Page 251 - IJB-10-4
P. 251

International Journal of Bioprinting                                   Cell viability in printing structured inks




            were developed. These models considered the cartridge   set, a pressure-based transient solver was employed to
            and nozzle specifications for CFD simulations (Figure 2).    calculate the solution, utilizing governing equations,
                                                                                                            37
            Properties of the equivalent homogeneous inks for   including continuity, momentum, and energy equations.
            both symmetric and core–shell inks were analyzed to   Post-simulation results were imported into the CFD-Post
            generate approximately the same fluid forces during the   package for visual representation and quantitative analysis
            printing processes.                                to evaluate the distribution of material phases at the nozzle
                                                               outlets, as well as fluid forces, including pressure, wall
            2.6. Simulation method overview
            To calculate the fluid forces affecting cell viability, a   shear stress, and shear stress at material phase interfaces.
            sequential approach encompassing pre-processing, solving,   2.7. Numerical simulation models
            and  post-processing was  employed for  CFD  simulation.   For fluid models, there are two categories corresponding to
            The ANSYS Workbench software, version 2020R2, was used   the printing of different structured inks and conventional
            with Fluent as the solver and CFD-Post package as the post-  printing. All fluid models were simplified due to their
            processor. To enhance computational efficiency, the fluid   inherent symmetry. The model corresponding to hepatic
            models were simplified by means of symmetry. During the   lobule analogue-like inks was simplified to 1/12 size, the
            pre-processing phase, models for fluid analysis, designed in   symmetric model was simplified to 1/2, while other models
            Parasolid format, were imported into the Design Modeler   were conventionally streamlined to 1/4. The respective 3D
            module. Subsequently, the corresponding mesh models   symmetric models were modeled, as illustrated in Figure
            were generated through the Mesh module, with a focus on   S2 (Supplementary File). For 2-symmetric inks, symmetric
            both computational accuracy and time efficiency. Inlets,   inks, and core–shell inks, the pink and green areas were
            outlets, pipe walls, and symmetry planes were then defined   constrained  as  ink  1  and  ink  2,  respectively.  In  other
            for the subsequent solving phase. In the solving stage,   cases, the pink, green, and blue-purple domains were all
            the  multiphase  flow  methodology  in  Fluent,  using  the   constrained as ink 1, with the distinction in their regions
            “Volume of Fluid” module and a laminar flow model, was   being the involvement of cells. All gray planes and black
            selected for  structured inks-based  methods.  This choice   planes indicate 3D symmetry planes for the respective fluid
            was made because the materials used in the 3D printing of   models. Domain in phase 1 corresponds to inlet 1, domain
            structured inks maintain laminar flow within the printhead   in phase 2 to inlet 2, and domain in phase 3 to inlet 3.
            channels during the extrusion processes.  Surface tension
                                            30
            modeling was implemented to capture interfaces between   In the process of model meshing, smaller elements
            different phases, and single-phase laminar flow models   typically result in more accurate finite element modeling,
            were chosen for simulating conventional printing involved   as an increased number of mesh elements  enhances
            in comparative and equivalent analysis. Once material   accuracy in calculating fluid behavior. However, this
            properties were defined and boundary conditions were   comes at the cost of higher computational resource,























            Figure 2. Fluid models for comparative and equivalent analysis of fluid forces. (A) Representative overall fluid domain illustrating the fluid models for
            analysis of fluid forces. (B) Domains of structured inks involving 2-symmetry and 4-symmetry (O ), vascularity (Q ), hepatic lobule analogue (Q ),
                                                                             1
                                                                                        2
                                                                                                            3
            symmetry (Q ), and core-shell structure (Q ), respectively. (C) Connecting part in the overall fluid domain (P). (D) Domains of conical parts in the
                     4
                                        5
            corresponding fluid domains involving 18G needle (M ), 27G needle (M ), and 32G needle (M ), respectively. (E) Combined modules of overall fluid
                                               1
                                                           2
                                                                          3
            models for comparative and equivalent analysis.
            Volume 10 Issue 4 (2024)                       243                                doi: 10.36922/ijb.2362
   246   247   248   249   250   251   252   253   254   255   256