What are the basics behind the design?

The device works by harnessing a pressure difference to generate electricity and consists of two fluid reservoirs separated by thin nanoporous membrane. When a pressure difference between the reservoirs drives water through the membrane, electric charge is carried with it. This well known effect is called the streaming current, and arises because the fluid flow transports charged counterions that accumulate near the charged inner surfaces of the membrane. Electrodes placed in either reservoir can simply collect the electricity. By using carbon nanotubes as the pores in the nanoporous membrane, we believe that hydrodynamic slip will enhance the electrical output and the efficiency.

Why is hydrodynamic slip so important?

Fluid that slips along the surface of a channel circumvents friction and dissipation. This should allow the efficiency of the energy conversion process to rise above 50%, 90% or even 95%, depending on the amount of slip. Without slip the efficiency will never be better than about 10%. In order to achieve slip, the inner surfaces of the fluidic device should be extremely smooth and hydrophobic. There are a few materials that satisfy these requirements. I think that the inner surface of a carbon nanotube is perhaps the best candidate.

In your recent work, you mention power densities of 0.72 and 1.2 Wm–2. What kind of applications would these values enable?
Those power densities are what we estimate would come out of two specific nanotube filters that were recently reported in the literature (assuming an applied a pressure of 1 atm). The values would likely be sufficient for some low-power electronic applications, like an mp3 player.

However, those filters had rather low densities of tubes running through them. Since the output power scales with the number of tubes, we would of course want to maximize that number in a practical device. If the density of tubes were increased to, say, a few percent of the membrane area, which does not seem unreasonable, then the output power density could rise above the level of several kW/m2. This would be roughly comparable to the output power density of a hydroelectric dam.

But I can imagine applications that are smaller in scale. A nanoporous membrane can be configured to harvest energy from almost any situation where useful mechanical work would otherwise be dissipated away, like in the shocks of a car or even the soles of shoes.

What are your next steps in exploring the technology?
Our top priority is to test the influence of hydrodynamic slip on streaming currents. The fundamental science must be understood before we can pursue device applications. To do this, I am teaming up with an engineering colleague of mine at Brown named Kenneth Breuer who is an expert in fluid slip. We plan to build some graphitic channels to test, but I am also teaming up with Jing Kong (at MIT) who is an expert on carbon nanotube growth.

The researchers presented their work in Nanotechnology.