Value chain · CLM-1110
In plain terms
The best near-infrared laser-to-electricity receiver so far turns 56 per cent of the light into electricity, but only when the light is 271 kilowatts per square metre, a thousand times brighter than the microwave design beam. At the level the authors call eye-safe, about 270 times lower, it turns about 30 per cent into a couple of milliwatts, while also carrying gigabit data.
The claim as recorded
A 2026 Fraunhofer ISE and Cambridge paper measures monolithic InGaAsP photonic power converters of 5.4 mm2 active area under 1455 nm laser light from 1 to 70 W/cm2: peak power conversion efficiencies of 48.1 per cent (one junction, at 7.1 W/cm2), 49.4 (two junctions, 13.9 W/cm2) and 55.9 +/- 2.0 per cent (four junctions, 27.1 W/cm2), the last stated as a new benchmark for 1.3-1.6 um receivers. On a 3 m eye-safe 1530 nm free-space link at 0.1 W/cm2 received, the same devices deliver 1.78, 1.86 and 1.58 mW (32.8, 33.3 and 29.3 per cent) while carrying 1.12, 2.23 and 2.00 Gb/s near short circuit, and 1.36 (two-junction) or 1.58 Gb/s (four-junction) at the maximum power point; 28 per cent of the optical power was lost in the link's lenses, collimation and propagation.
Evidence behind this: T1-peer-reviewed measured device results with stated uncertainty, from the device makers.
Scope and limits
A device-level result on a 5.4 mm2 chip under a laboratory laser; nothing here is a beaming, range or system efficiency. The 55.9 per cent is at 271 kW/m2, an irradiance no eye-safe or public-exposure regime permits; the eye-safe demonstration is at 1 kW/m2 over 3 m and returns about 30 per cent. The paper states the eye-safety basis by citation only (CLM-1112). The link efficiency is receiver-side; the 28 per cent optical loss is the authors' attribution, not a measurement by component. No thermal management or degradation figure is given for sustained operation at 27 W/cm2 (the I-V laser is pulsed, p.3). The p.7 body text gives the 1 J near-short-circuit data rate as 1.13 Gbit/s against Table 1's 1.12; the table's value is used here.
What it assumes
No assumptions are recorded against this claim.
Where it comes from
p.1, raw: "Our 4-junction InGaAsP photovoltaic receivers achieve 55.9%" and, with the fi of efficiency printed as a ligature, the squared unit as a superscript and the micron as the Greek letter (read at the render), "power conversion" efficiency "at 27.1 W/cm" 2 "(1.45" um ") and 2.23 Gb/s"; depunctuated (the page's hyphens are non-breaking): "Over a 3-m eye-safe link at 1.53" um, then raw: "we simultaneously achieve 1.58 mW harvested power and 1.58 Gb/s". p.3, normalised: "The active area for photodetection (mesa minus busbars) is Aact = 5.4 mm2"; "I-V characteristics were measured using a 1455 nm pulsed laser with variable irradiance from 1 W/cm2 to 70 W/cm2". p.5, render and raw: peak PCE values "48.1%", "49.4%" and "55.9%" with the errors "1.7%", "1.7%" and "2.0%", "were measured at irradiances of 7.1, 13.9 and 27.1 W/cm2, respectively"; normalised: "The highest PCE recorded for the 4 J device sets a new benchmark for this device class, consistent with the survey in Fig. 1(a)"; "The 28% power losses in the optical channel are attributed to channel losses due to lens absorption, imperfect collimation, and free-space propagation". p.6, normalised: "At PMP, 1 J, 2 J and 4 J PPCs deliver 1.78, 1.86 and 1.58 mW, respectively" with PCEs "32.8%", "33.3%" and "29.3%". p.7 Table 1, raw title: "Key FSO link measured parameters at 1.53" um "and 0.1 W/cm2 received optical power at the PPC"; data rates near short circuit 1.12, 2.23, 2.00 Gb/s. p.8, normalised: "we achieve 1.36 (1.58) Gb/s data rate while simultaneously delivering 1.86(1.58) mW"; depunctuated (the hyphen falls at a line break): "Near short-circuit operation, data rate as high as 2.23 Gb/s were demonstrated". Conversions: 27.1 W/cm2 = 271 kW/m2; 0.1 W/cm2 = 1 kW/m2; 1.58 mW over 5.4 mm2 at 0.1 W/cm2 received is 1.58 / 5.4 = 29.3 per cent, matching the printed PCE.
Basis: power conversion efficiency, per cent, at stated irradiance in W/cm2 and wavelength; electrical output in mW at a stated received irradiance
Related records
- Answers question
- QST-002 not published in this releaseQST-036 not published in this release
Cite this
energyaperture.com/c/clm-1110
CLM-1110, Energy Aperture, as of 2026.
Connected records
Other published records linked to this one. A list, not a judgement.
- Also from SRC-586
- CLM-1111Before this paper, the best near-infrared laser receivers reached about half efficiency, and only under very bright light; at gentler light...CLM-1112The paper says this colour of infrared is good because the air lets it through and the eye-safety rules allow more of it, but it cites...CLM-1113This paper does not describe a laser power satellite, though two NASA-funded designs for one exist from 1978-80. It does show how a laser...