Page 185 - IJB-7-4

Fu, et al.
sets of optimal printing parameters which have higher nozzle gauge. Material composition refers to the weight/
(>75%) probability to generate high fidelity PL 127 volume percentage (w/v %) of the PL 127 solution used
filaments. Traditional factorial experimental design is during testing. This parameter will likely have an effect
time-consuming, and the cost increases exponentially on print outcomes due to its impact on the viscosity
with increasing number of parameters and levels. of the material. Nozzle temperature is the temperature
Building physical models for the bioprinting process are that the nozzle and material inside it is held at while
also challenging due to the complex relationship between printing. It will likely affect prints due to the relation
disparate printing parameters (e.g., biomaterial properties between PL 127 temperature and viscosity. While
and process parameters) and print outcome. The SVM this parameter is controlled by the printer, its impacts
process optimization methodology was inspired by relate to the structure and material properties of PL
Aoyagi et al . In this study, we selected biomaterial 127. Therefore, it has been placed under the “material”
[27]
concentration, nozzle temperature, and printing path category of parameter, even though it may be seen as
height as three key parameters. A space-filling Design of both a “printing” and “material” parameter. Path height
Experiment technique was used to select only 12 training is the vertical offset between the printing nozzle and
data. A 3D process map was generated by the pairwise the print bed. During printing the material is stretched
probability prediction based on SVM model and the by different amounts depending on how high the path
validation on the unseen data points showed the model height is set. Nozzle gauge refers to the gauge number
generalized well on the parameter space. of the printing nozzle being used. Each nozzle has a
different inner and outer diameter.
2. Materials and methods
2.4. Rheological evaluation of PL 127
2.1. Preparation of materials
The viscosities of Pluronic inks were tested by a
Solutions of PL 127 were prepared by first cooling rheometer (R/S-CPS+, Brookfield, USA). The rheometer
deionized (DI) water in a 4°C refrigerator, adding Pluronic is equipped with a temperature control Peltier (0 – 135°C).
F-127 (Sigma-Aldrich, St. Louis, MO) powder to create A P50 plate (radius 25 mm) with 1 mm gap was used in
a large 30 w/v% sample, stirring using a magnetic stirrer, the plate/plate measuring system for all tests. For each
and then allowing the sample to homogenize fully in a test, 2 mL sample was loaded on 4°C plate to fill the gap
4°C refrigerator. Calculations for composition were based completely. Viscosities for all concentrations of Pluronic
on the final solution volume. For lower compositions, the inks were first measured at temperature ramp from 40°C
same sample of 30 w/v% PL 127 was then diluted down to 4°C for 15 min and constant shear rate at 1/s. The
using DI water, mixed, and again allowed to homogenize
fully in a 4°C refrigerator before testing began. This viscosities of all Pluronic inks at 23°C against shear rate
method was used to prevent any false affects appearing ramp were also tested from 0.01 to 100/s for 5 min.
in the data due to variations between batches of material. 2.5. Variable testing

2.2. Printing and measurement Baseline values were selected for each variable to be held
For all tests, an extrusion-based Bioprinter (BioMaker, constant while one category was varied independently.
SunP Biotech, Cherry Hill, NJ) was used along with the The selected values were a path height of 0.3 mm, nozzle
samples synthesized wit PL 127 powder (Sigma-Aldrich, gauge of 25, room temperature (23°C), and a composition
St. Louis, MO). This printer uses motor-based extrusion, of 30% PL 127. Values were varied in one category at a
as opposed to the also-common pressure-based extrusion time while all other categories were kept at their baseline
used in bioprinting. All CAD designs and slicing are values.
included in the software for this printer, and as such these 2.6. Model grid for printing
were the only software used to create and slice a grid
design for testing in this experiment. Measurements from The model grid used for printing was designed in the
microscope images of each print were taken using Fiji/ built-in software for the SunP Biotech Bioprinter. The
ImageJ. grid was a 0.6 mm tall square with three 0.2 mm layers
and a 6 mm side length. Theoretical line width of the infill
2.3. Parameter selection for evaluation lines was 0.4 mm.
Four parameters were selected for testing: two in the 2.7. Data collection
material property category and two in the printing
parameter category. The material parameters selected Before printing, material was pulled from the samples
were printing temperature and material composition, stored in a 4°C refrigerator into a 5 mL syringe. Syringes
and process parameters selected were path height and were allowed to come to room temperature for 10 min

International Journal of Bioprinting (2021)–Volume 7, Issue 4 181