Simulations on the interaction between grain boundaries and precipitates

该项目的目标是开发一种非常详细的理解和分析模型，以了解晶界和第二化学相的沉淀物之间的相互作用，即钉扎力或“曾格阻力”。这种力在材料的生产中非常重要，因为它控制着再结晶和晶粒长大等加工步骤中的组织演变，从而控制材料的性能。然而，尽管很重要，目前对钉扎力的物理理解是相当不完整的，这从以下事实可以看出：现有的模型只涉及关于特定沉淀物和晶界的很少的细节。例如，目前的模型根本不涉及关于析出物和晶界之间的界面的信息，或者关于晶界接触析出物的三重线的信息。同样，目前的模型没有考虑这样一个事实，即弯曲的晶界会比平坦的晶界同时接触更多的沉淀并与之相互作用。这一趋势被认为是Friedel位错钉扎模型成功的主要因素。与当今使用的非常详细的位错钉扎(强化)模型相比，晶界钉扎的影响仅以初级方式处理。这意味着，尽管它们很重要，但当前的齐纳阻力模型可能相当不准确。拟议的项目旨在为齐纳阻力开发全面的新模型，考虑多种潜在的重要影响。使用了一种多尺度方法，其中原子细节与更大尺度的模型相结合以得出同质化的结果。首先，在原子的基础上研究了晶界与一个沉淀物相互作用的细节。利用分子动力学，模拟了不同晶界从不同类型的沉淀物中剥离的过程。同样的脱钉过程也是用顶点模型模拟的，其中晶界和析出物界面由抽象的多面体平面表示。这些模型的直接比较，例如通过将顶点模型调整为原子模型，允许了解关于脱毛过程的所有细节，从而可以识别、量化和判断最重要的影响。其次，可以利用顶点模型来模拟晶界与真实的沉淀物阵列之间的相互作用。因此，推导出了高精度的齐纳阻力模型，包括再结晶和晶粒长大。这些可以在以后用于改进制造工艺和材料性能。

The goal of the project is to develop a very detailed understanding and analytical model for the interaction between grain boundaries and precipitates of secondary chemical phases, i.e. the pinning force or "Zenger drag". This force is of utmost importance in the production of materials because it controls the microstructure evolution and hence the material properties during processing steps involving recrystallization and grain growth. Still, in spite of the importance, the current physical understanding of the pinning force is rather incomplete, as can be seen from the fact that the existing models involve only very little detail about the specific precipitates and grain boundaries. For instance, current models involve no information at all about the interface between the precipitate and the grain boundary, or the triple line where the grain boundary touches the precipitate. Likewise, the current models do not account for the fact that a curved grain boundary will touch and interact with more precipitates concurrently than a flat one. This tendency is known to be the principal element in the success of the Friedel model for dislocation pinning. In comparison to the highly detailed dislocation pinning (strengthening) models in use today, the effect of grain boundary pinning has only been addressed in rudimentary ways. This means that in spite of their importance, the current Zener drag models are probably rather inaccurate. The proposed project is aiming to develop comprehensive new models for the Zener drag that consider a multitude of potentially important influences. A multi-scale approach is used in which atomistic detail is combined with a larger scale model to derive homogenized results. Firstly, the details of a grain boundary interacting with one precipitate is investigated on an atomistic basis. Using molecular dynamics, the detachment of various grain boundaries from various concrete types of precipitates is simulated. The very same depinning processes is also simulated using a vertex model, in which the grain boundary and the precipitate interface are represented by abstract, facetted planes. The direct comparison of these models, e.g. by adjusting the vertex model to the atomistic one, allows to understand all detail about the depinning process, so that the most important influences can be identified, quantified and judged in respect of importance. Secondly, the vertex model can then be utilized to simulate the interaction of a grain boundary with realistic arrays of precipitates. Hence, highly accurate models for the Zener drag are derived, both for recrystallization and grain growth. These can later be used to improve manufacturing processes and materials properties.