Page 73 - IJB-7-1
P. 73

Jing, et al.
                        A                                         B















           Figure 1. Electrohydrodynamic printing (EHDP) system and scaffold fabrication. (A) Schematic diagram of EHDP system. (B) Scaffold
           structure by fiber stacking.


               To establish a viable  manufacturing  process and   3. Results and discussion
           minimize  defect-related  loss in long time  fabrication,
           the  captured  EHDP  cone  and  jet  images  were  applied   3.1. Composite scaffold structure and tensile test
           to detect abnormal modes, as shown in Figure 2A. The   The  fabricated  scaffold  fibers  with  varied  material
           identification  task  in  this  EHDP  monitoring  system   compositions  may  influence  cell-scaffold  interactions.
           involves the application of an image processing algorithm   To investigate  this factor, we prepared scaffolds using
           to extract relevant features and a recognition algorithm to   the same structural parameters (pore size, fiber diameter,
           determine the modes of Taylor cones. This is very similar   and number of layers),  and fabricated  PCL, PCL/zein-
           to traditional  fabrication  process monitoring,  where   10, PCL/zein-20, PCL/gliadin-10, and PCL/gliadin-20
           feature extraction and selection are applied to determine   scaffolds for comparison.
           the input features of machining learning  methods for
           effective condition identification [24,25] .        (1) Composite scaffold structure analysis
               Figure 2B shows a standard cone with a straight jet
           for  EHDP  scaffold  fabrication.  In  the  EHDP  monitoring   The printed scaffolds consist of 12 layers of fibers with
           and identification system, various types of corona discharge   a thickness of about 65 – 85 μm. The thickness of PCL,
           were  reported.  As  shown  in  Figure  2C, the corona   PCL/zein, and PCL/gliadin scaffolds is stated in Table 1.
           discharge happened slightly below the needle tip at initial   The fiber diameter of top layer is approximately 8.9 –
           jetting process or during fabrication. Such discharge could   9.4 μm, which is close to typical cell size. Due to the
           be avoided by optimizing EHDP process parameters and   impact of ejected filament onto the previously deposited
           environmental  parameters.  Figure  2D shows the current   fiber  layers,  the  fiber  diameter  of  the  bottom  layers  is
           flow through the air from the region surrounding the jet to the   about twice the diameter of the top layer cross of all
           grounded substrate. This reduced the surface charge density   scaffolds. The porosity of these scaffolds is about 89 –
           and increased the jet’s lateral stability. Different from the   92.1%, which is in favor for exchanging nutrition and
           above-mentioned two scenarios, a huge Taylor cone with   waste. In Table 1, the structural parameters such as fiber
           serious discharge at the needle tip is shown in Figure 2E.   diameter, scaffold thickness, bulk density, and porosity
           To avoid possible damages of the EHDP fabrication system,   are  very  similar  for  all  the  scaffolds.  In  other  words,
           we need to properly vary the biomaterial ink properties,   the proposed biomaterial inks with current components
           process parameters, and environmental parameters.   and  portion  do  not  influence  the  morphology  of  these
               Overall,  this  monitoring  and  identification  system   scaffolds obviously.
           provides intuitive  information  of EHDP fabrication    As shown in  Figure  3, PCL and PCL/gliadin
           process and bridges  the  knowledge  gap between  the   scaffolds were observed using a scanning electron
           corona discharges and electrical  properties of the   microscope  (SEM,  JSM-6510,  JEOL,  Japan)  for
           composite materials. Researchers can evaluate new inks’   morphology analysis. The pore size was precisely defined,
           property and stability during long time printing and gain   and  fibers  were  precisely  stacked  in  a  layer-by-layer
           additional insights into fundamental mechanism causing   manner. For the fabricated scaffolds using PCL/zein and
           corona discharges. It can also assist new inks’ fabrication   PCL/gliadin inks, there was not much difference in terms
           process parameter optimization such as applied electric   of fiber diameter and cross-section structure. In general,
           field  between  the  nozzle  and  the  substrate,  solution   these scaffold parameters are independent of the material
           feeding rate, programmed stage speed and moving path,   compositions.  The  tensile  properties  of  these  scaffolds
           temperature, and humidity.                          were compared and analyzed in sub-section.

                                       International Journal of Bioprinting (2021)–Volume 7, Issue 1        69
   68   69   70   71   72   73   74   75   76   77   78