The lifetime of a fast reactor fuel pin is most strongly determined by the internal creep rupture strength of the cladding induced by the internal pressure of the fission gas at a temperature of around 700 °C. For 9Cr-ODS steel cladding, internal creep rupture data at 650, 700, and 750 °C are shown in Figure 15.30 Additionally, the best fit lines for hoop stress versus rupture time at each temperature are shown by solid lines. These results confirmed that creep rupture strengths in the hoop and longitudinal directions of cladding are almost the same, due to their equi-axed grains. The corresponding creep rupture curves for HT931 and austenitic PNC31632 are also presented for comparison. PNC316 is a typical austenitic clad­ding developed by JAEA in the fast reactor program. Notably, superior performance in rupture time is shown in 9Cr-ODS steel cladding. The slope of PNC316 is steeper, and there is a cross-over before 1000 h at 750 °C and before 10 000 h at 700 °C. The stress condition of the fast reactor fuel pin gradually increases due to the accumulation of fission gases and reaches around 120 MPa at its final service milestone of 75 000 h at 700 °C. In this stress range, it is obvious that 9Cr-ODS steel cladding is of advantage.

The ultimate tensile strength (UTS) of 9Cr-ODS ferritic cladding in the hoop direction as measured in a temperature range from room temperature to 850 °C, is shown in Figure 16, together with the corresponding data for the ferritic-martensitic stain­less steel (PNC-FMS)19 that is conventionally used as fast reactor fuel cladding. The strength of 9Cr-ODS steel is superior to that of conventional PNC-FMS. The uniform elongation that takes place from room temperature to 900 °C is also shown in Figure 16. In the temperature range from 400 to 700 °C at which a fast reactor is commonly operated, the measured uniform elongation exhibits adequate ductility. This advantage of superior elongation in the produced clad­dings can probably be ascribed to the pinning of dis­locations by oxide particles, which retard recovery and sustain work-hardening.

The invention of an electric furnace2 capable of reach­ing temperatures approaching 3000 °C by Acheson in 1895 facilitated the development of the process for the manufacture of artificial polygranular graphite. Detailed accounts of the manufacture of polygranular graphite may be found elsewhere.2-4 Figure 2 sum­marizes the major processing steps in the manufacture of nuclear graphite. Nuclear graphite consists of two phases: a filler material and a binder phase. The pre­dominant filler material, particularly in the United States, is a petroleum coke made by the delayed coking process. European nuclear graphites are typi­cally made from a coal-tar pitch-derived coke. In the United Kingdom, Gilsonite coke, derived from natu­rally occurring bitumen found in Utah, USA, has been used. Both coke types are used for nuclear graphite production in Japan. The coke is usually calcined (thermally processed) at 1300 °C prior to being

crushed and blended. Typically, the binder phase is a coal-tar pitch. The binder plasticizes the filler coke particles so that they can be formed. Forming pro­cesses include extrusion, molding, vibrational mold­ing, and isostatic pressing. The binder phase is carbonized during the subsequent baking operation (800-1000 °C). Frequently, engineering graphites are pitch impregnated to densify the carbon artifact, fol­lowed by rebaking. Useful increases in density and strength are obtained with up to six impregnations, but two or three are more typical.

The final stage of the manufacturing process is graphitization (2500-3000 °C) during which, in sim­plistic terms, carbon atoms in the baked material migrate to form the thermodynamically more stable graphite lattice. Nuclear graphites require high chemical purity to minimize neutron absorption.

Moreover, certain elements catalyze the oxidation of graphite and must be reduced to an acceptable level. This is achieved by selecting very pure cokes, utilizing a high graphitization temperature (>2800 °C), or by including a halogen purification stage in the manufac­ture of the cokes or graphite. Recently, comprehensive consensus specifications5,6 were developed for nuclear graphites.

The electronic hybridization of carbon atoms (1s2, 2s2, 2p2) allows several types of covalent bonded structure. In graphite, we observe sp2 hybridization in a planar network in which the carbon atom is bound to three equidistant nearest neighbors 120° apart in a given plane to form the hexagonal graphene structure. Covalent double bonds of both s-type and я-type are present, causing a shorter bond length than in the case of the tetrahedral bonding (s-type sp3 orbital hybri­dization only) observed in diamond. Thus, in its per­fect form, the crystal structure of graphite (Figure 3) consists of tightly bonded (covalent) sheets of carbon atoms in a hexagonal lattice network.7 The sheets are

weakly bound with van der Waals type bonds in an ABAB stacking sequence with a separation of 0.335 nm.

The crystals in manufactured polygranular graph­ite are less than perfect, with approximately one layer plane in every six constituting a stacking fault. The graphite crystals have two distinct dimensions, the crystallite size La measured parallel to the basal plane and the dimension Lc measured perpendicular to the basal planes. In a coke-based nuclear graphite, values of La ~ 80 nm and Lc~ 60 nm are typical.8 A combination of crystal structure bond anisotropy and texture resulting from forming imparts aniso­tropic properties to the filler coke and the manufac­tured nuclear graphite. The coke particles become preferentially aligned during forming, either with their long axis parallel to the forming axis in the case of extrusion, or with their long axis perpendicu­lar to the forming axis in the case of molding or vibrational molding. Consequently, the graphite arti­facts are often attributed with-grain and against-grain properties as in the American Society for Testing and Materials (ASTM) specifications.5,6 The degree of isotropy in manufactured graphite can be controlled through the processing route. Factors such as the nature of the filler coke, its size and size distribution, and the forming method contribute to the degree of isotropy. Nuclear graphites are typically medium or fine grain graphites (filler coke size <1.68 mm)5,6 and are considered near-isotropic. Fine grain graphites (grain sizes <100 pm) formed via isostatic pressing often exhibit complete isotropy in their properties.

In response to the recent renewed interest in high — temperature gas-cooled reactors, many graphite ven­dors have introduced new nuclear graphites grades. Table 1 summarizes some ofthe grades available cur­rently, although this list is not exhaustive. The graphite manufacturer is listed along with the coke type and comments related to the given graphite grade.

It would not be appropriate to continue without some discussion on stored (or Wigner) energy. The perfect crystal configuration is the lowest energy state for the graphite lattice. However, irradiation damage will considerably alter that configuration. Wigner38 predicted that the increased lattice vibra­tion due to heating would allow carbon atoms to rearrange themselves into lower energy states, and that in doing so energy would be released in the form of heat. Early experience in operating graphite­moderated plutonium production and research reac­tors at low temperatures in the United States, Russia, France, and the United Kingdom proved that this assumption was correct. The highest value of stored energy measured was ^2700Jg-1.15 If all of this were released under adiabatic conditions, the temperature rise would be 1500 °C. Fortunately, that is not the case. Furthermore, the accumulation of stored energy is insignificant above an irradiation temperature of ^300 °C, it is difficult to accidentally release the stored energy above an irradiation

Displacement cascade

Vacancy

loop /———— v

____________ v

Figure 15 Formation of interstitial and vacancy loops in graphite crystals. Modified from Simmons, J. Radiation Damage in Graphite; Pergamon: London, 1965.

temperature of ~-150 °C, and only limited self-sus­taining energy release of stored energy can be achieved in graphite irradiated below ^100 °C. Thus, stored energy is now of consideration in the United Kingdom only in the decommissioning of shutdown reactors such as the Windscale Piles and BEPO and other similar overseas systems, although there are graphite ‘thermal columns’ in some research reactors that may require periodic assessment.

The reason for this is the nature of the irradia­tion damage sites with respect to irradiation tempera­ture. In graphite irradiated in the early facilities,
at temperatures between about ambient and 150 °C, point defects associated with Frenkel pairs and small loops can diffuse only slowly through the lattice to form larger, more stable loops because of the low irradiation temperature. However, thermal annealing at temperatures above the irradiation temperature can readily release the stored energy, and under certain circumstances, this release can be self-sustaining over certain temperature changes. (A ‘rule of thumb’ tem­perature of 50 ° C above the irradiation temperature is often cited as a ‘start of release temperature.’ How­ever, this is misleading as a heat balance needs to be

900

700

^ 600

>s

P

500

c

(D

ф 400

о

oo

пі 300

£

200

100 0

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6

Fluence (1020 ncm-2 nvt)

Figure 16 The accumulation of stored energy as a function of fluence and temperature.

considered when assessing energy release rates. Thus, 50 °C above the irradiation temperature can be con­siderably overconservative.)

The accumulation of stored energy, measured by burning irradiated graphite samples in a bomb calo­rimeter, is given as a function of fluence and temper­ature in Figure 16. At low fluence, stored energy quickly accumulates reaching a plateau at high flu­ence. Many measurements were made in the Wind — scale Piles, BEPO, Hanford, and Magnox reactors that clearly illustrated this behavior.15

To fully understand the thermal stability of graph­ite containing stored energy, the most appropriate measure is the rate of release of stored energy measured using a differential scanning calorimeter (DSC) as illustrated in Figure 17.

A graphite sample is heated in the DSC usually at a constant rate of 2.5 °Cmin~ In simple terms, two runs are made and the heat capacity of the samples measured in each case. When the heat capacities from the two runs are subtracted, the energy release rate is easily obtained as a function of heating temperature. This can be compared to the specific heat of graphite as given in Figure 17. When the rate of release of energy is below the specific heat, energy needs to be added to continue the process. When the rate of release is above the specific heat, the process is self-sustaining. This behavior was used to ‘anneal’ the Windscale Piles; a ‘hit and miss’ strategy that ended in damage to the fuel cartridges and eventually a
‘metal uranium fire.’ (Contrary to ‘common folklore,’ the graphite did not burn in the Windscale incident. A limited amount of graphite was oxidized leading to enlargement of fuel and control channels but it was the metal uranium that burnt. Graphite is very diffi­cult to burn and requires large amounts of heat and oxygen or air, applied to crushed graphite in a flui­dized bed or in similar form.39)

The form of this rate of release curve is a func­tion of (1) the amount of stored energy in the sample, (2) the temperature the sample was irra­diated at, (3) the fluence the sample had been irradiated to, (4) the release temperature, and (5) the heating rate. Unfortunately, there are no com­prehensive datasets of these five parameters that allow a robust empirical model to be derived for assessing the stability of graphite containing stored energy. The models that usually exist take the worst — case rate of release curve and fit an Arrhenius type equation to the rate ofrelease curve.

where S is the stored energy remaining, t is time, T(t) is temperature in (K) as a linear function of time (T = at in the case of the DSC test and is nonlinear in most practical cases), Kis Boltzmann’s constant, and E(T S) is the activation energy as a function of the stored energy remaining and temperature and u is a frequency factor usually taken as 7.5 x 1013 s~140

3.0 1

2.5 —

0 7

U)

3,

2.0 —

Ф

c Ф "O

Ф

1.5 —

Vi

4—

О Ф

CO. „

1.0 —

Ф

(D

ч—

О Ф

°.5

DC

0.0

0 50 100 150 200 250 300 350 400 450 500

Temperature (°C)

It can be appreciated that the exact solution of eqn [30] requires a substantial amount of information from several rate of release curves from several samples, which is seldom available. Thus, a practical approach is usually taken, the simplest of which is to assume a single activation energy. However, this is not very satisfactory and more elegant approaches using vari­able or discrete activation energies can be found.41- Having derived a satisfactory model for the rate of release using a DSC, it then can be applied to a practi­cal situation using commercially available computer codes such as ‘user subroutine’ facilities.42

In assessing practical situations, it is important to use an energy balance that accounts for heat applied, heat generated by the release of stored energy, the heat capacity of the graphite itself, and heat lost to the surroundings. If the heat generation is intense and oxygen is available, the heat generated by graphite oxidation should be taken into account. However, the latter case should be unnecessary as a professional scientist or engineer would not design a system or process that would approach such conditions. It should be noted that irradiated graphite thermal conductivity and total stored energy are directly cor­related15; see eqn [31]. Therefore, the thermal con­ductivity will improve as energy is released.

1 Jg-1

The data that eqn [30] is based on is derived from rate of release curves obtained using a relatively fast heating rate. In dealing with irradiated graphite waste, much slower rates of heating are often required. Graphite samples taken from the Windscale Piles 40 years after the incident showed little change in the dS/dt curves,45 indicating that diffusion of atoms at around ambient temperature is extremely slow. Nevertheless, condi­tions relevant to any proposed encapsulation technique and repository will need to be accounted for in deter­mining if heat released from stored energy is an issue.

The rate of release curves given in Figure 17 are only to a temperature of around 450 °C. It had been observed, by comparing the energy released in the DSC with the energy released on a similar sample in a bomb calorimeter, that not all of the stored energy had been released in the samples heated to a maxi­mum of 450 °C in the DSC. It was found that on increasing the temperature to around 1600 °C, a sec­ond peak could exist46,47; see Figure 18.

It was observed that the ‘200 °C’ peak reduced in size and moved to a slightly higher temperature with increased irradiation, presumably as the irradiation induced defects became more stable, and the plateau between the two peaks increased in height and approached the specific heat value.15 The first of these phenomena could explain why it became more and more difficult to ‘anneal’ the Windscale Piles39 and the second had the implication that

eventually the rate of release curve would remain above the specific heat up to 1600 °C with the conse­quent safety implications. Fortunately, the second of these phenomena proved to be incorrect.

It is interesting to note that there is a correla­tion (eqn [32]) between the height of the plateau at ^400 °C and total stored energy.15

S

J g

1670

This equation, although not exact, was often used in reactor graphite sampling programs to avoid having to measure total stored energy.

Most of the discussion on stored energy above is relevant only to low temperature reactor systems, with graphite temperatures operating from ambient to 150 0C. When graphite is irradiated at higher tem­peratures, in practice above about 100 0C, the dS/dt does not exceed the graphite specific heat. One of the operating rules for the UK Magnox reactors was that the dS/dt, as measured on surveillance samples, should always be below 80% of the specific heat, which proved to be the case.

The performance of structural materials can strongly influence the blanket design. Especially, the opera­tion temperature window and expected lifetime are the key parameters. Increase in the upper operation temperature limit can enhance the blanket operation temperature and thus plant efficiency. Therefore, enhancing the high-temperature strength is the key issue for improving the performance of the blanket and thus the attractiveness of the fusion power sys­tems. For this purpose, efforts have been made to develop advanced vanadium alloys with potential use at higher temperature.

One of the relatively simple ways to enhance the strength of the alloy is to change the thermal and mechanical treatment of the alloys. Especially, for­mation of a high density of precipitates can strengthen the alloy. Figure 21 shows microstructure and hardness of V-4Cr-4Ti as a function of the temperature of reheating for 1 h after annealing at 1373 K for 1 h. The annealing at 1373 K dissolves most of the thin precipitates and the reheating can form new precipitates. By choosing an appropriate reheating temperature (873-973 K), the materials can be strengthened by the high density offine precipitates. However, the strengthening by this treatment will be lost at >973 K because of the coarsening of the pre­cipitates. To prevent the coarsening, cold work was applied to the specimens. Figure 22 shows the mini­mum creep rate for standard V-4Cr-4Ti and solution annealed, aged, and cold-worked V-4Cr-4Ti. Sup­pression of the creep rate occurred at 1073 K but only with relatively high stresses.45 Microstructural analysis showed that the suppressive role of cold — work-induced dislocations was lost during the creep deformation by the change in the nature of the dis­locations from sessile й(100) type to gliding й/2(111) type.46 Further efforts are being made, for example, to cold-work followed by aging (strain-aging-induced strengthening).

High-temperature strength of V-Cr-Ti alloys can be enhanced by increasing the Cr level. However, high Cr alloys have low ductility and fabricability issues. Recent detailed survey in V-xCr-4Ti alloys showed that the strength at high temperature increases with a small change in the DBTT with the Cr level at ^7%.47

High-strength vanadium alloys were made by addi­tion of Y, O, and N to vanadium followed by mechani­cal alloying (MA) and hot isostatic pressing (HIP). The addition of Y, O, and N was intended to enhance mechanical properties by dispersion of Y2O3 and YN and scavenging O and N from the matrix. Alloys pro­duced by optimization of the processes had small grains and homogeneously dispersed particles and showed higher tensile strength than those of NIFS-HEATs with moderate uniform elongation, both at room tem­perature and 1073 K as shown in Figure 23 48 Fine

10

grain and oxide dispersion increased high-temperature strength and inhibited formation of interstitial loops in the matrix by neutron irradiation because of the enhanced defect sinks. Thus, mechanically alloyed vanadium alloys have the potential to extend both low — and high-temperature operation limits.

Other efforts to improve high-temperature strength of vanadium alloys include strengthening by internal oxidation.49

4.12.8 Critical Issues

With the recent progress in the fabrication technology, the number of critical issues for the development of vanadium alloys for fusion reactors has been reduced. The remaining critical issues are thermal and irradia­tion creep, transmutant helium effects on high temper­ature mechanical properties, and radiation effects on fracture properties. The effect of helium, particularly, is still uncertain and can be evaluated precisely only with the use of 14MeV neutrons. This fact highly motivates the construction of a 14 MeV neutron source.

With the progress of the properties of vanadium alloys, the blanket concepts using the alloy become more attractive. Extension of the operation tem­perature window and lifetime of vanadium alloys contribute to the improvement of the quality of the blanket. Therefore, exploration of advanced vana­dium alloys from the current reference alloy is a valuable challenge for enhancing the expected per­formance, and then attractiveness, of fusion reactors.

When metals are subjected to displacive irradiation, especially at elevated temperatures, an intricate and coordinated coevolution of microstructure and microchemistry commences that is dependent pri­marily on the alloy starting state, the dpa rate, and the temperature, and secondarily dependent on vari­ables such as He/dpa rate and applied or internally generated stresses.

In general, the starting microstructure and micro­chemistry of the alloy determine only the path taken to the radiation-defined quasi-equilibrium state, and not the final state itself. If an alloy experiences enough displacements, it effectively forgets its start­ing state and arrives at a destination determined only by irradiation temperature and dpa rate. This quasi­equilibrium or dynamic-equilibrium state consists of microstructural components existing at relatively fixed densities and size distributions, but individual dislocations, loops, precipitates, or cavities at any one moment may be growing, shrinking, or even disap­pearing by shrinkage or annihilation.

The displacement process produces two types of crystalline point defects, vacant crystalline posi­tions (vacancies) and displaced atoms in interstitial crystalline positions (interstitials). These two defect types are both mobile, but move with different dif — fusional modes and at vastly different velocities, with interstitials diffusing much faster than vacan­cies. Therefore it is obvious that all diffusion-driven processes will be strongly affected by radiation. Both defect types have the ability to recombine with the opposite type (annihilation) or to form agglomerations of various types and geometries. These agglomerations and their subsequent evolution alter both the microstructure and elemental distribu­tion of the alloy.

It is important to note that interstitial agglomera­tions are constrained to be two-dimensional, while vacancies can agglomerate in both two-dimensional and three-dimensional forms. This dimensional dis­parity is the root cause of the void swelling phenom­enon covered in a later section.

The developing ensemble of various defect agglomerations with increasing dose induces signifi­cant time-dependent and dose-dependent changes in physical and mechanical properties, as well as resulting in significant dimensional distortion. Most importantly, under high displacement rates stainless steels and other alloys are driven far from equilibrium conditions as defined in phase diagrams, affecting not only phase stability but also all physical, mechanical, and distortion processes that involve phase changes in their initiation or evolution.

During irradiation, the phase evolution can be significantly altered, both in its kinetics and in the identity and balance of phases that form.46,47 Phases can be altered in their composition from that found in the absence of irradiation, and new phases can form that are not found on the equilibrium phase diagram of a given class of steels. In 300 series stainless steels these new or altered phases have been classified as radiation-induced phases, radiation-modified phases, and radiation-enhanced phases.48-51 These classifica­tions are equally applicable to phases formed in other classes of steel.

Radiation-induced alterations of microstructure and microchemistry occur because new driving forces arise that do not occur in purely thermal environ­ments. The first of these new driving forces is the presence of very large supersaturations of point defects, especially at relatively low irradiation tem­peratures (250-550 °C). Not only are vacancies present in uncharacteristically high levels, thereby accelerating normal vacancy-related diffusional pro­cesses, but interstitials are also abundant. Solutes that can bind with either type of point defect tend to flow down any microstructurally induced gradient of that defect, providing a new mechanism of solute segre­gation referred to as solute drag.52 This mechanism has been proposed to be particularly important for binding of smaller solute atoms such as P and Si, and sometimes Ni, with interstitials.

A second new driving force is the inverse Kirkendall effect 53 whereby differences in elemental diffusivity via vacancy exchange lead to segregation of the slowest diffusing species at the bottom of sink — induced vacancy gradients. This mechanism is par­ticularly effective in segregating nickel in austenitic Fe-Cr-Ni alloys at all sinks which absorb vacancies, leading to nickel-rich shells or atmospheres on grain boundaries and other preexisting or radiation — produced microstructural sinks. This type of segre­gation arises because the elemental diffusivities of Fe-Cr-Ni alloys are significantly different, with Dcr > DFe > DNi at all nickel levels.54-57

A third new driving force results from the action of the other two driving forces when operating on microstructural sinks that are produced only in irra­diation environments. These are Frank interstitial loops, helium bubbles, and voids that may have devel­oped from helium bubbles. Precipitates are often observed to form and to co-evolve on the surface of such radiation-induced sinks. Examples of typical radiation-induced microstructures in stainless steels are shown in Figures 12-15. These microstructural sinks have been implicated as participating in the evolutionary path taken by the precipitates and thereby influencing the microchemical evolution of the matrix.1,58-60

Minor solute elements such as Si and P have much higher diffusivities than those of Fe, Ni, and Cr and also participate in the segregation process. Addition­ally, these elements increase the diffusivities of the major elements Fe, Ni, and Cr.54

When the solute drag mechanism, operating between interstitials and smaller size Si and P atoms, combines with nickel segregation via the inverse Kirkendall mechanism, phases that are rich in nickel, silicon, or phosphorus often form (g0, G-phase and Ni2P for example), although in 300 series stainless steels these phases do not form thermally. Other phases that are normally stable in the absence of radiation (carbides, intermetallics) can be forced dur­ing irradiation to become enriched in these elements.1

The removal of nickel, silicon, and phosphorus from the matrix by radiation-induced precipitation exerts a large effect on the effective vacancy diffusiv — ity.57,61 On a per atom basis, phosphorus has been

shown to exert an even larger effect on the effective vacancy diffusivity57 and its removal into Ni2P and other precipitates has a strong influence on matrix diffusion. Silicon is the next most effective element on a per atom basis. As the effective vacancy diffusion coefficient falls with decreasing matrix levels of Ni, Si, and P, conditions for void nucleation become more favorable.

The radiation-induced evolution of diffusional properties has been strongly implicated in determin­ing the transient duration before void swelling accel — erates.1 This evolution often does not necessarily proceed by only one path but occurs in several inter­active stages. Some phases such as nickel phosphides and TiC, especially when precipitated on a very fine scale, are thought to be beneficial in resisting the evolution of nickel silicide type phases.59,62,63 It has been shown, however, that continued radiation — induced segregation eventually overwhelms these phases by removing critical elements such as Ni and Si from solution, causing their dissolution and replacement with nickel-rich and silicon-rich phases that coincide with accelerated swelling.63-65

In high-nickel alloys that normally form the y0 and y00 ordered phases, irradiation-induced segregation processes do not significantly change the identity or composition of the phases, but can strongly change their distribution, dissolving the original distribution but plating these phases out on voids, dislocations, and grain boundaries, with the latter often leading to severe grain boundary embrittlement.66,67

The original dislocation microstructure quickly responds to mobile displacement-generated point defects, increasing their mobility and leading to reductions in dislocation density and distribution in the cold-worked steels most frequently used for fuel cladding and structural components.1 These dislocations are quickly replaced by new micro­structural components, often at very high densities, with two-dimensional interstitial Frank loops first dominating the microstructure, then generating new line dislocations via unfaulting and interaction of loops. In well-annealed alloys there are very few pre­existing dislocations but the same radiation-induced loop and dislocation processes occur, eventually reaching the same quasi-equilibrium microstructure reached by cold-worked alloys.

At lower temperatures found in water-cooled test reactors especially, the microstructural features appear to be three-dimensional vacancy clusters or stacking fault tetrahedra and two-dimensional vacancy or interstitial platelets, which are probably also small dislocation loops. These ‘defect clusters’ at temperatures below ^300°C are usually too small to be easily resolved via conventional transmission

electron microscopy and are often characterized as either ‘black dots’ or ‘black spots.’ These dots are generally thought to be very small Frank interstitial loops.

The cluster and dislocation loop evolution is fre­quently concurrent with or followed by the loss or redistribution of preexisting precipitates. Most importantly, new radiation-stabilized precipitates at high density often appear with crystal structure and composition that are not found on an equilibrium phase diagram for austenitic steels.

As a consequence of these various processes the microstructure at higher doses often develops very high densities of crystallographically faceted, vacuum — filled ‘cavities’ called voids, thought to nucleate on helium clusters formed by transmutation, although residual gases in the steel often help nucleate voids at lower concentrations. Voids have frequently been observed in charged particle irradiations where no helium was introduced.

The void phenomenon is not a volume- conservative process and the metal begins to ‘swell’ as the microscopic voids in aggregate contribute to macroscopic changes in dimension, sometimes increasing the metal volume by levels of many tens of percent.

Concurrently, the dislocation microstructure responds to the local stress state, moving mass via a volume-conservative process designated irradiation creep. In general, irradiation creep is not a directly damaging process but it can lead to component failures resulting from distortion that causes local blockage of coolant flow or strong postirradiation withdrawal forces. Both swelling and irradiation creep are interrelated and are interactive processes that can produce significant distortions in component dimensions. Figure 16 shows some pronounced examples of such distortion.68,69

Eventually, the microstructural/microchemical ensemble approaches a quasi-equilibrium condition

Figure 17 High densities of nanocavities observed using highly under-focus conditions in a PWR flux thimble tube constructed from cold-worked 316 stainless steel. Reproduced from Edwards, D. J.; Garner, F. A.; Bruemmer, S. M.; Efsing, P. G. J. Nucl. Mater. 2009, 384, 249-255. The irradiation conditions were ~70 dpa and 330 °C, producing ~600 appm He and 2500 appm H. Note the high density of cavities on the grain boundary.

or ‘saturation’ state, usually at less than 10 dpa for mechanical properties but at higher doses for swelling. As a consequence, the mechanical properties tend to stabilize at levels depending primarily on tempera­ture and to a lesser extent on dpa rate. The two major deformation processes, swelling and irradiation creep, do not saturate but reach steady-state defor­mation rates when quasi-equilibrium microstructures are attained. This coupling of saturation microstruc­ture with steady-state behavior has been character­ized as ‘persistence.’70

Interestingly, the saturation states of each property change are almost always independent of the starting thermal-mechanical state of the material.1,70,71 If irradiation continues long enough, the memory of the starting microstructural state and the associated mechanical properties is almost completely lost. The only deformation-induced microstructural component that succeeds in resisting this erasure process is that of preexisting, deformation-induced twin boundaries.

If this quasi-equilibrium is maintained to higher neutron exposure no further change occurs in the steel’s mechanical properties. However, some slowly developing second-order processes are nonsaturable and are often nonlinear. Eventually, these processes force the system to jump toward a new quasi­equilibrium. These new states usually arise from either the microstructural or microchemical evolu­tion, with voids dominating the former and the latter involving continued segregation, continued transmutation, or a combination of these factors.70-72 A number of such late-stage changes in quasi­equilibrium state are discussed later in this paper.

Brown et al49 compared the swelling behavior of STA Nimonic PE16 and two cold-worked austenitic steels (M316 and Nb-stabilized FV548) which were irradiated in DFR as fuel pin cladding. Two PE16 clad pins were examined, which were irradiated to burn-ups of 6.1% and 21.6% of heavy atoms, corresponding to peak damage levels of about 17 and 80 dpa, respectively. Void concentrations and swelling were lower in PE16 than in the austenitic steels. Swelling data, void concentrations, and void diameters for the two PE16 pins examined by Brown et al. are shown in Figure 7. Note that Brown etal49 only showed trend lines for void concentration and void size in the less highly irradiated pin and com­pared the swelling tendencies of the two pins; the individual data points were not plotted and those shown in Figure 7 are previously unpublished data obtained by Sharpe. Brown et al. stated that the void concentration in PE16 decreased with increasing irradiation temperature but did not alter greatly with an increasing dose above ~17 dpa. It should be noted, however, that swelling measurements for the higher burn-up pin were restricted to temperatures

below 525 °C, so that a direct comparison of void concentrations in the two pins cannot be made at higher temperatures. Although there were fewer voids in PE16 than in the two steels, the voids

appeared to be homogeneously distributed and to have developed during the early stages of irradiation; once nucleated, the growth rate of voids in PE16 remained low. These observations are clearly contrary to early models which suggested that low swelling rates result from incomplete void nucleation and extended tran­sient regimes. Rather, in agreement with the more recent observations of Muroga et a/.,45,46 it appears that the swelling resistance ofPE16 is due to a combi­nation of a comparatively low saturation void concen­tration, which is reached at a relatively low displacement dose, and a low void growth rate. There does not appear to be any evidence of an accelerated swelling rate in PE16 once void nucleation is complete.

Additional data on void concentrations in neutron — irradiated PE16 are available from Cawthorne eta/.,8 Sklad et a/.,50 and Boothby.28 The results presented by Cawthorne et a/. for PE16 fuel pin cladding irra­diated in DFR to a peak fluence of 5.6 x 1026nm~2 (^28 dpa) differ from those shown in Figure 7 in that, although void number densities are similar for irradiations at ^380-520°C, void concentrations are about an order of magnitude higher at 350°C and 600-630 °C. Such discrepancies might arise from uncertainty and/or variability in irradiation tempera­tures. Another possibility is that void nucleation was incomplete at the higher irradiation temperatures in the lower burn-up pin examined by Brown et a/. Data from Sklad et a/. show an increase in void numbers in unstressed PE16 specimens irradiated in EBR-II at 500 °C from an average (for two differently heat trea­ted conditions) of about 4 x 1019 to 1.2 x 1020m~3 with increasing fluence from 1.2 x 1026 to

1. x 1026nm~2 (E > 0.1 MeV), that is, from ^6 to 20 dpa. In this case, the void concentration and overall swelling of ^0.2% at ^20 dpa remain below the levels shown in Figure 7 for the DFR-irradiated pin at ~ 17 dpa; this may reflect the effect of stress on swelling for fuel pin cladding.

Void swelling data determined from Transmission electron microscope (TEM) examinations of three heats of PE16 which were irradiated in the UK-1 rig in EBR-II are shown in Figure 8, which includes previously unpublished results for the low boron (4ppm) heat Z184 as well as data for heats DAA766 and Z260D (with 18 and 70 ppm boron, respectively) which were reported by Boothby.28 Data are shown for all three heats in the STA condition (ST 1020 °C and aged 4h at 750 °C) and for DAA766 in the OA condition (a multistage heat treatment that included aging at 900 °C, slow cooling to 750 °C, and then aging for 16 h at that temperature, resulting in the

O STA DAA766 □ OA DAA766 л STA Z260D ° STA Z184

д

E1020

1019

О

1018

300 350 400 450 500 550 600 650

Temperature (°C)
100

90

80

70

60 ф

50

га

40 30

>

20 10 0

300 350 400 450 500 550 600 650

Temperature (°C)

Figure 8 Swelling data, void concentrations, and void diameters for Nimonic PE16 samples irradiated in UK-1 rig in Experimental Breeder Reactor-II. Adapted from Boothby, R. M. J. Nucl. Mafer.1996, 230, 148-157; Unpublished data for Boothby, R. M. The Microstructure of EBR-II Irradiated Nimonic PE16; AEATRS 2002 (FPSG/P(90)23), with permission from AEA Technology Plc.

precipitation of TiC and an overaged g structure). Swelling data derived from the density measure­ments of STA PE16 heat DAA 766 from the same experiment are shown in Figure 4. An example of the

void distribution in the OA condition is shown in Figure 9. Note that the voids in neutron-irradiated PE16 tend to be cuboidal and that enhanced growth of voids attached to TiC precipitates (located at the site of a prior grain boundary) has occurred.

Neutron fluences and irradiation temperatures in the UK-1 experiment were similar to those for the first withdrawal of the AA-1 rig for which data is shown in Figure 2. Void concentrations for heats DAA766 and Z260D shown in Figure 8 appear to be less temperature-dependent than for the fuel pin clad­ding data shown in Figure 7. Void numbers are gener­ally lower than in the cladding at temperatures up to ^550 °C, but are intermediate between the results of Brown et a/.49 and Cawthorne et a/.8 for irra­diations at ^600 °C. Void concentrations for PE16 irradiated to fast neutron fluences (E >0.1MeV) of 9.4-12.3 x 1026nm~2 at 477-513 °C in the UK-1 ex­periment were very similar to those determined by Sklad et a/.50 for 4.0 x 1026nm~2 at 500 °C. The low boron heat Z184 showed atypical behavior, with a very high concentration of small voids and low swelling at 438 °C, but high swelling owing to increased void sizes at normal void concentrations at temperatures above 513 °C. It is probable that the effect of boron on swelling is related to the formation ofboron-vacancy complexes, which can give rise to the nonequilibrium segregation of boron in the presence ofquenched-in thermal vacan­cies as well as to radiation-induced effects.51

Some variability in the swelling response of Nimonic PE16 in PFR (Prototype Fast Reactor) components was reported by Brown and Linekar.52

Increased swelling in PE16 subassembly and guide tube wrappers in PFR compared to expectations based on the performance of DFR pin cladding appeared to be related to temperature fluctuations, particularly at temperatures below 400 °C during the early operation of PFR. Void concentrations were reported to be higher in the PFR components, and it was suggested (by Cawthorne, unpublished data) that this may have been due to the release of vacan­cies from vacancy loops which had formed during lower temperature excursions. In fact, the void con­centration reported by Cawthorne eta/.8 for DFR pin cladding irradiated at 350 °C was higher than the highest value reported for the PFR components by a factor of about 3, but this comparison was not made by Brown and Linekar. There were also indications of heat-to-heat variability and effects of the fabrication route on the swelling of PE16 wrappers in PFR. Nevertheless, swelling of PE16 wrappers, although higher than expected, remained low in absolute terms and did not give rise to any operational problems.

Although PE16 was originally selected as the ref­erence wrapper material for PFR and as an alterna­tive to cold-worked M316 steel for fuel pin cladding, PE16 was favored as a cladding material with 12%Cr ferritic-martensitic steel wrappers in subsequent subassembly designs.53 The 12%Cr steel was chosen as a wrapper material because of its superior swelling resistance, but its use was limited to relatively low temperatures owing to inadequate strength at the higher operating temperatures experienced by pin cladding. Design calculations for PE16 fuel pin clad­ding made by Cole54 indicated that cladding hoop stresses, which arise from the internal pressure from the gaseous fission products released from the fuel, were much lower than the yield stress of the material and were generally expected to remain below about 70 MPa. In addition, the void swelling and irradiation creep behavior of PE16 were considered to be well matched to the fuel swelling, so that fuel-clad inter­action stresses also remain low. Fuel pins with PE16 cladding successfully attained high burn-ups in PFR, with some 3500 pins exceeding dose levels of 100 dpa and 265 pins reaching maximum doses of 155 dpa.55 Very few failures of PE16 clad pins were recorded — three failures occurred in pins which had reached burn-ups over 17at.%, with one failure at 11.3 at.% burn-up which was believed to have resulted from a fabrication defect.56 In addition to the four PE16 cladding failures in PFR, Plitz et a/.57 recorded 14 failures in austenitic steel cladding, all at lower burn — ups than in PE16. The failures in PE16 cladding were regarded as benign and permitted continued opera­tion, with no significant loss of fuel into the primary circuit coolant. A peak burn-up of 23.2 at.%, corresponding to a peak dose in the PE16 cladding of 144 dpa, was achieved in PFR in an experimental fuel cluster. Postirradiation examinations of pins from this cluster and a high burn-up subassembly (18.9 at.%, with a peak cladding dose of 148 dpa) were carried out by Naganuma et a/.58 Maximum diametral strains of less than 1% were measured, attributable to the combined effects of void swelling, creep deformation arising from internal gas pressure in the pins, and small contributions from mechanical interactions between the fuel and cladding in the lower part of the pins.

4.06.4.1 Introduction and History of Mo and Mo Alloys

Molybdenum and its alloys are the perennial candi­dates for refractory metal alloy use in irradiation environments, due in part to their high melting temperature (2896 K), good thermal properties, high-temperature strength, and lower induced radioactivity (as compared to tantalum). The density of molybdenum (10.28 gcm~3) is also significantly lower than that ofTa and W, though greater than Nb. But like other refractory metal alloys, Mo can pres­ent difficulties in fabrication, low-temperature duc­tility, and low-temperature embrittlement from radiation damage. The TZM (Mo—0.5%Ti—0.1% Zr) and Mo-Re alloys were examined as part of the SP-100 and JIMO/Prometheus space reactor pro­grams, respectively, and offer additional benefits of improved high-temperature strength over the pure metal.5,19 Molybdenum and its alloys have also been examined for plasma facing and diverter components in fusion reactor designs due to the relatively low sputter yield, high thermal conductivity, and thermal compatibility with other structural materials.5-27,29-63 In addition, because ofthese benefits, Mo has also been examined for use as a grazing incident metal mirror in fusion diagnostic port designs.64,65

As in all other refractory metals, the mechanical properties are influenced by impurity concentra­tions, particularly through grain boundary weaken­ing. However, improvements in Mo ductility are achievable through grain refinement, impurity con­trol, and the addition of Re or reactive elements such as Ti and Zr. An upper limit to the acceptable level of C was also found to improve grain boundary strength. Low-carbon arc-cast molybdenum (LCAC-Mo) is one such example, in which oxygen impurities are reduced to tens of ppm, nitrogen to <10 ppm, and carbon to <100 ppm.66 Higher levels of C will result in reduced fracture toughness, unless additional reac­tive alloy additions are present in the alloy. The TZM alloy also incorporates a small level of carbon to produce Ti — and Zr-carbide strengthening.

Improvements in ductility and toughness through the ‘rhenium effect’ have been observed in Mo for some time,67-69 and generally occurs when Group Via metals are alloyed with elements from Group Vila and Villa metals.70,71 Explanations for this phe­nomenon range from enhanced mechanical twinning, reduced resistance to dislocation glide, reduction of oxygen at grain boundaries, and increased interstitial oxygen solubility. , Critical evaluation6 of the initial work that had suggested a maximum ten­sile ductility near 11-13 wt% Re78,79 was found to be inconclusive because of inadequate control of O and C impurity levels in the earlier studies. Higher con­centration alloys with 40-50 wt% Re have also been examined for use in the radiation environments. Alloys with Re concentrations up to 41-42% are single-phase solid-solution a-Mo, while those at higher levels incorporate the a-Re2Mo phase. Com­mercially available alloys include Mo-41Re and Mo-47.5Re (sometimes referred to as Mo-50Re).

Recently, introduction ofoxide dispersion strength­ened (ODS)-Mo through the incorporation of lantha­num oxide particles has been examined.80- 2 These alloys show great resistance to recrystallization and high-temperature deformation while maintaining low ductile-to-brittle transition temperatures (DBTT) partly because of their refined grain structure.83-85

The radiation effects database for Mo and its alloys is limited to scattered scoping examinations, which show little overlap in the experimental vari­ables such as material purity, alloying level, material thermomechanical history, irradiation conditions, and postirradiation test conditions. Where available, infor­mation on the physical and mechanical property changes to LCAC-Mo, TZM, Mo-Re alloys, and ODS-Mo will be reviewed.

Oxidation tests for 9Cr-ODS and 12Cr-ODS steels were performed using pickled specimens in a

controlled atmosphere of dry air. Weight measure­ment to evaluate the degree of oxidation was per­formed at intervals of 50, 100, 400, 1000, and 2000 h, at temperatures of 650, 750, and 850 °C. The results of the measured weight gain due to oxidation at 750 °C are shown in Figure 36.58 For 9Cr-ODS and 12Cr-ODS steels, the weight gain due to oxida­tion was quite small and comparable to that of PNC316 containing 17wt% Cr. Their weight gain is limited to below 0.1 mg mm~2. On the other hand, a quite large oxidation of 0.8 mg mm~2 was observed in PNC-FMS. The measured results on SUS430, which show a greater weight gain than that of ODS steels, show that advanced oxidation resistance is attained with ODS steels, even when compared to higher 17 wt% Cr containing stainless steel.

The element distribution obtained by Electron probe microanalysis (EPMA) showed a scale consisting

E

j? 0.8

c

о

0.6

"О

£______

0 500 1000 1500 2000 2500

Testing time (h)

of Fe-rich oxide in the outer layers and Cr-rich oxide in the inner layers. At the interface between ODS steel and the oxide scale, there was a thin layer (a few micrometers) of further Cr-enriched oxide. Raman spectroscopy measurement indicated that the outer Fe-rich and inner Cr-rich layers correspond to a-Fe2O3 and spinel type (Fe, Cr)3O4, respectively. It was also confirmed that a-Cr2O3 is formed at the matrix-scale interface.

0.10

E

0.08

C

0.06

■g

x

о

0.04

-Q

0.02

D)

0.00

In oxidation tests, Fe, which is a major constituent in steel, tends to be easily oxidized at an early stage, but further oxidation can be suppressed by the formation of a protective a-Cr2O3 layer. This a-Cr2O3 formation is generally controlled by the rate at which Cr is supplied to the reaction front. It is known that a high Cr content in steel, as well as an increasing diffusion flux through the grain boundary, that is, finer grains, accelerates both the Cr supply and the formation of a-Cr2O3. A short-term oxidation test, whose results are shown in Figure 37, was conducted to investigate the mechanism of suppressing oxidation in ODS steels.58 The decrease in oxidation in fine grain 12Cr-ODS ferritic steel can be attributed to the enhanced rate at which Cr was supplied throughout the accelerated grain boundary diffusion. In both cases of fine/large grains in 12Cr-ODS steels, Raman spectroscopy detected protective a-Cr2O3 at the interface between the matrix and scale. Comparing 12Cr-ODS large grain and PNC-FMS, the Cr content is similar, and the grain size is rather smaller in PNC-FMS. Never­theless, protective a-Cr2O3 cannot be detected by Raman spectroscopy, and oxidation is enhanced in PNC-FMS, implying that the suppression of oxidation in 12Cr-ODS with large grains could be due to the effects of the Y2O3 oxide particles themselves. Chen et al. showed some TEM images of Y-rich oxides on grain boundaries that may be part of the explanation.59

The properties and irradiation-induced changes in graphite crystals have been studied using both ‘natu­rally occurring’ graphite crystals and an artificial product referred to as highly orientated pyrolytic graphite (HOPG), formed by depositing a carbon substrate using hydrocarbon gas6 followed by com­pression annealing at around 3000 °C. HOPG is con­sidered to be the most appropriate ‘model’ material that can be used to study the behavior of artificially produced polycrystalline nuclear graphite. It has a density value near to that of a perfect graphite crystal structure, but perhaps more appropriately, it has imperfections similar to those found in the struc­tures that make up artificial polycrystalline graphite. A detailed description of the properties of graphite can be found in Chapter 2.10, Graphite: Properties and Characteristics.

4.11.2.1 Graphite Crystal Atomic Structure and Properties

In this section, the atomic structure of graphite crys­tal structures is discussed briefly, along with some of the properties relevant to the understanding of the
irradiation behavior of graphite. Graphite can be arranged in an ABAB stacking arrangement termed hexagonal graphite (see Figure 1). This is the most thermodynamically stable form of graphite and has a density of 2.266 gcm—3. The a-spacing is 1.415 A and the c-spacing is 3.35 A.

However, in both natural and artificial graphite stacking faults and dislocations abound.1

Irradiated and unirradiated Young’s modulus of nuclear graphite is usually measured either by an impulse or frequency method, giving the dynamic Young’s modulus (DYM). However, for use in com­ponent stress analysis assessments, the static Young’s modulus (SYM) is required. In addition, the UK irradiation creep modulus was originally defined using SYM, so there is a need to interconvert between the two measurements.

DYM is always higher in value than SYM. Unfor­tunately, SYM has historically been defined in vari­ous ways, that is, as the chord between the origin at zero stress and half, or in some cases two-thirds the failure strength in tension or bending.

There is limited amount of data on the SYM/ DYM ratio for UK nuclear graphite, as given in Figure 48.78 The data is for unirradiated graphite and for graphite irradiated in an inert atmosphere. The ratio for irradiated graphite is higher than for unirradiated graphite. At present, there are no
published data on this ratio for radiolytically oxidized graphite, although safety case requirements have rekindled effort in this area in the United Kingdom.

It has been shown79 that for fine-grained IG-110 nuclear graphite and PGX reflector graphite, as well as ASR-ORB baked carbon, the ratio of static to dynamic Young’s modulus strongly depends on the chord length chosen to define SYM.