport through the cap rock of a repository [1–4].

Plutonium, but also neptunium, americium, and curium, can be split directly by neutrons in the core of a nuclear reactor or transformed into other fissile nuclides (transmutation). The use of plutonium as a fissile material has been described in Chap. 8 (plutonium recycling). Besides generating energy, plutonium recycling entails the advantage that plutonium will no longer exist in the HLW (except for minor losses in chemical reprocessing and in refabrication of the MOX fuel). This applies similarly to the minor actinides (neptunium, americium, curium) when recycled like plutonium.

For this reason, studies have been performed internationally for a number of years to find out what would be the advantages of separating the longlived actinides and of their transmutation and incineration in nuclear reactors. These would be the advantages gained:

• The risk of human intrusion into a repository and the misuse of plutonium and neptunium (nuclear proliferation problem) would be non-existent.

• A safeguards concept extending over very long periods of time to monitor the HLW, especially the spent fuel elements in direct disposal, would not be necessary.

• The radiotoxic inventory of longlived HLW would be reduced drastically.

• The thermal load acting on the repository structures due to decay of actinides, e.g. plutonium and americium, would be diminished.

• The actindes could be transmuted and split, especially in reactor systems with fast neutron spectra, which would allow the generation of additional energy.

Especially breeder reactors with a fast neutron spectrum could exploit the vast potential of U-238 and Th-232 by way of the breeding process and fission of plutonium and U-233, thus ensuring energy generation for the world over many thousands of years.

However, some drawbacks are referred to as well [5, 6]:

• Recycling plutonium and neptunium causes problems in IAEA safeguards and problems of potential proliferation of plutonium and neptunium (plutonium economy).

• Larger quantities of radioactivity need to be handled in the nuclear fuel cycle than is the case in direct disposal of spent fuel elements.

• In a deep geological repository, the risk of radionuclide release after water intrusion is dominated by the C-14, Cl-36, and fission products with high solubility like Tc-99 and I-129 released first. On the other hand, the release of actindes and their contribution to a potential radioactive burden on the environment is not going to manifest itself until after about 10$^{6}$ years, and must be considered a lower risk of additional radiation exposure.

• Chemical methods of separating the actinides (plutonium, neptunium, americium, curium) need to be developed on a technical scale.

• Methods of fabricating new fuels must be developed which contain, in particular, plutonium, neptunium, americium, curium besides uranium and thorium.

• This adds to the cost of the fuel cycle and in some cases to the radiation exposure of workers.

9.2 Worldwide Inventories in Spent Fuel Elements of Uranium, Plutonium, Neptunium, Americium

The civil use of nuclear power has given rise to approximately 340,000 t of spent fuel elements with 2,300 t of plutonium by 2010. Roughly one third of these spent fuel elements were chemically reprocessed with the reprocessing capacity available worldwide (Fig. 9.1). By 2010, approximately half of this reactor-grade plutonium was processed into plutonium/uranium MOX fuel elements and is being recycled in nuclear reactors.

Figure 9.2 shows an IAEA estimate for the nuclear power installed world wide and of the quantities of reactor-grade plutonium the spent fuel elements contain. Moreover, the quantities of separated plutonium and MOX fuel elements are shown [7].

Fig. 9.1

Spent fuel discharged and spent fuel already reprocessed from nuclear reactors operating world wide [7]

Fig. 9.2

Nuclear power installed as well as plutonium stored in spent fuel elements, plutonium separated and plutonium in MOX fuel until 2010. Projections are also made until 2030 [7]

Figures 9.3 and 9.4 show IAEA estimates of the quantities of neptunium and americium generated in spent fuel elements and to be generated by 2030, respectively [7]. With one third of the fuel elements already reprocessed chemically, these quantities of neptunium and americium are to be found in liquefied or vitrified HLW. The quantities of curium generated or to be generated in the future in the case of spent LWR fuel elements are approximately a factor of 11 lower than those of americium.

Fig. 9.3

World wide neptunium stored in spent fuel elements or high active waste until 2030 [7]

Fig. 9.4

World wide americium stored in spent fuel elements or in high active waste until 2030 [7]

9.3 Radiotoxicity of HLW

Radiotoxicity is a measure of the health hazard posed by a radionuclide. It depends on the type and energy of the radiation emitted by the radionuclide and, in addition, on the resorptivity in the organism and the residence time of the radio-nuclide in the body. The radiotoxicity of a radionuclide is expressed by the effective dose in Sievert per Becquerel (Sv/Bq). For airborne radionuclides arising, for instance, in reactor accidents, the effective dose for incorporation and external exposure is important. For HLW in the repository and associated safety analyses, however, only the effective dose for ingestion is important. Table 9.1 shows the effective dose values for ingestion for some selected fission products and actinides [8].

Table 9.1 Effective dose values (Sv/Bq) for ingestion for selected fission products and actinides

Table 9.1 can be used to determine, from the known radionuclide quantities and their half-lives, the radiotoxicity of radionuclides in 1 tonne of spent LWR fuel as a function of time.

This is shown for the radiotoxicity for ingestion in Fig. 9.5 for 1 tonne of spent fuel of a PWR fuel element with a burnup of 40,000 MWd/t. The fresh fuel had an enrichment level of about 4% U-235. The contributions to radiotoxicity stemming from the fission products, neptunium, plutonium, americium, and curium can be seen. Moreover, the radiotoxicity of the radioisotopes of the decay chains of plutonium, americium, and curium is shown. The radiotoxicity of 1 tonne of natural uranium is represented by a line parallel to the abscissa at $2 \times 10^{4}$ Sv/t$_\mathrm{ HM}$.

Fig. 9.5

Radiotoxicity for ingestion of 1 tonne of PWR spent fuel with a burnup of 40,000 MWd/t (enrichment of fresh fuel 4% U-235) [8]

The radiotoxicity due to direct disposal of spent fuel elements is represented by the total sum of all contributions. Curium falls below the radiotoxicty of natural uranium after only approximately 200 years, while the fission products underrun this line of natural uranium not before 600–700 years. The americium isotopes reach this state at approximately $3 \times 10^{4}$ years, plutonium isotopes, afer $3 \times 10^{5}$ years [8].

9.4 Various Strategies of Partitioning and Transmutation with Incineration of Actinides

The influences of various possible strategies including partitioning of plutonium (plutonium recycling) and additional partitioning of americium and curium are shown in Fig. 9.6 [8]. These are the scenarios compared:

• Direct disposal of spent fuel elements. This strategy results in the highest radiotoxicity over very long periods of time.

• Partitioning and incineration of plutonium/uranium. This strategy of Pu/U recycling, with an incineration efficiency of 99 or 99.9%, leads to the radiotoxicity level of natural uranium being reached after approximately $6 \times 10^{4}$ years.

• Partitioning and incineration of U, Pu, Am, Cm with an efficiency of 99% results in underrunning the radiotoxicity level of natural uranium after some $3 \times 10^{4}$ years. Only partitioning and incineration of U, Pu, Am, Cm with an efficiency of 99.9% would cause the radiotoxicity level of natural uranium to be underrun after some 800 years.

• Incineration of curium could also be replaced by interim storage of this minor actinide for a period of about 200 years awaiting the $\upalpha$-decay of curium isotopes into plutonium isotopes and subsequent incineration in nuclear reactors (Sects. 9.7.6 and 9.8.1). However, this is an objective very difficult to achieve on a technical scale.

Fig. 9.6

Radiotoxicity for ingestion of 1 tonne of PWR spent fuel with a burnup of 40,000 MWd/t for different partitioning, transmutation and incineration strategies [8]

9.5 Chemical Separation of Actinides

9.5.1 Joint Chemical Separation of Plutonium and Neptunium from Spent Fuel

The chemical separation of uranium and plutonium from the fission products and minor actinides (reprocessing, PUREX process) was described in Sect. 7.2.1 above.

The chemical separation of uranium and plutonium is achieved in present-day large reprocessing plants, such as LaHague, France, or Sellafield, United Kingdom, by means of tributyl-phosphate (TBP) with an efficiency of approximately 99.8–99.9% [8–11].

In the PUREX process, neptunium together with uranium is put through the first separation stages and separated from uranium only in the uranium purificatioin step. The PUREX process thus can be modified in such a way that neptunium is separated together with plutonium. For this purpose, neptunium must be oxidized chemically from the pentavalent to the hexavalent state (Table 9.2) and can then be co-extracted with uranium (VI) and plutonium (IV) in the first separation cycle by means of tributylphosphate. This is achieved with nitric acid and by adding special agents, such as vanadium (V) compounds [8]. Purification of neptunium after co-extraction with uranium and plutonium can be achieved in the second uranium cycle [8, 9, 12].

This modified PUREX process was demonstrated by the Japan Nuclear Cycle Development Institute (JNC) and the reprocessing plant LaHague, France, with a separation efficiency of 99% [9, 12].

However, it is also possible to separate neptunium subsequently from the liquid HLW of present reprocessing plants [8].

Table 9.2 Valence or oxidation state of different elements [9]

9.5.2 Separation of Americium and Curium Together with the Lanthanides

Americium and curium as well as the lanthanides (Lns) are found in the trivalent state in the liquid HLW (Table 9.2) [9]. Therefore, they must be separated together. This can be achieved by the following chemical processes:

9.5.2.1 DIDPA Process

The DIDPA process was developed by the Japan Atomic Energy Research Institute (JAERI) [8, 13]. It is based on the organophosphorus agent, diisodecyl phosphoric acid (DIDPA). For this purpose, the liquid HLW must be reduced in nitric acid concentration from 2–3 to 0.5 mol/l. This can be done, e.g., by denitrating the HLW with formic acid. In this process, nitric acid is decomposed into gaseous products. A mix of DIDPA and TBP is then used to extract the actinides, including neptunium, and the lanthanides, and separate them from the remaining fission products. Americium, curium, and the lanthanides are then re-extracted together with 4 mol/l nitric acid. In this step, they are separated from the organic phase containing neptunium and residues of plutonium and uranium.

The nitric acid concentration is then reduced again to 0.5 mol/l [13, 14] and americium and curium are separated as described under Sect. 9.5.4.

9.5.2.2 The TRUEX Process

The TRUEX process was developed at the Argonne National Laboratory (ANL) in the United States [15]. It is based on CMPO, an organophosphorus extraction solvent. No pretreatment of the HLW is necessary, and the actinides and lanthanides can be extracted from HLW acidified with 0.7–5 mol/l nitric acid. However, as the efficiency of actinide and lanthanide separation is not sufficient, the Japan Nuclear Cycle Development Institute proposed a combination of the TRUEX and DTPA processes.

9.5.2.3 TRPO Process

The TRPO process for actinide and lanthanide separation was developed at the Tsinghua University, Beijing [16]. TRPO (trialkylphosphinoxide) dissolved in kerosene is used. Separation factors of 99.9–99.99% (laboratory scale) were achieved at a nitric acid concentration of 1 mol/l. However, as in the TRUEX and DIPA processes, an additional treatment step is required. As re-extraction from TRPO is carried out at a higher nitric acid concentration, another neutralization step is necessary for the separation of actinides and lanthanides.

9.5.2.4 DIAMEX Process

The DIAMEX process was developed by the French Commissariat à l’Energie Atomique (CEA) [17]. This chemical separation process employs dimethyl-dibutyl-tetradecyl malonamide (DMDBTDMA) dissolved in aliphatic hydrocarbon. The trivalent minor actinides and lanthanides are separated at a nitric acid concentration of 3–5 mol/l, while re-extraction is conducted at 0.1 mol/l. The DIAMEX process was successfully tested by the CEA in the ATALANTE facility at Marcoule on HLW with separation factors of 99.9% [12, 18].

9.5.3 Chemical Separation of Actinides from the Lanthanides

The chemical separation processes referred to above allow the trivalent lanthanides and americium and curium, respectively, to be separated only together as trivalent species. Special processes had to be developed to separate the lanthanides from americium and curium.

9.5.3.1 DIDPA–DTPA Process

In the second part of the DIDPA process selective stripping with diethylenetriaminepentaacetic acid (DTPA) separates americium together with curium from the lanthanides (fission products).

The DIDPA process was tested with HLW. All actinides (neptunium, plutonium, americium, curium) were recovered 99.9% [13, 14].

9.5.3.2 Chemical Separation with Dithiophosphinic Acids

In connection with the TRPO process of the Tsinghua University of Beijing, the CYANEX 301 solvent was used together with TBP. This achieved separation factors in excess of 99.9% for separation of the lanthanides from americium and curium. The process was developed further by the Jülich Research Center in Germany by modifying the CYANEX 301 solvent [19–21].

9.5.3.3 SANEX process

In connection with the DIAMEX process, extraction solvents based on bistriacylpiridine (BTP) were developed at the Karlsruhe Research Center and successfully tested by the Institute of Transuranium Elements in Karlsruhe, Germany, and the CEA in the ATALANTE facility in Marcoule. They resulted in the development of the SANEX separation process (Fig. 9.7) by the CEA, by means of which americium and curium, on the one hand, can now be separated from the lanthanides, on the other hand, with 99.9% efficiency [12, 22, 23].

Fig. 9.7

Chemical separation processes developed in Europe for partitioning U, Pu, Np, Am and Cm

9.5.4 Chemical Separation of Americium from Curium

Studies of the separation of americium from curium seem to be very successful. The TRPO process of the Tsinghua University, Beijing, is able to separate americium from curium with a separation factor of 99.9% [19].

With the TODGA and CyMe4BTBP agents separation of americium was achieved with an efficiency of 99% [24–26].

European research institutes developed the simplified GANEX process and EXAm or the LUCA processes to achieve this separation of americium from curium with high efficiency [12, 26–32]. Figure 9.7 shows the different processes developed in Europe. The efficient separation of americium and curium is absolutely necessary for a number of actinide fuel fabrication processes (Sect. 9.7).

9.6 Pyrochemical Methods of Separating Minor Actinides

In addition to the aqueous chemical separation methods described above, pyrochemical techniques have been under development for decades by which uranium, plutonium, and the minor actinides can be separated by electrolytic fractionation and reductive extraction, respectively.

9.6.1 The Integral Fast Reactor Pyroprocessing Process

This pyrometallurgical and electrochemical process is developed at Argonne National Laboratory (ANL) in USA in combination with the Integral Fast Reactor (IFR) programme [33]. It is an evolution of the pyroprocessing methods which had been utilized in the 1960s for the metallic fuel of the experimental breeder programme EBR-II [34]. These pyroprocessing technologies were improved by the development of electrorefining methods [35] for the separation of actinides from the fission products. The current improvements of the ANL pyropressing methods aim at separation efficiencies of $>$99.9%. However, the plutonium and minor actinides cannot be partitioned, since the plutonium and all other minor actinides (neptunium, americium, curium) in combination with about 30% uranium always remain together. From the nonproliferation point of view this is considered to be an advantage [33, 36].

The pyroprocessing method is a batch mode process, whereas the aqueous partitioning processes (Sect. 9.5) operate in a continuous mode.

After dismantling of the fuel assemblies and chopping of the fuel rods including claddings, the short fuel rod segments are loaded into perforated steel baskets and placed in the electrorefiner vessel (Fig. 9.8). The electrorefiner vessel is covered at the bottom by a thick layer of liquid cadmium (melting point 321$^\circ$C). This cadmium layer is again covered by a thick layer of an eutectic mixture of lithium chloride, LiCl, and potassium chloride, KCl, (melting point 350$^\circ$C) acting as electrolyte salt. The electrorefiner is operated at a temperature of about 500$^\circ$C.

The perforated steel baskets with fuel segments are lowered into the electrolyte salt and act as the anode. The actinides from the spent fuel are transported from the anode basket to two kinds of cathodes [a solid cathode and a liquid cadmium cathode (Fig. 9.8)] by means of an applied electrical current. Pure uranium is collected at the solid cathode and a mixture of uranium, plutonium, americium and curium is collected at the liquid cadmium cathode suspended in the electrolyte salt. The fission products remain in the electrolyte salt and collect in the liquid cadmium layer at the bottom.

After the desired amount of actinide materials has been collected, these deposites at the cathode are recovered in a high temperature vacuum furnace (cathode processor) by melting. Any volatile products are removed by vaporization. These include electrolyte salts and cadmium. They are collected in a condenser and recycled.

The metal ingots resulting from the cathode processing operation are rather free of impurities (traces of solid fission products remain) and sent to the injection casting station for fabrication of new metallic fuel rods (Sect. 9.7.5) [36, 37].

Fig. 9.8

IFR pyroprocessing scheme [36]

Due to the higher radiation resistance of the electrolysis in molten salts and remotely controlled technology, pyroprocessing of short-cooled spent fuel is possible. Cooling times as short as several months of the spent fuel seem possible compared to the present two years for the LMFR fuel cycle or seven years needed for aqueous reprocessing of LWR spent fuel.

9.6.2 Electro-Reduction and Refining of Spent UOX and MOX Fuel to Metallic Fuel

In addition to the pyroprocessing of spent metallic fuel also methods for electrolytic reduction and electrorefining of spent UOX and MOX fuel to metallic fuel were developed. Efficiencies for the electrolytic reduction of spent oxide fuel to metallic fuel of 99.7% were demonstrated [36, 37].

Similar research and demonstration experiments as at ANL were reported by CRIEPI (Japan) for electroreduction and refining as well as pyroprocessing of metallic plutonium/uranium fuel [38–40]. The pyroprocess can start with either spent LWR UOX or MOX fuel (lithium reduction) or with liquid HLW from spent LWR fuel reprocessing applying an intermediate step with denitration and chlorination. For the liquid cathode during electrorefining both a liquid cadmium and a liquid bismuth cathode are applied [38].

9.6.3 Actinide/Lanthanide Separation Using Aluminum

Conocar et al. [41] demonstrated that a one stage reduction process using molten fluoride salt (AlF$_{3}$-LiF) and an aluminum solid anode resulted in a separation of plutonium and americium of 99.3% from the lanthanides.

9.6.4 Pyro-Processing of Fast Reactors PuO$_{2}$/UO$_{2}$ Fuel in Russia

A similar pyroprocess approach is developed for fast reactors in Russia. Here, the PuO$_{2}$/UO$_{2}$ is the product instead of U-Pu-Zr Pu-metal. In the Russian DOVITA-process [42, 43] the mixed oxide fuel is converted into chlorides and separated by electrolysis in a melt of NaCl–KCl at 650$^\circ$C. The transuranium elements are precipitated sequentially as oxo-chlorides or oxides out of the NaCl–KCl melt by gassing with Cl$_{2}$/O$_{2}$ and adding Na$_{2}$CO$_{3}$.

9.7 Fuel Fabrication for Transmutation and Incineration of Actinides in Nuclear Reactors

The fabrication procedures of fuels for actinide transmutation and incineration are presently based on existing fabrication technologies for mixed oxide, e.g. (U/Pu)O$_{2}$ fuel or mixed nitride (U/Pu)N fuel. In principle the minor actinides can be mixed into the MOX fuel. However, due to the increased gamma and neutron radiation caused by americium or curium the fuel fabrication facilities will need heavy shielding.

Therefore, dust free aqueous processes like SOL-GEL techniques or other liquid–solid conversion processes for the fabrication of granulates or microspheres are applied. In addition infiltration methods are under development. In this case americium or curium nitrates are infiltrated in a porous medium, e.g. magnesium aluminate spinel pellets etc. [44, 45].

Metallic fuel has a better thermal conductivity than oxide fuel. Metallic U-Pu-Zr fuel with sodium bonding between the cladding and fuel is being developed by Argonne National Laboratory in the USA as part of the IFR program (Sect. 9.6.1).

Uranium free fuels (inert matrix fuel) on the basis of oxides like ZrO$_{2}$, Y$_{2}$O$_{3}$, MgO etc. are also developed in Europe and Russia [44, 46].

The ALFA fuel manufacturing facility for actinide bearing fuel elements is under construction at Marcoule, France [47].

9.7.1 Pellet Fabrication with SOL-GEL Microspherical Particles

In the SOL-GEL droplet-to-particle conversion process droplets are generated by passing a feed solution of nitrides of actinides over the edge of a cylindrical cone rotating at high speed. The droplets are collected in an ammonia bath where gelation occurs. After washing, drying and calcination, spherical particles of different size (20–150 $\upmu$m) are obtained. The spherical particles can be pressed and sintered to pellets [48, 49]. As an example the following actinide fuels can be produced with different enrichments

$$\begin{aligned} (\text{ U,Np})\text{ O}_{2}\;\text{ or}\;(\text{ U,Pu,Am})\text{ O}_{2}\;\text{ or}\;(\text{ U,Pu,Np})\text{ O}_{2} \end{aligned}$$

9.7.2 Fuel Fabrication by Vibrocompaction

Microspheres produced by the SOL-GEL process or small granulates produced by crashing can be rinsed into tubes of fuel rod claddings and be compacted by vibration. Smear densities of 85% of the theoretical density can be achieved by carefully selecting the grain sizes. Table 9.3 shows three particle sizes and their volume distribution applied to achieve a smear density of 85% theoretical density [45, 48–53].

Table 9.3 Particle size distribution for a smear density of 85% theoretical density

The remaining open porosity provides space for helium from alpha-particle decay and for gaseous fission products. This process was originally applied and is still used in Russia (VIPAC process) [49, 50] for MOX fuel fabrication of LMFBRs.

More recent developments in Europe use SOL-GEL microspheres (SPHE-REPAC process [48, 51, 52].

The granulates or microspheres must be produced from powders or actinide solutions having already the enrichment used in the fresh fuel.

9.7.3 Inert-Matrix Fuel

Inert matrix fuel is free of uranium or thorium. The actinides are distributed as a separate phase in a so-called inert matrix. Oxides, nitrides as well as metals can be considered as the inert matrix [44, 53, 54].

Oxides like ZrO$_{2}$, Y$_{2}$O$_{3}$, MgO, MgAl$_{2}$O$_{4}$ or Y$_{3}$Al$_{5}$O$_{12}$ were proposed as inert matrix [44, 53]. Different fabrication processes are considered:

• coprecipitation is based on the dissolution of the starting materials in nitric acid and the precipitation of all these materials. The resulting powder, after washing, drying and calcination is directly used for the pellet production.

• the mixing of particles and powder is based on the fabrication of microspheres or particles containing the actinides by the SOL-GEL technique followed by mixing these particles with the powder of the inert matrix.

Swelling of inert matrix materials during irradiation and induced $\upalpha$-particle production resulting from radioactive decay of, e.g. curium, may need particular attention [55, 56].

9.7.4 Infiltration Method

This process requires a porous medium in which the actinides can be infiltrated. The porous medium can be, e.g. a magnesium aluminate pellet formed by powder and subsequent calcination. Instead of porous pellets also porous beads produced by the SOL-GEL method can be used. In this case a higher loading of the beads with actinides can be achieved [44, 45].

9.7.5 Metallic Fuel

The most common fabrication process for metallic fuel is so-called remotely controlled injection casting of the U-Pu-An-Zr alloy [36]. In case of pyro-chemical reprocessing (Sect. 9.6.1) the product from the liquid cathode is already an alloy of uranium, plutonium and minor actinides. Uranium from the solid cathode is added to achieve the required fuel composition. This fuel batch is induction heated under vacuum and homogenized. Then the system is pressurized and the fuel alloy is injected into closed end molds which are rapidly cooled. The molds are removed, the fuel slugs cut to length and inserted into claddings with a small amount of sodium for bonding between the fuel and cladding [35, 36, 57, 58].

9.7.6 Intermediate Storage of Curium

Curium is a mixture of the isotopes Cm-242 (half-life 163 days), Curium-243 (half-life 29 years), Curium-244 (half-life 18 years) and Curium-245 (half-life 8,500years) and minor amounts of higher Cm isotopes. Curium 242 decays already almost completely to Pu-238 during reactor operation and subsequent cooling [59, 60].

Because of the high radiation and thermal loads during fabrication of curium containing fuel elements, it was proposed that curium should be stored until curium-243 and curium-244 have decayed to Pu-239 and Pu-240, respectively.

Curium solutions can be infiltrated into porous beads. These are calcined and sintered and then poured into vessels especially designed for interim storage over about 100–200 years. After this time period the fuel can be reprocessed and the separated Pu-239 and Pu-240 can be incinerated in FRs (see Sect. 9.4).

Curium transmutation in actinide fuel and subsequent irradiation in nuclear reactor cores would lead to extreme difficulties if aqueous reprocessing and subsequent refabrication were applied. The high neutron radiation and high alpha-particle heat production of Cm-244 are mainly responsible for these extreme difficulties.

This is different for pyrochemistry where subsequent metallic fuel refabrication to metallic fuel element is possible as described above. This process is performed entirely under remote handling (Sect. 9.7.5).

9.7.7 Irradiation Experience with Fuel Containing High Plutonium Contents or Neptunium and Americium

Experience is already available with fuel based on high plutonium contents for incineration of plutonium in FR burners [61]. Also, irradiation experience with neptunium and americium was obtained from experiments in JOYO and Phenix [62–65, 66, 67]. Whereas neptunium containing fuel behaves very similar to plutonium/ uranium mixed oxide fuel, americium needs more care for its fuel design. Neutron capture in americium leads to the build up of curium isotopes. The helium production from the decay of Curium-242 to Pu-238 and Cm-244 to Pu-240 needs special design provisions in order to avoid too high gas pressures in the fuel rod. In addition to the large experience with the irradiation of metallic fuel in EBR-II and FFTF in USA, experience was obtained with U/Pu/MA fuel in the French fast reactor Phenix [64].

Inert matrix fuel containing americium was tested in irradiation experiments in the Phenix reactor [64] and to very high burnup of 19 at.% in the Russian BOR 60 [46].

9.8 Incineration of Minor Actinides in Nuclear Reactors

9.8.1 Introduction

An LWR core containing fresh UOX fuel with an enrichment of 4.3% U-235 generates the following amounts of plutonium, neptunium, americium and curium (Table 9.4) after a fuel burnup of 51 GWd/t and 10 years cooling time of the spent fuel [68]:

Table 9.4 Generation of plutonium, neptunium, americium and curium of an LWR having a fuel burnup of 51 GWd/t

For the incineration of plutonium, neptunium, americium or curium, fuel elements containing minor actinides can be loaded into the cores of PWRs, FRs, ADSs or other nuclear reactors. Most of such investigations have been performed for PWRs, FRs and ADSs so far [69, 70].

Neutron absorption reactions lead to transmutation or fission of minor actinides. Fuel assemblies containing minor actinides can be distributed homogeneously over the core or be arranged heterogeneously at the periphery of the core. The presence of minor actinides influences the initial fissile enrichment, the safety parameters, e.g. Doppler coefficient and void coefficient and changes the decay heat and radiation characteristics of spent fuel elements [69, 70].

The amount of minor actinides to be loaded into the core of a nuclear reactor requires a detailed analysis of safety parameters. This leads to recommendations for upper limits. For homogeneous loading of minor actinides into the core of PWRs an upper limit of 1% for each of the minor actinides neptunium and americium was proposed. For large cores of LMFRs an upper limit of 2.5% of neptunium or americium was recommended [69].

The high radiation caused by curium makes recycling of this minor actinide very difficult except for the case of pyrochemistry combined with metallic fuel fabrication of the IFR (Sect. 9.7.5).

9.8.1.1 Curium Recycling in PWRs

In addition to the high radiation and heat loads (Sect. 9.7.6) during fabrication of curium containing fuel elements, recycling of curium in PWRs leads to the build up of Cf-252 [68, 69, 71]. This causes an increase of the decay heat and gamma radiation of spent fuel by a factor of 3 and an increase of the neutron radiation by a factor of 8,000 if compared to a spent MOX-PWR fuel assembly [72]. Therefore, such spent fuel containing Cf-252 (half-life about 2 years) would have to be stored intermediately for about 2 decades until Cf-252 will have decayed to sufficiently low concentrations to Cm-248.

Therefore, curium should be stored for about 100–200 years until the most important curium isotopes will have decayed to plutonium isotopes (Sect. 9.7.6) and the build up of Cf-252 can be minimized.

9.8.2 Transmutation and Incineration of Neptunium and Americium

Neptunium can be loaded homogeneously to the core fuel of PWRs and FRs. Neutron capture in Np-237 results in build up of Pu-238 with high spontaneous neutron radiation and high alpha decay heat power. There is presently a limit of about 5% Pu-238 in plutonium set by radiolysis during aqueous reprocessing and neutron radiation exposure during MOX fuel refabrication. Neutron capture in americium results in the build up of large amounts of curium (Cm-242, Cm-243 and Cm-244) which are strong neutron and alpha-particle emitters. In order to avoid deterioration of the safety parameters, heterogeneous loading of a certain number of americium containing fuel assemblies at the core periphery is preferred [69].

9.8.3 Neutronic Analysis for Potential Destruction Rates of Neptunium and Americium in PWRs and FRs (One Cycle Irradiation)

Neutronic analysis for destruction rates of plutonium, neptunium and americium for one cycle irradiation were reported by [69, 70, 73]. If neptunium and americium are admixed to the fuel together with plutonium, build up of Pu-238 via neutron capture in Np-237 or the decay of Cm-242 to Pu-238 (after neutron capture in Am-241) or the decay of Cm-244 to Pu-240 (after neutron capture of Am-243) decrease the destruction rate of plutonium compared to those reported in Sect. 8.1.2.3 (fissile fraction of plutonium 6%) for the case of recycling plutonium only.

Table 9.5 shows the destruction rates for a 1.3 GW(e) PWR and a 1.5 GW(e) FBR (burner) for the cases of

• plutonium only (fissile plutonium fraction 7.3% in PWR and 17.7% in FBR)

• plutonium and 1% neptunium (PWR) or 2.5% neptunium (LMFBR, homogeneously) (fissile plutonium fraction 10.2% in PWR and 17% in FBR)

• plutonium and americium 1% (PWR) or 2.5% (LMFBR) distributed heterogeneously at the core periphery.

These data were determined [69] with an accompanying analysis of all safety parameters for the fresh core only. This differs somewhat from the data given in Sects. 8.1.2.3 and 9.8.4 which are based on a determination of all safety parameters over the full burn up cycle. However, the results of Sect. 8.1.2.3 were based on a moderator to fuel ratio of 2.

The destruction rates for the case “plutonium only” are about 420 kg/GW(e)$\cdot$y in a MOX-PWR and about 570 kg/GW(e)$\cdot$y in an LMFBR. If 1% neptunium (PWR) or 2.5% neptunium (LMFBR) are admixed homogeneously to the MOX fuel the plutonium destruction rate decreases somewhat to 359 kg/GW(e)$\cdot$y in case of the PWR and 525 kg/GW(e)$\cdot$y in case of the LMFBR. The neptunium destruction rate is 85 kg/GW(e)$\cdot$y in a MOX-PWR and 78 kg/GW(e)$\cdot$y in an LMFBR.

If americium is loaded in special fuel assemblies at the periphery of the core of PWRs or LMFBRs the destruction rate is 39 kg/GW(e)$\cdot$y in case of the MOX-PWR and 110 kg/GW(e)$\cdot$y for the LMFBR case.

Slightly different results are also reported [72–75] for cases with higher than 1% loading of neptunium or overmoderated fuel assemblies in PWR cores. Also matrix fuel loaded with neptunium or americium can achieve somewhat higher destruction rates [73, 74].

Table 9.5 Destruction/production rates for plutonium, neptunium and americium [69]

9.8.4 Multi-Recycling of Plutonium, Neptunium and Americium in PWRs

Multi-recycling in PWRs of only plutonium or of plutonium with neptunium as well as plutonium with neptunium/americium was investigated in [68, 75]. As already described in Sect. 8.1.2.3 for multirecycling of plutonium in full MOX-PWRs the fraction of plutonium in the MOXfuel must be restricted because of the tendency to develop a positive moderator temperature coefficient. The required k$_\mathrm{ eff}$ is achieved by adding low enriched U-235/U-238 to the MOX fuel. Therefore the MOX fuel assembly structure of Fig. 8.4 with a fuel to moderator volume ratio of 2.5 was selected. The following results were given in [68]:

• Multi-recycling of plutonium By restricting the fraction of total plutonium to 10% and adding low enriched U-235/U-238 slightly increasing during multi-recycling up to 3.58% U-235, the incineration rates of plutonium would vary between 532 kg/GW(e)$\cdot$y (first cycle) and 420 kg/GW(e)$\cdot$y (10th cycle) (load factor 0.85 assumed) [68].

• Multi-recycling of plutonium together with neptunium By restricting the total plutonium/neptunium fraction in the fuel to 8% and adding low enriched U-235/U-238 with slightly increasing enrichment of up to 4.45% U-235 the incineration rates for plutonium would vary between 340 kg/GW(e)$\cdot$y (first cycle) and 290 kg/GW(e)$\cdot$y (10th cycle). Accordingly, the neptunium incineration rates would vary between 31 kg/GW(e)$\cdot$y (first cycle) and 16 kg/GW(e)$\cdot$y (10th cycle) (load factor 0.85 assumed) [68]. The Pu-238 isotopic fraction in the plutonium—would vary between 6% (first cycle) and 8.5% (10th cycle) [68]. These latter incineration rates are smaller—compared to Table 9.4—since both the fraction of total plutonium with 7.58% (first cycle) and that of neptunium with 0.42% (first cycle) are smaller than the 10.2% fissile plutonium and 1% neptunium used in the previous section.

• Multi-recycling of plutonium and neptunium/americium By restricting the plutonium/neptunium/americium fraction to 8% and adding low enriched U-235/U-238 with slightly increasing enrichment up to 6.61% U-235, the incineration rates of plutonium would vary between 235 kg/GW(e)$\cdot$y (first cycle) and 124 kg/GW(e)$\cdot$y (10th cycle). Accordingly the neptunium incineration rate would vary between 24.9 kg/GW(e)$\cdot$y (first cycle) and 6.8 kg/GW(e)$\cdot$y (10th cycle). The americium incineration rate would vary between 5 kg/GW(e)$\cdot$y (third cycle) and 10 kg/GW(e)$\cdot$y (10th cycle). In the first and second cycle there would be an americium production of 39.6 kg/GW(e)$\cdot$y and 8 kg/GW(e)$\cdot$y respectively (0.85 load factor assumed) [68]. The Pu-238 isotopic fraction in the plutonium would increase to 8.5% (first cycle) and 17% (10th cycle). This is mainly caused by neutron capture in Np-237 leading directly to Pu-238 or neutron capture in Am-241 leading to Pu-242 (15%) or Cm-242 which decays to Pu-238 (75%) [68]. These incineration rates are smaller—compared to Table 9.5 and 9.7—for the same reasons as mentioned already above for the case of multi-recycling of plutonium together with neptunium.

The most important reactivity (boron worth) and temperature coefficients (fuel Doppler coefficient and moderator temperature coefficient) for multi-recycling of plutonium only can be derived from [75] where these safety coefficients were reported for a restricted total plutonium content of 8 and 12%. These data are given in Table 9.6 for the subassembly shown in Fig. 8.4 with a fuel to moderator volume ratio of 2.5 (compared to the standard PWR UOX fuel element with a fuel to moderator volume ratio of 2).

Table 9.6 Boron worth (BW), fuel Doppler temperature coefficient (FDC) and moderator temperature coefficient (MTC) for standard UOX PWRs and MOX PWRs (multi-recycling)

The smaller boron worth coefficients for the MOX PWR would necessitate higher B-10 enrichment. The more negative fuel Doppler temperature coefficients (FDC) and moderator temperature coefficients (MTC) of the MOX PWRs guarantee good control and safety behaviour of these reactor types.

For MOX PWRs containing also neptunium and americium a careful analysis for these reactivity and temperature safety coefficients would be needed.

9.8.4.1 The Seed and Blanket PWR Using Plutonium and Thorium

This seed and blanket design concept [76] for PWRs uses the same fuel assemblies as a PWR. These fuel elements can be either quadratic as in Western PWRs or hexagonal as in Russian PWRs. The assembly consists of an inner seed assembly containing plutonium mixed oxide fuel, whereas the surrounding blanket assembly contains thorium dioxide as fertile fuel. During operation the plutonium in the inner seed elements is incinerated, whereas in the outer blanket elements U-233 is generated by neutron capture in Th-232.

Incineration rates of up to 800 kg of plutonium per GW(e)$\cdot$y were claimed [76].

9.8.5 Recycling of Plutonium and Minor Actinides in ADSs

Destruction rates of plutonium and TRU (neptunium, americium, curium) in a Pb/Bi (LBE) cooled 320 GW(e) ADS and in a sodium cooled 336 GW(e) ADS loaded with metallic fuel containing uranium, plutonium, neptunium, americium and curium from pyrochemistry (Sect. 9.6) are reported in [74, 77, 78]. The destruction rates are 670 kg/GW(e)$\cdot$y plutonium and 74 kg minor actinides (neptunium, americium, curium) in the LBE case and 796 kg/GW(e)$\cdot$y TRU (plutonium, neptunium, americium, curium) in the sodium cooled ADS.

9.8.6 Plutonium Incineration by Multi-Recycling in MOX-PWRs, FR-Burners and ADSs

Section 8.1.2.4 described already a strategy with MOX-PWRs incinerating the plutonium generated by a certain cluster of UOX-PWRs. If instead of this MOX-PWR strategy FRs or accelerator driven systems (ADSs) are loaded with the plutonium of UOX PWRs similar results are obtained. However, the plutonium incineration rates of FR-burners and ADS are higher than those of MOX-PWRs. Table 9.7 shows the higher incineration rates of FR burners, e.g. so-called CAPRA-FRs and the plutonium incineration rates of ADSs compared to those of MOX PWRs as assumed in [79, 80].

Table 9.7 Incineration rates of different plutonium burner reactors

Assuming these plutonium incineration rates, similar analyses as described in Sect. 8.1.2.4 can be performed. This leads to Fig. 9.9 which shows the plutonium inventories in the fuel cycle of a cluster of M = 8 UOX-PWRs operating in symbiosis with either MOX-PWRs or FR burners (CAPRA type) or ADSs. The inventories are normalized in tonnes per GW(e). The straight line represents the once through fuel cycle with direct spent fuel storage in a deep geological repository. The plutonium is accumulating as a function of time following an almost straight line in Fig. 9.9.

The plutonium recycle strategies for MOX-PWR FR-burner and ADS in the assumed scenario are represented by piecewise straight lines where each new line represents the introduction of an additional MOX-PWR or FR-burner (CAPRA-type) or ADS. The difference in tonnes of plutonium between the full line (once through cycle) and the lines for the UOX PWR strategy with MOX-PWRs or FR-burners (CAPRA) or ADS represents the plutonium inventory which is incinerated by these recycling burner reactors. As to be expected the FR-burners and even more the ADSs incinerate the plutonium produced by the UOX-PWRs more efficiently and faster than MOX-PWRs [79, 80].

Fig. 9.9

Balance of plutonium (normalized to 1 GW(e)) from a cluster of M = 8 UOX-PWRs with 10 GW(e) total power operating in either the once through direct spent fuel disposal mode (straight line). The polygon type lines show the plutonium inventory (normalized to 1 GW(e)) for the cases of UOX-PWRs operating in symbiosis with either MOX-PWRs or FR-burners (CAPRA-type) or ADSs [79, 80]

For the MOX-PWR some lines are shown as dotted lines from about 75 years on, because the coolant/moderator temperature coefficient, as an important safety related reactivity coefficient, could become intolerable (Sect. 8.1.2.3) for a fuel to moderator volume ratio of 2.0. For a fuel to moderator volume ratio of 2.5 the safety related reactivity coefficient would be acceptable (Sect. 9.8.4). Both the FR-burner and ADS strategies show a potential for incineration of all plutonium produced by the UOX-PWRs in a time frame of about 125 years (FR burners) or about 85 years (ADSs).

The results of Fig. 9.9 are theoretical examples following a certain strategy. In reality first the UOX-PWR in symbiosis with the MOX-PWR are started. Later FR-burners will follow when FRs will be deployed on large scale. This strategy might perhaps be followed by the introduction of several ADSs.

The results of strategies for incinerating neptunium and americium in MOX-PWRs, FR-burners or ADSs would be similar to those shown in Fig. 9.9 for plutonium. However, the masses involved for neptunium and americium would be smaller by about a factor of 15–20.

A strategy which is often called plutonium stabilization (no more increase of net plutonium production) [60, 81, 82] would terminate the replacement of LWR-UO$_{2}$ reactors by LWR-MOX reactors or CAPRA reactors after about 75 years and the introduction of ADSs after about 55 years (Fig. 9.9). The plutonium production and incineration would then remain contant (stabilization) for the following time. As the strategy of Fig. 9.9 is started with a cluster of M = 8 LWR-UO$_{2}$ reactors and up to the point in time of 75 or 55 years only three LWR-UO$_{2}$ reactors would be replaced by LWR MOX reactors, CAPRAs or ADSs. The fraction of Pu incinerating reactors for plutonium stabilization would be 3/8 or about 37%.

9.8.7 Influence of the Transmutation of Actinides on the Fuel Cycle and on the Waste Repository

9.8.7.1 Influence of Plutonium Recycling and Minor Actinide Recycling on Reprocessing and Fuel Refabrication

It was shown already in Sect. 7.4 for plutonium recycling in FBRs that the amounts of plutonium to be handled in the fuel cycle increases in comparison to the once-through case. Similarly, the amount of plutonium and of the minor actinides (Np, Am) increases in the fuel cycle for strategies with transmutation and incineration of these actinides. Plutonium and the minor actinides are collected after reprocessing of spent fuel in order to fabricate MOX fuel elements or other fuel subassemblies containing, e.g. oxides of neptunium and americium. This leads to a concentration of these actinides in fuel fabrication and reprocessing facilities applying the different chemical separation processes for actinide transmutation and incineration described in Sects. 9.5–9.7 [72, 81, 83, 84]. These amounts of actinides, their decay heat and their radiotoxicities are higher than those of uranium and plutonium in the once-through case.

The different nuclear characteristics for the calculation of the radioactivity (Ci/g), heat production (W/g) and radiotoxicity (using dose coefficients (Sv/Bq)) is given in Table 9.8 [72]. On the basis of these data the radioactivity, the heat production or the radiotoxicity of any composition of fresh (refabrication) or spent fuel (reprocessing) can be determined. High contents of Pu-238, Pu-241, Am-241, Am-242 or Am-242m lead to high specific activities of MOX and americium containing fuel (Table 9.8). High contents of Pu-238 and americium (as well as Cm-244) lead to relatively high internal heat production in the fuel. High contents of Pu-238 and Pu-242 (as well as curium isotopes) are responsible for relatively high neutron radiation.

Table 9.8 Properties of transuranium nuclides [72]

As already explained in Sect. 9.8.1 curium recycling is only feasible with pyrochemistry and related fuel refabrication methods in fast reactors.

Spent LWR MOX fuel can have a 6–7 times higher $\upalpha$-radioactivity than spent UO$_{2}$ fuel [60, 81, 82]. Detailed investigations for heterogeneous recycling of plutonium and of the minor actinides, neptunium and americium with the French CORAIL fuel subassembly were reported by [72, 83]. According to Table 9.8 the specific activities are mainly determined by Pu-238, Pu-241 and the americium isotopes Am-241 and Am-242m (Am-242 has a half-life of 16 h) as long as MOX fuel refabrication without curium is considered. The decay heat is determined by Pu-238 and the americium isotopes. The neutron emission is dominated by the plutonium isotopes Pu-238, Pu-240 and Pu-242 as long as MOX fuel mixed with neptunium and americium is considered.

Spent MOX and minor actinide fuel have only somewhat higher activity since the activity of the fission products is dominating that of the plutonium and minor actinide isotopes for the cooling periods of 5–10 years as long as curium is not recycled.

Present aqueous reprocessing technology is considered to be applicable up to Pu-238 contents of about 5% [83]. Present MOX fuel refabrication on the basis of glove box technology can be applied up to Pu-238 contents of about 4%.

As shown in the previous Sect. 9.8.4 recycling of plutonium, neptunium and americium can lead to Pu-238 contents in the plutonium of 8% and more. This means that not only present reprocessing technologies based on the PUREX process must be modified, but also new reprocessing facilities based on the chemical partitioning processes described in Sect. 9.5 must be developed. Also new refabrication facilities based on refabrication processes described in Sect. 9.7 must be deployed [72].

9.8.7.2 Influence of Transmutation and Incineration of Actinides on the Radioactivity, the Radiotoxicity, and on the Heat Load of Waste in a Deep Geological Repository

The different strategies for transmutation and incineration of plutonium and minor actinides have different radioactivity loads in Bq/TW(e)$\cdot$h and different radiotoxicity levels in Sv/TW(e)$\cdot$h (Figs. 9.10, 9.11) [72, 83]. The highest radioactivity levels and the highest radiotoxicity levels are shown for A1, the Once through strategy (direct disposal of spent PWR fuel) and for A2, the Once (mono) recycling strategy for Pu followed by direct disposal of the Pu-MOX spent fuel elements.

The next lower curve shows the multi-recycling strategy of Pu in fast reactors (FRs) and the lower curves for both radioactivity and radiotoxicity levels are represented by the Integral Fast Reactor (IFR) strategy with recycling of Pu and of the minor actinides (Np, Am, Cm) and the UOX-MOX-PWR with ADS strategy recycling also Pu and the minor actinides.

As during reprocessing also intermediate level waste (ILW) is produced (Sect. 7.5) these radioactivity levels are also indicated in Fig. 9.10. In Fig. 9.11 the radiotoxicity levels for High Level Waste (HLW), Intermediate level waste (ILW) and the remaining Reprocessed Uranium (RepU) are given for the above transmutation and incinceration strategies [72, 83, 84].

Fig. 9.10

Radioactivity (Bq/TW(e)$\cdot$h) as a function of time for different fuel cycle strategies (1 Ci$\wedge \over =$ $3.7\times {10^{10}}$ Bq) [72, 83–85]

Fig. 9.11

Radiotoxicity as a function of time for different fuel cycle strategies (HLW and ILW) [72, 83–85]

The different strategies for transmutation and incineration of plutonium and the minor actinides affect strongly the heat load of HLW packages, i.e. the contribution of the minor actinides (the heat load produced by the fission products is hardly influenced). Figure 9.12 shows the expected heat loads produced by the minor actinides in the deep geological repository for different transmutation and incineration strategies (the heat loads are given in TW(e)$\cdot$h [74, 83, 86, 87]. These units must be multiplied by a factor of 7.45 TW(e)$\cdot$h/GW(e)y (load factor 0.85) to obtain the real heat loads per GW(e)$\cdot$y) (7450 h per year if load factor is 0.85).

Fig. 9.12

Heat load produced by minor actinides in HLW for different recycling and incineration strategies [86]

The highest heat load is produced in HLW in case of the Once through cycle. Pu recycling once in PWRs provides only little improvement. Multiple Pu recycling in PWRs or FRs results in some improvement after about 1,000–10,000 years. However, Pu and Am recycling (with Cm disposed) results in a considerable improvement from about 100 years on. The highest improvements are obtained for Pu and MA (neptunium, americium and uranium) recycled in FRs or Pu and Am recycled in PWRs and Cm stored.

This improvement is about a factor of 100. It is, however, not only the radiotoxicity and the heat loads which are decreased by a factor of about 100 but also the masses of actinides which are decreased in the HLW packages by a factor of about 100. To achieve this goal the losses of actinides during reprocessing and refabrication must be not higher than about 0.2% [72].

These advantages achieved in the back end of the fuel cycles are, however, in contrast to higher masses of plutonium and actinides with higher radioactivity, radiotoxicity and decay heat to be handled during reprocessing and refabrication as mentioned already in Sect. 9.8.7.1.

9.8.7.3 Contributions of Cs-137 and Sr-90 to the Heat Load in a Deep Geological Repository

Cs-137 (half-life 30 years) and Sr-90 (half-life 29 years) are, besides americium, the main contributors to the heat load in the high level waste package during the first one hundred years. They determine the spacing between the high level waste packages and thereby the amount of HLW to be stored in a deep geological repository [86, 87]. It is therefore considered to separate Cs-137 and Sr-90 chemically from the liquid HLW and store Cs-137 and Sr-90 for about 100–200 years separately until they will have mostly decayed to stable Ba-137 and Zr-90 [72, 88, 89].

9.8.8 Transmutation of Long-Lived Fission Products

A number of radiologically important long-lived fission products have to be taken into account in the safety assessment of a deep geological repository. The following long-lived fission products have to be assessed: Tc-99, I-129, Sr-90, Cs-135, Cs-137, Se-79, Zr-93, Nb-94, Sn-126, Sm-151 and the activation products C-14 and Cl-36.

They are listed in Table 9.9 showing their half-lives, production during reactor operation in kg/GW(th)$\cdot$y and radiotoxicity in Sv/g.

Table 9.9 Long-lived fission products with half-life (years), production and radiotoxicity [90]

The objective for transmutation of long-lived fission products is to reduce their radiotoxicity significantly, before they have to be conditioned and sent as waste to a geological repository.

The transmutation rate can be characterized by the transmutation half-life which is a measure for the time needed to destroy by neutron capture half of the initial mass. It is defined as [90]

$$\begin{aligned} \text{ T}_{1/2}^\mathrm{ Tr} =3.171 \times 10^{-8} \cdot \frac{\ln 2}{\upsigma _\mathrm{ c} \cdot \upphi } \end{aligned}$$

$\upsigma _\mathrm{ c }$ microscopic neutron capture cross section in barn (1 barn = 10$^{-24}$ cm$^{2})$;

$\upphi$ neutron flux in n/cm$^{2}\cdot$s

This transmutation half-life $\text{ T}_{1/2}^\mathrm{ Tr}$ should be considerably smaller than the natural decay half-life T$_{1/2}$.

Table 9.10 Transmutability of long-lived fission products in fast and thermal neutron fields [90]

Table 9.10 presents an assessment whether or not these long-lived fission products can be transmuted in a PWR (thermal neutron spectrum), FR or ADS (fast neutron spectrum). The main parameters for this assessment are the capture cross sections of the long-lived fission products in the neutron flux of 10$^{14}$ n/cm$^{2}$ $\cdot$s of a PWR core or 10$^{15}$ n/cm$^{2}$ $\cdot$s of an FR or ADS core [90].

Some of these long-lived fission products, e.g. cesium exist as different isotopes like Cs-133, Cs-134, Cs-135, Cs-137 in spent fuel and would require isotope separation before transmutation.

Without application of isotope separation and considering that the production rate of Nb-94, Sn-126 and Sm-151 is small and the radiotoxicity of Pd-107 is very low, the practically remaining candidates for long-lived fission product transmutation are I-129 and Tc-99. The other fission products Se-79, Sr-90, Zr-93, Sn-126 and Cs-137 are considered nontransmutable with sufficient efficiency [90]. In addition the impact on long term radioactive dose release levels from a geological repository (Chap. 7) of the isotopes Nb-94, Pd-107 and Sm-151 are considered to be low [90].

Because of their high geochemical mobility I-129 and Tc-99 are major contributors to the biosphere release dose rates of a deep geological repository. I-129 and Tc-99 are well soluble in groundwater and are hardly adsorbed by the geological rocks (Sect. 7.6.3.2).

The extraction of Tc-99 from the HLW as TcO$_{4}$ by an advanced PUREX process is relatively easy [81]. I-129 can be extracted during reprocessing by silver impregnated filters [81, 91]. Several target materials for Tc-99 and I-129 have been studied. I-129 can be used in form of NaI, CaI$_{2 }$or as silver-iodide impregnated in silica [81, 91]. Tc-99 is mostly used in metallic form [81].

In special moderated subassemblies loaded at the periphery of the core, Tc-99 is transmuted by neutron capture to stable Ru-100, whereas I-129 is transmuted to Xe-130.

The transmutation rate for Tc-99 in special subassemblies of a PWR core is too low, whereas I-129 could be transmuted such that the I-129 produced in three PWRs can be destroyed.

In FRs with their fast neutron spectrum the transmutation rates for I-129 and Tc-99 can be much higher [70, 73, 90,92–94] than in LWRs. Special irradiation assemblies containing moderator rods (ZrH; written in short; in fact it is usually ZrH$_\mathrm{ x}$ with x $\approx$ 1.7) and rods with either BaI$_{2}$ or metallic Tc can be located at the radial boundary of the FR core. These assemblies would consist of either 37 or 127 special rods [92, 93].

9.8.8.1 I-129 Transmutation in FRs

Barium iodine, BaI$_{2}$, was chosen for its good chemical stability and manufacturing characteristics. In Table 9.11 the transmutation rates in % per year for I-129 are shown for a heterogeneous (moderator rods (ZrH) and BaI$_{2}$ rods are separated) and a homogeneous case (moderator ZrH and BaI$_{2}$ are mixed in rods) with 27 or 127 rods in the assemblies. In addition to full BaI$_{2}$ pellets in the rods also a case with hollow pellets (smear density over the pellet of 50%) was investigated. The heterogeneous case contains 27 BaI$_{2}$ rods and 10 separate ZrH moderator rods interspersed in the assemblies. For the homogeneous case the 127 rods contain the moderator ZrH mixed with BaI$_{2}$.

The support ratio is defined as the ratio of I-129 transmuted in the special I-129 assemblies to the quantity of I-129 produced in the driver fuel of the FR [93].

Table 9.11 Transmutation rate in % per year and support ratios for I-129 and different irradiation assembly cases [92–94]

The highest transmutation rates per year are attained for the homogeneous hollow pellet arrangement with 127 pins and 10% BaI$_{2}$. For another case with 67 BaI$_{2}$ and ZrH pins and 60 stainless steel even 9.5% as transmutation rate per year are reported [93]. However, these high transmutation rates belong to small initial inventories in the irradiation assemblies. Therefore the support ratios are relatively small. Small transmutation rates in % per year require several times recycling of the initial inventory. This will increase the chemical reprocessing losses to the HAW.

9.8.8.2 Tc-99 Transmutation in FRs

Technetium is difficult to mix with Zr metal. Therefore, a new irradiation pellet design was developed [93]. Between 55–124 thin Zr metal needles of about 0.3–1 mm diameter stuck in holes of an 18 mm thick ZrH pellet. This design with very thin Tc needles is chosen in order to avoid self shielding problems of the neutron flux in the absorbing material Tc-99. Table 9.11 shows the results of calculations for a homogeneous case as described above and for the thin Zr-needle cases (Table 9.12).

Table 9.12 Transmutation rate in % per year for Tc-99 and different irradiation assembly cases [92–94]

Again, the same observations hold for the high transmutation rates per year connected with the small support ratios (as discussed above for the destruction of I-129).

The irradiation time at the radial boundary of the FR core is limited by the radiation damage of the structural materials, e.g. austenitic steel. Dependent upon the transmutation rate achieved and on the possible irradiation time for transmutation, several recycle steps will be required. This increases the losses of I-129 and Tc-99 going to the HAW and to the deep geological repository.

9.8.9 Comparison of Possible Radiation Exposure Rates from Drinking Water in the Vicinity of a Deep Geological Repository for Different Incineration Schemes

Similar analysis for the potential radiation exposure in the vicinity of a deep geological repository—as were reported for direct spent fuel or vitrified HAW disposal in Sect. 7.7.3.3 and Figs. 7.32 and 7.33—were also performed for different plutonium and TRU recycle and incineration schemes [86]. Figure 9.13 shows the normalized radiation dose rates for four different fuel cycle scenarios

• once through fuel cycle or direct spent fuel disposal

• reprocessing of spent fuel, but recycling plutonium only once

• reprocessing and recycling plutonium in LWRs

• all fast reactor strategy with reprocessing and MOX fuel and plutonium minor actinides recycling.

Fig. 9.13

Human exposure due to radioactivity from a well near a deep geological repository for different fuel cycle options

All results are normalized to the peak radiation dose rate of the once through fuel cycle case with direct spent fuel disposal of Fig. 7.32. The cases Pu-recycling in LWRs and the case all fast reactors with Pu and minor actinide recycling lead to considerably lower radiation exposures in the vicinity of the deep geological repository. This is due to the incineration of plutonium and minor actinides.

However, cesium-135 might become a dominant contributor after about one million years to the release rate of radioactivity from the deep repository if cesium is not sorbed by the surrounding geological formations (granite, tuff, clay or salt) on its way from the high active waste package to a well with drinking water [84, 85].

9.8.10 Influence of the Transmutation of I-129 and Tc-99 on the Radiation Exposure from Drinking Water in the Vicinity of a Deep Geological Repository

The transmutation of I-129 and Tc-99 will decrease the amount of these LLFPs in the HLW to be deposited into the deep geological repository. Analyses [90] were performed for the removal of 95% of I-129 and Tc-99 and 5% losses going into the HLW. In this case the total radiation exposure by drinking water from a well 20 km away from the repository can be reduced by a factor of about 3 [90] during the first $3 \times 10^{5}$ years.

The above results show that only two long-lived fission products (Tc-99, I-129) could be destroyed by transmutation with reasonable efficiency. Other long-lived fission products are either nontransmutable or would need additional isotope separation. The improvement obtained by transmutation and destruction of Tc-99 and I-129 with regard to the estimated radiation exposure between 10$^{3}$ and 10$^{6}$ years caused by the HLW in a deep geological repository after water ingress (Sect. 7.6.3.3) would be limited.