Characterized dry electrodes for EMG using EIS and human studies; first-author IEEE EMBC 2024 paper.
This project grew out of our curiosity about finding cheaper, more accessible materials for dry electrophysiology electrodes—especially for EMG applications, where comfort and usability are key. Most dry electrodes today rely on silver/silver-chloride (Ag/AgCl), which performs well but can be expensive or difficult to fabricate at scale.
Zinc might seem like an unexpected choice, but it actually came from an earlier idea we explored at Triton NeuroTech, where we used pennies as low-cost electrodes in a few different projects (see Drone-EMG interface and EMG External Prosthetic Arm) . That experience planted the seed—if copper-zinc coins that were sanded down could work at a basic level, what could we do with purer zinc?
Zinc is cheap, widely available, and notably less soluble than silver, which could make it a good candidate for stable, gel-free skin contact. We hadn’t seen it widely adopted in professional settings, which made us curious: was there a good reason it hadn’t been used, or was it just unexplored? That question drove us to dig deeper and start testing.
We made two types of electrodes from zinc:
The disc versions were literally made by sanding down post-1982 pennies, while the CREs were CNC machined from zinc sheets. For comparison, we also made stainless steel versions of both (both versions CNCd). We coated the backs with hot glue(hot melt adhesive) to control the contact area and added connectors so we could test them easily.
After this paper was published, we also ended up developing smaller zinc electrodes (since area is a big consideration) and machining them with a CNC (to make the dimensions more consistent)
Smaller zinc electrodes. Simply CNCd, soldered, and crimped to fit to dupont connectors
We did two major tests:
EIS (Electrochemical Impedance Spectroscopy)
This measured how well the electrodes conducted over different frequencies. Lower impedance = better. Zinc consistently beat out stainless steel and even outperformed Ag/AgCl at certain frequencies (especially above 2 Hz).
EMG (Electromyography) recording
We stuck the electrodes to a volunteer’s arm and had him do movements like clenching or lifting. We compared signal quality—mainly looking at SNR (signal-to-noise ratio)—between the different metals and shapes. The zinc discs performed surprisingly well, hitting around 13 dB SNR. Stainless steel barely registered a signal.
We really hoped the tripolar CREs would be a game-changer, but... they weren’t. The SNR was about the same (and a bit more variable), and we didn’t see the spatial resolution benefit we expected. Maybe it’s because dry contact isn’t enough to see the difference—or maybe we just need better placement or signal processing. Still, it was worth trying.
We see this as a step toward making low-cost, DIY-friendly biosensors more viable. Whether for education, prototyping, or even accessible clinical tech, zinc might just be a dark horse worth betting on.
We’re excited to keep testing and refining—and we’d love to see others explore this too. If you're curious, you can read the full IEEE EMBC paper here or reach out!