Reliable operation of lab-on-a-chip systems requires precise and predictable management of multiple sample fluids. In our work, we use a Kirchhoff-based network simulation that solves the entire microfluidic system for inline prediction of required pressure settings for desired flow rates in less than 200 ms. Using image-based evaluation, the generated multicomponent flow patterns were combined with the predictions of the solver.

Our innovative approach allows us to predict pressure conditions for laminar uniform flow and multicomponent droplets and help support applications such as synthesis and micro reactions where droplet size and content are important. Unlike conventional CFD simulations, we can analyze an entire network within a few milliseconds.

We use the analogy between microfluidic networks and electrical circuits. This means that we can represent the microfluidic structures by nodes that separate and join fluids, and their connecting lines. As in Kirchhoff’s node analysis, the microfluidic networks are represented by an incidence matrix. To the desired flow rates for operation, the initial operating pressures are determined by our solution approach. The microfluidic network can be solved as a system of linear equations, which is a suitable representation for systems of low Reynolds numbers.

The generic solver can be adapted to any microfluidic system, since the solver is based only on the input of the manufacturing parameters and the desired flow rates. As a result, the operating parameters for laminar co-flow of multiple components or droplets of different sizes and compositions can be predicted.

This project has received funding from the European Union’s Framework Programme for Research and Innovation Horizon 2020 under the Marie Sklodowska-Curie Grant Agreement No. 860775.

In the image above:
Graph of a microfluidic network consisting of two input-output (IO) ports, one reference port (Ref), three nodes and six branching edges (lins). Example solution of the solver for a microfluidic network operated with two components to generate a laminar flow.