Planet Formation
Planet Gap Opening: How a Growing Planet Carves Its Orbit Clear
In 2014, the ALMA radio array turned toward a million-year-old star called HL Tauri, roughly 450 light-years away, and delivered an image that rewrote textbooks: a dusty disk sliced by a nested set of dark, concentric rings, like grooves on a vinyl record. Those dark rings are gaps — annular regions where the gas and dust density has plunged, most likely because unseen planets are gravitationally sweeping their orbital lanes clear.
Planet gap opening is the process by which a planet embedded in a protoplanetary disk exerts tidal (Lindblad and corotation) torques on the surrounding gas, pushing material away from its orbit faster than the disk's viscosity can refill it. When the planet is massive enough, the result is a depleted annulus — a gap — centered on the planet's semi-major axis. The phenomenon links planet formation, disk hydrodynamics, and the spectacular substructures now routinely imaged in nearby star-forming regions.
- TypePlanet-disk gravitational interaction
- RegimeGas-dominated protoplanetary disk
- Key criterionCrida et al. (2006)
- Threshold mass~10-100 Earth masses (Neptune to Jupiter)
- Governing balanceTidal torque vs. viscous torque
- Observed inHL Tau, TW Hya, PDS 70 (ALMA/VLT)
Interactive visualization
Press play, or step through manually. The visualization is yours to drive — try it before reading on.
Watch the 60-second explainer
A condensed visual walkthrough — narrated, captioned, under a minute.
What a Gap Is and the Physics Behind It
A protoplanetary disk is a rotating sheet of gas and dust, typically extending tens to hundreds of AU around a young star, with a mass on the order of 0.001-0.1 M_sun. Gas orbits at nearly the Keplerian speed, and its own viscosity (usually parameterized by the Shakura-Sunyaev alpha, with alpha ~ 1e-4 to 1e-2) slowly spreads material inward and outward.
An embedded planet perturbs this smooth flow. It launches spiral density waves at Lindblad resonances, where the gas orbital frequency beats against the planet's. These waves carry angular momentum: the outer disk gains it and is pushed outward, the inner disk loses it and drifts inward. The net effect is a repulsion of gas away from the planet's orbit.
- If viscous inflow refills the region faster than torques evacuate it, only a shallow density dip forms.
- If tidal torques win, the density collapses and a true gap opens, centered on the planet's semi-major axis.
A gap is therefore a balance point — not a hole punched instantly, but a steady state between clearing and refilling.
The Mechanism: Torque Balance and the Gap-Opening Criterion
Whether a gap opens is decided by a competition of torques. Three conditions are traditionally invoked:
- Thermal criterion: the planet's Hill radius R_H = r (M_p/3M_*)^(1/3) must exceed the disk's vertical scale height H. Since H/r = h is the aspect ratio (typically 0.03-0.1), this gives a thermal mass ratio q_th ~ h^3. For h = 0.05, q_th ~ 1.25e-4, about 40 Earth masses around a solar-mass star.
- Viscous criterion: the planet's tidal torque must exceed the viscous torque that refills the gap. This scales as q^2 > alpha h^2 roughly.
Crida, Morbidelli, and Masset (2006) unified both into a single semi-analytic relation:
3H/(4 R_H) + 50/(q Re) ≲ 1
where Re = r^2 Ω_p / ν is the disk Reynolds number and ν = alpha c_s H is the kinematic viscosity. The first term is thermal, the second viscous. A gap opens when the sum drops below ~1 — favored by higher planet mass, thinner (cooler) disks, and lower viscosity.
Key Quantities and a Worked Example
Consider a Jupiter-mass planet (M_p = 1 M_Jup, so q = M_p/M_* ≈ 9.5e-4 around a solar-mass star) at 5 AU in a disk with h = 0.05 and alpha = 1e-3.
- Hill radius: R_H = 5 AU × (9.5e-4/3)^(1/3) ≈ 0.35 AU, comfortably larger than H = 0.05 × 5 = 0.25 AU. Thermal term ~ 0.75/(4×0.35/5) — well below 1. Thermal criterion satisfied.
- Kanagawa gap depth: Kanagawa et al. (2015) showed the steady-state gap surface density follows Σ_gap/Σ_0 = 1/(1 + 0.04 K), with the single control parameter K = q^2 (h)^−5 alpha^−1. Plugging in: K = (9.5e-4)^2 × (0.05)^−5 × (1e-3)^−1 ≈ 9e-7 × 3.2e6 × 1000 ≈ 2900. Gap depth Σ_gap/Σ_0 ≈ 1/(1+116) ≈ 0.009 — a gap emptied to under 1% of the background. A deep, clean gap.
A Neptune-mass planet (q ~ 5e-5) in the same disk gives K ~ 8, so Σ_gap/Σ_0 ≈ 0.76 — a much shallower gap that removes only ~24% of the gas.
How Gaps Are Observed and Detected
Gaps became a mainstream diagnostic in 2014, when ALMA imaged HL Tauri at ~5 AU resolution and revealed at least seven bright rings separated by dark gaps. The 2018 DSHARP survey then showed that concentric gaps and rings are ubiquitous in bright disks, at radii from a few to over 100 AU.
- Dust continuum (mm): gaps appear as dark annuli; pressure bumps at gap edges trap drifting dust into bright rings.
- Scattered light (near-IR): instruments like VLT/SPHERE trace gaps in small grains at the disk surface.
- Kinematics: molecular-line velocity maps show 'kinks' and non-Keplerian wiggles from planet-driven pressure and flow perturbations.
The gold standard is a direct detection of the planet inside the gap. In 2018, Keppler et al. imaged PDS 70b — a ~5-10 M_Jup planet still accreting (seen glowing in H-alpha) within a wide dust gap; a second planet, PDS 70c, was later confirmed. PDS 70 remains the cleanest proof that planets carve the gaps we see.
Related Regimes: Type I, Type II, and Dust vs. Gas Gaps
Gap opening is one endpoint of a broader story of planet migration:
- Type I migration: low-mass planets (below q_thermal) do not open gaps; they feel a net corotation-plus-Lindblad torque and migrate rapidly, often inward, on ~10^5 year timescales.
- Type II migration: gap-opening planets become locked to the disk's slow viscous evolution and migrate on the much longer viscous timescale, ~10^5-10^6 years — historically thought to move 'with the gas,' though modern simulations show they can decouple.
A crucial subtlety is dust gaps versus gas gaps. Because dust drifts toward pressure maxima, even a modest planet that barely dents the gas can carve a strikingly deep dust gap. So an observed dust gap does not require a Jupiter — a Neptune-mass body may suffice. This decoupling (analyzed by Rosotti, Dipierro, and others) means gap width and depth in mm images must be modeled carefully before inferring a planet mass.
Significance, Famous Cases, and Open Questions
Gap opening is central to how planetary systems are assembled. Gaps set the migration mode and speed, throttle a planet's gas accretion (starving it once the gap deepens, helping fix final masses near ~1-10 M_Jup), and act as dust traps that concentrate solids and may seed the next generation of planetesimals.
- HL Tau (2014): the iconic first-light case; its gaps at ~13, 32, 64 AU imply multiple planets in a system only ~1 Myr old, challenging slow formation models.
- PDS 70 (2018): the first and still-best system with a planet directly imaged inside its gap.
- TW Hya, HD 163296, AS 209: DSHARP disks with clean multi-gap architectures.
Open questions remain sharp. Are all gaps planetary, or can some arise from ice lines, dead zones, or MHD winds and self-organization? How do we invert a gap's width and depth into a robust planet mass when viscosity (alpha) is poorly known? And why are so many gaps found at large radii (50-100 AU), where forming giant planets in situ is theoretically hard? These are among the most active problems in planet-formation astrophysics.
| Regime | Condition | Physical marker | Example planet |
|---|---|---|---|
| No gap (Type I) | q < q_thermal | Hill radius < disk scale height; density dip only | Earth-mass core |
| Thermal / partial gap | q ~ q_thermal = h^3 | Hill radius ~ scale height | Neptune (~17 M_Earth) in h=0.05 disk |
| Viscous-limited gap | 50/(q*Re) ~ 1 | Tidal torque overtakes viscous refilling | Saturn-mass in low-alpha disk |
| Deep gap (Type II) | Crida sum < 1 | Both thermal and viscous terms satisfied | Jupiter (q ~ 1e-3) |
| Very deep gap | K = q^2 h^-5 alpha^-1 >> 1 | Sigma_gap/Sigma_0 = 1/(1+0.04K) << 1 | PDS 70b (~5-10 M_Jup) |
Frequently asked questions
What is planet gap opening in simple terms?
It is the process by which a planet embedded in a young star's gas-and-dust disk clears an annular lane around its orbit. The planet's gravity launches spiral density waves that push nearby gas away faster than the disk's viscosity can refill it, leaving a depleted ring called a gap. These gaps appear as the dark rings in famous images like ALMA's HL Tauri disk.
How massive must a planet be to open a gap?
Roughly a Neptune-to-Jupiter mass, depending on the disk. The thermal criterion requires the planet's Hill radius to exceed the disk scale height, giving a thermal mass ratio q ~ h^3; for a disk aspect ratio h = 0.05 that is about 40 Earth masses around a solar-mass star. Lower viscosity and thinner, cooler disks lower the threshold, so even a Neptune can carve a deep dust gap.
What is the Crida gap-opening criterion?
Crida, Morbidelli, and Masset (2006) derived that a gap opens when 3H/(4 R_H) + 50/(q*Re) is less than or about 1, where R_H is the Hill radius, H the disk scale height, q the planet-to-star mass ratio, and Re the disk Reynolds number. The first term captures thermal (pressure) effects and the second captures viscosity; both must be small enough simultaneously.
How deep is a planet-opened gap?
Kanagawa et al. (2015) found the steady-state gap depth follows Sigma_gap/Sigma_0 = 1/(1 + 0.04 K), where K = q^2 * h^-5 * alpha^-1 combines mass, aspect ratio, and viscosity. A Jupiter in a low-viscosity disk (K in the thousands) can be emptied to below 1% of the background density, while a Neptune-mass planet leaves only a shallow dip.
Have astronomers actually seen a planet inside a gap?
Yes. In 2018, Keppler and collaborators directly imaged PDS 70b, a several-Jupiter-mass planet still accreting gas (glowing in H-alpha), sitting inside a wide dust gap of the PDS 70 disk. A second planet, PDS 70c, was later confirmed. This system is the strongest direct evidence that planets carve the gaps seen by ALMA.
Does every gap in a disk mean there is a planet?
Not necessarily. Planets are the leading explanation, but gaps can also form at ice lines where dust properties change, at dead-zone edges, or through magnetohydrodynamic winds and disk self-organization. Because dust drifts toward pressure maxima, a deep dust gap can also be carved by a surprisingly low-mass planet, so inferring a planet mass from a gap requires careful modeling.