Implantable and wearable ocular devices require a compact continuous power source that does not interfere with vision or regular physical activity. An enzymatic biofuel cell fits this description provided it has sufficient power density and stability. Two proof-of-concept contact lens lactate biofuel cell prototypes were designed and fabricated to investigate making enzymatic bioelectrodes that are flexible, have high surface area and conductivity and that could be integrated into a contact lens. The first prototype utilized a bilirubin oxidase biocathode and a lactate dehydrogenase bioanode. When tested in conditions similar to those found on the eye, this prototype produced more power (3 µW at ~0.2 V) than the second prototype but it did not sustain appreciable electrical activity for more than 4 hours due to poor cofactor/mediator immobilization. The second prototype used lactate oxidase at the anode and the device was stable for at least 24 hours of continuous operation but had lower power output (0.5 µW at ~0.2 V) than the first prototype because of lower electrode surface area (but increased biocompatibility). Both prototypes were limited by the oxygen reduction reaction (ORR) at the air-breathing biocathode. Oxygen concentration and transport are commonly blamed for low cathodic current but it is less understood how carbon nanotube (CNT) networking and CNT surface activity might affect catalytic current. For the second prototype, Monte Carlo and numerical simulations revealed that only 20% of the CNTs were networked to the current collector and, within that fraction, only ~5% of the CNT surfaces contributed to current output. These results have implications for enzymatic biofuel cells in general through two conclusions: 1) enzymatic electrode films that are thicker or having more CNTs may not be beneficial and 2) there is much opportunity to improve enzyme-CNT electron transfer.

Figure 1. Contact lens biofuel cell prototype with leads attached to it (right). On the left is a schematic to show that an electrode reaction (in this case oxygen reduction at a cathode) may be limited by poor connectivity between carbon nanotube clusters and the electrode.