Category Archives: Comprehensive nuclear materials

Tensile Behavior

Tensile behavior is determined by the irradiation — induced defect structure previously discussed. Aus­tenitic stainless steels will again be used for the example since they are typical of fcc alloys and in
many respects to other alloys (see Chapter 2.09, Properties of Austenitic Steels for Nuclear Reac­tor Applications and Chapter 4.02, Radiation Damage in Austenitic Steels). The behavior of other example classes of alloys will be discussed in later sections of this chapter. The tensile behavior characteristic of austenitic stainless steels is shown in Figure 2, where yield strength is plotted as a function of fluence and displacement level.9 Saturation in strength is clear with the saturation time becoming shorter as irradiation temperature is increased. At temperatures above about 500 °C, saturation is evi­dent, but in this case, strength decreases.

This decrease is a result of recovery of the cold — worked microstructure of the 20% cold-worked type 316 stainless steel presented in Figure 2. Figure 3

Подпись: Figure 3 Yield strength of type 316 stainless steel irradiated in the EBR-II. Reproduced from Garner, F. A.; Hamilton, M. L.; Panayotou, N. F.; Johnson, G. D. J. Nucl. Mater. 1981, 103 & 104, 803. Подпись: Figure 4 Strength properties of 20% cold-worked type 316 stainless steel irradiated in EBR-II. Reproduced from Allen, T. R.; Tsai, H.; Cole, J. I.; Ohta, J.; Dohi, K.; Kusanagi, H. Effects of Radiation on Materials; ASTM: Philadelphia, PA, 2004; ASTM STP 1447, p 3. Reprinted, with permission, from Effects of Radiation on Structural Materials, copyright ASTM International, West Conshohocken, PA.

shows yield strength resulting from the recovery of a cold-worked dislocation structure and the generation of a radiation-induced microstructure, resulting in a saturation strength independent of the initial condi­tion of the alloy. 0 Again, it is seen that the approach to saturation is faster with increasing temperature, with saturation achieved between 5 and 10 dpa at 538 and 650 °C, but 15-20 dpa is necessary to achieve saturation at 427 °C.

Saturation is observed in yield strength curves for fluences as high as 9 x 1022ncm~2 in a fast reactor (45 dpa), but more recent data show a hint of soft­ening above 50 dpa,11,12 and other fast reactor data have shown a reduction in strength even for dis­placements below 50 dpa, as shown in Figure 4.1 This could result from coarsening of the microstruc­ture or depletion of interstitial elements from the matrix due to precipitation. This effect is also observed in martensitic steels irradiated to high dpa levels in the FFTF, but in this class of alloys, recovery of the martensitic lath structure is also a factor.12 However, even in austenitic steels, it is difficult to attribute such softening with certainty to an irradiation effect because of the strong influence
of irradiation temperature on strength.14 Indeed, uncertainties in irradiation temperature are an inher­ent difficulty in neutron irradiation experiments.

Proposed Future Neutron-Irradiation Facilities

The proposed International-Fusion-Material- Irradiation-Facility (IFMIF) is an accelerator-driven neutron source that is based on the proton-stripping reaction.58,59 Neutrons are generated by a beam of 40 MeV deuterons that undergo a proton-stripping reaction when they interact with a flowing liquid lithium jet target. The resulting neutron beam has a spectrum with a high-energy tail above a peak around 14.60 As in the case of D, T, and spallation reactions, these neutrons are well above the thresh­old energy for n, a reactions; thus, IFMIF produces fusion like He/dpa ratios at high dpa rates. The nuclear reaction kinematics and limited neutron target source dimensions result in an IFMIF irradia­tion volume with large gradients over a high-flux region just behind the target. Two 125 mA beams on the Li target produce an «500 cm3 region with dpa rates of 20-50 fpy 1 (full power year) at He/ dpa «12appmdpa~. The medium flux region, from 1.0 to 20dpafpy~ , is much larger with a volume of «6000 cm3.

The Materials Test Station is a new spallation neutron source, proposed by Los Alamos National Laboratory, that is primarily intended to irradiate fast reactor materials and fuels.61 The LANSCE linear accelerator will produce a 1-MW proton beam to drive the spallation neutron source with a fast reactor like spectrum and a high-energy tail up to 800 MeV. The high-energy tail neutrons produce a He/dpa«6-13appmdpa~ close to that for a fusion first wall. The dpa rates are «7.5-15 fpy~ in a 200 cm3 irradiation volume and 2.5-12.5 fpy_1 in an additional volume of 450 cm3. An accelerator upgrade to 3.6 MW would increase these dpa rates to 20-40 fpy1 and 5-16 fpy, respectively.

In both cases, the limited volume for high-flux accelerated irradiations presents a great challenge to developing small specimen mechanical test meth — ods62,63 and experimental matrices64 that can pro­duce the database needed for materials qualification. The database will require irradiations over a range of temperatures for tensile, fracture toughness, fatigue, and creep property characterization. Indeed, it is clear that qualifying materials for fusion ap­plications will require a new paradigm of linking comprehensive microstructural characterization and physically based predictive modeling tools to multi­scale models and experiments of structure-sensitive properties as input into engineering models of mate­rials performance.

A variety of proposals have been made to develop volumetric D-T fusion devices such as the Fusion Nuclear Science Facility (FNSF), which would pro­vide a basis to test components and materials.65 In some cases, these devices would address a much broader array of issues, such as tritium breeding and extraction. In most cases, the fusion source would be driven by external energetic D beams. Discussing the details of such proposed devices is far beyond the scope of this chapter. However, we note that from a materials development perspective, such devices would be useful to the extent that they are steady state, operate with very high-duty factors, and pro­duce sufficient wall loading to deliver high He and dpa exposures.

Dislocation bias with size and modulus interactions

The modulus interaction has been discussed in Section 1.01.5.3. Treating both the material as well as the defect inclusion as elastically isotropic, the modulus interaction depends on two diaelastic polar­izabilities, aK and aG, for which values are provided in Table 11.

For this isotropic case, the modulus interaction for edge dislocations is58

W2 (r;’) = — (Ao + A2cos2′)(b/r)2 [145]

with

 

aK(1 — 2n)2 + |am(1 — n + n2)

[4p(1 — n)]2

 

[146]

 

Ao

 

Zedge = 1 + [ГС/(2ro)] — [142]

ln(R/ro) + m[rc/(2ro)]2

with m = 3 fits the exact results for 0 < rCJb < 6 as seen in Figure 26.

Very accurate results for the bias factors of edge dislocations can therefore be produced with eqn [142] for small capture radii and with eqn [141] for large capture radii. The two approximations transition extremely well at rc/b = 6. These approximations sug­gest a way to proceed when the interaction energy assumes a more complicated form than in eqn [133].

It is therefore tempting to see if an angular aver­age of the size interaction energy, eqn [134], could be used to evaluate at least approximately the bias factors. Obviously, the angular average of sinj is zero. The diffusion flux will wind around the dislocation as it approaches the core in order to avoid regions where the interaction energy W1 becomes strongly repulsive. Therefore, an average should only be taken over the angular range where W1 is attractive, that is, negative. So, ifwe replace W1(r, ‘) in eqn [134] with W1(r, p/2)/2 in the case of interstitials and with W1(r, 3я/2)/2 in the case of vacancies and evaluate the equation

Zd « — ‘n(R/rd) [143]

rd exp[b W1 (r)]d’n(r)

we obtain58

 

and

 

(aK — |am)(1 — 2n)2 + 4amn(1 — n)
[4p(1 — n)]2

 

[147]

 

A2

 

The perturbation theory of Wolfer and Ashkin58 with the sum of the size interaction [130] and the modulus interaction [145] gives the result

 

2Ao

kT

  image150

Zedge

 

[148]

 

This suggests that an effective capture radius can be defined as

 

and replacing rc with it in eqn [144] yields bias factors that include both the effects of size and mod­ulus interactions. Using the values given in Table 11 for the diaelastic polarizabilities, effective capture radii and new bias factors are obtained and presented in Table 15. Comparing these results with the corresponding ones in Table 14 shows that the mod­ulus interaction contributes to the net bias about 25% for fcc metals, and about 1o—15% for bcc metals.

 

Role of Crystal Structure

MD simulations23 predict the absolute level of defect production is not strongly affected by crystal struc­ture. Conversely, electrical resistivity studies of fis­sion neutron-irradiated metals suggest that the overall defect production is highest in HCP metals, intermediate in bcc metals, and lowest in fcc metals,1 1 which suggests that the anisotropic nature of HCP crystals might inhibit defect recombination within displacement cascades. TEM measurements of defect cluster yield (number of visible cascades per incident ion) in ion-irradiated metals have found that the relatively few visible defect clusters are formed directly in displacement cascades in bcc metals,122 whereas cluster formation is relatively efficient in fcc metals and variable behavior is observed for HCP metals.123 Faulted dislocation loops are often observed in irradiated fcc and HCP metals, but due to their high stacking fault energies most studies on irradiated bcc metals have only observed perfect loops.8,16,21,47,124 Since perfect loops are glissile, this can lead to more efficient sweeping up of radiation defects and accelerate the development of dislocation loop rafts or network dislocation structures in bcc materials. Figure 16 shows examples of the disloca­tion loop microstructures in bcc, fcc, and HCP metals with similar atomic weight following electron irradiation at temperatures above recovery Stage III.47 All of the loops are interstitial type with com­parable size for the same irradiation dose. However, significant differences exist in the loop configura­tions, in particular habit planes and faulted (Ni, Zn) versus perfect (Fe) loops. One significant aspect of loop formation in HCP materials is that differential loop evolution on basal and prism planes can lead to significant anisotropic growth.125-129

In general, defect accumulation in the form of void swelling is significantly lower in bcc materials com­pared to fcc materials, although there are notable exceptions where very high swelling rates (approach­ing 3% per dpa)130,131 have been observed in some bcc alloys. Pronounced elastic and point defect diffu­sion anisotropy12 can also suppress void swelling in HCP materials, although high swelling has been observed in some HCP materials such as graphite.1 It has long been recognized that ferritic/martensitic steels exhibit significantly lower void swelling than

image289

Figure 16 Dislocation loop microstructures in Fe, Ni, and Zn following electron irradiation at temperatures above recovery Stage III. The loops in Fe were perfect and located on (100) planes, and the loops in Ni and Zn were faulted and located on {111} and (0001) planes, respectively. Reproduced from Kiritani, M. J. Nucl. Mater. 2000,276(1-3), 41-49.

austenitic stainless steels.109,133,134Figure 17 com­pares the microstructure of austenitic stainless steel and 9%Cr ferritic/martensitic steel after dual beam ion irradiation at 650 °C to 50 dpa and 260appm He.135 Substantial void formation is evident in the Type 316 austenitic stainless steel, whereas cavity swelling is very limited in the 9%Cr ferritic/marten — sitic steel for the same irradiation conditions. Several mechanisms have been proposed to explain the lower swelling in ferritic/martensitic steel, including lower dislocation bias for SIA absorption, larger critical radii for conversion of helium bubbles to voids, and higher point defect sink strength.

Dislocation Loop Formation in Spinel and Alumina

1.05.2.3.1 Introduction to atomic layer stacking

Results of numerous neutron and electron irradiation damage studies suggest that two types of interstitial dislocation loops nucleate in a-Al2O3: (1) 1/3 [0001] (0001); and (2) 1/3(10І1){10І0} (see, e. g., the review by Kinoshita and Zinkle4). The first of these involves precipitation on basal planes in the hexagonal a-Al2O3 structure, while the second is due to precip­itation on m-type prism planes. In MgAl2O4, similar studies indicate that primitive interstitial dislocation loops also have two characters: (1) 1/6 (111) {111} and (2) 1/4 (110) {110}.4 Though the crystal struc­ture of spinel is cubic, compared with that of alumina, which is hexagonal, the nature ofthe dislocation loops formed in spinel is similar to those in alumina: {111} spinel loops are analogous to (0001) alumina loops; likewise, {110} spinel loops are analogous to {1010} alumina loops. We will first compare and contrast {111} spinel versus (0001) alumina loops and later discuss {110} spinel versus {1010} alumina loops.

Both spinel (111) {111} and alumina [0001] (0001) interstitial dislocation loops involve insertion of extra atomic layers perpendicular to the (111) and [0001] directions, respectively. These layers are either pure cation or pure anion layers. In both spinel and alumina, anion layers along (111) and [0001] directions, respectively (i. e., along the 3 direction in both structures), are close packed (specifically, they are fully dense, triangular atom nets), while the cation layers contain ‘vacancies,’ which are necessary to accommodate the cation deficiency (compared with anion concentration) in both compounds (these ‘vacancies’ actually are interstices; they are ‘holes’ in the otherwise fully dense triangular atom nets that make up each cation layer). Table 1 shows the arrangement of cation and anion layers in spinel and alumina, along (111) and [0001] directions, respectively.11 Both structures can be described by a 24-layer stacking sequence along these directions. Both spinel and alumina can be thought of as con­sisting ofpseudo-close-packed anion sublattices, with cation layers interleaved between the anion layers. The anion sublattice in spinel is cubic close-packed (ccp) with an ABCABC… layer stacking arrangement, while alumina’s anion sublattice is hexagonal close — packed (hcp) with BCBCBC. .. layer stacking. In both structures, between each pair of anion layers there are three layers of interstices where cations may reside: a tetrahedral (t) interstice layer, followed by an octahedral (o) interstice layer, followed by another t layer. In spinel, Mg cations reside on t layers, while Al cations occupy the o layers. In alumina, all t layers are empty and Al occupies 2/3 of the o layer inter­stices. In spinel, cation interlayers alternate between a pure Al kagome layer and a mixed MgAlMg, three — layer thick slab. In alumina, each interlayer is pure Al in a honeycomb arrangement.

Helium effects on fracture properties and He-induced embrittlement effects

The effects of He on fast fracture, typically charac­terized by shifts in the DBTT measured in CVN impact tests (AT), has long been a subject of
significant controversy. This controversy has been fueled by studies that were interpreted to suggest that even small to moderate amounts of He result in increases in DBTT14,34,240-242 However, it has been shown that at temperatures below about 400 °C embrittlement is primarily due to irradiation harden­ing (Asy), resulting from fine-scale irradiation — induced dislocation obstacles.20,21 The simplest rela­tion is hardening-shift relation, which is given by

AT = C Asy [18]

Here C depends on a number of variables but for irradiated FMS has an average value of «0.4 °C MPa-1 for subsized CVN tests. Thus, it is obvious that He would contribute to embrittlement of FMS to the extent that it contributes to hardening. However, as noted previously, He effects on hardening are minimal up to levels of about 500 appm. Further, most of the data on He effects on embrittlement are confounded by the experimental techniques, like Ni and B doping, or use of atypical fracture test methods. Irradiation embrittlement can also be induced by nonhardening mechanisms associated with changes in the local fracture properties that are controlled by coarse-scale microstructural features, like brittle trigger particles for cleavage, and segregation of ele­ments that weaken GBs.2 ,

The first data that clearly indicated a nonharden­ing role of He were generated in the early STIP experiments, showing a transition from ductile and cleavage fracture modes to extremely brittle IG fracture20,220 and somewhat larger than expected A T Analyses of a large database on irradiation hard­ening and embrittlement, including the STIP data,2 showed that He does not produce significant nonhardening embrittlement at less than about 500appm. However, above this rough threshold the hardening-shift coefficient C (=AT/Asy) increases due to weakening of the GBs associated with He accumulation, to the point where they became the preferred fracture path. The database was used to derive a simple semiempirical model for CVN AT for 300 °C irradiations as

C = 0.4 + 7 x 10-4(XHe — 500)(oC MPa-1) [19]

As shown in Figure 33 the model prediction (dashed curve)242a is remarkably consistent with SPNI and neutron data including more recent results. The STIP data are based on subsized CVN tests (KLST and 1/3 CVN) on different FMS irradiated in STIP — I-III up to about 17dpa at temperatures below 300 °C.19 The solid symbols are small punch test data converted to CVN A T The neutron data were taken from the literature,14,240-244 and these results

image435

Figure 33 DBTT shift as a function of irradiation dose for different FMS irradiated in STIP. Neutron-irradiation data are included for comparison. Reproduced from Dai, Y.; Wagner, W. J. Nucl. Mater. 2009,389, 288. The dashed curve is drawn according to the model prediction (eqn [19]).

 

Displacement (dpa)

image436
image437

Подпись:

Figure 34 A sketch showing the mechanisms for irradiation-induced hardening (increase of yield stress, Asy) and helium-induced grain boundary weakening effects (decrease in the intergranular fracture stress, s(g) that elevate the brittle to ductile transition temperature.

are also consistent with the analysis of a larger database.20

As schematically illustrated in Figure 34, the synergistic low-temperature hardening-helium embrit­tlement (LTHE) He threshold can be rationalized as follows. Cleavage fracture occurs when the stress concentrated at the tip of a blunting crack, Mcy, exceeds a critical local stress, c*, over a critical vol­ume needed to activate a brittle trigger particle.2 Here, M is a stress concentration factor. Likewise, brittle IG fracture occurs when the crack tip stress exceeds the critical local stress si*g over a sufficient volume needed to crack GBs. The c* is initially higher than sc*; thus, fracture occurs by transgranular cleavage (Figure 34(a)). However, c* decreases with increasing He GB accumulation, and beyond a bulk threshold level, ca. 500appm, c* falls below c* (Figure 34(b)). Thus, the grain boundary becomes the favored crack path. The c* continues to decrease with increasing He accumulation, resulting in an increasing increment of AT, even in the absence of additional hardening. The transition to IG crack paths is marked by a larger fraction of grain boundary facets on the fracture surface. Note that the continued increase in Ac ys with higher He was not recognized at the time that this simple model was developed, thus the new insight and expanded database will be used to refine the model.

Helium that is not clustered into bubbles is likely the most damaging condition, with a monolayer cov­erage producing essentially complete grain boundary decohesion. The actual amount and distribution of helium on GBs has not been established and is a function of the temperature and microstructure as well as bulk XHe. However, even at 400 ° C boundary bubbles are less than 1 nm in diameter. Assuming that grain boundary helium derives from regions in the
adjoining matrix and is located in spherical bubbles with equal numbers of mHe atoms and vacancies, the fractional grain boundary coverage can be estimated asfHe = tHeXHe/[10-4 mHe]; here, tHe is the thickness (mm) of the layers that feed helium to the grain boundary. Thus for example, fHe«0.25, assuming tHe = 0.25 mm and mHe = 5 and XHe = 500appm. Note that this tHe may be too large considering that denuded zones are not evidently observed at GBs in STIP samples. However, the data are not sufficient to reach firm conclusions, and a combination of models and mechanism experiments is needed to determine the partitioning of He to GBs for various microstruc­tures and irradiation conditions.

Other studies245,246 reached similar conclusions regarding the effect of He on grain boundary strength. Indeed, simple and direct evidence is provided by the brittle fracture stresses measured in the tensile tests cited previously, which decreased from «1850 to 1640 MPa with increasing He levels from 1250 to 2500 appm. These helium-degraded c* are well below the cleavage c* « 2000 MPa.

Embrittlement and AT are most properly evalu­ated by fracture toughness tests that are expected to show hardening-He synergisms that are similar to those measured in CVN tests. Figure 35 shows the estimated fracture toughness (Kjq) of various FMS after SPNI based on three-point bend tests on small precracked bars at test temperatures approximately equal to the irradiation temperatures.210,222 Note that

at high dose, Kjq decreases to less than 40 MPa Vm, close to lower shelf fracture toughness of FMS, even at the maximum irradiation temperature of 400 C.

Figure 35 also shows that the KJq of the T91 steel irradiated at LANSCE are degraded at lower doses (up to about 4.3 dpa)2 3 than in the STIP irradiations. This may be the result of the combination of the lower irradiation temperatures and higher helium generation rates in this case. Note that at 25 °C irradiation of T91 in STIP-I to 4.3 dpa also resulted in low KJq.

Figure 36 shows the predicted shifts in the master curve reference temperature (A T0) at 100MPaVm for FMS F82H (similar to that for Eurofer97) neutron irradiated at temperatures from 200 to 400 °C as a function of the square root of dpa. The corresponding

 

STIP SPNI AT0 data shown in Figure 35 are esti­mated by adjusting the measured KJc to 100 MPa Vm based on the master curve shape and further taking the unirradiated T0 as —100 °C.248 These approximate, but semiquantitatively correct comparisons show that the synergistic hardening-He mechanism also results in much larger fracture toughness AT0 when com­pared with neutron irradiation with low He. Most notably, the estimated AT0 for the 400 °C irradiation is of the order 700 °C.

Activation Volume for Self-Diffusion

When the crystal lattice is under pressure p, the self­diffusion coefficient changes and is then given by

DSD(T, p) —DSDexp(_QSD/kT) exp(_pVSD/kT) [24]

The activation volume VsD can be obtained experi­mentally by measuring the self-diffusion coefficient as a function of an externally applied pressure. Such measurements have been carried out only for a few

 

0

0 0.5 1 1.5 2

Experimental vacancy migration energy (eV)

Figure 10 Comparison of computed vacancy migration energies according to models by Flynn and Kornblit with measured values.

 

Table 5 Preexponentials for tracer self-diffusion

Metal

M

a (nm)

&d(K)

E m(eV

Experimental value

S&M

Flynn (m2 s 1)

Ag

107.9

0.409

229

0.66

4.5e-6

3.00e-6

5.90e-6

Al

26.98

0.405

430.6

0.61

4.7e-6

5.69e-6

1.08e-6

Au

197

0.408

162.7

0.71

3.5e-6

2.29e-6

4.16e-6

Cu

63.54

0.361

349.6

0.70

1.6e-5

3.55e-6

7.02e-6

Ni

58.71

0.352

481.4

1.04

9.2e-5

4.39e-6

9.18e-6

Pb

207.2

0.495

106.6

0.43

6.65e-5

2.11e-6

4.01e-6

Pd

106.4

0.389

278

1.0

2.1e-5

3.53e-6

6.46e-6

Pt

195.1

0.392

240

1.4

6.0e-6

3.11e-6

5.66e-6

Подпись: ♦ Seeger Mehrer □ Flynn

Подпись: ♦ ♦Подпись: [25]Подпись:

СЛ

10-5

c

0

C

О

CL

X

0

0

Q.

13

о

0

10-6

10-6 10-5 0.0001

Experimental preexponential D0 (m2s-1)

Figure 11 Comparison of preexponential factors for tracer self-diffusion as computed with two models and as measured.

metals, and it has been found that the activation volumes have positive values. Therefore, self-diffusion decreases with applied pressure. However, it has been noticed that the self-diffusion coefficient at melting appears to be constant, and this can be explained by the fact that the melting temperature increases in general with pressure. It follows then from the condition

d [ln DSD (p, Tm (p))] /dp|p=0 = 0

that

Q

Tm dp p=0

where T0 is the melting temperature under ambient conditions.

Brown and Ashby16 have used this relationship to evaluate the activation volumes for self-diffusion

for a variety of metals. Using more recent values for the pressure derivative of the melting tempera­ture by Wallace17 and Wang et a/.,18 one obtains activation volumes as shown in Figure 12. They are in reasonably good agreement with the experi­mental values where they exist. With the exception of Pt, the predicted values are also similar, giving an activation volume of about 0.85O for fcc metals, 0.65O for hcp metals, and around 0.4O for bcc metals.

image039Подпись: [26]Подпись:

image042
Подпись: Figure 13 An atom with its 12 nearest neighbors in the perfect fcc lattice, on the left, and a [001] self-interstitial dumbbell with the same nearest neighbors, on the right.

The equilibrium vacancy concentration in a solid under pressure p is given by

-ref

bT where VV is the vacancy formation volume. Since the self-diffusion coefficient is the product of the thermal vacancy concentration and the vacancy migra­tion coefficient, the activation volume for self-diffusion is the sum of two contributions, namely

Vsd = vf + vm = O + V^el + vm [27]

with V™ being the activation volume for vacancy migration.

If one takes the average of the predicted activation volumes shown in Figure 12, and the vacancy relax­ation volumes from Table 3, one obtains values for V™ listed in Table 6 and also shown in Figure 12.

KrOger-Vink Notation

It is usual for defects in ceramic materials to be described using a short hand notation after Kroger and Vink.6 In this, the defect is described by its chemical formula. Thus, a sodium ion would be described as Na, whatever its position in whatever lattice. A vacancy is designated as ‘V.’ The description is made with respect to the position within the lattice that the defect occupies. For example, a vacant Mg site is designated by VMg and an Na substituted at an Mg site is designated by NaMg. Interstitial ions are represented by ‘i’ so that an interstitial fluorine ion in any lattice would be Fi.

The charge on an ion is described with respect to the site that the ion occupies. Thus, an Na ion (which has formal charge +) sitting on an Mg site in MgO (which expects to be occupied by a 2+ ion) has one too few + charges; it has a relative charge of 1 — which is designated as a vertical dash, meaning that it is written as NaMg. An Al3+ ion at an Mg site in

image205

Figure 4 Disorder enthalpy is proportional to the gradient of a ln [n] versus 1/Tgraph.

MgO has too high a charge. Positive excess charge relative to a site is designated with a dot, thusAlMg. Similarly, a vacant Mg site in MgO is designated by VMg and an interstitial Mg ion in MgO byMg**. Finally, a neutral charge is indicated by a cross ‘x,’ so that an Mg ion at an Mg site in MgO is MgMg.

Ions such as Fe may assume more than one oxida­tion state. Therefore, in MgO, we might find both Fe2+ and Fe3+ ions on Mg sites, that is, FeMg andFeMg. It is also possible to encounter bound defect pairs or clus­ters. These are indicated using brackets and an indica­tion of the overall cluster charge; for example, an Fe3+ ion bound to an Na+ ion, both substituted at magne­sium sites, would be {FeMg: NaMg| . These cases are summarized in Figure 5.

Finally, defect concentrations are indicated using square brackets. Thus, the concentration of Fe3+ ions substituted at magnesium sites in MgO would be
[FeMg] . When we consider the role of hole and elec­tron species, these are represented as h* and e0, respectively.

High Temperature Embrittlement

High temperature helium embrittlement occurs at elevated temperatures (typically near or above 0.5 TM) when sufficient levels of helium are produced by nuclear transmutation reactions and mechanical stress is applied during irradiation. Intergranular fracture is induced by the transformation of grain boundary bubbles to voids, leading to breakaway growth, cavity coalescence, and rupture in the presence of mechanical stress.120,152,153,274-277 The application of tensile stress during high temperature irradiation induces migration of the helium to the grain bound­aries, where large cavities can be formed.120 In the absence of applied stress, the helium bubbles are distributed throughout the material. The observed tensile ductility due to helium embrittlement decreases with decreasing strain rate120,278 and decreasing

image307

Figure 29 Aligned cavity formation and grain boundary separation in Al2O3 following fast fission reactor irradiation to 12 dpa at 1100 K. Reproduced from Clinard, F. W., Jr.; Hurley, G. F.; Hobbs, L. W. J. Nucl. Mater. 1982, 108-109, 655-670.

stress120 (opposite of the behavior observed in many unirradiated metals and alloys), pointing out the importance of exposure time at elevated temper­ature on helium embrittlement. Figure 30 shows examples of the grain boundary microstructures of an Fe-Cr-Ni ternary alloy preimplanted with 160appm He during annealing at 750 °C with and without applied tensile stress.279 Cavity formation along the grain boundary is very limited in the absence of applied stress for annealing times up to 60 h, whereas pronounced grain boundary cavity swelling occurs for annealing times as short as 8h when ^20 MPa stress is applied. Evidence for high temperature helium embrittlement has been observed during tensile and creep testing of austen­itic stainless steel at temperatures above 550 °C (^0.45-0.5 TM) when the helium concentration exceeds ^30appm.255,265,277,280,281 Austenitic stainless steels containing fine dispersions of precipitates exhibit better resistance to helium embrittlement than simple Fe-Cr-Ni alloys, and microstructural investigations suggest that helium trapping at grain interior locations (thereby impeding the flow of helium to grain bound­aries) is an important factor.152,277,282-284 It has been observed that ferritic/martensitic steels exhibit better resistance to grain boundary helium cavity formation and growth compared to austenitic stain­less steels.274,285-287 This has been attributed to

several potential factors, including efficient trapping

Triple grain
junction

 

0MPa

 

/Z

0.1 ^m 18h

 

(a)

 

(b)

 

(c)

 

60 h

 

8h

 

YE-11560

 

lYE-11561

 

Matrix

 

19.6MPa

 

image308image309

image310

Triple grain junction

Figure 30 Effect of exposure time and applied stress during annealing at 750 °C on the formation of grain boundary cavities in Fe-17Cr-17Ni austenitic alloy preimplanted with 160 appm helium. Reproduced from Braski, D. N.; Schroeder, H.; Ullmaier, H. J. Nucl. Mater. 1979, 83, 265-277.

of helium in the ferritic steel grain interior by pre­cipitates and other features, a potentially larger criti­cal radius for conversion of helium bubbles to voids in ferritic steel, and lower matrix strength for ferritic steel compared to austenitic steel.119,274,286,288 The helium bubble densities observed in model Fe-Cr ferritic alloys and commercial ferritic steels following high temperature implantation are comparable to that observed in austenitic steels.11 Relatively good resistance to helium embrittlement compared to austenitic stainless steel has been observed in other bcc metals such as Nb and Nb-1Zr (no severe embrittlement observed for He concentrations up to 100-500 appm),289-291 whereas simple fcc metals such as pure copper are readily susceptible to helium embrittlement even at relatively high (tensile) strain rates at temperatures near 0.5 TM for He concentra­tions of 100-330 appm.292,293

Amorphization in Spinel and Alumina

Another response of materials to irradiation, not discussed up to now, is radiation-induced amorphiza — tion. Amorphization is a structural phase transforma­tion from a crystalline solid to a solid that lacks any long-range order. Typically, the material still main­tains a certain degree of short-range order, but as far as diffraction techniques can discern, any long-range, crystalline order is destroyed following an amorphi — zation transformation.

Amorphization is a metastable process in which material is forced into a glass-like structure, which under thermodynamic equilibrium would be a pro­hibited structure. Amorphization transformations are most prevalent under ambient or low temperature irradiation conditions, such that kinetic recovery mechanisms are not effective at annihilating atomic displacements produced by irradiation. Typically, above a critical temperature, amorphization can be avoided in an otherwise amorphizable material, due to thermal recovery processes.

Amorphization transformations can occur under both ballistic (displacive) and electronic (SHI) radia­tion damage conditions. Under ballistic conditions and depending on the material, amorphization can either occur within a single primary knock-on (PKA) ion track (or other irradiating particle track), or pro­ceed through the accumulation of defects due to overlapping of damage tracks. Amorphization tends to be detrimental to materials employed in radia­tion environments, because the crystal-to-amorphous transformation is usually accompanied by significant volume swelling, mechanical softening, and micro­cracking, to name but a few deleterious effects.

In ceramic materials, tendencies to radiation- induced amorphization are strongly dependent on crystal structure and chemistry, with the vast majority of ceramics exhibiting significant susceptibility to amorphization. One of the key properties that has been correlated quite well to amorphization resistance is ionicity: highly ionic compounds tend to resist amor — phization; highly covalent compounds tend to readily succumb to amorphization at relatively low doses.19

Both spinel and alumina are relatively ionic com­pounds, but interestingly both can be amorphized by both ballistic and electronic damage mechanisms. Single crystal MgAl2O4 spinel was found to amorphize under ballistic ion irradiation conditions at a peak displacement damage level of 25dpa (100 K irradia­tion temperature, 400 keV Xe2+ions)20 A similar result was obtained under in situ ion irradiation con­ditions (30 K irradiation temperature, 1.5 MeV Xe+ ions).21 The critical temperature, Ta for amorphiza — tion of spinel, has yet to be determined, but it is likely to be well below room temperature. (Only below Tc can the material be fully amorphized. Above Ta kinetic recovery dominates and the material is partially to fully crystalline.) Single crystal a-Al2O3 (sapphire) has been observed to amorphize by a ballistic damage dose of about 3.8 dpa (20 K irradiation temperature, 1.5 MeV Xe+ ions, in situ).2 This is a significantly smaller amorphization dose than that for spinel irra­diated under similar conditions. The critical tempera­ture, Tc, for amorphization of a-Al2O3 was estimated to be about 170 K. In both alumina and spinel, the radiation-induced amorphization transformation does not occur by direct, ‘in-cascade’ amorphization but by damage accumulation by overlapping cascades (dam­age tracks). Presumably, neither a-Al2O3 nor MgAl2O4 can be amorphized at ambient temperature or above using displacive radiation damage conditions. However, there is a report of amorphization of a-Al2O3 at a ballistic damage dose of 3-7 keV per atom.23

Under SHI irradiation conditions, where elec­tronic stopping predominates over nuclear stopping, both alumina and spinel undergo amorphization transformations, with significant concomitant volume swelling. In both materials, the transformation does not initiate until ion tracks are overlapped. In poly­crystalline a-Al2O3, the threshold for amorphization was found to be at an accumulated electronic energy deposition of about 1.5 GGy (85 MeV I7+ ions at ambient temperature; amorphization was found to a depth of ^4.5 pm, corresponding to energy deposi­tion cross-sections ranging from ^5 to 20 keV nm-1 per ion.24 In single crystal sapphire irradiated under similar conditions (90.3 MeV 129Xe at room temper­ature), amorphization was found to initiate at the sample surface at an accumulated electronic energy deposition of about 0.3 GGy.25 These authors also observed a correlation between swelling (as measured by surface ‘pop-out’) and amorphization. However, the swelling values obtained from their measure­ments are too large to be realistic (more than 50% volume swelling). Nevertheless, the swelling asso­ciated with SHI radiation-induced amorphization in alumina is substantial. Matzke26 observed ^30% free swelling in Al2O3 irradiated at ^420 K with 72 MeV I+ ions to fluences ranging from 1019 to 1021 ionsm-2 (5-500 GGy at the sample surface).

In MgAl2O4, amorphization and significant surface pop-out were observed in SHI irradiations at 370 K using 72 MeV I+ ions.27 The ion fluences where pop-out was observed were 1 x 1019 and 5 x 1019ionsm~2 (5.3 and 27 GGy, respectively, at the sample surface). The volumetric swelling asso­ciated with this crystal-to-amorphous phase transfor­mation was estimated to be ^35%.28 In summary, huge volume changes appear to be associated with SHI amorphization transformations in model cera­mics such as spinel and alumina.

This concludes the comparison and contrast of radiation damage effects in two model ceramic oxi­des, namely, a-Al2O3 alumina and MgAl2O4 spinel. To make this chapter on radiation effects in nuclear reactor relevant materials as comprehensive as possi­ble, we offer in the following section some notes on additional ceramic materials that are either important currently in nuclear reactor applications or have potential as advanced nuclear reactor materials with respect to future applications. In particular, we con­sider three representative ceramic materials, namely, urania, silicon carbide, and graphite.

Subsequent chapters treat these materials in more detail: uranium dioxide (Chapter 2.02, Ther­modynamic and Thermophysical Properties of the Actinide Oxides; Chapter 2.17, Thermal Proper­ties of Irradiated UO2 and MOX; and Chapter 2.18, Radiation Effects in UO2), SiC (Chapter 2.12, Prop­erties and Characteristics of SiC and SiC/SiC Composites and Chapter 4.07, Radiation Effects in SiC and SiC-SiC), and graphite (Chapter 2.10, Graphite: Properties and Characteristics; Chapter 4.10, Radiation Effects in Graphite; Chapter 4.11, Graphite in Gas-Cooled Reactors; and Chapter 4.18, Carbon as a Fusion Plasma-Facing Material).