26. Irradiation Induced Mechanical Property Changes: Hardening and Embrittlement

Transcription

1 26. Irradiation Induced Mechanical Property Changes: Hardening and Embrittlement 26.1 Introduction Changes in the Stress-strain curve after irradiation Hardening by Irradiation induced Obstacles Microstructure evolution and defect density Source hardening and friction hardening Hardening by depleted zones Precipitate hardening Deformation localization and loss of ductility Dislocation channeling Flow softening Loss of strain hardening Reduction of fracture toughness Change of plastic zone size Impact testing Pressurized thermal shock Charpy and NDT Monitoring pressure vessel embrittlement Helium embrittlement... 8 Table of contents page Chapter heading on p. 0: Bold - 16 font -centered dots from end of entry until page number on right dash (-) then title in font-12 italics Problems, then References 7/20/2011 1

2 26.1 Introduction Because core materials perform structural functions and provide containment for fission products, it is crucial to understand their response to mechanical loads or to deformation, i.e. their mechanical behavior. In particular it is important to determine the material s mechanical behavior after the material has been exposed to irradiation and its microstructure and mechanical properties have been altered as a consequence. The mechanical behavior of unirradiated materials was reviewed in Chapter 11. Exposure to the reactor environment alters material properties such that mechanical behavior is modified. For the materials of concern zirconium alloy fuel cladding, reactor internals and the pressure vessel all of which are metallic, most material property changes can be understood in terms of a stress-strain curve such as shown in Figure Irradiation causes hardening, reflected in an increase of the yield stress, embrittlement, reflected in a reduction of the uniform strain or the fracture strain, and a loss of fracture toughness, reflected in an overall decrease of the area under the stressstrain curve. In this chapter we discuss the mechanisms by which these irradiationinduced changes occur. Other mechanical property changes of importance caused by irradiation, such as increased creep rate, increased fatigue failure or enhanced stress corrosion cracking, are discussed in other chapters. We should also note that the exposure to the reactor environment can cause embrittlement by chemical processes related to corrosion, such as hydrogen ingress or stress-corrosion cracking. These are discussed in Chapters 22, 25 and Changes in the Stress-strain curve after irradiation It is useful to discuss the changes in mechanical properties using the stress-strain curves obtained in simple round bar tensile tests. It is also interesting to separate these into the different crystal structures of interest. We start with the fcc crystal structure. Figure 26.1 shows a sequence of tensile tests conducted on stainless steel before and after neutron irradiation. Both the irradiations and the mechanical tests were conducted near room temperature. In the unirradiated state the yield stress for this particular steel at room temperature is about 200 MPa. The most noticeable effect of irradiation is that at the highest dose (0.78 dpa) the yield stress is to nearly 700 MPa, an increase of over 300% over the unirradiated sample value. As will be seen in the following, the principal cause of this irradiation hardening is the creation by neutron irradiation of a high density of defect clusters that hinder dislocation motion and thus delay the onset of plastic deformation. As a result, yield stress increases with fluence until no more defect clusters can be created, such that a saturation yield stress is reached for that irradiation temperature. This saturation behavior is not shown in Figure 26.1, likely because the fluence is not high enough. 7/20/2011 2

3 In parallel with the yield stress increase, the ductility decreases: both the total elongation and the uniform elongation both decrease with increasing fluence. This ductility decrease is likely caused by various mechanisms of deformation localization that are present in the irradiated material or which operate at higher applied stress. One simple way of looking at this problem is to re-apply the Considere criterion (Ch. 10) for necking instability. As will be recalled, this criterion predicts necking when the ability of the material to workharden during deformation can no longer compensate the plastic deformation and the material starts to neck. Insert derivation of Considere The derivation above considers that the strain hardening coefficient n remains constant. Whereas in fact, the increasing defect density causes dislocation generation during deformation to become increasingly more difficult. Thus the work hardening mechanisms present in unirradiated materials are not as easy to activate for irradiated samples, which causes n to decrease. In that case a straight application of the Considere criterion would predict that ε U ~n, which means again that the ductility would decrease with decreasing n. The changes in the stress strain curves with neutron dose in this case are gradual, and the curves at various doses are reasonably self-similar - all show a well defined yield, work hardening and exhibit significant ductility. In fact, at the highest dose the ductility, although decreased from the initial value of 60%, is still 30%. Figure 26.1 Stress strain curves obtained during tensile testing of stainless steel at room temperature before and after irradiation. Include Byun reference 7/20/2011 3

4 Figure 26.2 Stress strain curves obtained during tensile testing of bcc steel at room temperature before and after irradiation. Figure 26.2 Ferritic steel Figure 26.2 shows stress-strain curves obtained in ferritic steel (bcc) alloys at various doses. Also, in this case the yield stress increases and both work hardening and ductility decrease with irradiation dose. However the stress strain behavior varies qualitatively with dose, and after 1 dpa, the stress strain curve is considerably different than that obtained when testing the material in the unirradiated state. As shown in Figure 26.2a, in the unirradiated state the post-yield behavior of bcc materials often exhibits a flat region which is followed by classical strain hardening, 7/20/2011 4

5 necking and fracture. The yield drop and the zero strain hardening region are related to macroscopic deformation localization, in which Luders bands develop within the gage section of the test sample. The stress required to cause deformation when bands are present, is less than the yield stress, thus, the yield drop. Also, while these bands are forming, additional deformation can take place without any additional applied stress. (This again illustrates the fact that it is necessary to examine the mechanical test sample and procedure to ensure that the physical phenomena reflected in the stress-strain curve are properly understood.) Once these bands propagate over the entire gauge length, then work-hardening can start to occur. The specimens exhibit considerable ductility (~20%) and fail in a ductile manner. As the irradiation dose accumulates in the sample to 10-4 and 10-2 dpa, both the yield stress and the ultimate tensile strength increase. Because the former increases faster than the latter, the work hardening (and thus the work hardening coefficient) also decreases. It is clear that at 0.01 dpa there is little work hardening left. The ultimate tensile strength is the same as yield (no significant uniform deformation can occur). Concomitantly, the sample elongation (both total elongation and uniform elongation) decreases. As the dose increases further to 0.1 and 1 dpa, further increases in yield strength are seen, accompanied by a flow instability that causes immediate failure upon yield. When the material yields, a drop in stress is observed, reflecting a flow stress lower than the yield stress (the stress to keep dislocations moving is less than that to get them started). In this case, little deformation occurs before failure and once the material yields it immediately undergoes macroscopic localization (necking). This is illustrated in Figure 26.2c, which shows that the ultimate tensile strength equals the yield stress at 0.01 dpa. Both uniform and total elongation drop precipitously at 0.1 dpa. Figure 26.1: Stress strain curves for non-irradiated and irradiated Zircaloy-4, tested at room temperature. 7/20/2011 5

6 Figure 26.3 shows the stress-strain curves obtained during a tensile test of recrystallized Zircaloy-4 before and after irradiation. In the non-irradiated condition, the yield stress is a little under 150 MPa, and considerable ductility exists. After the material is irradiated to 5 x n/cm 2, the yield stress increases to about 230 MPa. Although the total deformation before fracture is considerable, the strain hardening is considerably reduced. Finally, after irradiation to 8 x n/cm 2, the yield stress exceeds 350 MPa, and the ductility becomes less than 1%. It is clear, furthermore, that the area under the curve has been drastically reduced by exposure to irradiation, which directly relates to a reduction in fracture toughness. Similar changes in the stress strain curves of other metals are observed after exposure to irradiation, with particular differences. Figure Plot of yield stress versus irradiation dose 7/20/2011 6

7 Figure 26.5 Plot of n and elongation versus dose Fractography 26.3 Hardening by Irradiation induced Obstacles As illustrated in Figures a common observation in mechanical property changes of irradiated materials is the increase in yield stress, also referred to as irradiation hardening. The principal reason for this hardening is the development during irradiation of a high density of defect clusters of various types. This higher defect concentration reduces the spacing between defects, requiring greater dislocation bending around obstacles and thus a higher stress for dislocation motion to occur. Thus plastic deformation occurs at a higher stress, which is equivalent to saying the yield stress increases. As shown above, this hardening is normally accompanied by embrittlement which is the mechanical property change of greatest impact for engineering applications. 7/20/2011 7

8 The same microstructure changes that cause hardening will also directly or indirectly affect the available modes of plastic deformation, the overall uniform or localized nature of this plastic deformation, and the material elongation, (both uniform and total). Because ductility is often specimen and test-dependent and much more difficult to characterize than hardening, the latter is used in place of the former when characterizing irradiation induced changes in mechanical properties. The basic principle follows the calculation of the stress necessary to move a dislocation between obstacles. As the obstacles become more closely spaced, the stress to bend the dislocation enough that it can overcome the obstacles increases such that Δσ y is proportional to 1/l. Many different obstacles and combinations of obstacles can be formed during irradiation and their overall effect on hardening is reviewed in the following sub sections Microstructure evolution and defect density Source hardening and friction hardening Hardening by depleted zones Precipitate hardening Deformation localization and loss of ductility Dislocation channeling In irradiated metals the deformation when examined at the scale of the grain can be localized into dislocations channels Flow softening Loss of strain hardening => increased plastic instability Reduction of fracture toughness Change of plastic zone size 26.6 Impact testing Pressurized thermal shock Charpy and NDT Monitoring pressure vessel embrittlement Helium embrittlement 7/20/2011 8

9 cladding General idea is that irradiation -induced microstructure causes deformation localization and consequent loss of ductility Loss of fracture toughness by reducing size of plastic zone. Enhancement of creep (ch 27) Effect of irradiation on stress strain curve (increased sy, loss of work hardening and both uniform and total strain decrease) Mechanisms are the formation of effect cluster, IIP., Models of yield stress incerase Measurement is with Charpy Test/Master Curve Loss of fracture toughness Helium embrittlement Hydrogen embrittlement in Zircaloy Effect on stress-strain curve Relationship to microstructure (Russell-Brown theory of obstacle hardening) Embrittlement; hardening, fatigue effects Reduction of creep rupture time Deformation localization Pressure vessel embrittlement Mechanical Properties (of UO2) - Elastic properties - Fracture and Flow properties Creep Irradiation Hardening and Embrittlement Application to the Pressure Vessel In light water reactors, (LWR), the principal pressure boundary is the reactor vessel. The vessel contains the reactor core and maintains the pressure necessary to keep the coolant liquid at high temperatures. In the case of accidents the pressure vessel helps contain the fission products and holds the water from the emergency core cooling system. As such, it is a critical safety and licensing issue that the pressure vessel maintain its integrity in face of the most severe combinations of stressors in the design basis. Nuclear reactor pressure vessels are made of ferritic steel, with a stainless steel cladding. A PWR pressure vessel has a diameter of 4.25 m and is 23 cm thick. The most serious design accident for the pressure vessel involves a loss of coolant accident (LOCA) followed by a sudden recooling upon reinjection of coolant. This causes a Pressurized Thermal Shock (PTS).At the beginning of the reactor s life the pressure vessel material is ductile and able to withstand PTS. The concern is that, as the pressure vessel is exposed to neutron irradiation, it becomes embrittled and less able to resist PTS. The failure mechanism is the unstable propagation of an existing crack or flaw, under the application of the loads attendant upon PTS. The 7/20/2011 9

10 effect of neutron irradiation is to reduce the size of the critical flaw size required for unstable propagation, or conversely, to reduce the stress required for unstable propagation of a crack of a given size. Utilities applying for plant-life extension have to show that the radiation-induced embrittlement has not been enough to allow the maximum existing flaw to propagate in a brittle manner. Pressure Vessel Annealing and Re-embrittlement Models 7/20/

11 Irradiation Hardening and Loss of Ductility (Embrittlement) As a result of irradiation, metals exhibit an increase in the both the yield stress σ y and the ultimate tensile strength σ UTS. While the two values increase, they also become closer, which translates to a decrease in ductility. This manifests itself in the different responses that materials have to stress rupture tests before and after irradiation Irradiated high Notch Toughness Temperature Y.Nikolaev, A.Nikolaeva, Effects of Radiation on Materials 19 th Symposium, ASTM STP 1366, 2000, p.460 Charpy V-notch impact test h 2 h 1 7/20/

12 The energy absorbed in fracturing is given by E g ( h h ) abs = 1 2 The energy absorbed is a measure of fracture toughness, but has no mechanistic relationship to K IC or other material properties. E abs (J) Brittle 41 J upper-shelf energy lower-shelf energy Ductile NDT (DBTT) T (C) 7/20/

13 The advantage of the Charpy test is that it is simple and small and correlates with the fracture toughness. Empirical correlations have been developed to relate the yield stress with K IC, one example is given below: K σ IC y 2 5 σ y = Eabs σ 20 y This is an empirical correlation so each material has its own. Design Applications : σ Service Stress (applied, thermal, residual) K IC tough material K ISCC K IC weaker material c Dynamic Fracture Toughness Maximum allowable flaw size for steel with K IC Under dynamical conditions (high strain rate) we use a different measure of resistance to fracture, K ID K ID < K IC Once moving, the crack will arrest at K Ia 7/20/

14 Irradiation Hardening Neutron Irradiation increases the σ y (yield strength). Hard materials tend to be more brittle. Non-Irradiated ΔUSE Irradiated ΔDBTT(ΔNDT) The increase in yield stress is related to the stress necessary to have a Frank-Read dislocation source operating, That stress is the stress needed to have the dislocation bow between pinning points, limit is line tension E L 2 Gb Force per unit length due to stress F L = τ b Restoring force due to line tension = τ is maximum for R minimum 2 Gb R θ R R L min = 2 τ = Gb R τ FR = 2Gb L This equation is valid also for the obstacles formed during irradiation, because the dislocation has to either bow around them or cut through them (normally less favorable 7/20/

15 energetically). Each barrier will then have a given density, which can be calculated as follows: obstacles radius r 2r Obstacle density is N (cm-3). Number of obstacles in volume above is 2rN, which is also the number of intersections per unit area on the slip plane. If we now equate the inverse square of the average obstacle spacing with the density of intersections on the plane we get L = 1 2rN and thus the increase in obstacle density causes a corresponding increase in yield stress given by ( Δσ ) y k 2Gb = Lk where k stands for all possible obstacles to dislocation motion formed under irradiation (dislocation loops, precipitates, depleted zones, etc) 7/20/

16 Mechanisms of Hardening by Irradiation σ ( irrad) = σ ( non irrad) +Δσ y y y Δσ = ( Δσ ) + ( Δσ ) + ( Δσ ) + ( Δσ ) y y voids y def. clusters y ppts y loops In addition, irradiation creates helium by (n,α) reactions. Helium weakens the grain boundaries and reduces the elongation at fracture. This is embrittlement without hardening. Elongation to fracture (%) 8 annealing of depl.zones and low T defs. 6 onset of He embrittlement 4 matrix softening T (C) Atom maps showing the solute distribution in neutron irradiated KS-01 weld.a high number density of Cu-,Mn- Ni-,Si-and 7/20/

17 P-enriched precipitates is evident.a Cr-,Mn-,Ni-,Cu-,C-,N-,Si-and Mo-enriched feature is also evident. Auger et al. Journal of Nuclear Materials 280 (2000) 331}344 7/20/

18 Pareige et al. NIM B, /20/

19 Analysis of Pressurized Thermal Shock In the case of a loss of coolant accident it is possible for the core to become uncovered, and therefore with greatly diminished cooling. Because the decay heat produced by the fuel needs to be removed so the fuel does not melt, it is necessary to inject coolant into the core. However the injection of cooling water can cause steep thermal gradients, which can cause large tensile thermal stresses at the inner wall. In such a scenario a crack already present in the wall could propagate in an unstable manner causing fast fracture. σ x Flaw c Cold Leg T=280 C CORE Hot Leg T=320 C p=15.5 MPa Stainless Steel Cladding Pressure Vessel Figure 25.XX analysis of PTS At the beginning of life the critcal flaw size for brittle fracture is large, since the stress intensity factor for unstable crack propagation KIC is 7/20/

20 7/20/