A Numerical Model for Simulation and Design of Particle-Bed 3D-Printing Process

• Applicants: Professor Dr.-Ing. Dirk Lowke; Dr.-Ing. Ksenija Vasilic
• Subject Area: Construction Material Sciences, Chemistry, Building Physics
• Term: from 2019 to 2023
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 417019981

The present project addresses the emerging subject of Particle-bed 3D printing (PB3DP) and suggests an innovative numerical approach to simulate and predict the printing process. One of the most promising Particle-Bed 3D-Printing techniques is the selective paste intrusion method. It is based on a localised intrusion of a fluid (cement paste) into a bed of particles (aggregate), and the subsequent hardening. The main advantage of this technique is that the printed elements have a high resolution and almost no restrictions in freedom of form. Up to now, this method is applied successfully to print small and medium scale objects with compressive strength up to 70 MPa, but there is still lack of large scale implementation. To realise application in construction industry, fundamental questions need to be answered. These are related to the optimisation of the process as well as the material and granulate properties (such as rheological properties of the paste or permeability of the aggregate layer). Appropriate computational models, to describe and predict the printing process, are essential for a successful implementation. The aim of the present project is to study the process of the PB3DP numerically and to predict the propagation of the fluid through the particle bed. Based on experimentally determined input parameters such as the rheological properties of the propagating fluid and the permeability of the aggregate layer, the goal is to predict the final penetration depth, which determines the overall quality of the produced component (mechanical properties, durability and shape accuracy). The outcome of the project is a numerical tool, capable to predict the printing process. The tool will be able to deal with the flow prediction within the process as well as with its optimization. The originality and innovation of the proposed approach lies in two main attributes: (I) description of the particle bed as a porous medium and (II) consideration of structural build up caused by thixotropy. To achieve the above mentioned objectives, basic research regarding (a) the specific characterisation and control of the rheological properties of cement pastes for PB3DP, (b) the packing and permeability properties of the particle bed and (c) the numerical methods for the simulation and prediction of the printing process is necessary.

• DFG Programme: Research Grants
• International Connection: France
• Cooperation Partner: Professor Dr. Nicolas Roussel