(384b) Computational Printing of Cement-Based Pastes with Time-Dependent Rheology | AIChE

(384b) Computational Printing of Cement-Based Pastes with Time-Dependent Rheology


Mohammad, A. S. - Presenter, Tennessee Technological University
Biernacki, J. J., Tennessee Technological University
Recent advances in additive manufacturing (AM or 3D printing) of cement-based structures has led to new rheological and set (hydration) requirements for fresh cementitious pastes. 3D printing of cement-based materials requires four primary paste-related processing characteristics: (1) extrusion, (2) layer-by-layer deposition, (3) resistance to deformation under layering loads, and (4) time response of the printing paste, i.e. hydration rate and time-dependent rheological properties. Designing effective cement-based printing pastes involves optimization of paste yield stress, plastic viscosity and thixotropic response and design strategies and data correlating rheological properties to printability are among the existing research gaps. Thus, the objective of this study is to link relevant experimentally determined hydration kinetics and time-dependent paste rheological parameters, including yield stress, viscosity and structuration, to computational printing frameworks in an effort to correlate properties to printability metrics and to target optimal paste design parameters. Prior to conducting simulations, rheometric and calorimetric studies were done to define and extract rheological fluid properties as a function of extent of hydration. Computational fluid dynamics (CFD) simulations were then developed for printing of fresh cement paste in 2D and 3D geometries. Free-surface flow of square, spherical, and rectangular gobs of materials in 2-D geometries were simulated as a function of fluid properties linked to hydration kinetics. The importance of rheological characteristics and rate of hydrations (phase change) were quantified by deformation and rate of deformation of simulated cement paste gobs in both 3D and 2D geometries respectively. Cylinders and cone benchmark objects were also printed for validation.