Efficient Tesla valves for microfluidic applications

This research summary written by Sebastian Bohm describes highly efficient passive Tesla valves designed and characterized using high-precision pressure sensors for microfluidic applications.

The summary is based on the article “Highly efficient passive Tesla valves for microfluidic applications” by S. Bohm, H. B. Phi, A. Moriyama, E. Runge, S. Strehle, J. König, C. Cierpka and L. Dittrich published in the npg Microsystems & Nanoengineering journal in September 2022.

Introduction & Methods

Passive non-mechanical valves achieve a flow rectification effect completely without moving components and are, therefore, easy to manufacture, miniaturize, and integrate. Thus, they are attractive candidates for use in microfluidic systems.

The operating principle is based on an asymmetry in the flow profiles for the two flow directions, resulting in different flow resistances. In honor of the inventor, these valves are also called Tesla valves [1]. The efficiency of the valves can be characterized by the diodicity, which is defined as the ratio of the pressure differences for the backward and forward flow directions at a constant flow rate.

If Tesla valves are combined with a suitable pump mechanism, micropumps can be manufactured that function completely without moving components [2,3]. One challenge is that the diodicity depends strongly on the flow rate and is usually low for small flow rates. However, using modern sophisticated optimization methods, efficient valve geometries can be found that largely compensate for these disadvantages.

A particularly powerful optimization tool for Tesla valve design is topological optimization. In this process, the material distribution within a given design area is varied until a specified objective function becomes extremal. The result of the optimization, and thus the efficiency of the resulting valve geometries, is influenced by a variety of parameters, such as the shape of the design domain, the objective function used, and the execution of the optimization procedure itself.

In order to obtain particularly efficient valve geometries, the conventional optimization procedure was improved in several aspects: (i) The optimization was carried out in several stages, and the objective function was adapted so that (ii) the valves only show large and easy to manufacture structures, but that at the same time (iii) a small flow resistance in the forward direction results. Fig. 1 shows the material distribution after different optimization stages. The final resulting material distribution can be directly transferred to lithographic masks and the valves can be fabricated using conventional MEMS manufacturing techniques.

Aims

The primary objectives of the project were:

• Introduction and validation of a new optimization algorithm for designing efficient Tesla valves for small flow rates.
• Manufacturing and fluidic characterization of the designed Tesla valves.

Experiment setup

For the measurement of the diodicity, a constant flow rate is generated with a syringe pump, and the pressure difference is measured with two MPS 0 pressure sensors. Using the measured pressure differences, the diodicity can be determined directly.

The flow rate was increased in steps of 1 µl/s from 1 µl/s to 20 µl/s every 7 seconds. To consider only constant pressure differences, the values during the first and last second of each measurement interval were ignored, and the measured values of the remaining 5 seconds were averaged. The experimental setup is shown schematically in Fig. 2. With this simple setup, efficient and precise measurement of the diodicity of the valves is possible.

Materials

• Test chip with multiple different valve geometries
• Syringe pump
• Two Elveflow MPS 0 microfluidic pressure sensors
• Evaluation of electronics and PC for measurement data management

Key Findings and Conclusions 

A new multi-stage optimization algorithm for the design of efficient Tesla valves was introduced. The valves dimensioned with these features show a significantly increased diodicity, i.e., rectification effect, at low flow rates, as shown in Fig. 3, and are thus ideal valves for use in microfluidic systems with typically rather small Reynold numbers. The experimental investigations confirm the predictions of the simulations.

At small flow rates, the pressure differences are typically very small and thus hard to measure with high precision. However, with the help of the highly precise MPS0 pressure sensors, an accurate but also easy-to-implement measurement of the diodicity was possible even at low flow rates.

References