a Department of Chemical, Paper and Biomedical Engineering , Miami University , Oxford , OH , USA.
b Department of Statistics , Miami University , Oxford , OH , USA.
J Biomater Sci Polym Ed. 2018 Feb;29(3):195-216. doi: 10.1080/09205063.2017.1409022. Epub 2017 Dec 1.
Tissue engineering using three-dimensional porous scaffolds has shown promise for the restoration of normal function in injured and diseased tissues and organs. Rigorous control over scaffold architecture in melt extrusion additive manufacturing is highly restricted mainly due to pronounced variations in the deposited strand diameter upon any variations in process conditions and polymer viscoelasticity. We have designed an I-optimal, split-plot experiment to study the extrudate swell in melt extrusion additive manufacturing and to control the scaffold architecture. The designed experiment was used to generate data to relate three responses (swell, density, and modulus) to a set of controllable factors (plotting needle diameter, temperature, pressure, and the dispensing speed). The fitted regression relationships were used to optimize the three responses simultaneously. The swell response was constrained to be close to 1 while maximizing the modulus and minimizing the density. Constraining the extrudate swell to 1 generates design-driven scaffolds, with strand diameters equal to the plotting needle diameter, and allows a greater control over scaffold pore size. Hence, the modulus of the scaffolds can be fully controlled by adjusting the in-plane distance between the deposited strands. To the extent of the model's validity, we can eliminate the effect of extrudate swell in designing these scaffolds, while targeting a range of porosity and modulus appropriate for bone tissue engineering. The result of this optimization was a predicted modulus of 14 MPa and a predicted density of 0.29 g/cm (porosity ≈ 75%) using polycaprolactone as scaffold material. These predicted responses corresponded to factor levels of 0.6 μm for the plotting needle diameter, plotting pressure of 2.5 bar, melt temperature of 113.5 °C, and dispensing speed of 2 mm/s. The validation scaffold enabled us to quantify the percentage difference for the predictions, which was 9.5% for the extrudate swell, 19% for the density, and 29% for the modulus.
组织工程使用三维多孔支架已经显示出在受伤和患病组织和器官中恢复正常功能的潜力。在熔融挤出增材制造中对支架结构的严格控制受到高度限制，主要是由于工艺条件和聚合物粘弹性的任何变化都会导致沉积丝直径的明显变化。我们设计了一个 I-最优、分块实验来研究熔融挤出增材制造中的挤出物膨胀，并控制支架结构。该设计实验用于生成数据，将三个响应（膨胀、密度和模量）与一组可控因素（绘图针直径、温度、压力和分配速度）联系起来。拟合的回归关系用于同时优化这三个响应。将挤出物膨胀限制为接近 1，同时最大化模量和最小化密度。将挤出物膨胀限制为 1 可以生成设计驱动的支架，其丝径等于绘图针直径，并允许对支架孔径进行更大的控制。因此，通过调整沉积丝之间的平面距离，可以完全控制支架的模量。在模型的有效性范围内，我们可以在设计这些支架时消除挤出物膨胀的影响，同时针对适合骨组织工程的一系列孔隙率和模量。该优化的结果是，使用聚己内酯作为支架材料，预测的模量为 14 MPa，预测的密度为 0.29 g/cm（孔隙率≈75%）。这些预测的响应对应于绘图针直径为 0.6 µm、绘图压力为 2.5 巴、熔体温度为 113.5°C 和分配速度为 2mm/s 的因子水平。验证支架使我们能够量化预测的差异百分比，挤出物膨胀为 9.5%，密度为 19%，模量为 29%。
