# How Chips Are Using On-Chip Biofuel Cells for Self-Powered Implants

An episode of [The Hardware Podcast with Fexingo: Chips, Devices, and Electronics Engineering Conversations](https://invisiblenetwork.ai/podcast/1896799982), 14 Jul 2026, 8:40.

Audio: https://audio.fexingo.com/business/the-hardware-podcast/episode-0110.mp3

Episode 110 of The Hardware Podcast explores on-chip biofuel cells that generate electricity from glucose in the human body. Lucas and Luna discuss how researchers at the University of California San Diego have developed a chip that powers itself with a patient's own blood sugar, eliminating the need for bulky batteries in implantable medical devices. They examine the energy density of glucose, the efficiency of enzymatic fuel cells, and the challenges of biocompatibility and long-term stability. The hosts also consider potential applications beyond medicine, including environmental sensors and wearable electronics. A concrete look at how the boundary between living tissue and silicon is getting thinner. #BiofuelCells #ImplantableDevices #MedicalChips #GlucosePower #UCSanDiego #EnzymaticFuelCells #SelfPoweredChips #Biocompatibility #WearableTech #EnergyHarvesting #MedicalElectronics #Technology #HardwarePodcast #FexingoBusiness #BusinessPodcast #OnChipEnergy #BiomedicalEngineering #WirelessHealth Keep every episode free: buymeacoffee.com/fexingo

## Transcript

Lucas: So there's this fundamental tension in implantable medical devices: you want them small enough to not disrupt the body, but they need power. And batteries are bulky, they degrade, and eventually you have to replace them surgically.
Luna: Right, and that second surgery isn't trivial. It's invasive, it's risky, it costs a lot. So what if the chip could just power itself from something already in the body?
Lucas: Exactly. And that's where on-chip biofuel cells come in. We're talking about a chip that generates electricity from glucose — your own blood sugar. Researchers at UC San Diego published a paper last month on a chip that's essentially a miniature fuel cell integrated directly into the silicon, using enzymes to oxidize glucose and produce current.
Luna: And they actually demonstrated it in a live animal model, right? I read they implanted it in a rat and it powered a sensor for continuous glucose monitoring for over a week.
Lucas: That's the part that got my attention. It's one thing to show a reaction in a petri dish. It's another to have a chip running inside a living creature for seven days straight, transmitting data wirelessly. The chip itself was about two square millimeters — smaller than a grain of rice.
Luna: That's remarkable. But I have to ask — is the power output actually useful? I mean, a glucose fuel cell isn't going to run a pacemaker, right?
Lucas: No, not yet. The UCSD chip generated about 44 microwatts per square centimeter. That's enough for a low-power sensor, a small microcontroller, maybe Bluetooth Low Energy transmission in bursts. But you're not driving a motor or a high-power processor. The key is matching the application to the power budget.
Luna: So what applications are realistic? Continuous glucose monitors for diabetics — that's the obvious one. But what else?
Lucas: Think about any chronic condition where you want long-term biomarker tracking. Cardiac enzymes, lactate, even drug levels in the blood. If the chip can sense and transmit, and it never needs a battery swap, you've got a fundamentally different class of implant. One that could stay in for years.
Luna: And if today's tech conversation gave you something usable — maybe a new way of thinking about energy or sensors — consider supporting the show. It's what keeps us ad-free and digging into these deep hardware stories. You can find us at buy me a coffee dot com slash fexingo.
Lucas: Yeah, listener support really does make this possible. And speaking of long-term implants — the longevity question is fascinating. The enzymes in these fuel cells degrade over time. So the chip's lifespan is currently limited to weeks or months, not years. The UCSD team is working on engineering more stable enzymes, and also looking at using synthetic catalysts that don't degrade.
Luna: So you could potentially have a chip that lasts as long as the patient needs it. That changes the risk-benefit calculus for a lot of conditions.
Lucas: Exactly. And it's not just medicine. There's interest in using these chips for environmental sensors in remote locations — think a sensor in a river that generates power from organic matter in the water. Or wearable patches that harvest glucose from sweat. The same principle applies.
Luna: So let's talk about the engineering details. How do you actually build a fuel cell on a chip? I imagine it's not just sticking some enzymes on a piece of silicon.
Lucas: No, it's a multi-layer stack. You have the anode side where glucose is oxidized, releasing electrons. That's typically a carbon electrode coated with the enzyme glucose oxidase. Then you have a cathode where oxygen is reduced, completing the circuit. Between them is a proton exchange membrane — just like in a hydrogen fuel cell, but miniaturized.
Luna: And all of that has to be biocompatible. The body doesn't react well to foreign materials, especially if they're generating electrical current.
Lucas: Right, that's a huge constraint. The chip has to be encapsulated in a way that prevents immune response but still allows glucose to reach the anode. The UCSD team used a nanoporous gold membrane that lets glucose in but keeps larger immune cells out. It's a clever solution.
Luna: What about the voltage? I assume you need a certain voltage to run the electronics, not just current.
Lucas: Good point. A single biofuel cell produces around 0.5 to 0.7 volts. That's enough for some low-voltage circuits, but most microcontrollers need 1.8 volts or more. So you either need to stack multiple cells in series — which is hard to do on a tiny chip — or use a voltage booster. The UCSD chip used a charge pump to step up the voltage to 1.8 volts.
Luna: So the efficiency of the overall system depends on both the fuel cell and the power management. What's the overall efficiency from glucose to usable electrical power?
Lucas: It's low. Theoretical maximum for glucose oxidation is about 50 percent, but in practice these enzymatic cells are around 5 to 10 percent efficient. That's partly because the enzymes don't transfer electrons perfectly to the electrode — a lot of the energy is lost as heat. Researchers are working on better electron mediators to improve that.
Luna: Still, if you have a continuous supply of glucose in the blood, even low efficiency can work. The body has about 5 grams of glucose in the blood at any time, and it's constantly replenished.
Lucas: That's the beauty of it. You're tapping into a fuel source that's always there. Compare that to a battery that slowly drains and then you're done. The biofuel cell is essentially a renewable power source inside the body.
Luna: But there's a potential downside — what happens if the chip consumes too much glucose? Could it cause hypoglycemia? Or starve nearby tissue?
Lucas: That's a legitimate concern, but the power levels are so tiny that the glucose consumption is negligible. We're talking about nanograms per minute. The body's glucose regulation system easily compensates. In the rat study, there was no measurable effect on blood sugar levels.
Luna: So safety seems manageable. What about long-term stability? Enzymes denature over time, especially at body temperature.
Lucas: That's the biggest engineering challenge right now. The UCSD chip maintained about 60 percent of its initial power output after seven days. That's not bad for a first demonstration, but for a device that's supposed to last years, you need much better stability. Some groups are exploring using synthetic catalysts like Prussian Blue or conducting polymers that don't degrade.
Luna: But synthetic catalysts are less selective — they might react with other molecules in the blood, not just glucose. That could reduce efficiency or produce unwanted byproducts.
Lucas: Exactly. There's a trade-off between stability and specificity. Enzymes are incredibly specific — they only react with glucose — but they're fragile. Synthetic catalysts are robust but can be messy. The ideal solution might be a hybrid: an enzyme layer for specificity, coated with a protective polymer that extends its life.
Luna: So where do you see this technology in, say, five years? Will we have commercial implants powered by glucose?
Lucas: I think the first commercial products will be external, not implantable. Think wearable patches for continuous glucose monitoring that don't need a battery. That's a lower regulatory bar. Then, once the reliability is proven, we'll see implants for specific applications like cardiac monitoring or neurological sensing. I'd say 2030 before you see a glucose-powered pacemaker trial.
Luna: That timeline feels realistic. The hardware is basically here — it's the biocompatibility and longevity that need work. And the regulatory pathway, of course.
Lucas: On-chip biofuel cells are one of those technologies that feel like science fiction until you see the actual prototype running inside a living animal. It's a reminder that the boundary between silicon and biology is getting thinner every year.

Shows are listed from Apple Podcasts' charts and directory, and Podcast Index where it is used. Every show and episode belongs to its publisher and plays from the publisher's own feed; this site hosts none of them.