应用机器学习分析镁合金孪晶形核

镁（Mg）及其合金因其轻质、高强度重量比、良好的可回收性，成为了交通运输中结构组件的候选材料，且由于其可生物降解和相容性行为，在生物医学也应用广泛。

然而，孪晶变形导致了纹理镁合金的拉伸和压缩屈服强度以及工作硬化之间的显著差异，这种明显的塑性各向异性对锻制镁合金的延展性和成形性产生了负面影响。通过孪晶来实现塑性变形，包括两个连续步骤：孪晶核形成和孪晶生长（增厚）。

孪晶核形成是一种异质过程，发生在微观结构中应力集中的区域。一旦孪晶体形成，通常认为孪晶体增厚是由孪晶位错沿着孪晶面滑移介导的，而这一过程受到了孪晶面和方向上的解析剪应力的控制。从多晶体角度来看，用于解释单个晶粒中孪晶核形成的最常见标准是表观施密德因子（SF），其基于单个晶粒内的应力状态与宏观应力相同这一假设。

然而，最近的许多研究揭示，除了SF之外，其他微观结构特征也对孪晶核形成具有显著影响。因此，目前对导致孪晶核形成的潜在因素仍无共识，因为孪晶的生长不一定只发生在大晶粒、晶界或有有利取向的晶粒中。

Fig. 4 Distributions of some microstructural features.

来自西班牙IMDEA材料研究所的Javier Llorca教授课题组，通过电子背散射衍射对拉伸变形的Mg合金孪晶形核进行了研究，分析了晶粒的28个相关参数，将其分为四组不同的类别（加载条件、晶粒形状、表观施密特因子和晶界特征）。

他们训练了贝叶斯网络分类模型，研究发现较大的晶粒和具有较高施密特因子的晶粒更有利于孪晶的形成。此外，即使对于具有很低甚至负施密特因子的晶粒，如果其至少有一个较小的邻近晶粒和另一个（或相同的）较刚性的晶粒，也可能形成孪晶。同时，如果小晶粒具有较高的施密特因子，并且具有较低的基面滑移施密特因子以及至少一个具有易于变形的高基面滑移施密特因子的邻近晶粒，那么小晶粒的孪晶更有利。

这些结果揭示了许多体系之间的相互作用，例如给定晶粒与其相邻晶粒之间的刚度和大小差异，以获取抑制晶粒发生孪晶的手段。相关论文近期发布于npj Computational Materials 10: 26 (2024)。手机阅读原文，请点击本文底部左下角“阅读原文”，进入后亦可下载全文PDF文件。

Editorial Summary

Magnesium (Mg) and its alloys have emerged as promising candidates for structural components in transport, owing to their light weight, high strength-to-weight ratio, and good recyclability, as well as in biomedical applications due to their biodegradable and biocompatible behavior. However, twinning deformation leads to a large difference between the tensile and the compressive yield strengths and work hardening of textured Mg alloys, and this marked plastic anisotropy has negative effects on the ductility and formability of wrought Mg alloys. The accommodation of plastic deformation by twinning involves two successive steps: twin nucleation and twin growth. Twin nucleation is a heterogeneous process that takes place in regions with large stress concentrations in the microstructure, such as grain boundaries. Once the twin has been formed, it is generally accepted that twin thickening is mediated by the glide of twinning dislocations along the twin planes, and this process is controlled by the resolved shear stress on the twin plane and direction. From the polycrystal viewpoint, the most common criterion used to explain twin nucleation in one grain is the apparent Schmid factor (SF), based on the hypothesis that the stress state in one grain is identical to the macroscopic applied stress. However, many recent works have revealed that other microstructural features, besides the SF, also have remarkable influence on twin nucleation. Thus, there is still no consensus on the underlying factors leading to twin nucleation because extension twins not necessarily occur in all large grains, at all GBs, or in all grains with favorable orientations. 

A group led by Prof. Javier Llorca, from IMDEA Materials Institute, Spain, studied twin nucleation in textured Mg alloys by means of electron back-scattered diffraction in samples deformed in tension along different orientations, 28 relevant parameters, categorized in four different groups (loading condition, grain shape, apparent Schmid factors, and grain boundary features). They trained Bayesian network (BN) model and found that twin nucleation is favored in larger grains and in grains with high twinning Schmid factors, but also that twins may form in the grains with very low or even negative Schmid factors for twinning if they have at least one smaller neighboring grain and another one (or the same) that is more rigid. These results reveal the role of many-body relationships, such as differences in stiffness and size between a given grain and its neighbors, to assess extension twin nucleation in grains unfavorably oriented for twinning.

原文Abstract及其翻译

Application of machine learning to assess the influence of microstructure on twin nucleation in Mg alloys (应用机器学习评估微观结构对镁合金孪晶形核的影响)

Biaobiao Yang, Valentin Vassilev-Galindo & Javier Llorca

Abstract Twin nucleation in textured Mg alloys was studied by means of electron back-scattered diffraction in samples deformed in tension along different orientations in more than 3000 grains. In addition, 28 relevant parameters, categorized in four different groups (loading condition, grain shape, apparent Schmid factors, and grain boundary features) were also recorded for each grain. This information was used to train supervised machine learning classification models to analyze the influence of the microstructural features on the nucleation of extension twins in Mg alloys. It was found twin nucleation is favored in larger grains and in grains with high twinning Schmid factors, but also that twins may form in the grains with very low or even negative Schmid factors for twinning if they have at least one smaller neighboring grain and another one (or the same) that is more rigid. Moreover, twinning of small grains with high twinning Schmid factors is favored if they have low basal slip Schmid factors and have at least one neighboring grain with a high basal slip Schmid factor that will deform easily. These results reveal the role of many-body relationships, such as differences in stiffness and size between a given grain and its neighbors, to assess extension twin nucleation in grains unfavorably oriented for twinning.

摘要该文章通过电子背散射衍射对拉伸变形的Mg合金孪晶形核进行了研究，分析了晶粒的28个相关参数，分为四组不同的类别（加载条件、晶粒形状、表观施密特因子和晶界特征），样本多达3000多个晶粒。这些信息被用来训练监督式机器学习分类模型，以分析微观结构特征对Mg合金中延伸孪晶的影响。研究发现，较大的晶粒和具有较高施密特因子的晶粒更有利于孪晶的形成。此外，即使对于具有很低甚至负施密特因子的晶粒，如果其至少有一个较小的邻近晶粒和另一个（或相同的）较刚性的晶粒，也可能形成孪晶。此外，如果小晶粒具有较高的施密特因子，并且具有较低的基面滑移施密特因子以及至少一个具有易于变形的高基面滑移施密特因子的邻近晶粒，那么小晶粒的孪晶更有利。这些结果揭示了许多体系之间的相互作用，例如给定晶粒与其相邻晶粒之间的刚度和大小差异，以获取不利于晶粒发生孪晶的因