3D Printing Mudrocks: Experiments in Validating Clay as a Build Material for 3D Printing Porous Micromodels

3D printing technologies have the ability to turn digital 3D designs into tangible, lab-testable objects. While 3D printing in plastic and granular materials is quite common now, new equipment has been developed that can 3D print using paste-like materials (e.g., clays, Portland cement, foods). This study characterized simple, core-plugsized models to evaluate whether this technology can be applied to producing 3D-printed analogs for mudrocks. Most models produced were designed as solid, 25-mm-diameter cylinders, 25 mm tall in two different types of clay (Limoges Clay and ISU Clay). Models were printed on a Delta WASP 60100 with the Delta WASP low-density material extruder kit. Models displayed negligible dimensional loss after desiccation but shrank considerably after first firing (8 to 12% loss for both height and diameter). Mass loss was 8 to 11% after the first firing. A second firing yielded 5 to 6% loss for height and diameter, 0.1 to 0.2% mass loss. Models produced from Limoges Clay reduced from similar to 39% porosity after desiccation to similar to 7% after first firing to similar to 1% after the second firing. ISU Clay reduced its porosity less with firing going from similar to 36 to similar to 23 to similar to 10%. Models survived mercury porosimetry up to 33,000 psi (similar to 230 MPa) with no signs of deformation. Pore-throat-size distributions became more monomodal after each firing. For Limoges Clay, the modal pore-throat size lessened after each firing reaching functionally zero after the second firing. ISU Clay's modal pore-throat size increased after the first firing, before reducing after second firing-though not down to the modal sizes of the desiccated sample. Pore-throat-size distributions were similar to those reported for tight sandstones and shales, suggesting that the method outlined in this study could be used to create analogous pore structures for laboratory experiments with the caveat that surface physics (e.g., wettability) of the models would need to be assessed to understand to what extent it reproduces the properties of natural rock surfaces.