Direct measurement of isothermal flow stress of metals at elevated temperatures and high strain rates with application to Ta and Ta [sbnd]W alloys

Abstract

A technique is developed for measuring the flow stress of metals over a broad range of strains, strain rates, and temperatures, in uniaxial compression. It utilizes a recent, enhanced version of the classical (Kolsky) compression split Hopkinson bar [1], in which a sample is subjected to a single stress pulse of a predefined profile, and then recovered without being subjected to any other additional loading. For the present application, the UCSD's split Hopkinson bar is further enhanced by the addition of a new mechanism by means of which the incident and transmission bars of the split Hopkinson construction are moved into a constant-temperature furnace containing the sample, and gently brought into contact with the sample, as the elastic stress pulse reaches and loads the sample. The sample's temperature is measured by thermocouples which also hold the sample in the furnace. Since straining at high strain rates increases the sample's temperature, the sample is allowed to attain the furnace temperature after each controlled incremental loading, and then is reloaded, using the same stress pulse and strain rate. The technique also allows for checking any recoveries that may occur during unloading and reloading. Using several samples of the same material and testing them at the same strain rate and temperature, but different incremental strains, an accurate estimate of the material's isothermal flow stress can be obtained. Additionally, the modified Hopkinson technique allows the direct measurement of the change in the (high strain-rate) flow stress with a change of the strain rate, while the strain and temperature are kept constant, i.e., the strain rate can be increased or decreased during the high strain-rate test. The technique is applied to obtain both quasi-isothermal and adiabatic flow stresses of tantalum (Ta) and a tantalum-tungsten (Ta W) alloy at elevated temperatures. These experimental results show the flow stress of these materials to be controlled by a simple long-range plastic-strain-dependent barrier, and a short-range thermally activated Peierls mechanism. For tantalum, a model which fits the experimental data over strains from a few to over 100%, strain rates from quasi-static to 40 000/s, and temperatures from −200 to 1000°C, is presented and discussed.