For materials known to creep at high stresees bydiffusion-controlled generation and movement of dislocations, adebate has been continuing for over a decade on the mechanismsresponsible for creep under low-stress conditions. Evidence hasbeen obtained to support the elegant diffusional creep theories,which are based on stress-directed vacancy flow withoutdislocation movement. This evidence relies mainly on(a) experimental results for copper, which indicate that the creeprate can vary with stress, temperature and grain size in themanner predicted by diffusional creep theories, and(1)) the observation that, in certain particle-hardenedmagnesium alloys, 'denuded zones' have been detected on grainboundaries normal to the tensile creep stress.However, creep rates well above those predicted by diffusionalcreep theories have been reported for single crystals and coarse-grain polycrystals, demonstrating that creep must be able tooccur as a result of dislocation creep mechanisms, even at lowstresses. Furthermore, analyses of the shapes of creep curvesrecorded at high stresses offer reasonable predictions of thecreep rates determined experimentally at low stresses inindependent investigations completed for magnesia and low-alloy ferritic steels, suggesting that essentially the samedislocation creep processes are rate controlling at high and lowstress levels. For these reasons, detailed assessments have beenmade of case studies widely quoted in favour of diffusionalcreep theories, indicating that this evidence is inconclusive. Inparticular, it appears that the 'denuded zones' are developedduring fabrication and heat-treatment of the magnesium-zirconium hydride alloys rather than during creep exposure,which would account for the fact that such zones are rarelyobserved during low-stress creep exposure of commercialparticle-strengthened alloys. Even so, there is no evidence toprove that diffusional creep processes cannot occur. In thecontext of the present unresolved debate, it is perhaps salutary tonote that the evidence most quoted for and against diffusionalcreep concepts was obtained prior to 1970. Clearly, arequirement exists for new comprehensive ranges of criticalexperiments designed specifically to clarify the mechanismsdominant during creep of different materials over extendedstress ranges at high temperatures.
展开▼
