Monte Carlo random walk simulation as a complement to experimental and theoretical approaches: Application to mass transfer in fish muscle tissue.

摘要

Mass transfer processes in food systems, such as solute infusion, are poorly understood because of their complex nature. Food systems contain porous matrices and a variety of continuous phases within cellular tissues. Mass transfer processes are generally not pure diffusion: often convection, binding and obstructing diffusion will occur. Monte Carlo (MC) simulation has been increasingly used in life science and engineering to elucidate molecular transport in biological systems. However, there are few articles available discussing MC simulation in food processing, especially mass transfer. The main goal of this study was to show the inherent simplicity of the MC approach and its potential when combined with traditional experimental and theoretical approaches to better describe and understand mass transfer processes. A basic framework for MC---random walk simulation applied to a diffusion problem---is developed in this project. Infusion of two sizes of dextran macromolecules in fish muscle cells is used to apply the MC framework in combination with Fluorescence Recovery After Photobleaching experiments. Effective diffusivity coefficients within cells, considering the degree of obstruction due to the myofibrilar matrix, are assessed. Then, the results are used as input in a mathematical model that was developed for theoretical simulation of mass transfer in the multi-cellular tissue. Diffusivity values obtained by the MC framework had an SD of +/-0.02 [mum2/s] around the true value of 0.25 [mum2/s]. MC results for degree of obstruction were 0.29 and 0.34 for dextran FD10S and FD20S, respectively, and the De values were 23.7 and 11.2 [mum2/s]. The statistical error in the estimation of De was estimated to be [22.8-24.6] and [9.7-12.7] (95% CI), where average experimental values of 24.3 [mum2/s] for FD10S and 11.4 [mum2/s] for FD20S were captured by the respective interval. The theoretical model showed a significant influence of the cell membrane characteristics and tissue porosity in both the degree of solute penetration and the solute distribution between intra- and extra-cellular space. The combined approach was successfully applied to a diffusion problem. Overall, it is expected that the present work will contribute towards the application of MC simulation in the field of Food Science and Engineering.