James Wimshurst was a 19th-century engineer and inventor who devised a fascinating machine for generating high voltages. The machine that now bears his name consists of two counter-rotating discs fitted with metal patches.
The discs are in contact with two metallic brushes connected to a pair of metal spheres separated by a small gap. The brushes pick up charge from the discs, which builds on the spheres so that, eventually, a spark jumps across the gap between them. The mechanism that charges the spheres is electrical induction, which amplifies any small charge on the metal patches.
In the 19th and early 20th centuries, physicists and engineers used Wimshurst machines and similar devices to power x-ray machines and even particle accelerators. But today, these machines are rare, used only in science museums and classrooms to demonstrate the principles of electrostatics.
Today that could change thanks to the work of Maria Napoli and colleagues at the University of California, Santa Barbara, who have reinvented the Wimshurst machine for the 21st century. These folks have created a microfluidic version that can harvest energy from the environment and turn it into usable power.
In the new device, droplets of mercury in oil flow through a channel carved into a sheet of PDMS (polydimethylsiloxane) plastic. The channel carries the mercury droplets past each other in opposite directions, just like the counter-rotating discs in a conventional Wimshurst machine.
Electrodes embedded in the microfluidic channel carry away charge as it builds up. But instead of creating sparks, this charge can be used as power. Napoli and co calculate that a centimeter-scale circuit with channels just 300 micrometers wide, with mercury drops flowing at a rate of 10 millimeters per second, could generate about 12 microwatts of power.
The team has built a proof-of-principle device to test the idea. The microfluidic Wimshurst machine consists of a main channel just five centimeters long, carrying a few cubic millimeters of mercury. It generates a small fraction of the theoretical maximum power—just four nanowatts.
But this doesn’t faze the team. Microfluidics allow a range of improvements that aren’t possible with solid-state devices, such as changing the width and separation of the channels and better controlling the droplet size and distribution.
“Calculations indicate that straightforward improvements to the geometry should be capable of increasing the output power of a single-channel device by up to three orders of magnitude,” they say.
What’s more, several channels can easily be run in series or parallel to generate even more power. And a big advantage of this kind of microfluidic device over other energy harvesters is that it does not have to run at a resonant frequency.
Napoli and co study the potential output of this kind of device driven by a diaphragm pump in the heel of a boot. Assuming a person can walk at the rate of one step per second, a pump two centimeters in diameter could provide enough flow for 250 parallel microfluidic channels, which together would produce an output of about 10 milliwatts, say Napoli and co.
That’s about enough to power the laser in a DVD drive, and a promising amount for various low-energy communication devices and sensors currently under development.
“There is thus good reason to expect that a scaled-up version of our prototype device could be portable, practical, and sufficiently powerful for a variety of energy harvesting uses,” they conclude.
Of course, there are challenges ahead. An important question hangs over the durability of such a device given the pounding that boots go through during their lifetime. But that is a development question the team can work on.
Perhaps it won’t be long before the best way to recharge a phone will be to pull on a pair of pumps and go for run. Wimshurst would surely be amazed.
Ref: arxiv.org/abs/1803.02454 : Energy Harvesting with a Liquid-Metal Microfluidic Influence Machine