C. González-Fernández, J. Gómez-Pastora, I.H. Karampelas, E. Bringas, I. Ortiz
The Ohio State University,
Keywords: magnetophoresis, microfluidics, CFD, fabrication
Summary:Continuous flow bioseparators allow magnetic particles that are properly functionalized so that they get attached to dangerous pathogens. Specifically, the unique properties of magnetic particles enable their separation from blood, after fulfilling their role of pathogen capture, in a continuous process. This could be achieved inside microfluidic systems that display multiple advantages such as low sample consumption, high throughput when optimally parallelized, laminar flow conditions (and thus coflowing of miscible fluids without mixing) and the feasibility of integrating multiple steps onto one chip. However, the optimization of the system with an emphasis on the available chip fabrication methods and the potential limitations of these methods regarding the final cross-sectional area as well as the chip length remain less explored. In this study, we introduced a computational approach to address different system configurations in order to optimize the performance of a multiphase continuous-flow microfluidic device and thus select the appropriate fabrication method depending on the cross section and length desired. In these systems, the beads, suspended in human whole blood and a buffer stream are continuously injected through different inlets of a Y-Y shaped microchannel. The application of an external magnetic field by a permanent magnet allows the deflection of the particles and their collection in the co-flowing buffer stream. The optimization of the system involves the treatment of high contaminated blood volumes while delivering both maximum beads recovery and impaired blood quality. In order to address this technological challenge, we introduced a Computational Fluid Dynamics (CFD)-based Eulerian-Lagrangian approach using the commercial software FLOW-3D, which was linked to a Fortran code to combine a magnetic and fluidic analysis so as to accurately describe the bead trajectories. The parametric analysis focuses on the impact of different variables, such as the shape of the cross sectional area and its dimensions and device length, which in turn, determines the fabrication method to be employed. The model results were experimentally validated through fluorescence microscopy for two of the chips under study. Both experimental and computational results suggest that when the shape of the cross section is changed, this has a tremendous effect on the velocity profile developed within the channel, and thus in the flow rates that can be used to achieve maximum particle recovery. Thereby, when considering U-shaped cross sections, as obtained by conventional photolithography and wet etching fabrication techniques, the velocities for obtaining beads sequestrations higher than 95% are one order of magnitude lower than when rectangular cross sections are analyzed. Also, by increasing 5 times the chip length, the separation efficacy increases as much as 68% for all the devices, allowing the use of higher flow rates and thus enhancing the system throughput. Finally, for comparing the magnetophoretic microfluidic systems, dimensionless numbers that take into account key operating variables and parameters are introduced. The theoretical and experimental methodology developed in this work provide insight into the rational design of magnetophoretic-microfluidic devices for biomolecule separations and can be applied to a broad range of magnetically-enabled microfluidic applications.