Viscosimeter on a Microfluidic Chip

Abstract

In this work, a viscosimeter implemented on a microfluidic chip is presented. The physical principle of this system is to use laminar parallel flows in a microfluidic channel. The fluid to be studied flows side by side with a reference fluid of known viscosity. By using optical microscopy, the shape of the interface between both fluids can be determined. Knowing the flow rates of the two liquids and the geometrical features of the channel, the mean shear rate sustained by the fluid and its viscosity can thus be computed. Accurate and precise measurements of the viscosity as a function of the shear rate can be made using less than 300 microL of fluid. Several complex fluids are tested with viscosities ranging from 10(-)(3) to 70 Pa.s.

On the flow topology inside droplets moving in rectangular microchannels

“…In Fig. 3a, the trend in increasing v d /v c with increasing Ca was 35 moderately significant for a < w (black and red), giving r = 0.6779, p = 0.0312. For the plug droplets a > w (green and pink), a highly significant correlation was observed: r = 0.7928, p < 0.0001.…”

“…A channel depth of 150 µm was achieved by spin-coating 30 SU8 2025 (MicroChem, USA) at 75 µm onto a 3-inch silicon wafer (Compart Technology Ltd, UK), according to the protocol supplied by MicroChem, including preheat and soft bake steps, twice. After UV light exposure in an MJB4 mask aligner (Suss MicroTec, Germany), the master was hard baked for 30 min and 35 developed in 1-methoxy-2-propyl acetate (Sigma-Aldrich). A commercially available PDMS kit (Sylgard 184, Dow Corning Co.) was poured on top of the master, at a mixing ratio of 10:1 w/w pre-polymer to cross-linker.…”

“…The monodispersity of the droplets generated in the microfluidic device ensured that the ensemble correlation 11,28,29 of 30 -50 image pairs was sufficient to minimise outliers. The normalised 35 median test 30 was applied to identify outliers, which frequently appeared at the edges of the data field; the invalid vectors were replaced by the median of the surrounding vectors. Finally, the vector field was smoothed with a 3 3 median filter.…”

“…It should be noted that according to Guillot et al, 35 the interface shear viscosity term can be neglected when considering the shear stresses at the liquid-liquid interface, as the interface shear 25 viscosity tends to be substantially smaller than the fluid viscosity in the presence of surfactant monolayers at the w/o interface. It is clear that the boundary conditions dictate the droplet flow topology to a large extent and the viscosity ratio λ = (∂u/∂y) c /(∂u/∂y) d plays a key role.…”

“…On the other hand, high-throughput cell manipulation requires encapsulation at high speed, which might generate a high shear environment inside the droplets and could potentially be harmful to the viability of certain cells. 8,9 In recent years, the dynamics of microdroplet flows have been 35 quantified using µPIV, which is an established optical method to characterise microscale flows by tracking the motion of particles seeded in the flow. 10, 11 A number of such studies on the droplet flow topology of liquid-liquid systems can be found in the literature.…”