Robocasting of advanced ceramics: ink optimization and protocol to predict the printing parameters - A review

Abstract

Direct-Ink-Writing (or robocasting) is a subset of extrusion-based additive manufacturing techniques that has grown significantly in recent years to design simple to complex ceramic structures. Robocasting, relies on the use of high-concentration powder pastes, also known as inks. A successful optimization of ink rheology and formulation constitutes the major key factor to ensure printability for the fabrication of self-supporting ceramic structures with a very precise dimensional resolution. However, to date achieving a real balance between a comprehensive optimization of ink rheology and the determination of a relevant protocol to predict the printing parameters for a given ink is still relatively scarce and has been not yet standardized in the literature. The current review reports, in its first part, a detailed survey of recent studies on how ink constituents and composition affect the direct-ink-writing of ceramic parts, taking into account innovative ceramic-based-inks formulations and processing techniques. Precisely, the review elaborates the major factors influencing on ink rheology and printability, specifically binder type, particle physical features (size, morphology and density) and ceramic feedstock content. In the second part, this review suggests a standardized guideline to effectively adapt a suitable setting of the printing parameters, such as printing speed and pressure, printing substrate, strut spacing, layer height, nozzle diameter in function of ink intrinsic rheology.

Keywords: Ceramics; Feedstock features; Glass; Printing parameters; Robocasting.

Figure 1

Representation of different apparatus configurations used for extruding the high-solid-loading viscous pastes, where the mechanical displacement in screw-driven, the rotation in piston-dispensing and the air pressure in pneumatic dispensing, respectively, deliver the driving force to ensure a continuous flow of the ink through the nozzle [9].

Figure 2

Schematic diagram illustrating the three main mechanisms for particle stabilisation: steric and electrostatic interactions are represented with black lines and negative red charges respectively. Colloidal inks involve an electro-steric stabilization mechanism, which gives rise to polyelectrolyte complexes. The green lines represent the polymer chains within gel-embedded inks. Adapted with permission from [18].

Figure 3

Optical images of geometrically complex B₄C printed ceramic parts using robocasting: a) green and dried B₄C parts shaped by robocasting, b) in situ image of complex circular gear during printing adapted with permission from [55]. Optical images of Si₃N₄ printed structures using robocasting: c) cellular structure and d) honeycomb structure adapted with permission from [65].

Figure 4

SEM micrographs showing powder morphology and size used for ink preparation: a) zirconia granules and b) zirconia particles derived from the crushed granules after planetary and rotational milling, c) silicon carbide platelets incorporated in aqueous silicon carbide paste designed for robocasting. Adapted with permission from [80] and [81] respectively.

Figure 5

Amplitude-sweep curves showing the variation of storage modulus G′ and loss modulus G′′ in function of stress for: a) pastes containing Al₂O₃ platelets at various content of 20, 30, 40 vol% in relation to the total ceramic content of 51, 50,45 vol% and b) paste containing 28 vol% of BaTiO₃ platelets at different sizes and amounts. c) Shear stress and viscosity curve showing the viscosity instability due to pastes produced from granules. a), b) and c) images are adapted with permission from [79, 83] and [80] respectively.

Figure 6

A schematic summarizing the influence of grain size, shape on paste rheology and post-treatment process.

Figure 7

a) Colloidal particles in aqueous suspensions can form a non-continuous, weak, or strong network depending on the particle volume fraction. b) Printability map for GO suspensions including three key rheology parameters: at rest structure or network stiffness (G_LVR′), flow stress, σ_f (oscillation stress at the crossover point, when G′ = G′′), and flow transition index FTI (the ratio between the flow stress and the yield stress, σ_f/σ_y). The stiffness (G_LVR‘) of the GO networks increase as sf increases following a power relationship up to 2.8 vol%. The trend changes abruptly at 3 vol%. At concentrations below 0.8 vol% the network across flakes is weak, resulting in very small values of σ_f and G’, in this range FTI values do not display a clear trend. From 0.8 vol%, considered as the ‘network threshold’, the FTI clearly decreases revealing the ‘brittle’ character of the networks. Printable concentrations, between 2 vol% and 3 vol%, exhibit an FTI value <20 with small uncertainties and a G’/σ_f ratio ≥20. Adapted with permission from [15].

Figure 8

a) Flow ramps of Al₂O₃ and SiC inks compared to the stock gel. b) Viscosity temperature sweep of the F127 showing the reversible gelation between 10 and 20 °C. c) DMA of the hydrogel gel and SiC and Al₂O₃ inks at 1 Hz. (d) Estimated viability map of different inks printing at different temperatures incorporating the major limiting effects during printing adapted with permission from [66].

Figure 9

Map of the relationships between robocasting variables including the properties of the initial ceramic paste, the post-printing variables, and the final properties of printed ceramic samples.

Figure 10

Optical photographs of Si₃N₄ green parts (cuboidal or beam geometries) with porous grid-like or solid architectures; adapted with permission from [69].

Figure 11

Contour plots of (left) mean arithmetical roughness Sa (μm) vs. infill (IN) and printing speed (PS), and (right) dimensional error (%) vs. infill (IN) and layer height (LH); adapted with permission from [95].

Figure 12

B₄C green parts produced using the 584 μm (A0-2) and 406 μm (B0-2) extrusion orifices; better resolution and shape retention were achieved with the finer extrusion nozzle. Printing parameters: 47 vol% B₄C paste, 80 psi (550 kPa) syringe pressure, 4 mm/s deposition speed, and 440 μm layer height (A); and 47 vol%, 30 psi (200 kPa), 4 mm/s, and 260 μm, respectively (B). Adapted with permission from [56].

Figure 13

Robocasting protocol flowchart, from starting powder to final sintered and post-processed ceramic part.