Subcutaneous power supply by NIR-II light

The study addresses the challenge of powering implantable medical devices (IMDs), which currently rely on batteries with limited lifetimes requiring surgical replacement. Many IMDs demand power from microwatts to hundreds of milliwatts. Contactless power transfer (CPT) could enable in situ recharging through tissue without surgery, but existing CPT approaches—electromagnetic coupling (300 MHz–20 GHz) and visible-light photovoltaics—are limited by shallow penetration depth and/or low permissible exposure, yielding insufficient subcutaneous power. NIR-II light (1000–1350 nm) forms a biological transparency window with reduced tissue absorption and scattering, allowing deeper penetration (~≥20 mm) and higher safe exposure (up to ~1.0 W cm−2). However, photovoltaic approaches in this band are hampered by poor exciton separation. The authors propose leveraging NIR-II photothermal heating coupled to thermoelectric conversion to achieve safe, efficient subcutaneous power transfer sufficient for high-consumption IMDs.

Prior CPT efforts include wireless power transfer via electromagnetic waves and visible-light photovoltaic systems, which face strict exposure limits (~≤0.02 mW cm−2 for RF/microwave; ~≤0.2 mW cm−2 for visible) and shallow penetration, limiting practical output power at depth. NIR-II has been widely explored for biomedical imaging and photothermal therapy due to lower tissue absorption/scattering and minimal damage, with reports demonstrating deep-tissue photothermal effects and imaging. Photovoltaic conversion in NIR-II has seen limited success due to poor exciton separation. Comparative analyses and cited works indicate existing CPT demonstrations for subcutaneous use generally provide lower power at ~10 mm depth than required for many IMDs, motivating a new approach that exploits NIR-II photothermal absorption and thermoelectric conversion.

Concept and modeling: The device is a photo-thermal-electric (PTE) converter implanted beneath skin comprising: an upper layer (Fresnel lens and air gap with quartz separator), a photothermal (PT) layer deposited on a thermoelectric (TE) generator, a bottom layer (cooling fin plus phase change material, PCM), all enclosed in a biocompatible PMMA shell. A one-dimensional steady-state heat transfer model is developed to maximize TE output by balancing photothermal input, upward losses, and downward heat transfer: (1) PPT − Ploss = Ptrans + Pop, with Pop = [S(t)2 (t1−t2)2]/R(t). Upward loss combines convection and radiation, approximated as Ploss = A1 H1 (t1 − t0). Downward transfer is Pbottom = Ptrans = A2 H2 (t2 − t3). The temperature difference across the TE is t1−t2 = [PPT − H1 A1 (t3 − t0)] / [H1 A1 + H2 A2]. Photothermal input is PPT = ηPT τ Plight, where τ is tissue transmittance, ηPT the PT efficiency. The model and finite-element simulations (Ansys) guide reducing H1 A1 (upper losses), increasing H2 A2 (downward heat transfer), raising PPT, and keeping t3 low to maximize ΔT across the TE. PT layer design: A multilayer selective absorber on Ni substrate: 80 nm SiO2 (anti-reflection), 90 nm Ni/Al2O3 cermet (strong NIR-II absorption, low emittance), and Ni substrate. This yields high NIR-II absorption and low thermal radiation, achieving ~94% photothermal conversion efficiency. IR imaging confirms low emittance and uniform heating under NIR-II irradiation. Upper layer: A PMMA Fresnel lens focuses NIR-II to increase Plight; an air gap provides thermal insulation; a transparent quartz separator with anti-reflection coatings (>99% NIR-II transmission) suppresses convection and radiation across the gap, reducing Ploss. The air gap thickness d is optimized (d = 8.6 mm) balancing convection onset (dc) and optical focusing. With the separator, Δt0 at the lens surface remains <5 °C above 37 °C under 0.3 W cm−2 for 10 min, and <10% of NIR-II is attenuated by the upper layer while >70% of heat loss is reduced. Bottom layer: To enhance H2 A2 and keep t3 low, a copper cooling fin increases heat exchange area and is embedded in PCM for thermal buffering. Myristyl alcohol (PCM, phase transition near body temperature) is modified with 5 wt% oxidized carbon black to increase thermal conductivity (with some loss of specific heat). Finite-element analysis indicates modified PCM lowers t3 by ~6 °C versus pure PCM; adding fins further improves TE voltage stability and magnitude. An aluminum alloy base within the resin shell isolates PCM when melted. Fabrication: PT stack deposited by magnetron sputtering (Ni base), RF co-sputtering (Ni/Al2O3), and SiO2 deposition. Commercial TEG (TEG-28706), PMMA Fresnel lens, and copper fin are assembled into a 3D-printed resin shell with an aluminum alloy base. Modified PCM is prepared by dispersing acid-treated carbon black in myristyl alcohol at 50 °C with ultrasonication. Characterization: Photothermal efficiency measured via cyclic IR thermography under on–off NIR-II laser (30 W) irradiation (ambient 25 °C). Material thermophysical properties (specific heat by DSC; thermal conductivity by Hot Disk TPS2500). Optical spectra (UV–Vis–NIR, FTIR). Electrical performance (open-circuit voltage, internal resistance, short-circuit current) recorded by electrochemical workstation. Thermal safety assessed by IR imaging and H&E histology. Electronics: A step-up energy harvesting circuit (key chip BQ25504) boosts Bio-PS output to ~5.3 V with <20% conversion loss, enabling charging of small batteries/supercapacitors and powering loads requiring higher voltage. In vitro/in vivo tests: In vitro performance measured with tissues placed over the device at 37 °C (3.5 mm pigskin, ~8.5 mm rabbit abdomen tissue, 20 mm pork). In vivo, Bio-PS implanted in rabbit abdominal cavity; NIR-II irradiation at LPD up to 1.0 W cm−2 for ≤10 min per ANSI Z136.1-2007; outputs recorded; safety assessed via skin temperature monitoring and histology; functional demonstrations include powering a pacemaker and charging a wireless endoscopic camera.

• NIR-II window and safety: NIR-II enables deeper tissue penetration (≥20 mm) and higher permissible exposure (up to ~1.0 W cm−2) versus visible and RF, with theoretical endurance energy ~60× higher than visible or RF for the same tissue.
• Photothermal layer: PT layer exhibits ~94% photothermal conversion efficiency with strong NIR-II absorption and low thermal emittance.
• Thermal management: Upper layer reduces heat loss by >70% while transmitting ≥90% of NIR-II; Δt0 at the outer surface remains <5 °C above baseline under 0.3 W cm−2 for 10 min. Bottom layer (modified PCM + fin) lowers t3 and enhances TE voltage stability.
• Electrical performance: With upper layer only, open-circuit voltage under 0.3 W cm−2 is ~1.12 V; with both upper and bottom layers (Bio-PS), Voc increases to ~2.21 V under the same conditions.
• Output power (in vitro, tissue-covered): Bio-PS outputs ~195 mW through 3.5 mm pigskin, ~28 mW through ~8.5 mm rabbit abdominal tissue, and ~6 mW through 20 mm pork. Packaging improves energy output efficiency nearly 20-fold over a bare PTE converter.
• In vivo rabbit demonstration: With Bio-PS implanted in the abdominal cavity under NIR-II irradiation through 8.5 mm skin, device outputs ~20 mW. Skin temperature under 1.0 W cm−2 rises to ~45 °C after 200 s; H&E staining shows no evident tissue damage or inflammation; animals remained healthy over 1 month.
• Functional demonstrations: Directly powered 648 LEDs through 3.5 mm pigskin under 1.0 W cm−2 for 10 s. Via step-up circuit, recharged a battery powering a wireless endoscopic camera; after 2 h charging, the camera recorded and transmitted a ~5-minute video via Bluetooth. A battery-free pacemaker connected to Bio-PS altered the rabbit ECG upon irradiation, indicating successful powering.
• Comparative ranking: Bio-PS delivers among the highest reported subcutaneous CPT power at ~10 mm depth, exceeding 10 mW and reaching >200 mW in shallow tissue scenarios.

The proposed NIR-II-driven PTE approach overcomes limitations of prior CPT methods by leveraging the biological transparency window to deliver higher safe optical power through tissue and converting it efficiently via a selective photothermal absorber and a thermoelectric generator. Thermal management with an optimized upper layer (Fresnel focusing, air gap, quartz separator) minimizes upward heat loss while ensuring tissue safety, and a bottom layer (fin + PCM) enhances the temperature gradient across the TE generator. The resulting system achieves sufficient power for clinically relevant IMDs, including pacemakers and cameras, with in vivo validation in rabbits. The findings directly address the need for non-invasive, deep, and higher-power subcutaneous energy delivery, demonstrating output power that meets or exceeds thresholds (~>10 mW) for a broad range of IMDs. Histological assessments indicate minimal thermal damage under the tested protocols, supporting the potential clinical relevance of the method.

This work introduces a subcutaneous power supply strategy that combines NIR-II photothermal absorption with thermoelectric conversion in a multilayer, biocompatible package (Bio-PS). Guided by a heat transfer model and thermal management design, the system delivers high output power at clinically relevant depths, achieving up to 195 mW ex vivo and 20 mW in vivo through 8.5 mm of tissue, sufficient to power or recharge high-consumption IMDs. Demonstrations include powering a pacemaker and operating a wireless endoscopic camera. The approach represents a promising paradigm for contactless, non-invasive recharging of implanted devices with deep tissue penetration and minimal biological damage. Future work may focus on further miniaturization, long-term chronic implantation studies, optimization of thermoelectric materials and packaging for enhanced efficiency, and translation toward human-scale applications under clinical safety standards.

• Exposure constraints: In vivo NIR-II irradiation was limited to ≤10 minutes per cycle at up to 1.0 W cm−2 (ANSI Z136.1-2007), and charging strategies used on–off cycles with 10-minute intervals to reduce burn risk.
• Thermal elevation: Skin temperature could rise to ~45 °C at 1.0 W cm−2 within ~200 s; while histology showed no evident damage, thermal management remains critical, especially for longer or repeated sessions.
• Animal model and duration: Efficacy and safety were demonstrated in rabbits with a one-month observation period; broader, longer-term, and multi-species/human studies are needed to assess chronic performance and biocompatibility.
• Tissue thickness and conditions: Performance was characterized under specific tissue thicknesses (3.5–20 mm) and controlled conditions; results may vary with different anatomical sites, tissue optical properties, and motion.
• External circuitry: Some applications required a step-up converter (with <20% energy loss) and intermediate energy storage, which introduces efficiency penalties and system complexity.