H21A-1351
Using Microfluidics for Visualisation of Displacement Mechanisms on Pore Network Models

Tuesday, 15 December 2015
Poster Hall (Moscone South)
Alireza Gerami Kaviri Nejad1, Peyman Mostaghimi2, Ryan T Armstrong2, Mehdi Rafeie2 and Majid Ebrahimi Warkiani2, (1)University of New South Wales, Sydney, NSW, Australia, (2)University of New South Wales, Sydney, Australia
Abstract:
We use microfluidic methods for studying displacement mechanisms of immiscible fluids including drainage and imbibition in porous media at the pore scale. We use soft lithography method to make 4’ diameter silicon wafer to be used as a mould of our designs. We have fabricated a range of microfluidic chips based on a range of patterns including pore junction with unequal throats, junction of throats with different coordination numbers, and junction of tortuous throats. We also fabricate more complex networks as a combination of the mentioned simple patterns. Decane and water are used as the wetting and non-wetting phases, respectively. Using high-resolution microscopy, we visualise the displacement processes and the movement of the interface between two fluids at different saturations. We initially test to micromodel chip for modelling drainage into a pore junction and compare our results with the prediction by the Young-Laplace equations. Then we focus on the sequence of pore filling and effects of pore space geometry, tortuosity and the injection rate. We use plasma treating to vary the contact angle and then study the effects of wettability on interface movement. Using accurate pump and pressure controller, we measure pressure drop across the micromodels at different time. By image processing of fluids distribution in the microfluidic chip, saturation of both phases can be estimated. Then, we plot relative permeability versus saturation curves for different pore space geometries. Our results can be used for validation of numerical two phase flow simulation and also we provide novel suggestions for modifying the equations of motion in pore network models.