The elimination of the B2 phase in a β-solidifying high Nb-containing TiAl alloy with β/B2 and γ phases was investigated using different heat treatments, with a focus on understanding the phase transformations and lamellae formation during the process. The phase transformation and lamellae formation during B2 phase elimination differs from that observed in conventional TiAl alloys. During the holding stage of heat treatment, the β/B2 phase is replaced by the α phase through primary phase transformations of β→α and γ→α. Lamellae formation occurs within both α and γ grains during cooling, initiating 30−40 °C below the annealing temperature. This lamellar structure was formed via two main mechanisms: nucleation at grain boundaries followed by growth into the grain, and direct precipitation and growth within the grain. The orientation relationship between the γ phase and its adjacent α phase is (111)γ//(0001)α and 
, with a coherency between the phases characterized by a misfit of approximately 1.7%.
To address the zero-sample challenge in preparation parameter design for newly developed alloys, a novel machine learning strategy that integrates basic dataset construction with Bayesian optimization, was proposed. The impact of basic sample dataset construction methods, optimization benchmarks and multi-objective utility functions on Bayesian optimization was investigated. It was found that the combination of orthogonal design, linear benchmark, and shifted multiplicative utility function exhibits the best optimization performance. The strategy was then applied to a new Cu−Ni−Co−Si alloy with ultra-low Co content (0.7 wt.% Co), previously designed by our research team. Rapid optimization of six preparation parameters in the two-stage deformation and aging process of the zero-sample alloy was achieved through only 23 experiments. The measured ultimate tensile strength and electrical conductivity of the new alloy were 878 MPa and 44.0%(IACS), respectively, reaching the comprehensive performance level of the Cu−Ni−Co−Si alloy (C70350 alloy) containing 1.0−2.0 wt.% Co.
To improve the overall magnetic properties of Sm(CoFeCuZr)z sintered magnets, a dual-alloy sintering process that involves mixing high-iron, low-copper powders with low-iron, high-copper powders was systematically investigated. The results demonstrate that this method significantly improves the Cu-lean phenomenon at the grain boundaries, achieves multiscale uniform microstructures, greatly enhances the pinning field strength, and ultimately produces a high-performance dual-alloy magnet with a maximum energy product ((BH)max) exceeding 240 kJ/m3 and an intrinsic coercivity (Hcj) exceeding 2400 kA/m. In particular, when 35 wt.% of low-iron, high-copper alloy powder is incorporated, the dual-alloy magnet achieves a remanence of 1.13 T, Hcj of 2691.2 kA/m and (BH)max of 248 kJ/m3. To evaluate the overall magnetic performance, the sum of Hcj (in kA/m) and (BH)max (in kJ/m3) is used as a combined parameter, yielding a value of 2939.2. Compared with single-alloy magnets of the same composition, the dual-alloy sintering process yields magnets with a more uniform elemental distribution and superior magnetic properties.
