Agglomeration applying high intensity mixing (mechano-fusion) – An integrated approach towards the synthesis of tailored hetero-aggregates combining experiments with 2D and 3D structural characterization via image analysis and stochastic modeling

异质聚集体的特殊性质来源于聚集体内的粒子-粒子接触。触点的数量以及触点处的微观材料相互作用对宏观应用性能起着决定性的作用。例如输运性质，如电导率、离子导电率或热导率。第一个资金期调查的是混合体的混合，因此也就是团聚体的几何三维结构。该项目应用了介于研磨和高强度混合过程之间的机械融合过程，以从初级纳米粒子系统生成异质聚集体。骨料的合成是由于动态的解聚和再团聚过程，其中机器中的高剪切应力和屈服应力导致了骨料中稳定的颗粒-颗粒接触。该方法能够生成具有异质聚集体涂层的核心粒子和主体异质聚集体。这些研究将量化初级颗粒的分散和材料特性，这是确定集料设计所必需的。异质聚集体被提交给2D和3D表征方法，以量化单个聚集体结构内初级粒子的混合状态。纳米和微米计算机断层扫描(CT)数据的图像分析提供了对聚集体的详细3D洞察。该分析使用统计数据挖掘的方法来确定异质聚集体的结构描述符向量的多变量概率分布，并使用空间随机建模来创建虚拟(但真实的)聚集体，即模拟聚集体，以生成用于应用机器学习算法的大型数据库。相关的表征方法，例如二维原子力显微镜和µ-CT的结合，允许交叉检查是否也可以从平面预测三维结构集合体的性质，即混合状态的二维测量。第二个资助期还将考察原子水平，特别是不同材料相互作用的异质接触。机械融合过程中的强烈剪切和屈服应力能够在这些接触点上定制材料焊接，产生或增强宏观性能。最后，将得到定量的过程-结构-性质关系，将过程参数与真实和模拟的异质聚集体的三维形态及其宏观物理性质关联起来。

The special properties of hetero-aggregates derive from the particle-particle contacts within the aggregate. The number of contacts as well as microscopic material interaction at the contacts play a decisive role regarding macroscopic application properties. Examples are transport properties like electric, ionic or thermal conductivity. The first funding period investigates the mixing and therefore the geometric 3D structure of the aggregates. The project applies the mechano-fusion process, which is situated between grinding and high intensity mixing processes, to generate hetero-aggregates from primary nano-particle systems. The aggregate synthesis occurs due to a dynamic de-agglomeration and re-agglomeration process, where the high shear and yield stresses in the machine lead to stable particle-particle contacts in the aggregate. The process is able to generate both core particles with a hetero-aggregate coating and bulk hetero-aggregates. The investigations will quantify the disperse and material properties of the primary particles, which are required for a defined aggregate design. The hetero-aggregates are submitted to 2D- and 3D-characterization methods to quantify the state of mixing of the primary particles within the individual aggregate structures. Image analysis of nano- and µ-computer tomography (CT) data provides detailed 3D-insight into the aggregates. The analysis uses methods of statistical data mining to determine the multivariate probability distribution of structural descriptor vectors of hetero-aggregates, and spatial stochastic modeling to create virtual (but realistic) aggregates, i.e. simulated aggregates, to generate a large data base to apply machine learning algorithms. Correlative characterization methods, e.g. the combination of atomic force microscopy at 2D with µ-CT, allow to cross check whether it is also feasible to forecast 3D structural aggregate properties from planar, i.e. 2D measurements of the mixing state. The second funding period will also have a look at the atomic level, especially at the hetero-contacts, where different materials interact with each other. The intensive shear und yield stressing in the mechano-fusion process is able to tailor the material welding at these contact points generating or enhancing macroscopic properties. Finally, quantitative process-structure-property relationships will be derived, correlating process parameters with the 3D morphology of real and simulated hetero-aggregates and their macroscopic physical properties.