• Hot deformation behavior of a novel biomedical β-Ti alloy, Ti-5Mo-20Nb-10Zr-5Ta-1Fe, is elucidated for the first time. • Three constitutive models (Arrhenius, multiple linear regression, and ANN) are developed and compared, with the ANN showing the best predictive performance. • An optimized hot working window (780–820 °C / 0.001–0.0024 s −1 ) and instability regions are identified via hot processing maps. • Dynamic recrystallization (CDRX & DDRX) mechanisms and their dependence on processing parameters are thoroughly elucidated. Hot deformation behavior of a newly designed Ti-5Mo-20Nb-10Zr-5Ta-1Fe β-titanium alloy was analyzed over a temperature interval of 700–900 °C and across strain rates of 0.001 to 1 s −1 . The Arrhenius constitutive model with strain compensation, multiple linear regression model, artificial neural network (ANN) model, and hot processing map have been established for the alloy. Electron backscatter diffraction (EBSD) analysis of the alloy’s microstructural development revealed the impact of microstructure on the alloy’s high-temperature plastic flow characteristics. The results demonstrate that the highest values of power dissipation efficiency (η) are found within 780–820 °C and 0.001–0.0024 s −1 , which represent the favorable conditions for hot working. Under these conditions, the deformation mechanism is mainly dynamic recrystallization (DRX), including continuous dynamic recrystallization (CDRX) and discontinuous dynamic recrystallization (DDRX). Unstable deformation primarily occurs in the regions of high strain rate and low temperature, where deformation bands (DB) serve as the principal deformation mechanism. As η increases, the dominant mechanism shifts from DB to dynamic recovery (DRV), ultimately evolving into the coexistence of CDRX and CDRX/DDRX. This study conclusively identified the most suitable range of hot deformation conditions for the alloy and established a theoretical basis for the subsequent optimization of its processing parameters.
Ma et al. (Sun,) studied this question.