Computational Fluid Dynamics-Based Design of a Microfabricated Cell Capture Device

Journal of Chromatographic Science 2015;53:411– 416 doi:10.1093/chromsci/bmu110 Advance Access publication September 8, 2014 Article Computational Fluid Dynamics-Based Design of a Microfabricated Cell Capture Device Gabor Jarvas1,2, Marton Szigeti1,3, Laszlo Hajba1, Peter Furjes4 and Andras Guttman1,3* 1 MTA-PE Translational Glycomics Research Group, University of Pannonia, Veszprem, Hungary, 2CEITEC—Central European Institute of Technology, Brno, Czech Republic, 3Horvath Laboratory of Bioseparation Sciences, University of Debrecen, Debrecen, Hungary, and 4 Institute for Technical Physics and Materials Science, Research Centre for Natural Sciences—HAS, Budapest, Hungary *Author to whom correspondence should be addressed. Email: Received 14 July 2014 A microfluidic cell capture device was designed, fabricated, evaluated by numerical simulations and validated experimentally. The cell capture device was designed with a minimal footprint compartment comprising internal micropillars with the goal to obtain a compact, integrated bioanalytical system. The design of the device was accomplished by computational fluid dynamics (CFD) simulations. Various microdevice designs were rapidly prototyped in poly-dimethylsiloxane using conventional soft lithograpy technique applying micropatterned SU-8 epoxy based negative photoresist as moulding replica. The numerically modeled flow characteristics of the cell capture device were experimentally validated by tracing and microscopic recording the flow trajectories using yeast cells. Finally, we give some perspectives on how CFD modeling can be used in the early stage of microfluidics-based cell capture device development. Introduction Cell sorting is a crucial part of blood sample preparation and thus has particular importance in biomedical sciences. In circulating tumor cell (CTC) research, effective cell capture is an absolute necessity, because blood represents an extremely heterogenic sample, thus reduction of complexity is of high importance (1, 2). Flow cytometry, which is the traditional way of cell sorting and capture, is based on optical detection of cells encapsulated in droplets passing in front of a detector with high speed (3). The major drawbacks of flow cytometry are the requirement for presorting (filtering, centrifugation and rinsing), long processing time, need highly trained service personnel and the requirement for large sample volumes. In rare cell analysis, the latter one is a limiting factor since for the time being, the detection limit of flow contemporary cytometers is around a hundred cells, while the typical number of CTCs in blood range from one (if any) to several dozen per 10 mL (4, 5). To overcome the abovementioned issues, fluorescently activated cell sorting (6), dielectrophoretic sorting (7), electrokinetic isolation, inertial separation, controlled pressure sorting (8, 9) and magneticactivated particle-based (4) methods have been proposed (10). Microfabricated chips with bioaffinity surfaces represent an additional specific class of cell capture devices with the possibility of integration to further processing compartments, such as digestion, derivatization, cleaning, etc. These devices usually consist of an array of microposts coated by cell-specific antibodies with the usual footprint of a microscope plate (11 – 13). Because of their relatively large size, it is challenging to integrate them into complex lab-on-a-chip systems. The capture compartment topography could feature well-ordered pillars with uniform diameter (5, 11, 12) or randomly sized and randomly positioned posts (13). Because microfabricated cell capture devices (MCCDs) mostly process and analyze minute amount of blood samples (14, 15), their capture efficiency is critical and could be estimated by computational fluid dynamics (CFD) approaches. Characteristic geometrical dimensions of microfabricated cell sorters and target cells are in the same order of magnitude (16), 10 mm. Because of the narrow channels, the surface-to-volume ratio of microfabricated cell sorters is very high, posing unusual engineering challenges, which further justify the need of computerassisted design. While numerical modeling and simulation of microfluidic systems is primarily considered as a design tool, it can also be used to support experimental data interpretation (17). From the viewpoint of bioanalysis, CFD is a developer tool to help quickly achieve an optimal design of custom-made devices at low cost with a minimal number of actual experiments. Poly-dimethylsiloxane (PDMS) is a silicon-based organic polymer, which is frequently used in rapid prototyping of lab-on-a-chip or microfluidic devices due to its biocompatibility, chemical and biological resistance, transparency, easy pattern transfer and low cost (18, 19). Numerous types of inflow liquid spreader designs have been published to aim maximized flow throughput, while keeping footprint and shear stress in the distribution channels at minimum level, and offering a uniform flow field along the device. Viovy and coworkers (20) improved the traditional tree-like inflow design, in a way that the distribution microchannels were subsequently divided into two subchannels with equal lengths and widths. The resultant new flow distributor applied subchannels with unequal lengths and widths according to the Hele-Shaw approximation. Please note that in this arrangement the fluid spreader took up two-third of the chip footprint. Another type of the microfluidic cell sorter was developed by Nora Dickson et al. (13) with a rather simple flow distributor in which the incoming flow was equally divided to four parts, covering only 25% of the functional surface of the microdevice. In this study, a fluid distributor with an extremely small footprint was used to minimize the nonfunctional area of the MCCD. This modified disc-section shaped distributor offered lower uniformity than common channels. However, it showed no significant effect on cell capture efficiency because just the maximum value of the share rate was defined as design criteria and not its distribution. In this paper, we report on the design, microfabrication and validation of a novel minimal footprint MCCD with special # The Author 2014. Published by Oxford University Press. All rights reserved. For Permissions, please email: emphasis on fluid flow engineering. First, numerous alternative chip strategies were designed by altering the size and layout of the micropillars as well as the type of flow spreader at the inlet part. The most promising designs were further investigated by means of numerical simulations. The modeled layouts were microfabricated using a standard soft lithography technique, including SU-8 master replica formation followed by PDMS molding and oxygen plasma-enhanced bonding. The flow inside the devices was validated using manual flow pattern tracking to evaluate the fluid dynamics performance of the developed cell sorters. Modeling The applied microfabric (...truncated)