PSI - Issue 77

Valeria Lemkova et al. / Procedia Structural Integrity 77 (2026) 279–291 Valeria Lemkova and Florian Schaefer / Structural Integrity Procedia 00 (2026) 000–000

282

4

A tension stress state is assumed to result in particle or interfacial fracture leading to a better particle erosion or to promoted crack initiation even during the manufacturing. Comparable moduli of ceramic particles and matrix leads to an increased influence of the local stress state at sharp edges and should result in particle erosion tending to rounding and a evolution of the microstructure near the interface that reflects the local stress state.

2.2. Materials processing by HPT

As early as 1935, Bridgman Bridgman (1935) recognized that a combination of a high hydrostatic pressure on disc-shaped samples and a heavy shear deformation results in an enormous increase in the strength of a material. During HPT, two anvils with a cavity for the pressed green body in the case of the powder route or for a bulk metal disk are twisted against each other. The high shear straining results in dynamic recrystallization and a continually grain refinement down to a grain size below 100 nm. The equivalent strain ε eq is given by:

2 · π · n · r t · √ 3

ε eq =

n denotes the number of rotations applied, t the sample thickness assumed to be constant and r the distance from the center of the disk Bridgman (1935); Sabirov et al. (2005). A pressure of 7.6 GPa and a counter-rotation of 0.6 rpm was applied during manufacturing the specimen material. The equivalent strain at 3 mm from the specimen center with a diameter of 8 mm was varied between ∼ 35 (2 rotations) and ∼ 900 (50 rotations). If necessary, due to severe hardening or cracking during HPT and in order to achieve a good mixing between matrix an embedments, HPT was also performed at 300 ◦ C . A300 ◦ C step was always finished with 5 post-rotations at room temperature afterwards, to achieve comparable grain sizes. This additional step is included in the sample designation. For example, the addition 50T + 5R denotes that the sample has undergone 50 rotations at 300 ◦ C (T)and then 5 rotations at room temperature (R). Prior to the powder HPT, the ceramic particles were pre-dispersed in a two step process: • The particles and the metal powder were mixed together with 1 wt% of the ceramic by mortaring. • The pre-mortared powder is dissolved with the solvent ethylene glycol and dried at 175 ◦ under continuous homogenization by stirring. Isochronal heat treatments were conducted in ambient air from 150 ◦ C to 650 ◦ C . Below150 ◦ C no influence of thermal loading on the mechanical properties of nc materials from the HPT powder route is known. The temperature was increased in steps of 15 − 25 ◦ C for 15 min and Vickers hardness HV 0.05 was probed ex situ as an indicator for the evolution of the grain size. The hardness was probed 3 times at the same disk radius for comparable test results. The final grain size was estimated using electron backscatter di ff raction (EBSD, Nordlys nano detector with the Aztec, Oxford Instruments). All scanning electron microscope (SEM) measurements were performed with a Zeiss Sigma VP field-emission SEM. 2.3. Characterization of Thermal Stability

2.4. Microstructural Evolution

In situ heat treatments in the SEM were performed analogous to the ex situ treatments to reveal not only the evolution of the grain size but also the evolving grain size distribution and homogeneity. Hereto the specimens were mounted to a heating stage (Kammrath & Weiss) with a special designed heat shield using a high temperature Ni paste (plano GmbH). High resolution images were taken using the in lens detector of the SEM system. The setup was pre-heated for 1 h at 150 ◦ C to ensure a consistent average temperature. The temeperature was increased in steps of 25 ◦ C at maximum. Each step lasted 30 min. This procedure was continued until 500 ◦ C . This temperature was held