A multiaxial theory of creep deformation for particle-strengthened metal-matrix composites is derived. This derivation is based on the observation that there are two major sources of creep resistance in such a system. The first, or metallurgical effect, arises from the increased difficulty of dislocation motion in the presence of particles and is accounted for by a size- and concentration dependent constitutive equation for the matrix. The second, or mechanics effect, is due to the continuous transfer of stress from the ductile matrix to the hard particles and the corresponding stress redistribution is also incorporated in the derivation. Both power-law creep and exponential creep in the matrix, each involving the transient as well as the steady state, are considered. The constitutive equations thus derived can provide the development of creep strain of the composite under a combined stress. The multiaxial theory is also simplified to a uniaxial one, whose explicit stress-creep strain-time relations at a given concentration of particles are also given by a first- and second-order approximation. The uniaxial theory is used to predict the creep deformation of an oxide-strengthened cobalt, and the results are in reasonably good agreement with the experiment. Finally, it is demonstrated that a simple metallurgical approach without considering the stress redistribution between the two constituent phases, or a simple mechanics approach without using a modified constitutive equation for the metal matrix, may each underestimate the creep resistance of the composite, and, therefore, it is important that both factors be considered in the formulation of such a theory.
High-Temperature Creep Resistance Resulting from Grain Morphologies in P/M Tungsten Fine Wires