Institute of Fundamental Technological Research

This paper explores an anomalous size effect in thermal residual stresses occurring in the alumina matrix of Al2O3/Cr sintered composite upon varying the particle size of the chromium reinforcement. When a coarse chromium powder (45 µm mean particle size) is used the average residual stress in the alumina phase after cooling is compressive in accordance with the classical Eshelby solution. However, in the case of a fine chromium (5 µm mean particle size) it switches to tension. This effect, detected by photoluminescence piezospectroscopy, is also confirmed by X-ray and neutron diffraction experiments. As the classical micromechanics models are incapable to capture it, a finite element model is developed with the actual composite microstructure being reconstructed from the microtomography images. It is shown by numerical simulations that the anomalous size effect is associated with the complex microstructure of the composite fabricated with the fine chromium powder. It is also pointed out that the temperature dependence of the coefficients of thermal expansion of the matrix and the reinforcement affects the residual stress levels.

thermal residual stress, metal-ceramic composites, size effect, microcomputed tomography, finite element analysis

This work describes optical measurements of processing-induced thermal residual stresses in an alumina matrix reinforced with chromium particles. This ceramic-metal composite is manufactured by the powder metallurgy method comprising powder mixing in a planetary ball mill and consolidation by hot pressing. Two different chromium powders (5 μm and 45 μm mean particle size) are used, while the average alumina particle size is kept constant (1 μm). The residual stresses in aluminum oxide are determined by applying two optical methods: photoluminescence piezo-spectroscopy (PLPS) and Raman spectroscopy (RS). Both experimental techniques reveal a chromium size effect in the residual stress measurements. When the fine chromium powder (5 μm) is used, the average residual stress in the ceramic phase is tensile (unusual effect), whereas for the coarser chromium powder (45 μm) it becomes compressive. The PLPS measurements of the hydrostatic residual stress component in the ceramic phase yield the values of 0.290 and −0.130 GPa for samples with 5 μm and 45 μm chromium powders, respectively. In the RS experiments, the corresponding stress component in the alumina equals 0.351 GPa for the composite with 5 μm chromium and −0.158 GPa for that with 45 μm chromium powder. These values indicate that the residual stress in the alumina reinforced with 5 μm chromium is approximately twice higher than that in the alumina reinforced with 45 μm chromium. Finally, the validity of the results obtained with the optical techniques is confirmed by the neutron diffraction measurements.