Optimization of an ad hoc Realized Space Frame Structured RVE for FEM Modeling of Nanoporous Biopolymeric Scaffolds Obtained by Supercritical Fluids Assisted Process

• Published in: Chemical engineering transactions 2016-05, Vol.49
• Main Authors: F. Naddeo, L. Baldino, S. Cardea, A. Naddeo, E. Reverchon
• Format: Article
• Language: English
• ISSN: 2283-9216
• DOI: 10.3303/CET1649029

In this work an hypothesis of modeling nanofibers network of Poly-L-lactide (PLLA) scaffolds loaded with hydroxyapatite (HA) nanoparticles, suitable for tissue engineering applications, is presented to investigate the mechanical properties by FEM analysis. Scaffolds were produced by Supercritical CO2 drying of polymeric gels. FEM modeling of nanoporous biomaterials involves computational problems such as: the reproduction of the nano-morphology by means of different techniques, such as molecular dynamics simulations that hardly lead to results adherent to the experimental evidence; the reverse engineering to extrapolate geometric information; the identification of a periodic representative volume element (RVE) to reach more coherent results and to reduce the simulation computational effort. Basic modeling assumptions are: a) polymer particles are small enough to exhibit Brownian motion; b) scaffold SEM images show that the porous structure consists of curved fibers that depart from punctiform nuclei realizing a space frame; c) scaffold experimental compressive tests show that the porous material behaves as a soft isotropic material. On the basis of these assumptions, a parametric algorithm that creates a cubic RVE, showing a nanofibers network and having the same porosity of the real material, has been written; RVE size has been optimized on the bases of its material isotropic degree measured by an ad hoc created iterative algorithm for generating a rod spaceframe; RVE mechanical behavior has been optimized by curving appropriately each fiber according to the experimental data and on the basis of SEM imaging diagnostic. Linear FEM simulations on mechanical behavior have given qualitatively and quantitative satisfactory results when compared to the experimental ones.