Predicting flows through microfluidic circuits with fluid walls

Cyril Deroy,Nicholas Stovall-Kurtz,Federico Nebuloni,Peter R Cook,Edmond J Walsh,Cristian Soitu

doi:10.1038/s41378-021-00322-6

Cyril Deroy, Nicholas Stovall-Kurtz + Show 4 more

Open Access

https://doi.org/10.1038/s41378-021-00322-6

Copy DOI

Abstract

The aqueous phase in traditional microfluidics is usually confined by solid walls; flows through such systems are often predicted accurately. As solid walls limit access, open systems are being developed in which the aqueous phase is partly bounded by fluid walls (interfaces with air or immiscible liquids). Such fluid walls morph during flow due to pressure gradients, so predicting flow fields remains challenging. We recently developed a version of open microfluidics suitable for live-cell biology in which the aqueous phase is confined by an interface with an immiscible and bioinert fluorocarbon (FC40). Here, we find that common medium additives (fetal bovine serum, serum replacement) induce elastic no-slip boundaries at this interface and develop a semi-analytical model to predict flow fields. We experimentally validate the model’s accuracy for single conduits and fractal vascular trees and demonstrate how flow fields and shear stresses can be controlled to suit individual applications in cell biology.

Highlights

Flow is important in many biomedical applications, as cell survival and behavior depend critically on it.[1]
A thin layer of cell-growth medium (i.e., DMEM) plus 10% fetal bovine serum (FBS) in a virgin Petri dish is overlaid with FC40, and the tip of a needle held by a 3-way traverse (a ‘printer’) is lowered through the FC40 until it is located just above the medium (Fig. 1a)
We develop a method to measure height (Fig. 2) and find that the addition of FBS or serum replacement (SR) to the medium creates a solid medium:FC40 interface that induces no-slip boundary conditions; this defines the boundary conditions for our model (Fig. 2)

Summary

Introduction

Flow is important in many biomedical applications, as cell survival and behavior depend critically on it.[1]. It is essential in organ-on-a-chip devices[2–4] where cells are perfused continuously to mimic in vivo conditions and when studying the effects of transient shear stress on cells.[5]. Equations based on pipe flow can be adapted for conduits with arbitrary cross-sections[9] and to model vascular circuits with branches of varying widths.[10–13]. As solid walls restrict access, open microfluidics are being developed where parts of walls are replaced by interfaces with air or immiscible liquids[14–16] and equations describing rivulet[17] multiphase[18] and droplet-based flows[19] through conduits with free surfaces have been described The aqueous phase in conventional microfluidic devices is typically surrounded by solid walls (made, for example, of polydimethylsiloxane, PDMS), and flows through them can usually be predicted.[6–8] For example, equations based on pipe flow can be adapted for conduits with arbitrary cross-sections[9] and to model vascular circuits with branches of varying widths.[10–13] As solid walls restrict access, open microfluidics are being developed where parts of walls are replaced by interfaces with air or immiscible liquids[14–16] and equations describing rivulet[17] multiphase[18] and droplet-based flows[19] through conduits with free surfaces have been described

Methods

Results

Conclusion