Luigi Catacuzzeno
University of Perugia

Energy harvesting from electrically polarized biological cells: a theoretical study

Microenergy harvesting has lately been attracting special attention in the field of human implantable biomedical microdevices. Currently, these devices are powered by batteries, with the concern of limited lifetime, side effects, large size. Investigation is therefore presently strong in developing methods to harvest energy for implantable devices directly from the ambient environment. Energies that are today more commonly considered include thermal energy (human heat), kinetic energy (body motion), chemical reaction energy. Here we propose scavenging energy from the electric field (potential difference) existing across excitable cell membranes, and present a theoretical study to predict the amount of energy that can be scavenged from this source.
By modeling a biological cell membrane as an RC electrical circuit, we simulated the power that can be generated by a model cell incorporating the ion channels and transporters typical of a hippocampal neuron.
The model cell included Na+, K+ and Cl--selective ion channels as the pathways for passive ion fluxes, and the Na+/K+ pump that actively (using high-energy ATP) moves Na+ and K+ ions in opposite directions across the membrane against their electrochemical potential, to maintain the transmembrane ionic gradients. In order to assess the feasibility of using a biological cell to provide electrical power, we connected the intracellular side of the cell to a variable resistance, and calculated the power produced as the product of the current flowing through the resistance, and the potential difference maintained by the cell at steady-state. Our model indicates that the cell absorbs a chemical power of about 50 pW to fuel all the Na+/K+ ATPases, while only 1 pW is made available as current passing through the implanted resistance, resulting in a chemical to electrical energy conversion efficiency of about 2%. Notably, we found that the power produced by the model cell can be sensibly increased using human skeletal muscle cells which have a much higher number of ion channels and transporters and larger cell dimensions than most neurons, and are optimized to store and use chemical energy needed for movement.
Our simulations indicate that the Na+/K+ ATPases contained in a single skeletal muscle cell absorb a power of about 70 nW for their activity, and can give back to a connected device a power of about 5 nW, with a conversion efficiency of about 7%.
In conclusion, our simulations suggest that skeletal muscle cells can generate an amount of power sufficient to fuel currently available integrated microcircuits, and may thus represent a possible source of energy for future human implantable microsystems.