PSI - Issue 40

Sergey Smirnov et al. / Procedia Structural Integrity 40 (2022) 378–384 Sergey Smirnov, Marina Myasnikova / Structural Integrity Procedia 00 (2022) 000 – 000

380

3

Fig. 1. The microstructures of the model materials: (a) the composite; (b) the complexly alloyed brass, ×400 .

According to the approach, the composite volume at the microscale is modeled by a cube with a 30 µm edge size, which represents the aluminum matrix with embedded silicon carbide particles (Fig. 2). The microstructural properties of the metal matrix composite were chosen according to metallographic investigation (Pugacheva et al. (2016)). The structurally inhomogeneous microvolume is surrounded with a buffer layer. The layer has smeared macroscopic mechanical properties of the composite and dilates evenly from the microvolume borders. The volume thickness is equal to the linear microvolume size. The rheological properties of commercially pure aluminum were set as strain-hardening curves according to compression tests with cylindrical specimens at the macroscale (Smirnov et al. (2017)). Testing was performed with a strain rate of 1 s − 1 at 300 °C (the experimental studies used the equipment installed at the Plastometriya shared access center of the IES UB RAS). To develop a computational model of brass at the microscale, we dealt with structurally inhomogeneous cells (microcells) being an array of structural-phase constituents specified as a system of interconnected regions, with their geometries, dimensions, and properties corresponding to those of actual structural constituents of the brass tested. The initial image of the brass structure was determined by the results of quantitative metallographic analysis with the application of the Monte Carlo statistical method. The coordinates of the metallographic section points in whose vicinity the structure was photographed were selected as random values (Fig. 3a). The images of the microstructure were digitized and further used to develop the geometrical models of deforming areas. The area was chosen to be approximately 140×160 µm in size, and it was enclosed in a buffer layer with smeared material properties. Altogether 10 photographs were treated, and this ensured adequate sampling for subsequent statistical averaging of the results. The buffer layer was composed of eight identical homogeneous and isotropic cells, their sizes being consistent with that of the central microcell (Fig. 3b).

Fig. 2. The computational model of the three-dimensional metal matrix composite.