Monthly Ultrasonic Charging for Decade-Life Implants
OverviewAnalysisSolutions
Complete
·Feb 2, 2026The Core Insight
Monthly charging collapses the storage requirement from impossible to trivial
- Reframe from '10 years of storage' to '30 days of storage with monthly replenishment.' At 50 μW average: 30 days × 24 hours × 50 μW = 36 mWh.
- At 1000 Wh/L, that's 0.036 cm³—not 4.4 cm³.
- The 'impossible' density requirement vanishes when you stop over-engineering for full autonomy that the problem statement doesn't require.
Viability
Solvable
- The physics works; components exist; this is systems integration and qualification, not fundamental research.
Key Decision
If monthly charging is acceptable and you need deployment within 3 years, pursue ultrasonic. If you're building next-generation architecture and have 5+ year horizon, the biofuel cell paradigm could obsolete storage-based approaches entirely.
Solution Paths
01NEEDS DEVELOPMENT
Integrated Ultrasonic Power System
0.31 cm³ system with proven components; blocked by integration/qualification, not physics
02NEEDS VALIDATION
CGM-Derived Biofuel Cell
Fastest path to 'body as battery'; blocked by interstitial O₂ uncertainty
Recommendation
- If this were my project, I'd place two parallel bets with very different risk profiles.
- First, I'd pursue the ultrasonic path aggressively.
- Monday morning, I'd email Astellas business development about licensing the iota ultrasonic platform.
- While waiting for that response (probably slow), I'd order PZT-5H samples and Cymbet thin-film batteries to start bench integration.
- The physics is proven; this is just engineering execution.
- Within 6 months, I'd have a benchtop demonstration that validates the volume budget.
- That gives me negotiating leverage with Astellas and confidence to commit $2-5M for custom ASIC development.
- Second, I'd spend $200K to answer the O₂ question for biofuel cells.
- This is the cheapest, fastest way to validate or kill the paradigm-shifting approach.
- I'd fabricate Pt/Ir test electrodes, run them in a potentiostat at 20-40 μM O₂ and 5 mM glucose, and measure power output for a week.
- If I see >15 μW/cm², I'd call Dexcom R&D the next day to propose a partnership.
- If I see <5 μW/cm², I'd have my answer and focus entirely on ultrasonic.
- The mistake to avoid is over-engineering the first attempt.
- The problem statement allows monthly charging—I'd take that gift rather than heroically trying to achieve full autonomy on the first device.
- Ship the ultrasonic version in 3 years.
- If the biofuel cell validates, that becomes the next-generation architecture for 5 years from now.
- One thing I'd do that might surprise people: I'd call the MIT Schroeder group about the ionic gradient work.
- It's probably too early for product development, but understanding the reference compartment problem better might reveal a solution the 2017 paper didn't pursue.
- Academic collaborations are cheap and occasionally goldmines.