ACTA

In the present investigation, the microstructure and hardening of an IF steel after one thermal cycle (heating to 1050°C and holding for 30 min followed by 10°C/s cooling to room temperature) under hydrostatic pressure of 1-5 GPa were studied. Experimental results show that typical lath martensite was induced, giving rise to significant hardening from 80 to 780 HV. The lath martensite shows hierarchical packet-block-lath structure and obeys the classical K-S orientation relationship between martensite and austenite. High pressure influences the propensity and size of martensitic variants. As the pressure increases from 1 to 5 GPa: 1) single-variant blocks gradually replace those with dual-variants of same Bain group; 2) twin-related variants become predominant; 3) variants decrease their thickness from micron- to nano-scale. The hardening mechanism was analyzed assuming a linear additivity of carbon-independent contributions from (block) boundary strengthening and (forest) dislocation strengthening. High pressure was proposed as an effective method to widely tune the martensitic transformation independent of alloy element, showing potential scientific and technological importance.

ACTA

Hydrogen is a common impurity or artificially added element in magnesium alloys and is generally known to assist crack propagation along grain boundaries. However, how hydrogen atoms interact with atoms of other alloying elements segregated to grain boundaries and its impact on grain boundary cohesion of Mg alloys are still unclear. In this work, interactions between hydrogen and each one of the 18 alloying elements commonly used in Mg alloys in {10-12} twin boundary, and their impacts on interfacial cohesion of the {10-12} twin boundary are investigated using first-principles calculations. It is found that hydrogen is unlikely to bond with Al, Sn, Bi, Zn, or Ag, while H prefers to bond with solute such as Li, Ca, Mn, Zr, Y, La, Pr, Nd, Sm, Gd, Tb, Dy, or Ho in the Mg lattice and in the twin boundary. As a result, hydrogen is unlikely to segregate to a {10-12} twin boundary that has the presence of Al, Bi, Zn, or Ag, while H atoms will be attracted to a twin boundary to form a hydrogen-solute cluster around the solute such as Li, Ca, Mn, Zr, Y, or Nd that is already present in the twin boundary. The strengthening or weakening effects of segregated solute atoms and their interactions with hydrogen atoms on the twin boundary cohesion are found to be closely related to the electronic interactions of atoms of Mg with those of the segregated solutes. Mn, Zr, Y, or Nd can significantly enhance the twin boundary cohesion, in the absence of any hydrogen segregation in the boundary, because of the strong interactions between the d electrons of atoms of the solute element and the p electrons of Mg atoms. However, electron transfer from the solute atoms to hydrogen atoms will reduce the strengthening effect, making the boundary less resistant to fracture. This work uncovers the hydrogen-induced interfacial fracture from the perspective of the interactions of hydrogen and alloying elements, which is important for understanding hydrogen embrittlement in Mg alloys.

ACTA

ACTA

ACTA

Though physics-based crystal plasticity models are able to accurately predict the material response under complex loading conditions, the high computational cost limits their engineering applications. Machine learning techniques, widely used to understand and predict data trends, can provide advantageous computational efficiency over conventional numerical techniques. In the current work, an artificial neural network (ANN) model is proposed and trained based on the datasets generated by the physics-based viscoplastic self-consistent (VPSC) model on the material behavior of OFHC copper. The ANN model successfully captures the loading path-dependent micromechanical behavior of the copper polycrystals with arbitrary texture, even beyond the bounds of the generated dataset. The proposed ANN model, without any major loss in accuracy, can greatly improve the computational efficiency. It can be envisioned that the engineering application of the physics-based model becomes promising through alternatively using the ANN model, which treats the underlying complex physics as a blackbox.

ACTA

向上滑动阅览英文摘要

A two-steps tempering-partitioning process is designed for redistributing carbon and austenite stabilizer such as manganese in a cold rolled medium-manganese steel. After tempering at 450℃ for 0.5h and 1h, a large number of metastable Mn-rich M₁₂C carbides and high density of NiAl-type nanoparticles were found, and the segregation of carbon at the dislocations/interfaces around NiAl precipitates was revealed by 3D atom probe tomography (APT). The competition between the M₁₂C carbides and the dislocations/interfaces for carbon inhibits transformation of the metastable M₁₂C carbides to more stable carbides. During the following partitioning at 630℃ for 0.5h, the metastable Mn-rich M₁₂C dissolved and the fine reversed austenite nucleating on the metastable M₁₂C carbides exhibits a sharp gradient Mn distribution. The tailored metastable austenite inherited from metastable M₁₂C carbide shows strong enrichment of Mn and C (Mn: 18.1 at.%, C: 1.56 at.%), dispersed distribution (9.8 μm⁻²) and fine size (50-200 nm). With the nature of both hard particle and transitionable phase, the austenite couples Orowan strengthening and transformation induced plasticity (TRIP) assisted work hardening. Outstanding dynamic mechanical properties (yield strength: 1350 MPa, total elongation: 30%) were achieved due to this microstructure. Here we report a strategy for the design of ultrahigh-strength steels by making metastable austenite serve as a strengthening second phase in addition to the TRIP effect.

Vol. 214，1 Aug. 2021, 117019