A 2013 experiment by Paul A. Williams and collaborators mounted a 25-mm dielectric mirror on a commercial force-restoration balance. Removing its pan-weight compensation spring allowed horizontal force sensing with a horizontal laser path. They measured mirror optical reflectance and varied incidence from about to . A 1071-nm laser, capable of 530 W, was switched at 0.1 Hz. Digitized mass-equivalent readings were converted to force using ; quadrature demodulation isolated the switched signal from drift. [ref]
Measurement calibration used a 20-g mass, with checks down to 1 mg. The optical forces lay below that checked range, limiting the accuracy claim. An angle scan tested the predicted cosine dependence at fixed intercepted power. Replacing the mirror with a carbon nanotube coated absorber increased absorbed power while reducing the predicted optical force; no modulation peak was detected. This bounded gas-heating artifacts under those conditions, rather than proving their universal absence. The fitted slope was about 7% below . Inferring laser power requires calibrated force, optical reflectance, radiometric absorptance and angle of incidence, with measurement uncertainty for each. [ref]
Published in Optics Letters 38(20), 4248–4251.
Experimental article jointly credited to Paul A. Williams, Joshua A. Hadler, Robert Lee, Frank C. Maring and John H. Lehman. Journal publication is 15 October 2013 in Optics Letters 38(20), pp. 4248–4251; NIST hosts the manuscript and bibliographic record rather than replacing the human byline.
Experimental article jointly credited to Paul A. Williams, Joshua A. Hadler, Robert Lee, Frank C. Maring and John H. Lehman. Journal publication is 15 October 2013 in Optics Letters 38(20), pp. 4248–4251; NIST hosts the manuscript and bibliographic record rather than replacing the human byline.
Cited passages: Williams and collaborators, Use of radiation pressure for measurement of high-power laser emission, equation (1): force expressed using intercepted optical power and the incidence-angle cosine; Williams and collaborators, pp. 2–3, Figures 1–3 and equation (2): preparation, apparatus and readout; Williams and collaborators, pp. 2–4, Figures 3, 5–7: calibration limits, angle and absorber controls, and reported slope discrepancy.