Kinematic characterization of the advancing liquid–gas interface in microfluidic components combining digital processing of microscope images with curve matching technique

Abstract

A method based on flow visualization to measure the advancing planar liquid–gas interface velocities of unsteady two-phase microflows is presented. A high-frequency CCD camera connected to a microscope is used to acquire flow images, which are digitally processed to define mathematically the interface locations at each frame of the acquired images with accuracy of ±½ pixel. Curve matching technique including a function to define the optimal matching path is applied to the parameterized interface locations previously obtained to determine the advancing interface velocities. The continuity equation is included in the algorithm to evaluate the velocity profile that minimizes the mass imbalance. The method is appropriate for open curves typical of moving interfaces inside confined microdevices and is applied to the filling process flows at constant rate, imposed by a syringe pump, in two constant height microchannels with different sizes for validation purposes. The results yielded by the methodology prove that it is accurate in determining planar flows with no vertical velocity component with mass balance errors below 0.07%. When a relevant vertical velocity component is present, the technique is hardly applicable to such flows since it is unable to capture the interface movement in that direction, yielding errors of mass balance that can go up to 41.6% in the case of a triangular microvalve with varying height.