Density-gradient-assisted centrifugal microfluidics: an approach to continuous-mode particle separation.

Centrifugal microfluidics has been recognized as a promising pumping method in microfluidics because of its simplicity, easiness of automation, and parallel processing. However, the patterning of stripe flow in centrifugal microfluidics is challenging because a fluid is significantly affected by the Coriolis force, which produces an intrinsic secondary flow. This paper reports a technical and design strategy for centrifugal microfluidics called "density-gradient-assisted centrifugal microfluidics." The flow behavior is observed with the presence of a density gradient and without a density gradient in two concentrically traveling phase flows. As a result, clear stripe flow pattern is observed with a density difference of 0.05 g/cm ³ between water and a percoll solution at a flow rate of 11.8 μl/s (7 ml/10 min) and spinning speed of 3000 rpm. In contrast, without a density gradient, it is necessary to reduce the flow rate and spinning speed to 0.1 μl/s and 1000 rpm, respectively. This paper also presents the use of a density gradient to assist in focusing resin (polystyrene) particles on the boundary of a stripe flow pattern that consists of water and percoll with different densities. Moreover, the density-based separation and sorting of particles in a mixed particle suspension is demonstrated. Polystyrene is selectively focused on the boundary, but silica particles are separated from the focused trajectory due to a difference in density. The separated particles are continuously sorted into different reservoirs with polystyrene and silica separation efficiencies of 96.5% and 98.5%, respectively. The pumping, stripe flow pattern formation, particle concentration, and sorting are simultaneously realized by applying a density gradient and centrifugal force. Therefore, this principle can realize a very simple technique for label-free particle separation by just spinning a disk device and can be applied in other applications by the use of the density-gradient assistance.