ON DISLOCATION LINK LENGTH STATISTICS FOR STRAIN-HARDENING AND RECOVERY DURING ELEVATED-TEMPERATURE PLASTIC-DEFORMATION

Tensile tests are performed for an AISI 310 type stainless steel over a range of test temperatures and strain rates. The experimental result s generally show that the strain hardening behaviour of the 310 stainl ess steel has two distinctive regions, namely: (i) a low-temperature r egion, in which strain hardening decreases linearly with strain and is almost independent of the strain rate; and (ii) a high-temperature re gion (where the temperature is higher than about half of the melting p oint), in which strain hardening is affected by strain rate, and the s train hardening coefficient increases with an increase in strain rate at a given temperature. For the 310 stainless steel, the low-temperatu re region is from 298 to about 873 K, and the high-temperature region is above 873 K. The dynamic effects of strain hardening and recovery p rocesses during elevated temperature plastic deformation of the 310 st ainless steel are then analyzed by means of dislocation link length st atistics. The main findings are: (i) The strain hardening coefficient for an elevated temperature tensile test is given by theta = H - 2A(0) (-1)psi(t) R/epsilon, where H is strain hardening coefficient for low- temperature plastic deformation without recovery, A(0) a numerical con stant about unity, psi(t) is dependent on dislocation structure during deformation, R recovery rate, and epsilon strain rate; (ii) for stead y state deformation, the relationship between flow stress, sigma, and dislocation density, rho, namely sigma = alpha(1)M mu b rho(1/2), can be deduced from this analysis, where alpha(1) is a constant, M the Tay lor factor, mu shear modulus, and b the Burgers vector; (iii) the disl ocation annihilation rate, rho(a), has a stronger dependence on stress than recovery rate, R, and strain rate, epsilon(s). The dislocation a nnihilation rate, rho(a), is proportional to the dislocation density, rho, in the manner rho(a) proportional to rho(m), where m = 2 to 3 is a constant. Direct comparison of these new results from dislocation li nk length statistics is made with the experimental results for the 310 stainless steel. Agreement is good between the analysis and the exper imental results.