Tailored Low-Carbon Footprint Cementitious Material for 3D Printing (2024-09)¶

Despite the potential of 3D printing to save material by using bespoke optimized structures, the printable mixtures available are still expensive and carbon-intensive due to their high clinker contents. This led us to develop a formulation methodology [1]. This method shows that the 2K printing process allows for the use of Portland limestone cement (PLC) mixtures that maximize the performance-toenvironmental footprint ratio (i.e., clinker content). In this context, this research presents a novel development, delving into the intricacies of adjusting the water and superplasticizer contents in different dry mixes to fit the rheological assessment of 2K processes and measuring the resulting mechanical properties. An essential finding revolves around determining an optimal level of looseness within the mix, thereby necessitating a specific volume of liquid in the formulation. Furthermore, the research investigates the influence of PCE-based superplasticizers across various mix compositions, facilitating an initial estimation of the superplasticizer content required for a particular formulation. From the experimental mechanical results, the model proposed in [2] is then calibrated and can be used to select an optimal PLC-based formulation tailored to the needs of a specific project. The main hypothesis of our approach uses the compaction index 𝐾 proposed by de Larrard [3] as the primary parameter to express the material process suitability. Through large-scale 3D printing tests using different water quantities for a specific dry mix formulation, we determine the minimum water dosage needed to prevent blockage and ensure printability. The corresponding compaction index 𝐾 for this specific process can then be identified. Then, the superplasticizer type and dosage need to be fine-tuned to achieve the correct consistency and corresponding open time. To do this, we experimented with different dosages of three PCE-based superplasticizers with different adsorption capacities and determined a suitable proportion of each PCE for a specific dry mix formulation. This optimal dosage was then kept constant, and a rheological test campaign (slump test and rheometer) was conducted on different mixes, varying the proportions of cement and different limestone fillers while adjusting the superplasticizer content to target a 100 Pa yield stress consistency. Using linear regression, we can determine the amount of superplasticizer needed for any mix to achieve the desired "printable" rheology. We carried out a mechanical characterization campaign on 15 different PLC mortar mixes, utilizing 3-point bending tests and compression tests on cubic specimens, to calibrate the Chidiac model [2]. This model integrates physical and chemical phenomena, such as average granulometry and chemical interactions, into a comprehensive framework that accounts for hydration degree, chemical composition, packing density, and chemical interactions according to the cement type. Calibration is achieved through an optimization process that minimizes the standard error between model predictions and experimental compressive strength values, thereby validating the model's accuracy and applicability in predicting concrete strength based on formulation parameters. With this model calibrated, we are able to compute the mechanical features of the mix during the formulation process and choose the mix strength according to its future application. On this research, we improve our methodology to formulate printable mortar [1] by estimating the water and superplasticizer quantity, and adding a predictive model of mechanical features to be able to choose specific formulas according to their future use, as follow. Selection: Find available and local resources. Characterization: the particle size distribution and the packing density of each component is needed to compute the model [1,3]. Computation: by using sing the compressive packing mode: compute the packing density of the mix for all the component proportions - Powders proportion selection: different criterion for printability are presented on [1]. Thanks to the Chidiac Model, select a printable mix based on his mechanical features. Liquid proportion selection: with the adjustment of the compaction index K made on this work: compute the water quantity and estimate the amount of superplasticizer based on linear regression.