3.1 Neptunium Isotopes

Table 9.11 lists the isotopes of neptunium important in nuclear technology and some of their important nuclear properties.

J36Np. The isotope 236 Np is formed in reactors by (n, 2n) reactions in 231 Np. It undergoes beta decay, with a half-life of 22 h, to form 236 Pu.

237Np. The isotope 237Np is formed in considerable quantities in reactors, by the nuclide chains initiated by (n, 7) reactions in 235 U and by (n, 2n) reactions in 238 U. Neutron capture by 237 Np leads through 238 Np to 238 Pu, which is the principal alpha-emitting constituent of plutonium in power reactors. To produce 238 Pu for use as a heat source for thermoelectric devices, neptunium has been recovered from irradiated uranium to form target elements for further irradiation in reactors. Commercial processes designed for this recovery are discussed in Chap. 10.

In normal reprocessing of irradiated uranium fuel neptunium appears in the high-level wastes. Because of its long half-life of 2.14 X 10* years, 237Np persists in these wastes long after most of the fission products and other actinides have decayed. It undergoes alpha decay in the 2n + 1 decay chain to form 233 Pa, which subsequently decays to 233 U, to 229 Th, and thence to 225 Ra and its decay daughters. Because of its half-life and the radiotoxicity of its daughters, 237Np is the source of important long-term toxicity in high-level wastes. If the radionuclides in these wastes ever become dissolved in groundwater, the chemistry of neptunium is such that it may not be as effectively retarded by sorption in geologic media as are the other actinides in these wastes.

238Np. The isotope 238Np is the 2.1-day beta emitter formed by neutron capture in 237Np. With the availability of separated 237 Np from fuel reprocessing, 238 Np is easily made by irradiation of the 237Np target. It has displaced 239Np as a tracer for chemical studies. The high capture-to — fission ratio for 237 Np results in only a relatively small contamination by fission products, which are easily removed chemically [K2].

Source: C. Keller, The Chemistry of the Transuranium Elements, Verlag Chemie, Weinheim, 1971.

239Np. The isotope 239Np is formed by neutron capture in 238U or by decay of 243 Am. The latter method is the easiest for laboratory preparation, if separated americium is available. Reactor — produced americium will not produce pure 239 Np, however, because of the presence of241 Am, which decays to 237Np.

Dissolver off-gases are processed for recovery of oxides of nitrogen, known collectively as NO*, in step 15. This is necessary to prevent corrosion of downstream equipment and acid contamination of the environment, and it reduces HN03 makeup. The preferred procedure is to pass the hot dissolver off-gases, which contain water vapor, nitric acid vapor, N2 04, N02, NO, N20, and added air or oxygen, successively through a downdraft condenser and a water scrubber.

In a downdraft condenser gases and condensed dilute nitric acid flow concurrently down the condenser wall. In this way the leanest gas is contacted with the maximum volume of coolest condensate, thus improving absorption. The following reactions take place:

2N02(?)-N204C?)

2N02 (or N204) (0 + H2O(0 — HNO3(0 + HN02(1)

3HNO2(0 — H2 0(/) + HN03(/) + 2NO(?) and 2NO(?) + 02(?) -> 2N02 (?)

The first three equilibrium reactions are rapid, but the fourth reaction, which eventually proceeds to completion, is slow even with excess oxygen and determines the extent of absorption.

Gases leaving the downdraft condenser are passed through a bubble-plate or packed water scrubber, where additional absorption of NO* takes place. Laboratory studies [М2] indicate that the nitrogen oxide content can be reduced to from 0.1 to 0.5 percent with a residence time of 2 min in the condenser and water scrubber. The design of scrubbers for recovery of nitrogen oxides is described in standard texts, e. g., [Р5].

The nitrogen oxide content of dissolver off-gases can be further reduced to 10 ppm by adding NH3 to the gases leaving the absorber and passing the mixture over a hydrogen mordenite catalyst [P4], which reduces NO* to N2 and H2 O.

Oxidation of pentavalent neptunium by nitric acid. Oxidation of pentavalent neptunium to hexavalent by nitric acid requires catalysis by nitrous acid. The kinetics of this reaction have been studied by Siddall and Dukes [S16], Swanson [S24] and Mouline [М9]. Siddall and Dukes reported that the reaction was first order in neptunium concentration, independent of nitrous acid concentration if greater than 5 X 10"s M, and depended on temperature T (K) and nitric acid molarity xH as can be represented by Eq. (10.33):

kN = 7.03 X 10’7 е7062/гл$280 min’1 (10.33)

fcfj is the specific rate constant in the equation

/Jy

^ = -*N(x5-;t?) (10.34)

where xs is the aqueous molarity of Np(V) and x° is its equilibrium molarity.

Swanson’s results appear very different. He reported that the first-order reaction rate constant in Eq. (10.34) was independent of nitric acid concentration and proportional to nitrous acid concentration:

kN = fc,[HN02] (10.35)

Values of к і are 46 Af-^min-1 at 24°C and 250 M~’ — min-1 at 46°C. However, at concentrations of nitric and nitrous acid used in reprocessing, values of kN from Eqs. (10.33) and (10.35) are not far apart.

Mouline’s experiments partially explain the apparent discrepancy. When the nitrous acid molarity is less than that of neptunium, the rate is proportional to nitrous acid concentration. At nitrous acid/neptunium concentration ratios above unity, the rate is independent of nitrous acid concentration. Values of the first-order rate constant observed by Mouline at 35°C for the latter condition are compared below with ones calculated from Eq. (10.33) correlating Siddall and Duke’s data.

First-order constant fc^, min

Nitric acid molarity, xH Average observed, Mouline Eq. (10.33)

2 0.013 0.0127

3 0.059 0.048

4 0.14 0.124

These reaction rates are too low to explain the appreciable extraction of neptunium obtained in the short-residence-time HA contactors used at Hanford and elsewhere. Swanson [S24] reported that radiolysis reaction products of TBP and nitric acid present in Purex solutions increased the neptunium oxidation rate and provided a possible explanation. He found that the oxidation rate could be increased several orders of magnitude by adding a synthetic “rate-accelerating material” (RAM) produced by reacting the aciform of nitropropane, C2H5(CH)(NO)(OH), with nitric and nitrous acids, and recommended addition of such a catalyst to Purex feed if increased neptunium extraction were desired.

Oxidation of pentavalent neptunium by pentavalent vanadium. Oxidation of pentavalent neptunium by pentavalent vanadium proceeds at a practical rate without catalyst. Dukes [D4] found that the rate of reaction could be represented by

=-kv[H+]J[V02+]*5 (10.36)

at

with values for the specific rate constant ky given in the second column of Table 10.24. The third column gives values for the first-order rate constant ky [H+] 2 [V02+] for conditions to be recommended in the HA contactor, 2.5 M HN03 and 0.01 M V02+. The rate is much greater than the rate with nitrous acid catalysis and is high enough for a practical process. Srinivasan et al. [S20] extracted more than 90 percent of the neptunium in laboratory mixer-settler experiments with V02+ as oxidant.

Reduction of neptunium. To separate neptunium from plutonium in the Purex process, plutonium is reduced to inextractable Pu(III) while neptunium is reduced from extractable

Source: E. K. Dukes, “Oxidation of Neptunium (V) by Vanadium (V),” Report DP-434, 1959.

Np(VI) through inextractable Np(V) to extractable Np(IV). Reduction to Np(V) is rapid, but reduction to Np(IV) is slow, probably because of need to remove oxygen from Np02+. Of the three reductants considered, ferrous iron reacts most rapidly, but must be present in such great excess for complete reduction to Np(IV) that one of the stronger reductants, tetravalent uranium or hydroxylamine, is preferred.

Reduction with tetravalent uranium. Newton [N3] found the rate of reduction of hexavalent neptunium to pentavalent to be rapid and given at 25°C by

_ d[^g(V^ = 21 7[Np(VI)] [U(IV)] (10.37)

The rate of reduction of pentavalent neptunium to tetravalent is much slower. Shastri et al. [S10] made an extensive study of the reduction of Np(V) by U(IV). In one series of experiments at 25°C, [H*] =0.1, and an ionic strength of 0.6 M, the rate of increase of Np(IV) molarity x4 could be represented by

y-= (0.039*5 + 0.26*4 )[U(IV)] (10.38)

^*min

The rate varied inversely as [H+]2. At hydrogen ion molarity [H+] and when U(IV) is present in sufficient excess to remain effectively constant during reduction, the integrated rate equation with *4 = 0 at t = 0 is

In the process example to be used in Sec. 7.7, where [H+] = 0.06, [U(IV)] = 0.044 and *4/*! = 0.99; t = 35 min.

Reduction with hydroxylamine. No comparable rate data for reduction of Np(V) by hydroxyl­amine are available. Barney [B2] reported that the initial rate of reduction of Pu(IV) by 0.1 M hydroxylamine nitrate (HAN) was about one-fourth the initial rate of reduction of Pu(IV) by U(IV) at 25°C. If the same ratio applies to reduction of Np(V) by HAN or U(IV), a reaction time of around (4X35)= 140 min might be required. Use of hydroxylamine would have the advantage of not requiring reduction of uranium to U(IV) and its subsequent recycle.

The spray column, shown schematically in Fig. 4.29, is the simplest of the contactors. The heavy aqueous phase enters the top of the vertical cylinder through a distributor and flows downward under gravity, usually as the continuous phase. A distributor at the bottom of the column disperses the entering organic phase into small drops, which rise through the continuous heavy phase and collect in a layer at the top. Coalescence of the dispersed phase drops and axial circulation and mixing of the continuous phase result in relatively low efficiency of contacting. Very tall columns may be required to obtain only a few theoretical stages.

Because of absence of internal structure, spray columns are sometimes selected for

liquid-liquid separation when suspended solids are present, as in the extraction of zirconium from hafnium in acidic thiocyanate solutions, wherein solid thiocyanate polymers tend to form (Chap. 7).

9.1 Uranium Refineries

Table 5.27 lists the principal uranium refineries of the Western world and their feed and products. In all these refineries except Allied Chemical’s, the sequence of operations follows some or all of the steps shown in Fig. 5.21, in which uranium ore concentrates are first purified by solvent extraction and then converted to the materials of principal practical importance, uranium dioxide, uranium metal, or uranium hexafluoride. The steps in these refining operations will be described in process sequence in Secs. 9.2 through 9.6.

In Allied Chemical’s uranium refinery the sequence of process operations is reversed, with conversion to UF6 preceding purification, and with UF6 as the sole purified product. The Allied Chemical process will be described briefly in Sec. 9.7.

Purified thorium is usually produced in the form of an aqueous solution of thorium nitrate or crystals of hydrated thorium nitrate. The principal forms in which thorium is used in nuclear systems are the oxide Th02, the carbide ThC2, the fluoride ThF4, the chloride ThCl4, or the metal. Conversion to oxide, fluoride, chloride, and metal are discussed in this section; production of thorium carbide was discussed in Sec. 5.3.

1.14 Conversion of Thorium Nitrate to Th02

Three methods that have been used to convert thorium nitrate to Th02 are as follows:

1. Thermal denitration,

Th(N03)4-4H20 hea‘ > Th02 + 4HN03 + 2H20

2. Precipitation of thorium hydroxide from aqueous solution with NH3,

Th(N03)4 + 4NH3 + 4H20 -*• Th(OH)4 + 4NH4N03 followed by ignition of the hydroxide,

Th(0H)4 heat > Th02 + H20

3. Precipitation of thorium oxalate with oxalic acid,

Th(N03)4 + 2H2C204’2H20 -> Th(C204)2-2H20 + 4HN03 + 2H20

followed by ignition of the oxalate in air,

Th(C204)2 -2H20 + 02 heat—> Th02 + 4C02 + 2H20

Precipitation with NH3 has been used to prepare a colloidal sol of Th(0H)4 for formation into small spheres of Th(OH)4 gel, in the so-called sol-gel process, followed by ignition to small Th02 spheres [Zl] to be incorporated in fuel elements for an HTGR (Chap. 3, Sec. 7.3).

Precipitation with oxalic acid followed by ignition to oxide has the advantage of separating thorium from several impurities (uranium, iron, and titanium) that remain in nitric acid solution. Following is a brief description of the process developed at Iowa State College [W2], pp. 70-75, for the U. S. Atomic Energy Commission (ЛЕС) and used on a production scale by the National Lead Company at Femald, Ohio [C6], pp. 150-152. To an aqueous solution of Th(N03)4 containing about 200 g thorium/liter and 0.5 N in HN03 at 60°C is added about 105% of the oxalic acid, H2C204*2H20, needed to convert all thorium to the oxalate. The solution is stirred for about 5 min to complete precipitation. This results in an easily filtered crystalline precipitate of Th(C204)‘2H20. This is filtered on a vacuum filter and washed with about half the feed solution volume of distilled water at 35°C. The precipitate is dried in a twin-screw drier with a jacket temperature of 120°C and a screw temperature of 154°C to a water content of 10 w/o.

The dried oxalate is converted to oxide in an externally fired rotary kiln, with counterflow of air. The exit gas temperature is controlled at 820°C. This produces a reactive, free-flowing oxide containing less than 0.5% carbon and about 0.5% moisture.

The long-term radioactivities of neptunium, americium, and curium in the high-level reproces­sing wastes from the uranium-fueled water reactor are shown in Fig. 8.7. Except for 241 Am and e7Np, these curves are also applicable to unprocessed discharge fuel. The curves 241 Am and M7Np have been calculated for 0.5 percent of the plutonium in discharge fuel to appear in the wastes, so that there is not sufficient 241 Pu to significantly increase the amounts of 241 Am and

»’This effective cross section is greater than the cross section for thermal neutrons because of resonance absorption in 236 U.

231Np during the decay periods. The high activities of americium persist for thousands of years and are greater than the fission-product activity after a few hundred years of storage.

The radioactivities of the plutonium radionuclides in the high-level wastes from fuel reprocessing are shown as a function of storage time in Fig. 8.8 [Р1]. Because the initial plutonium quantities are due only to the small fraction, e. g., 0.5 percent, of the plutonium that is lost to these wastes in reprocessing, larger quantities appear after a few years due to the decay of americium and curium. The 238Pu increases with time because of the decay of 24201 Am and 242 Cm, 239 Pu increases from the decay of 243 Am and 243 Cm, and 240Pu increases due to the decay of 244 Cm. Therefore, even though the total actinide activity in these wastes is dominated by plutonium after the americium has decayed, the plutonium in the wastes at this time is due mainly to the earlier decay of americium and curium and not to the small fraction of plutonium lost to the wastes in fuel reprocessing.

The ingestion toxicity indices of the actinides in the wastes are shown as a function of decay time in Fig. 8.9 [P2]. Because the actinides are nonvolatile and because the wastes are expected to be geologically isolated, ingestion toxicity is probably a more important measure than inhalation toxicity. During the first 600 years the total toxicity index is controlled by the fission products, mainly 90 Sr. It is thereafter controlled by 241 Am and 243 Am, followed by

Figure 8.7 Radioactivity in cur­ium, americium, and neptunium as a function of decay time. (Amount in the wastes produced annually by reprocessing fuel dis­charged from a 1000-MWe uran­ium-fueled PWR.) these wastes. In Fig. 8.10 the toxicity indices are shown relative to the ingestion toxicity of the ore [P2]. The ore toxicity is due mainly to the 236 Ra, which is in secular equilibrium. Also shown are the relative toxicity indices for the uranium mill tailings, which contain 230Th and 226 Ra separated from the uranium ore, and for the depleted uranium from isotope separation, neglecting the likely later use of this uranium as fuel for breeder reactors. Because the uranium ore ingestion toxicity is dominated by 226 Ra, all of this toxicity is transferred to the mill tailings and is preserved for over 100,000 years because of the long half-life of 230Th. The tailings toxicity then decays to a lower value due to the residual uranium, e. g., about 5 percent, which remains with the mill tailings.

The ingestion toxicity of the high-level waste decays to a level below that of the initial ore after the fission-product period of about 600 years, and it ultimately decays to a toxicity that is a fraction of a percent of the toxicity of the original ore consumed to generate these wastes.

Because in the LWR fuel cycle most of the uranium in the ore appears in the depleted uranium from isotope separation, this depleted uranium if not used as breeder fuel, will slowly build up its decay daughters and 226Ra toxicity. Ultimately, a toxicity level within a few percent of that of the original ore will be reached.

The toxicity indices are not measures of hazards, in part because they take no account of the barriers that isolate these wastes from the biosphere or of the behavior of different radioactive elements with respect to these barriers. However, the long-term toxicities of the high-level reprocessing wastes are due to radium, which is the same element that controls the ore toxicity. The long-term radium toxicity of the reprocessing wastes is considerably less than the radium toxicity of the ore. It seems reasonable that high-level wastes can be geologically

Figure 8.8 Radioactivity in pluton­ium in high-level wastes as a func­tion of decay time (in wastes pro­duced annually by reprocessing fuel discharged from a 1000-MWe uranium-fueled PWR).

isolated so that the waste material has less access to the environment than the radium in the natural ore. Therefore, it is likely that the longer-term hazards from geologically isolated high-level wastes will be less than those already experienced due to the naturally occurring uranium minerals. The period of greatest importance in high-level waste management is probably the earlier, 600-year period of high fission-product toxicities.

Americium oxides. Keller [K2, КЗ] reports three stoichiometric’binary oxides of americium: AmO, Am2 03, Am02. The dioxide Am02 is the most stable of the americium oxides. It crystal­lizes with the cubic fluorite structure of all the actinide dioxides. It can be formed as a dark brown powder, stable up to 1000°C, by heating trivalent americium nitrate, hydroxide, or oxalate in oxygen to 700 to 800°C. Americium dioxide is readily soluble in mineral acids. Hydrogen reduction of the dioxide yields Am2 03.

Americium monoxide AmO is observed as a surface layer on americium metal if oxygen is present during preparation.

Americium halides. Americium forms binary halides in the oxidation states of III and IV. The trifluoride is prepared by hydrofluorination of Am02 with HF at 400 to 500°C and by precipita-

tion from aqueous solutions. Fluorimtion of AmF3 or Am02 with F2 at 400 to 500°C yields AmF4. Similar compounds are formed with chlorine, bromine, and iodine.

The general use of TBP as extractant in reprocessing nuclear fuel is due to its selectivity for the actinides, its reasonably good stability against radiolysis and reaction with nitric acid, its nonflammability, and its ready availability at low cost. Because its density is close to water’s and because of its high viscosity, for reprocessing TBP is diluted with a less dense, less viscous hydrocarbon. The diluents most inert to nitric acid and radiation are straight-chain paraffins. The diluent now usually chosen is a mixture of normal paraffins, mostly n-dodecane, because it is commercially available and provides a reasonable compromise between the desired low viscosity and high flash point. Maximum capacity of pulse columns is obtained at TBP concentrations between 20 and 30 v/o, the composition usually used in processing irradiated natural or slightly enriched uranium. TBP concentrations in the range of 2.5 to 7.5 v/o are used in processing fully enriched uranium or plutonium as one of the measures to avoid criticality.

Physical properties of pure TBP have been given in Table 4.5. Physical properties of n-dodecane and a 30 v/o solution of TBP in n-dodecane are summarized in Table 10.13.

When TBP and a hydrocarbon such as n-dodecane are mixed, a slight volume increase takes

* Given concentrations.

Table 10.12 Concentrations in scrubbing section, uranium decontamination unit 2D

Stage number, m 1_____________ 2_________________ 3^

Organic concentration, mol/liter

place. For TBP contents between 15 and 45 v/o, the increase is about 0.2 percent of the volume of the separate constituents. In precise work it is thus necessary to specify whether v/o TBP is referred to the sum of the volumes of the separate constituents or to the mixture, as is done in this text. Then, the molarity of TBP is related to its volume percent by

The viscosity t? of mixtures of TBP and hydrocarbons is given within 20 percent by

log Vmix = v log r? TBP + (1 — u) log т? нс (Ю.6)

where и is volume fraction TBP.

The mutual solubilities of water and TBP-dodecane mixtures are given in Table 10.14. The volume change when water dissolves in TBP-dodecane mixtures is negligible.

The aqueous raffinate from the first extraction cycle of light-water reactor (LWR) fuel reprocessing has an original volume of up to 5 m3/MT^ of heavy metal. It is concentrated by evaporation, and the residues of the evaporated raffinates from further extraction cycles may be combined with the concentrate. The result of these operations is the HLW concentrate that will be transferred to the waste management section of the reprocessing plant.

Volume. The volume reduction factor that can be achieved depends strongly on the bumup and on the cooling time the fuel has experienced. According to the length of the cooling period, either the heat generation of the resulting HLW or its content of dissolved solids including fission products and process chemicals are the factors limiting the degree of concentration. Most of the volume reduction is achieved in the HLW evaporator. Some further

11 MT = 1 metric ton = 1 megagram (1 Mg).

evaporation usually takes place in the storage tank, where the temperature of the solution is kept below 60° C. According to the wide range of parameters the specific volumes of the HLW concentrate may range from 0.4 to more than 1 m3 /МТ of heavy metal reprocessed. A value of

0. 6 m3/MT may be used in the design of reprocessing plants to calculate tank volume requirements.

Chemical composition. In a well-designed reprocessing scheme the amount of process chemicals that would appear in wastes will be kept as small as possible. Then the bulk of dissolved solids includes fission products, uranium and plutonium losses, neptunium, and the transplutonium elements. Table 11.2 shows the amounts of fission-product elements (more than 0.1 percent contribution) present in the waste from reprocessing 1 MT of heavy metal from spent LWR fuel with 30,000 MWd/MT bumup, at the time of and 6 years after discharge from reprocessing. According to widely used design parameters, reprocessing is assumed to take place 150 days after reactor discharge. An aging time of 6 years can presently be envisaged prior to solidification of the liquid HLW. In practice, spent fuel will be aged much longer, and storage of liquid HLW may be shorter.

With a specific HLW concentrate volume of 600 liters/MT of heavy metal, the total fission product concentration will be on the order of 50 g/liter and the actinide concentration on the order of 10 g/liter.

Table 11.2 Amounts of fission-product elements and actinide elements in the waste from 1 MT LWR uranium fuel (30,000 MWd/MT bumup) at discharge from reprocessing (150 days cooled fuel elements) and 6 years after discharge (contributions of more than 0.1 percent) according to ORIGEN

Corrosion products play a minor role and make up to about 1 percent of the total solid content of the HLW solution. If gadolinium is used as homogeneous poison for criticality control, it will significantly increase the total solid content of the waste.

The optimum HN03 concentration for tank storage is in the range of 2 to 4 M, determined by the corrosion behavior of stainless steel. The aqueous raffinate stream (HAW) leaves the extraction process with 1 to З M HN03. A good deal of the nitric acid is stripped in the HAW evaporator or destroyed by a reductant (Chap. 10), and more by subsequent water vapor distillation. The final adjustment may be performed in the storage tank, taking advantage of radiolysis, which removes HN03 effectively when the radiolytic products are swept by air.

Table 11.3 shows typical HAW concentrate data as they have been observed in pilot-plant operations and as they are envisaged for commercial operation.

Radioactivity and heat generation. The total radioactivity and the heat generation of the waste solution up to 100 years is essentially a function of the fission-product concentration and of their age. Then the actinides start to contribute significantly. The computer program ORIGEN [B2] permits the calculation of all relevant data such as radionuclide activities, ingestion hazards, element concentrations, heat generation, and neutron generation for fission products and actinides as a function of age. Figure 11.1 shows radioactivity concentrations and heat generation of the waste from 1 MT of heavy metal with 30,000 MWd/MT burnup up to an age of 106 years. Typical maximum radioactivity concentrations and specific heat generations for a freshly filled tank and reprocessing of 150-day-old fuel are of the order of 103 Ci and 10 W/liter, respectively. After about 500 years radioactivity and heat generation have decayed by a factor of more than 1000. The heat generation will then be insignificant.

Table 11.4 shows the contributions (more than 0.1 percent) of individual fission-product nuclides to the total fission-product activity and of individual actinide nuclides to the total actinide activity. After 6 years only nine fission products or fission-product mother/daughter pairs contribute significantly. After 100 years, 90 Sr/90 Y and 137Cs/l37mBa make up to 98 percent of the fission-product activity, and among the actinides 238 Pu, 241 Am, 243Am/239Np, and 244 Cm are responsible for 90 percent of the total alpha radioactivity.

Hazard indices. The radioactivity of the waste is no direct measure of its radiotoxicity or its hazard. When we assume ingestion as the most likely path of incorporating radioactivity from

waste, we may characterize the hazard by the ingestion hazard index. This is defined for an individual radionuclide as radioactivity divided by the radioactivity concentration limit for drinking water (general public) and has the dimension of a volume. It may be understood as the volume of water required to dilute a given quantity of radioactive material so that drinkable water will be obtained. For a mixture of radionuclides, such as radioactive waste, the ingestion hazard index is the sum of the ingestion hazard indices of all radionuclides present. The hazard index characterizes the potential hazard rather than the actual risk associated with the waste. It does not give credit for the various barriers between waste and humans.

Figure 11.2 shows the ingestion hazard index of HLW as a function of time up to 10® years. After about 500 years the actinide radiotoxicity clearly dominates.

HLW from advanced fuel cycles. Advanced fuel cycles that will be considered are plutonium recycling and the LMFBR fuel cycle. There is little difference from LWR waste as far as fission products are concerned. Actinide concentrations, however, are considerably different, as shown in Chap. 8 for spent fuel.

To discuss other properties of advanced fuel-cycle waste, such as chemical composition, is of little use. As yet, the reprocessing technology for advanced fuel cycles is in a very premature state of development.

□adding hulk. Cladding hulls as collected from the chop-leach head end are radioactive due to activation products in the zircaloy and to fission products and actinides from (U, Pu)02 adsorbed at the inside of the hulls. The principal activation products are 60 Co, 125Sb/12SmTe, and 63Ni. Their total activity is on the order of 103 AiCi/g zircaloy after 6 years and about 100 times lower after 100 years. The fission-product activity after 6 years is of the same order of magnitude and is dominated by 137Cs, 90Sr, and tritium. The residual (U, Pu)02 after leaching is estimated to be of the order of 0.1 percent of the charge.

A number of processes are under development for consolidation and volume reduction of the hulls. Melt densification has received much effort, resulting in volume reductions by a factor of 6. Other processes seriously considered are mechanical compaction and consolidation in concrete. In any case, final disposal of consolidated hulls will be similar to that of other HLW. Thus, hulls (and dissolver sludge) will add considerably to the volume of waste eventually to be disposed of with essentially the HLW technology.

2.1 Projections of HLW Generation

Projections of future waste generation depend strongly on projections of nuclear power generation and of reprocessing capacity. Therefore they have a high degree of uncertainty.

Nudear capacity forecasts have considerably decreased over the last years, and it is presently an open question when reprocessing plants will go on stream. Table 11.5 shows the most recent estimate of the amount of solidified HLW to be accumulated at a federal repository in the United States [B4]. There may be a shift of time scale, but this will not greatly affect the general conclusion from this table: Early in the next century on the order of 104 m3 of solidified waste with on the order of 103 MT of actinides and on the order of 1010 Ci of total radioactivity will be collected in the repository and probably be buried underneath the United States.

To put these numbers into perspective, they should be compared with natural radioactivity already contained in the crust of the earth. We consider a layer of soil all over the United States that is 1 m thick. Its volume is of the order of 1013 m3. With an average uranium concentration of 3 ppm, its uranium content is very roughly З X 107 MT, corresponding to a total radioactivity of 1010 Ci. This corresponds to about 1500 MT of J34U in the 1-m layer, a nuclide whose relative ingestion hazard resembles that of 239 Pu within one order of magnitude.

This simple calculation shows that a 1-m layer of the United States contains about as much long-lived radioactivity and actinides as nuclear industry will put beneath several hundred of those layers within the next 30 years. This illustration does not mean that HLW represents no serious and long-lasting hazard potential, but it emphasizes that the amounts of radioactive material dealt with in waste management are not at all alien to nature.