Curve-Based Infill Pattern Optimization for 3D Printed Polymeric Scaffolds for Trabecular Bone Applications

Abstract

Additive manufacturing technology, specifically material extrusion, offers great potential for scaffold manufacturing in tissue engineering. This study presents a novel methodology for the design and optimization of 3D printed polymeric scaffolds to enhance cell viability, thereby promoting improved cell proliferation for tissue engineering applications. Different infill patterns, including gyroid, parallel sinusoidal, and symmetric sinusoidal, were evaluated to determine their impact on cell proliferation and tissue regeneration. To overcome the limitations of existing slicer software, a novel open-source software called FullControl GCode Designer was utilized, enabling the creation of customized infill patterns without restrictions. VOLCO software was employed to generate voxelized 3D models of the scaffolds, simulating the material extrusion process. Finite element analysis was conducted using Abaqus software to evaluate the mechanical properties of the different designs. Additionally, new scripts were developed to evaluate the interconnectivity and pore size of the voxelized models. A factorial design of experiments and a genetic algorithm (combined with Kriging metamodels) were applied to identify the optimal configuration based on optimization criteria (keeping the mechanical stiffness and pore size within the recommended values for trabecular bone and maximizing the surface and interconnectivity). Biological testing was conducted on polylactic acid scaffolds to preliminarily validate the effectiveness of the modeling and optimization methodologies in this regard. The results demonstrated the agreement between the optimization methodology and the biological test since the optimum in both cases was a symmetric sinusoidal pattern design with a configuration resulting in a structure with 53.08% porosity and an equivalent pore size of 584 µm. Therefore, this outcome validates the proposed methodologies, emphasizing the role of pore surface area and interconnectivity in supporting cell proliferation. Overall, this research contributes to the advancement of AM technology in tissue engineering and paves the way for further optimization studies in scaffold design.

Keywords: FEA; additive manufacturing; modeling; optimization; scaffold; tissue engineering.

Figure 1

Methodology for modeling and optimizing sinusoidal scaffolds.

Figure 2

Design parameters for sinusoidal path generation in FullControl.

Figure 3

Differences among sinusoidal infill patterns. (a) First layer of cosine pattern. (b) Two layers of cosine pattern. (c) Four layers of cosine pattern. (d) First layer of sine pattern. (e) Two layers of sine pattern. (f) Four layers of sine pattern. (g) First layer of cosine and sine combination pattern. (h) Two layers of cosine-sine pattern. (i) Four layers of cosine-sine pattern.

Figure 4

Geometric differences of 9 × 9 × 9 mm sinusoidal scaffolds depending on the values of variables. (a) Scaffold with a layer height of 0.15 mm (front view). (b) Scaffold with a layer height of 0.3 mm (front view). (c) Filament layout with an amplitude of 0.5 mm, 5 cycles, and 5 filaments per layer. (d) Filament layout with an amplitude of 0.5 mm, 5 cycles, and 9 filaments per layer. (e) Filament layout with an amplitude of 0.5 mm, 9 cycles, and 5 filaments per layer. (f) Filament layout with an amplitude of 0.1 mm, 5 cycles, and 5 filaments per layer.

Figure 5

Difference between filament configurations. (a) Parallel filaments or sinusoidal. (b) Symmetric filaments or reflected sinusoidal.

Figure 6

Pore size calculation process.

Figure 7

Pore size determination. (a) 3D voxel matrix and its Z projection. (b) Equivalent pore diameter (red circle).

Figure 8

Optimization diagram of scaffolds with curve-based infill patterns: parallel sinusoidal scaffold (blue), symmetric sinusoidal scaffold (yellow), and gyroid (red).

Figure 9

Optimization process.

Figure 10

Optimal and non-optimal scaffolds geometry. (a) REF_op. (b) GYR_op. (c) SIN_op. (d) REF_nop. (e) GYR_nop. (f) SIN_nop.

Figure 11

Optimal and non-optimal 3D printed scaffolds. (a) REF_op. (b) GYR_op. (c) SIN_op. (d) REF_nop. (e) GYR_nop. (f) SIN_nop.

Figure 12

Models pore size compared with the optimal value.

Figure 13

Comparison of the pondered values of the objectives, ordered according to the weighted average value.

Figure 14

Comparison criteria for optimal model determination. S = surface area, I = interconnectivity, P = penalty.

Figure 15

Metabolic activity of human bone marrow mesenchymal stem cells cultured on the different groups of scaffolds determined by the CCK-8 assay at day 5 (* p < 0.05, ** p < 0.01, and *** p < 0.001).

Figure 16

Optimal sample configuration. (a) Geometry. (b) Z-projection. (c) Non-deformed scaffold in the XZ plane. (d) Displacement of the deformed scaffold in the XZ plane.