A film thin as a postage stamp, pushed by a ground-based laser array, accelerated to a fraction of light speed over twenty years toward Proxima Centauri—that is the vision of Breakthrough Starshot. The sail carries no fuel; it rides entirely on radiation pressure. A preprint posted to arXiv on June 2, 2026 (arXiv:2606.04052, not yet journal-published) by Shen Chao and Li Jiaze of Harbin Institute of Technology does the detailed accounting for such a sail—and finds a 'brake line' near 0.75 times the speed of light that has largely gone unexamined.
Radiation pressure is three forces, not one
1. Direct impact of incident photons—the strongest component. 2. Specular reflection—photons bounce back off the mirror-like sail, pushing it forward. 3. Diffuse scattering—no real surface is a perfect mirror, so some light scatters in all directions. The weakest term, long treated as a rounding error.
The first slowdown: Doppler redshift
The faster the sail flies, the redder the chasing laser appears in the sail's frame; photon energy and arrival rate drop, weakening all three components:
| Speed | Doppler factor | |-------|----------------| | 0.1c | 0.90 | | 0.2c | 0.82 | | 0.5c | 0.58 | | 0.75c | 0.38 |
At 0.75c the light is redshifted by 62%, and thrust shrinks accordingly.
The paper's headline: aberration flips the weakest term
The deeper effect is relativistic aberration. In the sail's own frame, diffusely scattered light goes equally forward and backward. But in the ground frame, all scattered photon directions are tipped forward. At 0.75c, backward-emitted photons are confined to a cone only 41.4° from the direction of travel (versus 5.7° at 0.1c). Since photons carry momentum forward, the sail receives a reaction force—backward.
Using a one-sided Lambertian scattering model, the authors derive a clean critical value: β = 3/4 exactly. With a blackbody-radiation model, the critical value shifts to β = 2/3. Beyond the line, the weakest force reverses into drag. Notably, another calculation in the paper shows the sail exceeds 0.75c after about 5 characteristic acceleration times τ—so the brake line sits squarely on the course of a maximal-acceleration profile.
The necessary cold water
The paper itself supplies the caveats, and so must any honest retelling:
- Net thrust never turns negative. All three components summed, the sail keeps accelerating; it just works harder for each unit of thrust.
- The drag is small. Diffuse-scattering drag peaks at roughly 0.8% of the momentum flux (around 0.87c).
- Breakthrough Starshot's official cruise target is only 0.15–0.2c, far below the brake line; at that speed diffuse scattering still acts as ordinary thrust. The 0.75c threshold governs hypothetical faster designs, not current plans.
- 2026-06-02 — Preprint posted to arXiv by the HIT Shenzhen–Harbin team.
- 2026-06-08 — First coverage by Universe Today.
- 2026-09-02 — ScienceDaily syndication sparks a second wave of discussion.
Proper academic credit
Radiation-recoil drag from re-emission is not new: Parkinson's 2018 model of the Starshot system already contained a Poynting-Robertson-like drag term, and it appears in this paper's references. The genuine novelty is threefold: the clean three-component decomposition of radiation pressure, the tensor derivation using a one-sided Lambertian model, and the exact critical speed. As for media claims that "a metasurface sail can use aberration to self-stabilize at the beam center," no such result appears in the paper—that was the reporter's embellishment.
Publication timeline
Perspective
How fast is 0.75c? The fastest object humanity has built, the Parker Solar Probe, used the Sun's gravity to reach about 190 km/s—just 0.06% of light speed. The brake line seems almost laughably distant. But this decimal-precision ledger reminds every star-chart drawer that in a light sail's account book there are no rounding errors: even the weakest force will switch sides when relativity gets involved.