--- title: "The drone was not recharged in flight. The real result is what 38.49% leaves out" locale: "en" url: "https://irz.fr/en/articles/laser-drone-3849-efficiency-en" markdown_url: "https://irz.fr/en/articles/laser-drone-3849-efficiency-en.md" category: "tech" tags: ["drone", "laser", "energy", "photovoltaics", "charging"] published_at: "2026-08-23T16:21:00.000Z" author: "Hugo Marchal" translation: "https://irz.fr/fr/articles/drone-laser-3849-rendement-fr.md" --- # The drone was not recharged in flight. The real result is what 38.49% leaves out A new receiver converts 38.49% of green laser light into electricity. That is a strong result, but its denominator begins after several losses a complete system still has to absorb. The number that travels best is **38.49%**. Researchers at the Civil Aviation University of China and Tsinghua University obtain it from a receiver illuminated with green light at 1.2 W/cm².[1](https://doi.org/10.1016/j.matlit.2026.100066) Retold too quickly, the result becomes a drone “recharging in flight” and never needing to land, even though the published experiment stops well before that scenario. The receiver sits inside a model wing under 520 nm illumination, with airflow simulating cooling in flight, and its electrical output drives a propeller.[1](https://doi.org/10.1016/j.matlit.2026.100066)[2](https://www.optica-opn.org/Home/NewsRoom/2026/August/Lasers_Promise_Power_Solution_for_Flying_Drones) Outdoor testing on a real aircraft appears only as the next step.[2](https://www.optica-opn.org/Home/NewsRoom/2026/August/Lasers_Promise_Power_Solution_for_Flying_Drones) This distinction makes the result easier to read correctly: **where do the 38.49% begin and end?** ## After the laser The published efficiency begins with light already reaching the receiver, comparing the electrical output of the photovoltaic-thermoelectric tandem with incident optical power. At 1.2 W/cm², the modified device produces 461.86 mW/cm² and reaches 38.49%.[1](https://doi.org/10.1016/j.matlit.2026.100066) Outside that denominator sit the electricity used to generate the beam, losses through the optics and atmosphere, spillover from imperfect pointing, and later conversion or battery-storage losses. > **The path of a real watt** > - ground electricity → laser: 1 > - laser → atmosphere and pointing: 2 > - received light → electricity: this is where the 38.49% sits: 3 > - electricity → battery, motors and payload: 4 > Han et al. mainly characterise step 3, not wall-to-battery efficiency for the complete system. DARPA offers a useful comparison through PRAD. In 2025 the POWER programme delivered more than **800 W of electrical power** over 8.6 km, while shorter-range tests exceeded **20% from optical power leaving the laser to electrical power leaving the receiver**; the agency stresses that efficiency was not the primary objective.[5](https://www.darpa.mil/news/2025/darpa-program-distance-record-power-beaming) This remains short of wall-to-wall accounting, although the denominator already includes the path of the beam that sits upstream of the Chinese receiver's 38.49%. ## Heat wins The new receiver mainly addresses a less cinematic problem than tracking a moving aircraft: **conversion efficiency falls as the illuminated cell heats up**. Under 1.2 W/cm² irradiation, the authors measure a steady surface temperature near **85.3°C**.[1](https://doi.org/10.1016/j.matlit.2026.100066) More optical power brings more energy to harvest, but at the same time pushes the photovoltaic material into a hotter and less favourable operating condition. Their receiver stacks two conversion mechanisms: the CsPbBr₃ perovskite layer turns light into current, while the thermoelectric element underneath recovers part of the temperature gradient otherwise left as waste heat.[1](https://doi.org/10.1016/j.matlit.2026.100066) The researchers also add **Sb₂Se₃ antimony-selenide nanorods**, using their low thermal conductivity as a barrier while the paper reports better crystal formation and charge transport.[1](https://doi.org/10.1016/j.matlit.2026.100066) At 1.2 W/cm², the tandem reaches 34.65% without this addition and **38.49%** with it.[1](https://doi.org/10.1016/j.matlit.2026.100066) > Illustration: Laser-energy receiver fitted into a model wing with airflow used for cooling. The prototype is more than a solar cell attached to an aircraft: an internal wing channel uses airflow to maintain the thermal gradient needed by the tandem receiver. Credit: [Y. Han, X. Han et al. / Matter & Light](https://doi.org/10.1016/j.matlit.2026.100066). The wing itself becomes part of the energy circuit. The team designs a channel that cools the cold side of the thermoelectric device and uses simulations to check aerodynamic integration.[1](https://doi.org/10.1016/j.matlit.2026.100066) ## Hit the wing Even an excellent receiver contributes nothing when the beam misses its active area. A separate paper published in July 2026 by Politecnico di Torino and ORIS addresses this side of the problem. Its architecture uses RTK positioning for coarse target acquisition and receiver-side optical feedback for fine corrections.[3](https://doi.org/10.3390/aerospace13080664) The experiments characterise propagation, spot size, pointing accuracy and closed-loop steering in laboratory and outdoor tests.[3](https://doi.org/10.3390/aerospace13080664) Positioning can bring the beam into the right region, after which the optical loop has to keep the spot on the active receiver as distance, vibration and movement change; navigation accuracy alone cannot guarantee energy transfer. Pointing loss is unforgiving because, relative to a solar panel illuminated across a broad part of the sky, this receiver gets no benefit from light spilling beyond its active surface. Changes of aircraft attitude add further losses by altering projected area and making illumination less uniform.[3](https://doi.org/10.3390/aerospace13080664) Optical tracking is therefore part of energy efficiency, not merely a navigation problem. ## Flight already happened The history of laser-powered aircraft also reaches well beyond 2026. In 2012, Lockheed Martin and LaserMotive powered the Stalker UAV from a laser during an outdoor test; the same technology had previously kept the aircraft operating for 48 hours in a wind tunnel.[6](https://spectrum.ieee.org/uav-flies-on-laser-light) In April 2026, PowerLight, which grew from LaserMotive, says it delivered laser power to a UAS **during real flight** in a US demonstration.[4](https://powerlighttech.com/press-release-powerlight-achieves-industry-first-wireless-power-beaming-to-a-fielded-military-uas-in-flight/) That answers the dynamic-integration question that the new Chinese receiver has not yet tested. > Illustration: PowerLight demonstration showing its ground system and a receiver-equipped drone. PowerLight's April 2026 test provides the element the Chinese receiver paper does not yet have: transmission to a real UAS in flight. Its public release does not provide a complete end-to-end efficiency figure. Credit: [PowerLight Technologies](https://powerlighttech.com/press-release-powerlight-achieves-industry-first-wireless-power-beaming-to-a-fielded-military-uas-in-flight/). The comparison has a useful limit: PowerLight demonstrates dynamic integration, but its public release gives no detailed energy accounting that can be compared directly with the receiver's 38.49%.[4](https://powerlighttech.com/press-release-powerlight-achieves-industry-first-wireless-power-beaming-to-a-fielded-military-uas-in-flight/) The projects therefore answer different questions, one about thermal management and optical conversion inside the receiver, the other about maintaining energy delivery to a moving aircraft. ## A beam in airspace Safety belongs inside the architecture too. In the US, the FAA notes that intentionally aiming a laser at an aircraft is a federal crime and treats laser illumination as a serious aviation risk.[7](https://www.faa.gov/about/initiatives/lasers/laws) A controlled energy-transfer installation has a different purpose, yet the physical issue remains the passage of an energetic beam through shared airspace. Real systems therefore need line-of-sight supervision and shutdown when the path is no longer safe, a class of interlock that PowerLight includes in its public architecture.[4](https://powerlighttech.com/press-release-powerlight-achieves-industry-first-wireless-power-beaming-to-a-fielded-military-uas-in-flight/) > Illustration: DARPA PRAD receiver designed to convert a distant laser beam into electrical power. PRAD spreads the incoming beam onto photovoltaic cells. In 2025 DARPA received more than 800 W at 8.6 km, with both ends of the test on the ground. Credit: [DARPA](https://www.darpa.mil/news/2025/darpa-program-distance-record-power-beaming). Fog, cloud, rain or a physical obstruction add a dependency that an onboard battery avoids: beaming moves part of the energy mass off the aircraft **in exchange for** sufficient visibility and external tracking infrastructure. ## The denominator The Chinese work remains important once the promise of an “infinite drone” is removed and attention returns to the component actually tested. A lightweight receiver has to survive high optical intensity without degrading, keep producing under airflow and fit into an aircraft without cancelling the aerodynamic benefit. The perovskite-thermoelectric tandem offers a credible answer to that portion of the system.[1](https://doi.org/10.1016/j.matlit.2026.100066)[2](https://www.optica-opn.org/Home/NewsRoom/2026/August/Lasers_Promise_Power_Solution_for_Flying_Drones) A laser energy network will still make sense only when the complete accounting works, from the electricity drawn on the ground to the useful power stored or consumed on board, including generation and propagation losses, receiver mass, tracking cost and the share of a mission spent in a geometry where transfer remains possible and safe. The **38.49%** answers one part of that accounting very well; this is already useful, and very different from 38.49% for the whole machine. ## References 1. [Han et al., Sb2Se3 nanocrystals enable efficient perovskite-thermoelectric tandem devices for laser-powered unmanned aerial vehicles](https://doi.org/10.1016/j.matlit.2026.100066) 2. [Optics & Photonics News, Lasers Promise Power Solution for Flying Drones](https://www.optica-opn.org/Home/NewsRoom/2026/August/Lasers_Promise_Power_Solution_for_Flying_Drones) 3. [Sfasciamuro et al., Laser-Based Far-Field Wireless Power Transfer for UAV Recharging](https://doi.org/10.3390/aerospace13080664) 4. [PowerLight Technologies, wireless power beaming to a UAS in flight](https://powerlighttech.com/press-release-powerlight-achieves-industry-first-wireless-power-beaming-to-a-fielded-military-uas-in-flight/) 5. [DARPA, program sets distance record for power beaming](https://www.darpa.mil/news/2025/darpa-program-distance-record-power-beaming) 6. [IEEE Spectrum, UAV flies on laser light](https://spectrum.ieee.org/uav-flies-on-laser-light) 7. [FAA, Laser Incidents](https://www.faa.gov/about/initiatives/lasers/laws)