The internal creep rupture properties of the manu­factured 12Cr-ODS steels at 700 °C are shown in Figure 20.36 Increasing the yttria and titanium con­tents improves the internal creep rupture strength (F4 > F3 > F2 > F1). The uni-axial creep rupture strength for F4 is also plotted; there is the strength anisotropy between the uni-axial and internal hoop directions. This strength anisotropy can be associated with the slightly elongated grain structure shown in Figure 19.

The stress-strain rate relationship was investi­gated for ODS ferritic claddings to evaluate the creep deformation mode. The results of the analyses are given in the log-log plot in Figure 21.36 In general, the creep strain rate in the steady-state con­dition is expressed using applied stress o as:

" = As” [5]

where n is the stress exponent and A is the temperature-dependent coefficient.37 In the case of

1000

ra

CL

СЛ

CL

о

о

X

the uni-axial creep mode, a significantly high stress sensitivity of n = 43.7 appears. This stress exponent value is typical for an ODS alloy.37 The applied stress that initiates the strain is clearly located around 250 MPa; this stress corresponds to the so-called threshold stress for deformation. On the other hand, the stress exponent, n, is 10.4 for the internal creep
mode of F4, and a higher strain rate is found even below a stress of 200 MPa. A transverse section of this specimen shows finely equi-axed grains of 5-10 pm (Figure 19). Apart from pinning the gliding dislocations due to oxide particle-dislocation interaction, the defor­mation mechanism associated with grain morphology may be the dominant factor that induced accelerated strain in the hoop stress mode of the tubular specimen.

In order to characterize the high temperature strength of manufactured 12Cr-ODS steel cladding, its strength mechanism was evaluated from the view­point of the interaction between Y2O3 particles and dislocations. This interaction could be formu­lated by the void-hardening mechanism proposed by Srolovitz,38 in which oxide particles were re­placed by voids. The oxide particle-hardening stress sp can be evaluated by the following equation based on Scattergood and Bacon’s equation,39 which takes into account the interaction between the branches of the bowed-out dislocation around a Y2O3 particle:

sp/G = AMb/(2nX)[ln(D/r0) + B], [6]

for screw dislocation,

A =(1 + v sin2′)cos ‘/(1 — v),

B = 0.6

for edge dislocation,

A = [l — v sin2’/(1 — v)] cos ‘,

B = 0.7

where G is the Shear modulus, v is Poisson’s ratio, M is the Taylor factor, b is the magnitude of Burgers vector, and r0 is the inner cut-off radius of the dislo­cation core. The value of ‘ is the critical angle at which the dislocation detaches from the particles. This value was estimated to be ‘ = 46° for screw dislocations and ‘ = 19° for edge dislocations. Fur­ther, l is the average face-to-face distance between particles on a slip plane and is given as a function of the average particle radius rs and the average center — to-center distance ls between the particles by

l = 1.254 — 2rs, [7]

where the averages are calculated by considering the size distribution of the particles. The factor 1.25 is the conversion coefficient from regular square distri­bution to random distribution.40 The characters ls and rs represent the results of the measurement of oxide particles by means of TEM. D is the harmonic mean of 2rs and l. The values of l were calculated, and the oxide particle-hardening stress was estimated by

substituting l, M = 3.0,41 n = 0.334, b = 2.48 x 10 10 m, and G = 50 600 MPa, at 700 °C.

Figure 22 shows the results of analyses in rela­tion to the face-to-face distance between particles.36 The oxide particle-hardening stress levels estimated by using the aforementioned equations at 700 °C are represented by vertical bars, with the upper and

Face-to-face distance between particles, l (nm)

lower bars derived from an estimate of edge and screw dislocations, and with the uncertainty of r0 ranging from b to (3 x b). The measured stress in the uni-axial mode of the F4 specimen is shown by an open circle. These results imply that the higher oxide particle-hardening stress for specimen F4 is due to its shortened face-to-face particle distance l of 70 nm. The lower band represents the stress corresponding to a strain rate of 10~9s_1 in the internal hoop direc­tional mode. For the F1 specimen, as a stress level corresponding to a strain rate of 10~9s_1 approaches the oxide particle-hardening stress, the strong anisot­ropy tends to disappear. However, for the F3 and F4 specimens with a shortened distance between par­ticles, stress levels for a strain rate of 10~9s_1 in the hoop direction are degraded from the oxide particle­hardening stress. The strong anisotropy still remains in the F4 specimen. The accelerated deformation in the internal hoop direction could be the result of grain boundary sliding, since finely equi-axed grains with a small size of 5-10 pm are formed, and the grain bound­aries occupy a large fractional area in the transverse cross-section of the F4 specimen (see Figure 19). Based on these results, it seems to be difficult to control internal creep rupture strength by recrystallization processing in 12Cr-ODS steel cladding.