Natural dysprosium contains a mix of seven stable isotopes 156, 158, 160, 161, 162, 163, and 164 with natural abundances of 0.06%, 0.10%, 2.34%, 18.9%, 25.5%, 24.9%, and 28.2%, respectively. The low abundance of 156Dy and 158Dy are such that only the latter five make a significant contribution to neu­tron capture in a dysprosium poison rod. 164Dy has the highest thermal capture cross-section (Figure 10) and combined with 28.2% initial abundance, it dominates the overall neutron capture rate. On capturing a neu­tron, 164Dy changes to 165Dy, with a small cross­section. The presence of the 160, 161, 162, and 163 isotopes, each of which has a small but nevertheless significant neutron capture cross-section, causes dys­prosium to have a high residual absorption penalty. Moreover, the residual absorption does not burn out because for all the isotopes, a neutron capture event generates the next higher dysprosium isotope and a significant residual cross-section remains.

The first application of dysprosium burnable poi­sons is likely to be in advanced CANDU heavy water moderated reactors. A new fuel design consisting of a bundle of 42 fuel rods arranged in three annular rings around a central uranium oxide/dysprosium

oxide (UO2/Dy2O3) poison rod. The dysprosium plays an important role in helping to ensure a more negative void reactivity coefficient and more negative total power reactivity coefficient, as well as con­trolling the neutron flux shape across the core, while maximizing the power density of the reactor (which in turn has a direct impact on the economics of reactor operation). The void coefficient is the reactivity change in response to steam void formation in the core. The power coefficient is the change in reactivity following a perturbation to reactor power. Both of these coefficients are important in ensuring that the inherent response of the reactor in transient conditions is safe. The presence of a persistent neu­tron absorber, such as dysprosium, helps to remove thermal neutrons in the presence of steam void and thereby helps to maintain a negative void coefficient.

Graphite can survive sudden thermally induced loads (thermal shock), such as those experienced when an arc is struck between the charge and the tip of a graphite electrode in an electric arc melting furnace, or on the first wall of a fusion reactor. To provide a quantitative comparison of a material’s resistance to thermal shock loading, several thermal shock figures of merit (D) have been derived. In its simplest form, the Figure of merit (FoM) may be expressed as

K Sy

a E where K is the thermal conductivity, sy the yield strength, a the thermal expansion coefficient, and E is the Young’s modulus. Clearly, graphite with its unique combination of properties, that is, low ther­mal expansion coefficient, high thermal conductivity, and relatively high strain to failure (s/E), is well suited to applications involving high thermal shock loadings. Taking property values from Table 1 for Toyo Tanso IG-43 and for POCO AXF-5Q gives FoM values of D = 99923 and D = 67 875, respec­tively (from eqn [13]). Another FoM takes account of the potential form of failure from thermally induced biaxial strains, Dth, and may be written as

K Sy

aE(1 — n) where K is the thermal conductivity, sy the yield strength, a the thermal expansion coefficient, E the Young’s modulus, and n is Poisson’s ratio. Larger values of Dth indicate improved resistance to thermal shock. Using the values above and dividing by (1 — n) from eqn [14] gives FoM values of Dth = 124 904 and 84 844 for IG-11 and AXF-5Q, respectively. The thermal shock FoM, Dth, has been reported48 for several candidate materials for fusion reactor first wall materials (see Chapter 4.18, Carbon as a

Fusion Plasma-Facing Material). Wrought beryl­lium has a value of ~1 x 104, pure tungsten a value of ^0.5 x 105, and carbon-carbon composite material ~1 x 106. If the thermal shock is at very high tem­perature, the material’s melting temperature is a key factor. Again, graphite materials do well as they do not exhibit a melting temperature; rather they pro­gressively sublimate at a temperature higher than the sublimation point (3764 K).

2.13.4.1 Thermal Conductivity

It is appropriate to discuss thermal and electrical conductivity as coupled phenomena. Thermal con­ductivity is considered a sum of phonon and electron contributions to conductivity. The phonon contribu­tion to thermal conductivity should decrease with temperature, as atomic vibrations inhibit phonon
transport. The contribution to thermal conductivity due to electrons is calculated by the Wiedemann — Franz law,41 according to

LT

P

where ke is the electronic thermal conductivity, L is the Lorentz constant (2.44 x 10~8W O K~2), T is absolute temperature, and P is electrical resistivity. Generally, electrical resistivity of metals increases with temperature; in transition metal carbides, elec­tron thermal conductivity increases with tempera­ture. At low temperatures heat is mainly conducted by phonons, which are scattered strongly by con­duction electrons.42-44 At intermediate tempera­tures, both electrons and phonons contribute to thermal conductivity, but in the transition metal carbides the electronic component is dominant. Phonon scattering by carbon vacancies becomes important above about 50 K, contributing to a decrease in thermal conductivity with increasing temperature. At high temperatures, thermal conduc­tivity increases approximately linearly with temper­ature. The temperature dependence of electronic thermal conductivity is plotted in Figure 14; this was computed from the Wiedemann-Franz law and

a linear fit to the electrical resistivity measurements of Taylor45 and Grossman46: p = 0.79T + 36.3.

Experimental measurements of thermal conduc­tivity of ZrCx as a function of temperature between

1.8 and 3400 K are also plotted in Figure 14. The overall trend is a steep increase of thermal con­ductivity with temperature up to 50 K, followed by a slight decrease in an intermediate temperature range

F~ 70 60	T	T	T
50
40
30
20
10
0
0

Figure 14 Thermal conductivity of ZrCx as a function of temperature.

(up to 100-1000 K) and then a more gradual increase up to the melting temperature. Room-temperature thermal conductivity has been reported between 20 and 40 W m-1 K-1, meeting or exceeding that of Zr metal.47 A source of experimental scatter in thermal conductivity is sample porosity, which is not always reported by authors.

Room temperature thermal conductivity is also a strong function of C/Zr ratio (Figure 15). Storms and Wagner35 measured thermal diffusivity of hot — pressed ZrC064-1 (0.01-0.1 wt% O) by the laser flash method,48 computing thermal conductivity from sample density and heat capacity according to

k = a dCp [6]

where k is thermal conductivity (W m-1 K-1), a is thermal diffusivity (m2 s — ), d is density of the sample (kgm — ), and Cp is heat capacity (Jkg-1K- ). As described in Section 2.13.3.4, Cp was available for ZrC096 but not for other compositions and Cp versus x was estimated by assuming that it was parallel to that of NbCx. A maximum room temperature ther­mal conductivity of 45 W m-1 K-1 occurs at near­stoichiometric compositions, with a steep drop-off as carbon atoms are removed from the lattice. Further reduction of the C/Zr ratio below approximately

ZrC09 has little effect on thermal conductivity, which approaches a constant value of 10 W m-1 K — .

From a fit to literature electrical resistivity measure­ments and the Wiedemann-Franz law, Storms and Wagner calculated the composition dependence of the electronic component of thermal conductivity as

ke = 1.05 x 103 0.00382 + — [7]

e 55 + 950(1 — x)

where x is the C/Zr ratio, and a Lorenz number of 3.5 x 10-8 V2 K-2 was used (by assuming that the ther­mal conductivity in the low-carbon region was entirely electronic). By taking the difference between their experimentally measured thermal conductivities and their calculated electronic thermal conductivities, Storms and Wagner expressed the phonon thermal con­ductivity as a function of composition by the equation

0.007

(1 — x)2 where x is the C/Zr ratio. As plotted in Figure 15, electronic thermal conductivity is dominant for highly nonstoichiometric ZrCx, while lattice or phonon conductivity makes a larger contribution in near­stoichiometric ZrCx. The effect of a decrease in C/Zr ratio is proposed by Avgustinik et al.49 to reduce the

50
E ‘Й >. >  о га TO c о о га E ф
0.6	0.7		0.9	1.0

connectivity of the lattice while introducing vacancies and increasing the concentration of nonlocalized elec­trons. The net effect is an increase in phonon scattering and a decrease in conductivity with deviation from stoichiometry.

Storms and Wagner also studied the effect on thermal conductivity of tripling the oxygen content in ZrCo.64-o.682 from 0.042 to 0.125-0.13 wt%. They found that thermal conductivity was affected little by varying oxygen content in the low-carbon region but asserted that 0.6 wt% O in ZrC0.92450 produced a more noticeable effect. They suggested that impuri­ties which substitute for carbon (i. e., O or N) reduce the vacancy concentration and have the same effect on thermal conductivity as an increase in C/Zr ratio. The effect of impurities on thermal conductivity is correspondingly more pronounced for ZrC0.9-1.0. Too few measurements of well-characterized near­stoichiometric samples are available to assess this phenomenon more conclusively.

Neshpor eta/.1 measured room-temperature ther­mal conductivity of 85-95% dense sintered ZrC06-0 9 containing 1.4 wt% nitrogen by a steady-state heat — flow method, repeating this study with Avgustinik et a/.49 after decreasing N content to 0.05 wt%. Other room-temperature measurements by heat

flow or thermal diffusivity measurements42,49-55 are
consistent with the trend established by Storms and Wagner, but by covering only one composition, or compositions only below the drop-off at ZrC09, the individual studies fail to capture the true trend.

Figure 530 shows some examples of BWR fuel assem­blies. BWRs have 110-140 mm square full-core height assemblies which, unlike their PWR counter­parts, are contained within thick-walled channel boxes of zirconium alloy. They contain arrays of 6 x 6 to 10 x 10 fuel elements, usually with eight elements acting as tie rods that screw into upper and lower tie plates. Some of the element posi­tions are occupied by unfueled water-filled tubes (called water rods) or water channels and are used to control local flux peaking. Element separation is maintained by grid spacers that are attached to the water rods and evenly distributed along the entire length. The square duct is attached to a top-end fixture, relative to which the remainder of the sub­assembly may slide. The bottom-end fitting has a mechanized orifice to control flow in the subassem­bly and this is located in the core grid plate. The upper end fixture has a handle for loading and unloading against which the hold-down bars rest to prevent levitation.

There are no absorber elements in BWR assem­blies and reactor control is achieved by having cruciform-shaped absorber blades throughout the core which move vertically in the clearance between

GNF

Areva Nuclear fuel GNF2

sets of four subassemblies. Power peaking is mini­mized on the local scale by having fuel elements with different enrichments and burnable poisons (gener­ally Gd2O3) dispersed within each assembly. Various fuel design improvements have been adopted, such as a debris-filtering structure for better reliability, opti­mized distribution of water channels, fissile material with partial length fuel rods and burnable poison use to improve fuel cycle economy and to extend reactor cycle length.

2.15.3.2.2 VVER fuel assembly

Figure 630 shows an example of a VVER fuel assem­bly. The VVER uses hexagonal fuel assemblies of 3200-4690mm length and 145-235mm width. The assembly is used such that it is contained in a hexag­onal shroud, but shroudless assemblies are available for the VVER-1000.30

2.15.3.2.3 CANDU reactor fuel

Figure 730 shows an example of a CANDU fuel bundle. Twelve fuel bundles fit within each fuel chan­nel that is horizontally aligned in the reactor core.

2.15.3.2.4 AGR fuel

AGR fuel assemblies typically have 36 rods contained within a graphite sleeve. Twenty fuel assemblies are placed in a skip inside a flask.

2.15.3.2.5 LWR MOX fuel assembly

Plutonium recycling has so far been limited to partial loading in LWR cores. A primary design target of the MOX fuel assembly is compatibility with the UO2 standard fuel assembly. In the neutronic design for partial loading of LWR cores, significant thermal neutron flux gradients at the interfaces between the MOX and UO2 fuel assemblies have to be considered. The increase in thermal neutron flux in the direction of an adjacent UO2 assembly is addressed by a grada­tion in the plutonium content of the MOX fuel rods at the edges and corners of the fuel assembly. There are three typical rod types for PWR MOX fuel assemblies. Optimized BWR fuel assemblies are more heterogeneous: wider water gaps and larger water structures within a BWR fuel assembly result in MOX fuel assembly designs with an increase in the number ofdifferent rod types. Examples ofMOX fuel assembly designs are shown in Figure 8.2 There are plans for recycling weapons grade plutonium in PWRs in the United States.33

Plenum

spring

Enriched UO2 fuel pellets

Fuel rod

Natural

uranium

axial

blanket

Zirc-4

cladding

Zirconium

diboride

integral

fuel

bundle

absorber

Figure 6 Example Westinghouse VVER-1000 fuel assembly. Reproduced from Tarlton, S., Ed. Nucl. Eng. Int. 2008, 53, 26-36.

The 100% MOX cores permit an increase in the amount of plutonium under irradiation at a reduced level of heterogeneity of the core. An advanced boil­ing water reactor (ABWR) to be constructed in Ohma, Japan, will be the first plant with an in-built 100% MOX core capability.

2.15.3.2.6 FBR fuel assembly

Figure 92 shows an example of an FBR fuel assembly. FBR fuel assemblies have a hexagonal fuel rod arrangement with small gaps provided by a wire spacer, helically wound around each of the fuel pins or by hexagonal grid spacers. The fuel bundle is

Spacer pad (0.8t, 0.6t), □ Bearing pad (1.32t)

^w2 ^

(2 Bearing pad (2 Sheath

(3) End plate

(4) UO2 pellet (2 Spacer pad (2 End plug

encased in a wrapper tube, in order to form a sodium flow channel for efficient cooling and to prevent fuel failure propagation during an accident.

Austenitic or ferritic steels or nickel alloys are selected as materials for structural components because of their good compatibility with sodium and their ability to cope with high temperatures and high levels of fast neutron exposure. These features of FBR fuel assembly design result from the unique design require­ments of the FBRs, including the hard neutron energy spectrum, compact core size, high power density, high burn-up, high temperature, and plutonium breeding. The fuel structure and actual fuel design vary with the reactor scale, design targets, and the design methodol­ogy. Table 3 summarizes the fuel assembly design specifications of the SUPERPHENIX, BN-600, and MONJU.34

2.15.3 Uranium Oxide Production

Uranium oxide has become the primary fuel for the nuclear power industry today. As of April 2010, there
are some 438 commercial nuclear power reactors operating in 30 countries, with a total capacity of 374 000MWe.* Most of these reactors are of the LWRs, AGRs, or the CANDU reactor types, and they are fuelled with sintered pellets of UO2 contain­ing natural or slightly enriched uranium.

2.15.4.1 Uranium Oxide Powder Production

Prior to UO2 pellet fabrication, the enriched uranium feed, UF6, is converted to UO2 powder. Although a number ofconversion processes have been developed, only three are used on an industrial scale today. Two of these are wet processes: ADU and ammonium uranyl carbonate (AUC) and the third is a dry process.

The selected conversion process and its process parameters strongly influence the characteristics of UO2 powder and the resulting UO2 pellets.

2.15.4.1.1 ADU process

The ADU process has been widely used for many years. It uses ADU as an intermediate product in

■ ■ X X X X x|x X X X X X X X ■ ■ ■ X X X X X ■ X • • • X X • • X X X X X • • • l« • X X X X X X X X • • Co) I* • X X X X X X X X • • • • • X X X X X • • X X • • • X ■ X X X X X ■ ■ ■ X X X X >ф< X X X X X X X ■ ■

и Fuel rod, 3.7 wt% Pu m Fuel rod, 5.2 wt% Pu □ Fuel rod, 8.2 wt% Pu m Guide tube @1 Instrumentation tube

Ц 2.5wt% 235U Q2.8wt% Pu I I 3.8 wt% Pu

I I 5.1 wt% Pu 5.1 wt% Pu (part length)

Figure 8 Example light water reactor mixed oxide of uranium and plutonium fuel assemblies. The upper is pressurized water reactor design of the 17 x 17-24 type with a fuel assembly averaged plutonium concentration of 7.2% Pu. The lower is boiling water reactor design of the 10 x 10 -9Q type with a fuel assembly averaged plutonium concentration of 5.4 wt% Pu. Reproduced from IAEA. Status and Advances in MOX Fuel Technology, Technical Reports Series No. 415; IAEA: Vienna, 2003.

a two-step process. First, UF6 is vaporized and injected into an ammonia solution. UF6 hydrolyzes and precipitates as ammonium diuranate (NH4)2U2O7. The ADU precipitate is collected on filters and dried to get the ADU powder.

UF6 + 2H2O! UO2F2 + 4HF

2UO2F2 + 6NH4OH! (NH4)2U2O7 + 4NH4F + 3H2O

Secondly, the ADU powder is calcined and then reduced to UO2 with hydrogen.

(NH4)2U2O7 + 2H2 ! 2UO2 + 2NH3 + 3H2O

The properties of the resulting UO2 are strongly depen­dent on the processing parameters of precipitation,
calcinations, and reduction and equally on material contents, and reacting temperatures. For example, the amount of NH3 is critical in the precipitation step: too much will yield gelatinous ADU which is difficult to filter; if there is too little then the result­ing UO2 powder will be difficult to press and sinter into pellets.

2.15.4.1.2 AUC process37

In Europe, the AUC process is widely used for fabri­cating UO2 fuels. The precipitation of AUC is done in a precipitator, filled with demineralized water. The vaporized UF6, CO2, and NH3 are added as gases through a nozzle system. Reaction occurs according to the following equation:

UF6 + 5H2O + 10NH3 + 3CO2 ! (NH4)4{UO2(CO3)3g + 6NH4F

Fuel assembly Fuel pin

The AUC precipitates in the form of yellow single crystals. The grain size depends on the precipitation conditions. Instead of UF6, uranyl nitrate solution can also be used as a feed material.

The AUC precipitate is filtrated and washed with a solution of ammonium carbonate and methyl alco­hol. Then, the AUC powder is pneumatically trans­ferred to a fluidized-bed furnace, decomposed, and reduced to UO2 with hydrogen according to the following equation.

(NH4)4{UO2(CO3)3} + H2

! UO2 + 4NH3 + 3CO2 + 3H2O

The transformation of AUC to UO2 gives rise to desirable UO2 powder properties: it is free-flowing and has a high sintering activity.

The resulting UO2 powder is made chemically stable by a slight oxidation to about UO210.

2.15.4.1.3 Dry process38

The dry process was developed in the late 1960s and is widely used today. UF6 is vaporized from steam or hot-water-heated vaporizing baths, and vaporized UF6 is introduced into the feed end of a rotating kiln. Here, it meets and reacts with superheated steam to give a plume of uranyl fluoride (UO2F2). UO2F2

passes down the kiln where it meets with a counter­current flow of steam and hydrogen and is converted to UO2 powder. The reaction sequence follows the equations below.

UF6 + 2H2O! UO2F2 + 4HF
4UO2F2 + 2H2O + 2H2 ! U3O8 + UO2 + 8HF

U3O8 + 2H2 ! 3UO2 + 2H2O

The UO2 powder resulting from dry processes is of low bulk density and fine particle size. Therefore, granulation before pressing and the employment of a pore former process are usual during the pellet fabri­cation process.

A dry process has preferable advantages: the pro­cess is simple and the equipment is compact; the criticality limitation is less required; and liquid waste treatment is not necessary.

Because of the complex characteristics of the irra­diated fuel, the thermal conductivity is often deduced from correlations using global parameters summariz­ing the state of the fuel: for instance, the burnup and the irradiation temperature. Some fuel characteristics are implicitly taken into account. For instance, the grain size is a parameter having a small impact on the conductivity of fresh fuels but a large impact on the conductivity of irradiated fuels. This is because grain boundaries are sinks for point defects and there­fore a smaller grain size induces reduced concentra­tions of point defects.

2.17.2.3.1 Irradiation temperature

The irradiation temperature has an impact on the state of the fuel. During long irradiations, tempera­ture has an effect on the microstructure of the fuel and on the concentration of radiation damage accu­mulated. During a short annealing (transient or laboratory annealing), the effect is mainly restricted to the radiation damage concentration change and to the redistribution of some fission products: pre­cipitation of atoms that were distributed as single atoms and formation of fission gas bubbles. How­ever, the definition of this parameter is vague because the irradiation temperature at a given radial position is not constant throughout the irradiation and the real relevant parameter is the irradiation temperature history. The out-of-pile measurements must be interpreted by considering the irradiation temperature at end of life (EOL).

J. Lamon

CNRS/National Institute of Applied Science, Villeurbanne, France © 2012 Elsevier Ltd. All rights reserved.

SEP Societe Europeenne de Propulsion SiC/SiC SiC matrix composite reinforced by SiC fibers

2.12.1 Introduction

Silicon carbide is composed of tetrahedra of carbon and silicon atoms with strong bonds in the crystal lattice. This produces a very hard and strong ceramic with outstanding characteristics such as high thermal conductivity, low thermal expansion, and exceptional resistance to thermal shock and to corrosion in aggressive environments at high temperatures. How­ever, this implies a few inadequate characteristics for structural applications, such as low fracture tough­ness, high sensitivity to the presence of microstruc­tural flaws, brittle behavior, and lack of reliability. Reinforcing with continuous SiC-based fibers allows these weaknesses to be overcome. The composite SiC/SiC that is obtained is damage tolerant, tough, and strong, and it can be insensitive to flaws and notches. The concept of composite material is very powerful. Composites can be tailored to suit end — use applications through the sound selection and arrangement of the constituents. Ceramic matrix com­posites (CMCs) reinforced with continuous ceramic or carbon fibers are of interest in thermostructural applications.1-4 They are lightweight and damage tol­erant and exhibit a much greater resistance to high temperatures and aggressive environments than metals or other conventional engineering materials.

CMCs can be fabricated by different processing techniques, using either liquid or gaseous precursors. The chemical vapor infiltration (CVI) method can produce excellent SiC/SiC composites with a highly crystalline structure and excellent mechanical prop — erties.5 The quality of the material obtained by the polymer impregnation and pyrolysis (PIP) method is insufficient. A novel processing technique (nanopow­der infiltration and transient eutectic-phase proces­sing, NITE) was claimed to achieve good material quality.5-7

The SiC/SiC composites prepared using the CVI method and reinforced with the latest near­stoichiometric SiC fibers (such as Hi-Nicalon type S and Tyranno-SA3 fibers) appear to be promising candidates for nuclear applications7-12 because of their high crystallinity, high purity, near stoichiome­try and radiation resistance of the р-phase of SiC, as well as excellent resistance at high temperatures to fracture, creep, corrosion, and thermal shock. Studies on the р-phase properties suggest that CVI SiC/SiC composites have the potential for excellent radiation stability.3 CVI SiC/SiC is also considered for ap­plications as structural materials in fusion power reactors because of low neutron-induced activation characteristics coupled with excellent mechanical properties at high temperature.1

The CVI technique has been studied since the 1960s.13-19 It derives directly from chemical vapor deposition (CVD).13-15 In very simple terms, the SiC-based matrix is deposited from gaseous reactants on to a heated substrate of fibrous preforms (SiC).15 CVI is a slow process, and the obtained composite materials possess some residual porosity and density gradients. Despite these drawbacks, the CVI process presents a few advantages: (1) the strength of reinfor­cing fibers is not affected during the manufacture of the composite; (2) the nature of the deposited mate­rial can be changed easily, simply by introducing the appropriate gaseous precursors into the infiltration chamber; (3) a large number of components; and (4) large, complex shapes can be produced in a near-net shape.

Development of CVI SiC/SiC composites began in the 1980s when SEP (Societe Europeenne de Propul­sion), Amercorm, Refractory Composites, and others began to develop equipment and processes for produc­ing CVI components for aerospace, defense, and other applications. The development of CVI SiC/SiC com­posites has been inspired by the poor oxidation resis­tance of their predecessor CVI C/C composites. CVI SiC/SiC components have been produced and tested. SNECMA (formerly SEP) is at the forefront of this technology and has demonstrated satisfactory compo­nent performance in engine and flight tests.

The mechanical properties of SiC/SiC compo­sites depend on the fiber-matrix interface. Pyrocar — bon (PyC) has proved to be an efficient interphase to control fiber-matrix interactions and composite mechanical behavior.20 But PyC is sensitive to oxida­tion at temperatures above 450 °C. A few versions of high-temperature-resistant CVI SiC/SiC composites have been produced. In order to protect the PyC interphase against oxidation, multilayered inter­phases and matrices have been developed.3,21 Multi­layered matrices contain phases that produce sealants at high temperatures, preventing oxygen from reach­ing the interphase.22 This composite is referred to as CVI SiC/Si-B-C. Oxidation-resistant interphases such as BN or multilayered materials can also be coated on the fibers. An ‘oxygen getter’ can be added to the matrix to scavenge oxygen that might ingress into the matrix (enhanced CVI SiC/SiC).

The mechanical behavior of CMCs displays sev­eral typical features that differentiate them from the other composites (such as polymer matrix compo­sites, metal matrix composites, etc.) and from homo­geneous (monolithic) materials. These features are due to heterogeneous and multiscale composite microstructure and the respective properties of the constituents (interphases, fiber, and matrix). The main characteristics of CVD SiC, CVI SiC/SiC, and NITE-SiC/SiC are reviewed in this chapter. Features of mechanical behavior of SiC/SiC are dis­cussed with respect to microstructure, on the basis of the large amount of work done on CVI SiC/SiC.

The measurement of other properties of irradiated ZrC is limited and often contradictory. Some evi­dence for an increase in mechanical strength with irradiation is available. Andrievskii eta/.164 irradiated sintered ZrC098 at 423 and 1373 Kwith a fast neutron fluence of 1.5 x 1020 cm—2, and found more substantial strengthening at low-temperature irradiation than at high temperature (bend strength increased by 28% vs. 4%, microhardness by 12% vs. 7.3%, and Young’s modulus by 1.2% vs. no increase). Yang eta/.175 irra­diated hot-pressed commercial ZrC0.99 at 1073 K with 2.6 MeV protons to a fluence of 1 x 1019 or

2.3x 1019cm—2 (0.7 or 1.5 dpa), and found Vickers hardness increased after irradiation, with a slightly more pronounced increase at higher fluence (12% increase at 0.7 dpa vs. 14% increase at 1.5 dpa). Inden­tation toughness also increased 79% after 1.5 dpa, but scatter was large. Absent from the literature are stud­ies of irradiation-induced creep of ZrC.

Electrical and thermal conductivity, sensitive to defect concentration, have also been studied. In general, electrical resistivity was found to increase with irradia­tion and thermal conductivity to degrade. Koval’chenko and Rogovoi165 irradiated ZrC098 at 323 Kwith a ther­mal neutron fluence of 1 x 1019—1.5 x 1020cm—2, and resistivity increased by 17-167%, increasing with flu­ence, versus an unirradiated 60 pQ cm. The authors attribute the increase to point defect formation, but low initial lattice parameter suggests high O and N impurity content in any case. Following the same irradiation by Andrievskii eta/.164 described in the pre­ceding paragraph, resistivity was found to increase, with the effect less pronounced for higher irradiation tem­peratures: a 481% increase was measured at 423 K, and a 51% increase at 1373 K, versus an unirradiated 43 pQ cm. In unirradiated ZrC*, resistivity increases as the C/Zr ratio decreases, and Andrievskii eta/.166 found that the increase in resistivity following irradiation at 413 K in a fast neutron fluence of 1 x 1019cm—2 was more pronounced for compositions closer to stoichiom­etry. A 6% increase in resistivity was measured for ZrC07 versus a 213% increase for ZrC0 94.

Thermal conductivity was studied by David eta/.178 following irradiation at 298 K with 28.5 MeV Kr ions to a fluence of 1 x 1016 or 6 x 1016cm—2. The authors distinguished between thermal conductivity degrada­tion due to inelastic and elastic collisions, with inelastic damage in ZrC calculated to occur in the first 3.3 pm into the surface and elastic damage initiating at a depth where dpa increases to 20% of the maximum damage, continuing for 1.4 pm below the inelastic damage. A modulated thermoreflectance microscopy technique was employed to characterize subsurface thermal con­ductivity degradation. Elastic collisions were deemed considerably more damaging than inelastic, reducing thermal conductivity from 20 W m—1 K—1 before irra­diation to less than 1 W m—1 K—1. In the inelastic dam­age region, thermal conductivity of 10 W m—1 K—1 was measured after 1 x 1016 cm—2 fluence and 5 W m—1 K—1 after 6 x 1016cm—2 fluence.

In the United Kingdom, over the past 25 years, exten­sive work has been carried out on the manufacture of MOX fuel under the support of the UK Fast Reactor Development Program.51

Based on these experiences, the SBR process was developed by the British Nuclear Fuels plc (BNFL) to fabricate MOX pellets for LWRs. The process was originally developed in the 1980s by BNFL-UKAEA (United Kingdom Atomic Energy Authority). Figure 22 shows the flow sheet for the SBR process.

In the SBR process, three kinds of feed materials, PuO2 powder prepared by the oxalate precipitation method, UO2 powder prepared by the ADU process, and dry recycled scrap powder are prepared to get the desired plutonium concentration in the initially mixed powder. These powders are milled completely using an attritor mill (a photograph is shown in MacLeod and Yates51), an off-the-shelf mill widely used in the pharmaceutical industry. The attritor mill provides good blends with a homogenized plutonium distribution in a short blending time and can be operated continuously.6 The milled MOX powder must be granulated in order to provide a free-flowing, dust-free feed to the pelletizing press to ensure uniform die filling and good compaction.51 In the milling step, the lubricant and Compo pore former are added in order to control the pellet density and obtain character­istics similar to those of the UO2 pellets produced by BNFL from IDR UO2 powder.66 In order to condition the milled MOX powder to form granules prior to pelletizing and sintering, a spheroidizer is introduced instead of the precompaction granulation equipment commonly used.6 The spheroidizer is used in a powder agglomeration process and was invented by SCK’CEN (Studiecentrum voor Kernenergie — Centre d’Etude de l’energie Nucleaire) in the 1970s to fabricate a fuel kernel, the pit of coated particles fuelling high temper­ature reactors.6

In the SBR process, the binder that is commonly used in the conventional MOX fuel manufacturing process is not used. As a result, the dewaxing step of the green pellets prior to sintering is not needed and the process is similar to the current UO2 fuel fabrica­tion process in this respect. The processing time is short and the equipment can be stacked so that the powder can be discharged by gravity from the feed dispensing and dosing glove box through the proces­sing equipment into the hopper of the pelletizing press. The simple sequence of one attritor mill and one spheroidizer, utilized in the Manufacturing Demonstration Facility, was made more sophisticated for the Sellafield MOX Plant by the addition of one homogenizer and one more attritor mill.68 This expansion allowed the size of the powder lot to be increased from 50 kg MOX to 150 kg MOX with additional benefits such as reducing the number of quality control points and operating with a larger quantity of fuel with uniform plutonium isotopic composition.6 After conditioning in the spheroidizer, the powder is pelletized into green pellets using a hydraulic multipunch press, and then green pellets are sintered at temperatures of up to 1750 °C under an atmosphere of Ar + 4% H2 mixture gas without heat treatment in a dewaxing furnace.67 An automatic pellet inspection system is adopted for monitoring each pellet diameter, pellet surface, and end surfaces after centerless grinding.51 The MOX pellets pro­duced by the SBR process have a mean grain size of about 7.4 pm with a standard deviation of 0.6 pm, and mean pore diameter is about 5 pm.68

With regard to general corrosion and oxidation, stainless steels with 16-18% Cr passivate and have good resistance to aqueous corrosion and various types of other acidic or corrosive environments at room temperature and up to about 200-300 °C.2 Additions of molybdenum give type 316 better resis­tance to pitting and acidic attack. Effects of stress can aggravate corrosion resistance, and types 304 or 316 processed to have Cr-carbides precipitated along grain boundaries can suffer from stress-corrosion­cracking (SCC), which causes grain-boundary cracking at reduced ductility to embrittle the steel. Lower carbon steels (304LN, 316L) tend or reduce or elimi­nate SCC, as do the stabilized stainless steel grades such as 321 and 347, which form TiC or NbC carbides to prevent Cr-carbide precipitation at grain bound­aries. Exposure to supercritical water at 300 °C and above can be very corrosive, and cause oxidation of
austenitic stainless steels.16 Generally, 300 series aus­tenitic stainless steels have minimal oxidation in air at 500 °C and below, but oxidation and the protective behavior of chrome-oxide scales become a concern at 550-600 °C and above. Finally, 300 series steels such as types 304 and 316 tend to show little or no corrosion and behave quite well in liquid-metal sodium envir­onments at 650 °C and below. More detailed informa­tion on austenitic stainless steels and their corrosion behavior in aqueous environments, oxidation at ele­vated temperatures, and behavior in liquid metals such as sodium is available in other chapters of this publication, or elsewhere.

Ultimate failure generally occurs after saturation of matrix cracking. The fibers break when the applied load is close to the maximum. Matrix damage and ultimate failure thus appear to be successive phenomena.

The ultimate failure of a tow of parallel fibers involves two steps:

• a first step of stable failure and

• a second step of unstable failure.

During the first step, the fibers fail individually as the load increases. In the absence of fiber interactions, the load is carried by the surviving fibers only (equal load sharing). Fiber interactions cause tow weaken­ing. The ultimate failure of a tow (second step) occurs when the surviving fibers cannot tolerate the load increment resulting from a fiber failure. At this stage, a critical number of fibers have been broken.

The ultimate failure of a longitudinal tow coated with matrix also involves a two-step mechanism and global load sharing when a fiber fails. In the presence of multiple cracks across the matrix and associated interface cracks, the load-carrying capacity of the matrix is tremendously reduced or eliminated. The matrix-coated tows behave like dry tows subject to the typical stress field generated by the presence of matrix cracks. The ultimate failure of a matrix-coated tow occurs when a critical number of fibers have failed. This mechanism operates in the tows within textile CVI SiC/SiC composites. The ultimate failure of the composite is caused by the failure of a critical number of broken tows (>1) depending on the stress state: ~1 under an axial tension, >1 in bending.

It is worth pointing out that the failure mechanism ofCVI SiC/SiC composites differs from that observed in polymer matrix impregnated tows, where local load sharing prevails when a fiber fails. In these composites, the fibers fail first. Therefore, the uncracked matrix is able to transfer the loads.

2.12.6.2 Reliability

The ultimate failure of CVI SiC/SiC composites is highly influenced by stochastic features. As fibers are brittle ceramics, they are sensitive to the presence of flaws (stress concentrators) that are distributed randomly. As a consequence, the strength data exhibit significant scatter, as illustrated by Figure 7.39’40 The figure shows that the magnitude of the strength and scatter decrease from single fibers to tows, then to infiltrated tows, and finally to woven composites.

0. 012 0.01 0.008

0­0.006

M

0.004

Q

0.002 0

0 500 1000 1500 2000 2500

Stress (MPa)

Figure 7 Strength density functions for SiC fibers (NLM 202), SiC fiber tows, SiC/SiC (1D) minicomposites, and 2D SiC/SiC composites.

As a result of the previously mentioned two-step failure mechanism, the ultimate failure of an entity is dictated by the lowest extreme of the strength distri­bution pertinent to its constituent: that is, tows versus filaments, infiltrated tows versus fibers, and 2D com­posites versus infiltrated tows. The lowest strength extremes correspond respectively to the critical num­ber of individual fiber breaks («17% for the SiC Nicalon™ fibers and for the SiC Hi-Nicalon™ fibers) and to the critical number of tow failures (>1). The gap between tows and SiC infiltrated tows results from the method ofstrength determination: the critical number of individual fiber breaks was taken into account for tow strength determination, whereas the strength of infiltrated tows and composites was under­estimated because the total cross sectional area of the specimens was used.

The flaw populations are truncated during the successive damage steps, which leads to a homoge­neous ultimate population of flaws.40 This process of progressive elimination of flaws governs the trends in the ultimate failure. The tensile stress-strain curves obtained on a batch of several CVI SiC/SiC test specimens coincide quite well (Figure 5), whereas the strength data exhibit a certain scatter (Figure 5). This scatter is limited (Figure 8). Dependence of composite strength on the stressed volume is not significant (Figure 8). Furthermore, dependence on the loading conditions is not so large (Figure 9): for instance, the flexural strength is 1.15 times as large as the tensile strength40,41 when measured on specimens having comparable sizes (Figure 9).

The Weibull model is not appropriate to describe the volume dependence of strength data,40 as the weakest link concept is violated. However, the

1

Ulti

permission from Springer.

Weibull modulus (m) can be extracted from the sta­tistical distribution of the strength data: m is in the range of 20-29. This value provides an evaluation of the scatter in strength data. It reflects a small scatter.