Poynting–Robertson Drag

Sunlight as a headwind

A dust grain on a stable circular orbit around the Sun is, by Newton's reckoning, supposed to stay there forever. Gravity supplies exactly the centripetal force, the orbit closes on itself, and nothing is lost. Yet the inner Solar System is conspicuously short of dust: the zodiacal cloud you can see as a faint cone of light after sunset is thin, young, and constantly disappearing. Something is removing the small grains far faster than the age of the Solar System would allow. That something is Poynting–Robertson (PR) drag — the tiny but relentless braking force that starlight exerts on anything moving through it.

The intuition is a headwind. Stand still in falling rain and the drops come straight down; start running and they appear to slant into your face. A dust grain orbiting at v ≈ 30 km/s sees sunlight the same way: photons that travel radially outward from the Sun appear, in the grain's frame, to arrive tilted slightly forward by the aberration angle v/c. When the grain absorbs that light and re-radiates the heat evenly in all directions, the net momentum it picks up has a small backward (anti-orbital) component. That backward push is a drag. It removes orbital angular momentum, the orbit shrinks, and over thousands of years the grain spirals into the Sun.

How it works, step by step

There are two equivalent ways to see the effect, and it helps to hold both.

Aberration picture. In the grain's rest frame, the incoming sunlight is not purely radial — it is tilted toward the direction of motion by an angle θ ≈ v/c. The grain absorbs this slightly forward-tilted radiation, so the absorbed momentum has a component opposing the orbital velocity. After absorption the grain re-emits thermally, but in its own frame that emission is isotropic and carries away no net momentum. The leftover is a tangential drag plus the familiar radial radiation-pressure push.

Re-emission picture. Equivalently, work in the Sun's frame. The grain absorbs radial momentum (pure outward push, no tangential part). But the grain is moving, so the photons it re-emits — isotropic in its frame — are beamed slightly forward in the Sun's frame. Emitting forward-beamed photons is like firing a rocket forward: the grain recoils backward. The recoil is the drag. Both pictures give the same answer because they are the same physics in two reference frames.

The key is that the drag is tangential. A radial force (gravity, radiation pressure) changes the orbit's size and shape but conserves angular momentum about the Sun. A tangential force does work against the orbital motion and bleeds angular momentum away. Lose angular momentum on a bound orbit and you must drop to a smaller, lower-energy orbit — the grain sinks inward. PR drag also damps eccentricity, so an initially elongated orbit circularises as it shrinks.

Quantitative analysis

Robertson's 1937 relativistic treatment gives the radiation force on a grain of geometric cross-section A intercepting a stellar flux S (W/m²), to first order in v/c:

F = (S A Q_pr / c) [ (1 − v_r/c) r̂ − v/c ]

The first term in the bracket is radiation pressure (outward, along r̂); the v/c term is the PR drag, opposing the velocity vector v. Q_pr is the radiation-pressure efficiency, of order unity for grains larger than the wavelength of sunlight. It is convenient to package the radial part with gravity using the dimensionless ratio

β = F_rad / F_grav = (3 L_* Q_pr) / (16 π G M_* c ρ s)
  ≈ 0.57 Q_pr / (ρ[g/cm³] · s[µm])   for the Sun

Both radiation pressure and gravity fall as 1/r², so β is a constant for a given grain — independent of distance. Grains with β > 0.5 released from a circular parent orbit are immediately unbound (they leave as hyperbolic "beta meteoroids"); grains with β < 0.5 stay bound and slowly spiral in under the drag term.

For a grain on a circular orbit, the tangential drag removes angular momentum at a steady rate, and the orbital radius a(t) decays. Integrating the equation of motion gives the time to spiral from radius a₀ all the way to the Sun:

t_PR = a₀² c² / (4 G M_* β)          (compact form)
     ≈ a₀² c² s ρ / (3 L_*)          (substituting β and Q_pr ≈ 1)

The second form is the costed relation worth memorising: the inspiral time scales as a₀² · s · ρ / L_*. It grows as the square of the starting distance, linearly with grain radius s and density ρ, and inversely with stellar luminosity L_*. Larger, denser grains around fainter stars live longest; small grains close in vanish quickly. Numerically, for circular orbits around the Sun the compact form reduces to a clean rule of thumb:

t_PR ≈ 400 · (a₀ / AU)² / β   years

Worked example: a millimetre grain at 1 AU

Take a silicate grain with radius s = 0.5 mm (so diameter ≈ 1 mm), density ρ = 3000 kg/m³, on a circular orbit at a₀ = 1 AU, with Q_pr ≈ 1. First compute β:

β ≈ 0.57 / (ρ[g/cm³] · s[µm])
  = 0.57 / (3.0 · 500)
  = 3.8 × 10⁻⁴

The grain is far below the β = 0.5 blow-out threshold, so it is firmly bound and will spiral in rather than be ejected. Now the inspiral time:

t_PR ≈ 400 · (a₀/AU)² / β
     = 400 · (1)² / (3.8 × 10⁻⁴)
     ≈ 1.0 × 10⁶ ... wait — use the size form directly:

t_PR ≈ a₀² c² s ρ / (3 L_*)
     = (1.496×10¹¹)² · (3×10⁸)² · (5×10⁻⁴) · 3000 / (3 · 3.83×10²⁶)
     ≈ (2.24×10²²)(9×10¹⁶)(1.5) / (1.15×10²⁷)
     ≈ 3.0×10³⁹ / 1.15×10²⁷ · ...
     ≈ 3 × 10¹¹ s
     ≈ 1 × 10⁴ years

So a millimetre grain at 1 AU spirals into the Sun in roughly 10,000 years — the headline number for this effect. Compare that to the 4.6-billion-year age of the Solar System: the grain's lifetime is shorter by a factor of nearly half a million. Any millimetre dust we see at 1 AU today cannot be primordial; it was produced recently and is on its way to destruction.

Scale the same grain up and down to feel the dependences. A 10-micron grain (s = 5 µm) at 1 AU has β ≈ 0.04 and falls in about a thousand years; a sub-micron grain may have β > 0.5 and be blown out instead of spiralling in. Move the millimetre grain out to the asteroid belt at 3 AU and the a₀² factor multiplies the lifetime by nine, to ≈ 90,000 years — still a geological eyeblink. This is the engine that keeps the inner Solar System dust-poor.

Variants and regimes

PR drag is one term in a richer competition. Which process controls a grain's fate depends on its size and where it lives.

• Blow-out regime (β > 0.5). The smallest grains, typically below ~1 µm for silicates, feel a radiation-pressure push that exceeds half the Sun's gravity. Released from a parent on a circular orbit, they are immediately on unbound hyperbolic trajectories and stream out of the Solar System as "beta meteoroids" rather than spiralling in.
• PR-drag regime (β < 0.5, low collision rate). Bound grains from microns to millimetres spiral inward on the timescales above. This is the regime of the Solar System's tenuous zodiacal cloud, where collisions are rare and PR drag wins.
• Collision-dominated regime. In dense debris disks around other stars, grains shatter in mutual collisions before PR drag can move them. The collisional lifetime is shorter than t_PR, so dust grinds down a cascade in place rather than spiralling in. Whether a disk is "PR-dominated" or "collision-dominated" is set by its optical depth.
• Solar-wind (corpuscular) drag. The solar wind exerts an analogous momentum drag — the same v/c geometry, but with protons instead of photons. It adds roughly a 20–35% correction to photon PR drag for typical grains and dominates for the very smallest particles where the wind's effective cross-section exceeds the photon one.
• Resonant trapping. Grains spiralling inward can be captured into mean-motion resonances with planets, which halts or slows their decay and piles them into a ring. The Earth sits in such a resonant dust ring, a leading-trailing asymmetry detected by IRAS and COBE.

Common pitfalls and misconceptions

• Confusing PR drag with radiation pressure. Radiation pressure is the radial (outward) push; PR drag is the much smaller tangential (v/c) braking. Radiation pressure changes orbit size by lowering the effective gravity (via β); only the tangential drag removes angular momentum and forces a true inspiral. They are different terms of the same force.
• Thinking the grain is "pushed in" by light. Light pushes outward. The inward spiral happens because the grain loses angular momentum and therefore must drop to a smaller orbit — an indirect, dynamical consequence, not a direct inward shove.
• Assuming bigger means faster. The opposite is true. Because deceleration ∝ 1/s, smaller grains spiral in faster. Boulders and asteroids feel PR drag too, but their lifetimes vastly exceed the age of the universe, so it is irrelevant for them (the size-dependent Yarkovsky effect dominates large-body drift instead).
• Forgetting the v/c smallness. The drag is of order v/c ≈ 10⁻⁴ relative to radiation pressure. It is genuinely tiny per orbit — but it acts every orbit for millions of revolutions, and the cumulative effect is decisive.
• Treating the zodiacal cloud as primordial. Since PR clearing is so fast, the cloud cannot be a 4.6-Gyr-old relic. It is a steady state continuously fed by comets and asteroid collisions; turn off the supply and it would fade in tens of thousands of years.

Observational status and applications

PR drag is not a paper effect; its fingerprints are all over the inner Solar System and other planetary systems.

• The zodiacal cloud's steady state. The very existence of a thin, smoothly distributed interplanetary dust cloud — rather than either a thick disk or nothing — is the signature of PR removal balanced against cometary and asteroidal supply. Modelling by Nesvorný and collaborators (2010) concluded that Jupiter-family comets supply the bulk of the cloud, precisely because PR drag fixes the removal rate.
• Earth's resonant dust ring. IRAS (1983) and COBE/DIRBE mapped a ring of dust trapped in mean-motion resonance with Earth, with a density enhancement trailing the planet — direct evidence of grains caught while spiralling sunward under PR drag.
• Collected interplanetary dust particles. Stratospheric collection of IDPs and analysis of cosmic spherules confirm a steady rain of grains drifting in from the asteroid belt and comets, consistent with PR-driven transport to 1 AU.
• Debris disks around other stars. Whether a disk shows a dust-depleted inner hole or a filled-in disk depends on the PR-versus-collision balance; PR drag is a standard ingredient in modelling systems like Vega, Fomalhaut, and β Pictoris.
• Spacecraft and small-body design. The same v/c radiation force perturbs the orbits of small satellites, balloon and solar-sail craft, and is a known non-gravitational term in the trajectory modelling of asteroids and reflective spacecraft.

Inspiral times across the Solar System

Approximate Poynting–Robertson inspiral times from a circular starting orbit, for silicate grains (ρ ≈ 3000 kg/m³, Q_pr ≈ 1) around the Sun. Times scale as a₀² · s, so doubling distance quadruples the lifetime and doubling size doubles it.

Grain radius s	β	Start radius a₀	Inspiral time t_PR	Regime
0.2 µm	> 0.5	1 AU	— (blown out)	Beta meteoroid
1 µm	~ 0.19	1 AU	~ 2,000 yr	PR inspiral
10 µm	~ 0.019	1 AU	~ 2 × 10⁴ yr	PR inspiral
100 µm	~ 1.9 × 10⁻³	1 AU	~ 2 × 10⁵ yr	PR inspiral
1 mm (s = 0.5 mm)	~ 3.8 × 10⁻⁴	1 AU	~ 1 × 10⁴ yr*	PR inspiral
1 mm (s = 0.5 mm)	~ 3.8 × 10⁻⁴	3 AU (belt)	~ 9 × 10⁴ yr	PR inspiral
1 cm	~ 3.8 × 10⁻⁵	1 AU	~ 10⁷ yr	Slow drift
1 m	~ 3.8 × 10⁻⁷	1 AU	> 10⁹ yr	Yarkovsky takes over

*The widely quoted "~10,000 years for a millimetre grain at 1 AU" figure uses the size form t ≈ a² c² s ρ / (3 L_*) with s ≈ 0.5 mm; the slightly different β-based row above reflects rounding in the prefactor. Both land within a factor of a few of 10⁴ years — the point is the timescale, which is tiny compared to the age of the Solar System.

Why it matters

Poynting–Robertson drag is a small effect with an outsized role. It is the reason the inner Solar System is not buried in the debris of 4.6 billion years of cometary and asteroidal grinding: every millimetre grain at 1 AU is gone within ten thousand years, swept into the Sun by a headwind of its own absorbed light. The zodiacal cloud we see is the thin, ever-renewed froth left over from that balance between supply and PR removal. Around other stars the same v/c braking decides whether a debris disk keeps a clean inner hole or fills with sunward-drifting dust. It is a textbook demonstration that even the momentum of light, acting through nothing more than relativistic aberration, can reshape the architecture of a planetary system over astronomical time.