Brain power. A new implantable fuel cell could allow for a new wave of bionic implants.

B. I. Rapoport et al., PLoS ONE, 7 (12 June)

Fuel Cell Runs on Brain Power

For centuries, humans have used the power of their brains to solve complex mysteries, create radical new inventions, and devise wondrous works of art. But now, scientists have developed a technology that enables us to use our brains to actually harvest electrical power.

Researchers at the Massachusetts Institute of Technology have designed an implantable glucose fuel cell that can generate electricity from the cerebrospinal fluid around the brain. The results of these efforts, published this week in PLoS ONE, show that a few hundred microwatts of power could be harvested from glucose within the cerebrospinal fluid with no adverse physiological effects.

A few hundred microwatts is certainly not enough energy to power, say, a pair of Google glasses, or, for fans of The Matrix, an entire alternate reality. But it is enough to fuel futuristic, highly efficient brain implants that could allow paralyzed patients to regain the ability to move their arms and legs.

"We envision [the fuel cells] powering brain-machine interfaces for paralysis in the medium-term, or in the longer-term, those for blindness or deep-brain disorders," says team leader and computer scientist Rahul Sarpeshkar.

Currently, most experimental brain-machine interface devices are powered through inductive power transfer, in which power is transferred wirelessly, or single-use batteries that must be surgically replaced after several years. The new, micrometer-scale fuel cell is primarily composed of platinum, a material with a good track record of being safe in the body. Sarpeshkar posits that the new fuel cell could work for decades without being replaced, though he emphasizes that animal and human tests will be necessary to prove his contention.

The miniature fuel cell, shaped very similarly to a computer chip, was tested in a saline solution simulating cerebrospinal fluid, at sizes of 1 square millimeter and 2 square millimeters. It functions by oxidizing glucose from the brain's cerebrospinal fluid at the surface of an activated platinum anode and converting oxygen to water at the surface of a network of single-walled carbon nanotubes embedded at the cathode end of the cell. Electrons are stripped from the glucose, and the fuel cell uses them to generate electricity.

To the best knowledge of the researchers, this was the first time that cerebrospinal fluid was used as a medium for an implantable fuel cell. They viewed it as a promising niche environment because it is under minimal immune system surveillance, the fluid is practically devoid of cells, and glucose levels are comparable to that of blood. The fuel cell also utilized an oxygen gradient to prevent oxygen from reaching the anode and thus possibly creating electrochemical short circuits.

To gauge the fuel cell's safety, the researchers first evaluated whether brain glucose levels would be adversely affected. Using data from their earlier testing and available knowledge about the replenishment rate of glucose in the cerebrospinal fluid, they determined that their fuel cell would consume glucose at only 2.8% to 28% of the rate at which cerebrospinal fluid glucose is replenished, not nearly enough to cause adverse effects. The researcher s also analyzed the fuel cell's oxygen consumption to ensure that it wouldn't destabilize brain oxygen levels. The calculations revealed that it would not disrupt oxygen equilibrium within the cerebrospinal fluid.

Adam Heller, a chemical engineer at the University of Texas, Austin, says the device looks promising but that it still needs to be tested in an animal model. Sarpeshkar's team is currently planning for animal and human testing in the coming years and already has several candidate regions of the brain and spinal cord targeted for possible implantation.