Two gentlemen at the Institut d’Electronique, de Microélectronique et de Nanotechnologie (IEMN) at the University of Lille in France, have demonstrating an interesting proof of concept of a new mechanical switching mechanism that relies on the interplay between mechanical and capillary forces as well as how “wet” a drop can get.

The concept is based upon the fact that the wetting properties of a droplet – that is, how easily it spreads out on a surface – can be changed capacitively. Placing a drop on a thin insulator with a semiconductor region below it, and touching the drop above with a conductor, makes the droplet part of a capacitor. The droplet becomes one plate, the semiconductor – p-type in their experiment – is the other plate. Modulating the voltage to which the p-region is biased changes the wetting properties of the droplet, making it sink down or rise up as the bias is modulated. This is “electrowetting.”

Now, by shining plain-old white light on the semiconductor – nothing fancy, no “lasers” – you further modulate the charge in the semiconductor, further affecting the wetting. Presumably this only works with a transparent insulating layer. This is referred to as “photoelectrowetting.” Gesundheit.

Here’s how you make a switch, then. You take a MEMS cantilever and place the droplet right under the tip (or close enough to have some leverage). The cantilever is a spring and wants to stay straight. But once in contact with the droplet, the capillary force makes it “stick” to the surface of the drop. So when the droplet rises or falls, then the cantilever rises and falls with it – as long as the restoring spring force doesn’t overcome the capillary force and break the connection.

Putting it all together, the idea is that, in principle, given this setup, you should be able to use the cantilever as a relay (some portion of it being able to contact another conductor), using white light to activate the switch. The good news is that, because the switching mechanism is capacitive, it makes an extremely low-power switching mechanism (as they point out, it’s like the gate of an MOS transistor, qualitatively).

You can find their complete paper here.