Development of a methodology for the quantitative prediction of the damage in to the diffusion layers of reactive air brazed BSCF-Steel-Joints

• Applicants: Dr. Markus Apel; Professor Dr.-Ing. Christoph Broeckmann
• Subject Area: Metallurgical, Thermal and Thermomechanical Treatment of Materials ; Mechanical Properties of Metallic Materials and their Microstructural Origins
• Term: from 2017 to 2022
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 392944287

Perovskite type ceramics like Ba0,5Sr0,5Co0,8Fe0,2O3-d (BSCF) are promising materials for the use as oxygen transport membranes (OTM) for oxygen separation from air. Supplying oxygen by membrane technology within a high temperature process can be more efficiently than other air separation techniques. For reliable working of the membranes at working temperatures of around 800°C, a gas tight joint between ceramic and metals is required. Reactive air brazing (RAB) was identified as a suitable technique for this material combination. Other brazing techniques are difficult to realize in the case of BSCF due to a thermodynamic instability of the material in vacuum. In the case of RAB, wetting of the silver copper braze occurs due to interface reactions between insito built CuO and BSCF. These reactions are required for wetting. The higher the CuO-content in the braze alloy is, the better the wetting is, although, at the same time a change of microstructure in the ceramic is visible, characterized by Cu-Co-Oxides at the grain triple points combined with micro cracks. Additionally, a reaction layer between steel and braze can be observed. Mechanical tests showed that both diffusion layers decrease the mechanical strength of the joints. The object is the development of a methodology for a quantitative prediction of damage inside of the diffusion layers of reactive air brazed ceramic steel joints.Hence, the first step should be a thermodynamic modelling of the system. Because an entire modelling of the overall system is not possible for ceramic / braze alloy / steel due to the complexity, the modelling is divided into four zones. With this data, structural analyses of the microstructure (grain structure, phase portion division and distribution) should be done in the reaction layer as well as in the braze - in its temporal development very near to reality. The calculated structures are a basis for the micromechanical simulation to predict the joint qualities and the damage development in the joints. Thereby, the infiltration of BSCF with the silver braze to generated variations of the qualities of the BSCF interface will be considered in particular. The interactions between the joining partners, the braze alloys of different composition and the different process parameters can be modelled by the simulation. The following micromechanical simulation is based on these simulated microstructures as well as on experimentally 2D and 3D structures (3D EBSD). The aim of the micromechanical simulation is the prediction of the damage development during brazing. Therefore, optimum brazing process parameters (Cu content, temperature, process) can be determined for the special case. The methodology should be transferable on other brazed joints. For example, similar reactions appear with other perovskite type ceramics (e.g., LSCF). With such a methodology, process parameters during brazing can be correlated directly with the mechanical qualities of the joints.

• DFG Programme: Research Grants