Ngoc H. Pham

Predicting the stress distribution within scaffolds with ordered architecture

Current tissue engineering technologies involve the seeding of cells on porous scaffolds, within which the cells can proliferate and differentiate, when cultured in bioreactors. The flow of culture media through the scaffolds generates stresses that are important for both cell differentiation and cell growth. A recent study [Appl. Phys. Lett. 97 (2010), 024101] showed that flow-induced stresses inside highly porous and randomly structured scaffolds follow a three-point gamma probability density function (p.d.f.). The goal of the present study is to further investigate whether the same p.d.f. can also describe the distribution of stresses in structured porous scaffolds, what is the range of scaffold porosity for which the distribution is valid, and what is the physical reason for such behavior. To do that, the p.d.f. of flow-induced stresses in different scaffold geometries were calculated via flow dynamics simulations. It was found that the direction of flow relative to the internal architecture of the scaffolds is important for stress distributions. The stress distributions follow a common distribution within statistically acceptable accuracy, when the flow direction does not coincide with the direction of internal structural elements of the scaffold.

Flow Recovery Downstream from Nanoposts Grown at the Wall of a Microchannel

In this article, hydraulic responses to a linear array of finite-length nanoposts attached at the bottom wall of square microchannels under viscous flow conditions are investigated with lattice Boltzmann simulations. Different configurations of the array are considered because these changes can directly contribute to flow pattern deformation. Simulation results indicated that the flow structure strongly depends on nanopost height and space between two adjacent nanoposts in the nanopost line but not on the Reynolds number in the range examined. However, if nanoposts are grown far apart from each other, a fully developed velocity profile can be recovered at sufficiently long distance downstream and an empirical correlation for calculating the recovery length is proposed. In the studied low Reynolds number regime, energy loss due to friction drag was found to be inversely proportional to an appropriately defined Reynolds number that accounts for both the channel size and the nanopost aspect ratio.

Chapter 23 – Lattice Boltzmann Methods for Bioengineering Applications

Abstract Computational techniques have gained wider acceptance and application within biomedical and bioengineering applications. Among the most potent numerical techniques for simulating biological flows in complex geometries is the lattice Boltzmann method (LBM). It is appropriate for cases where the implementation of boundary conditions can be difficult when applying other methods and in cases where parallelization of the computations is needed to simulate large systems. In this work, we report on the LBM methodology and applications, drawing mainly from our research on using LBM to simulate flows in scaffolds and perfusion bioreactors. The flow-induced stresses can be predicted with LBM, and vital information about the successful culture of cells can be generated. In addition to the simulation of flow, the LBM can be used in conjunction with particle-based techniques to simulate mass transfer in the flow field. The combination of LBM with Lagrangian particle tracking or Lagrangian scalar tracking is also highlighted.

Effect of spatial distribution of porous matrix surface charge heterogeneity on nanoparticle attachment in a packed bed

In this study, the effect of spatial distribution of the porous matrix surface heterogeneity on nanoparticle deposition is numerically explored using lattice Boltzmann simulation methods and tracking of individual particles with Lagrangian algorithms. Packed beds with four different patterns of surface charge heterogeneity, on which favorable surfaces for particle attachment are located at different locations, are generated. The heterogeneity is binary, so that the porous surface can either accommodate nanoparticle attachment or not. It is found that the heterogeneity pattern has a stronger effect when the rate constant for particle attachment is high, when the particle size is small, and/or when the fraction of the surface area that is favorable to attachment is about 0.5. At fixed conditions, the heterogeneity pattern with randomly and uniformly distributed active surface area is the most favorite for particle attachment, compared to those where the active surface areas are banded perpendicularly to the...