Gate 1: freeze the physical question
Specify the wavelength, beam geometry, material stack, electrical states and moving assembly. Define the desired force or torque change and what would distinguish it from known artifacts. Choose a passive reference and a comparable electrical dummy before selecting a favorable data window.
Gate 2: qualify the optical states
Measure the angular distribution of reflected and transmitted power for each state, plus incident power and absorption. Use these measurements to predict the mechanical effect with uncertainty. Preserve raw optical data and fabrication metrology. A numerical design is a hypothesis until the sample’s optical response is known.
Gate 3: qualify the force measurement
Calibrate with a known force and measure drift, noise and cross-axis sensitivity. Keep external illumination and beam catchers outside the moving assembly. Plan pressure sweeps, beam-off tests, electrical dummy cycles, illumination reversals and temperature logging. Include possible gas-mediated forces, outgassing and electrostatic attraction.
Gate 4: run the comparison
Use randomized or balanced switching sequences with a predefined analysis. Compare the optical prediction and measured force change within the combined uncertainty. Do not interpret a residual automatically as new physics. A disagreement first triggers checks of the model, instrument and unaccounted momentum paths.
Gate 5: establish a reason to continue
Replicate on a second sample and, eventually, an independent instrument. Compare against the passive architecture for a specified application. Report useful force per watt, control range, heat, moving mass and failure modes. Continue to mission engineering only after the result is repeatable and the system-level benefit is clear.
Who can move this forward
We welcome nonconfidential contributions from researchers in infrared photonics, graphene fabrication, optical metrology, precision force measurement and spacecraft control. A useful introduction identifies the relevant evidence, available capability and the specific assumption or measurement the contributor can test. No institutional relationship or funding award is implied.