Membraneless Fuel Cell Is Tiny, Versatile

CHAMPAIGN, Ill. — A fuel cell designed by researchers at the University of Illinois at Urbana-Champaign can operate without a solid membrane separating fuel and oxidant, and functions with alkaline chemistry in addition to the more common acidic chemistry.

“The system uses a Y-shaped microfluidic channel in which two liquid streams containing fuel and oxidant merge and flow between catalyst-covered electrodes without mixing,” said Paul Kenis, a professor of chemical and biomolecular engineering and a researcher at the Beckman Institute for Advanced Science and Technology.

Fluids flowing through channels of microscale dimensions behave differently than fluids flowing through the much larger pipes found in home plumbing systems, Kenis said. “At the microscale, there is no turbulence. This laminar flow means streams of fuel and oxidant can pass side by side without having a physical barrier in between.”

This configuration offers several advantages over PEM-based fuel cells, including fewer parts and simpler design. It also means that membraneless fuel cells are compatible with alkaline chemistry.

http://www.sciencedaily.com/releases/2005/03/050325160631.htm

Kenis, a prof of chemical and biomolecular engineering is using "real kidney and liver physiology" -- in the past the older generation of fuel cell membranes used "artificial kidney" physiology. (The first generation of separators were derived from artificial kidney dialysis separators).

Angstrom level chemistry versus micron level chemistry.

It's pretty neat to design a system where two steams of fluid are touching, and yet can be relied on to not mix.

that's the way some biological systems work. (When it gets to intra-cellular transport I am really on shaky ground; I just completed several credits of biochem in John Hennessey's program to retrain "old fart" engineers in life science engineering -- but we haven't hit modern physiology yet).

Cool. There's some interesting stuff going on at the interfaces of biology and computer science.

He took over from Condi Rice as Provost at Stanford -- just as the tech bubble burst -- and had some ideas about the medicine-engineering interface.

Or, if you prefer, it's a flow regime of large Knudsen number.

Turbulent mixing on such scales of length & velocity won't happen.

Are you saying it just doesn't have time to develop turbulence?

I was just sort of teasing ;-)

Fluid dynamics defines several quantities that are used to describe the type of fluid flow regime. The Knudsen number (http://en.wikipedia.org/wiki/Knudsen_number) relates the 'mean free path', i.e. the distance a particle will travel before bumping into anouther, to the representative length scale, or size, of the region where the fluid is flowing.

The Reynolds number (http://en.wikipedia.org/wiki/Reynolds_number) is velocity times length divided by viscosity. This characterizes the fluid flow independent of the fluids. Example: an airplane model immersed in water can exhibit the identical behavior as the real airplane in air, provided you pick your water velocity and model size to give the same Reynolds number as for the real airplane in flight.

At very low R (under say 1500), there's no turbulence, and little mixing goes on, aside from diffusion. So, in microfluids you have very small length scale (a channel maybe a few hundreds of nanometers across) and very low velocity, so R is very small. So it's not that there's no time for turbulence, it's just that it won't happen.

Another way to characterize such flow is via the Knudsen number. Low Knudsen number indicates a traditional fluid. High Knudsen number indicates the flow of free molecules in which normal fluid-like behavior is rare, since it'll act like a stream of BB's instead of a fluid. Microfluids (especially flows of gases through microscopic channels) often exhibit highish Knudsen numbers, so they don't act like normal fluids do.

Let me know if this helped, I can try again but I don't know your background, and can try to be more or less technical.