Theories of Creep in Ceramics

Abstract Mathematical models that have been proposed for creep in ceramics are described. Emphasis is on models involving grain boundary motion (sliding or flow). In Lifshitz models the crystalline grains elongate with strain; the elongation results from diffusion, slip, or solution and precipitation. In Rachinger models the grains do not elongate during creep. The sliding strain can be accommodated by viscous flow of a glassy phase at the grain boundaries, or if there is no boundary glass by diffusion or slip in superplastic models. Sliding of a glass-free boundary can result in cavitation, cracking, or formation of boundary dislocations or triple point folds. Most models of ceramic creep at high temperatures predict a steady state (stage II) creep rate that depends on the applied stress, grain size, and temperature. A general equation for the creep rate as a function of these factors, as well as the elastic modulus and a diffusion coefficient, is used to compare models. The models give different exponents for the functional dependence of creep rate on grain size and strain and different temperature dependencies. These differences are compared in tables, and the main mechanistic features of the models are described in the text. The purpose of this review is to describe creep models rather than to compare them with experimental results or to select the most applicable models. There are few critical experimental tests that allow selection of the most accurate models; such experiments are suggested as the next step in choosing between the models for specific experimental results.