Darwin Instability

The Darwin Instability is a secular tidal instability in close binary stars, planetary systems, and compact binaries that occurs when the orbital angular momentum of the system is insufficient to maintain the component bodies in synchronous rotation—specifically, when the orbital angular momentum J_orb falls below three times the sum of the rotational spin angular momenta J_spin (J_orb < 3J_spin). First derived by the English astronomer and mathematician Sir George Howard Darwin in 1879 while investigating the fission and tidal evolution of viscous self-gravitating fluid bodies, the instability represents the point at which inward orbital shrinkage releases less angular momentum than is required to spin up the stars to match the rising orbital frequency. Once a binary crosses the Darwin threshold, tidal torques can no longer restore synchronous equilibrium, triggering a runaway tidal inspiral that culminates in stellar merger (producing blue stragglers and luminous red novae such as V1309 Scorpii) or the engulfment of ultra-short-period hot Jupiters by their host stars.

Consider a close binary system consisting of two bodies with masses M₁ and M₂ (total mass M = M₁ + M₂, reduced mass μ = M₁M₂ / M), moments of inertia I₁ and I₂ (total moment of inertia I = I₁ + I₂), and spin angular velocities Ω₁ and Ω₂ aligned perpendicular to a circular orbit of semi-major axis a and Keplerian orbital frequency n = √(GM / a³).

In an isolated system with negligible stellar-wind magnetic braking, dissipative tidal friction conserves the total angular momentum J_tot while monotonically decreasing the total mechanical energy E_tot into internal tidal heat (Ė_tot ≤ 0):

J_tot = J_orb(a) + J_spin(Ω₁, Ω₂) = μ √(GMa) + I₁Ω₁ + I₂Ω₂

E_tot = E_orb(a) + E_spin(Ω₁, Ω₂) = −(GMμ / 2a) + (1 / 2)I₁Ω₁² + (1 / 2)I₂Ω₂²

Equilibrium Condition (Tidal Synchronization)

Expressing the orbital angular momentum in terms of the orbital frequency n = G^1/2M^1/2a^−3/2 via a = (GM / n²)^1/3 gives J_orb(n) = μ(GM)^2/3n^−1/3 and E_orb(n) = −(1 / 2)μ(GM)^2/3n^2/3. To find the stationary states of energy E_tot subject to fixed total angular momentum J_tot = const, we apply the method of Lagrange multipliers or eliminate n(Ω₁, Ω₂) using dJ_tot = (∂J_orb / ∂n)*dn* + I₁dΩ₁ + I₂dΩ₂ = 0. Because dE_orb / dJ_orb = n, the first differential of total energy is:

dE_tot = n dJ_orb + I₁Ω₁ dΩ₁ + I₂Ω₂ dΩ₂ = I₁(Ω₁ − n)*dΩ*₁ + I₂(Ω₂ − n)*dΩ*₂

Stationary equilibrium (dE_tot = 0) therefore requires full corotation (tidal synchronization) and circularization:

Ω₁ = Ω₂ = n

Second Variation and the Minimum Angular Momentum Threshold

Whether this synchronous state is a stable local energy minimum (d²E_tot > 0) or an unstable energy saddle/maximum (d²E_tot < 0) depends on the second derivative of E_tot along the constant-J_tot surface with Ω₁ = Ω₂ = Ω:

d²E_tot / dΩ² = I − I² (dn / dJ_orb)_n=Ω

Since J_orb(n) ∝ n^−1/3, differentiating with respect to n gives dJ_orb / dn = −J_orb / (3n), or dn / dJ_orb = −3n / J_orb. Substituting J_spin = IΩ = In into the second derivative yields:

d²E_tot / dΩ² = I [ 1 − (3*In*) / J_orb ] = I [ 1 − (3J_spin) / J_orb ]

Thus, a synchronized circular binary is secularly stable (d²E_tot / dΩ² > 0) if and only if:

J_orb > 3 J_spin (equivalently, a > a_Darwin = √(3I / μ))

At the threshold separation a_Darwin = √(3(I₁ + I₂) / μ), J_orb = 3J_spin and the total angular momentum attains its absolute minimum value capable of sustaining synchronous rotation:

J_crit = 4 J_spin(a_Darwin) = (4 / 3^3/4) (GM³ μ³ I)^1/4

If J_tot < J_crit (or if a synchronized binary evolves inward past J_orb = 3J_spin), no synchronous equilibrium exists (or the inner synchronous branch is an energy maximum). Any perturbation where the orbit shrinks slightly causes the orbital mechanics frequency n ∝ a^−3/2 to increase three times faster in angular-momentum terms than the stars can spin up. Because n > Ω, tidal bulges on the primary lag behind the companion, exerting a retrograde torque on the orbit that extracts still more orbital angular momentum to spin up the star, accelerating the inward spiral in a runaway feedback loop until the bodies collide and merge.

Contact Binary Minimum Mass Ratio and V1309 Scorpii

In W Ursae Majoris (W UMa) contact binaries, both stars share a common convective envelope whose moment of inertia is dominated by the primary star: I₁ = k₁²M₁R₁², where k₁ ≈ 0.20–0.28 is the dimensionless radius of gyration. As Frederic Rasio (1995) showed, combining the Roche-lobe volume relation R₁ / a = f(q) with the Darwin stability limit μa² ≥ 3I₁ defines a minimum mass ratio q_min = M₂ / M₁ ≈ 0.07–0.09 below which no contact binary can remain tidally stable.

In September 2008, the contact binary V1309 Scorpii erupted as a luminous red nova. Archival OGLE photometry recorded six years of pre-outburst light curves showing a contact binary with period P ≈ 1.4 days whose orbital period was decaying exponentially at an accelerating rate—providing the first direct, real-time observational confirmation of a binary star undergoing Darwin-instability runaway merger.

Hot Jupiter Tidal Inspiral

Most known hot Jupiters (M_p ≈ 10⁻³ M_☉, a ≈ 0.02–0.05 AU) possess orbital angular momentum J_orb smaller than three times the spin angular momentum J_spin = I_*Ω_* that their host star would have if spun up to synchronous corotation (Ω_* = n). Consequently, hot Jupiters with J_tot < J_crit have no stable synchronous state at all, or sit in a metastable pseudo-synchronous state that decays inward on a timescale governed by the stellar modified tidal quality factor Q′_*, as directly measured in the decaying orbit of WASP-12b (Ṗ_orb = −29 ± 2 ms yr⁻¹, inspiral lifetime P / |Ṗ| ≈ 3.2 Myr).

Why is the critical threshold J_orb = 3J_spin rather than J_orb = J_spin?

The factor of 3 arises directly from Kepler's third law (n² = GM / a³). Because orbital angular momentum scales as J_orb ∝ a^1/2 while orbital angular frequency scales as n ∝ a^−3/2, the relation between orbital angular momentum and orbital frequency is J_orb ∝ n^−1/3. Differentiating log n = −3 log J_orb + const shows that a 1% decrease in J_orb causes a 3% increase in orbital frequency n, requiring J_orb to be at least 3 times larger than J_spin = In so that the angular momentum shed by the shrinking orbit is sufficient to spin up the stars by 3%.

Can a binary system with J_orb > 3J_spin still evolve into the Darwin Instability?

Yes. Even if a binary starts safely in the stable synchronous regime (J_orb > 3J_spin), two evolutionary processes drive it across the threshold: (1) as the primary star evolves off the main sequence toward the subgiant branch, its radius R₁ and moment of inertia I₁ = k₁²M₁R₁² expand, raising 3J_spin; and (2) magnetized stellar winds (magnetic braking) strip angular momentum from the star's spin, forcing tides to drain J_orb to maintain synchronization until J_orb drops below 3J_spin.

Why is the Earth–Moon system immune to the Darwin Instability?

In the present-day Earth–Moon system, the Moon's orbital angular momentum (J_orb ≈ 2.89 × 10³⁴ kg m² s⁻¹) is roughly 4.9 times Earth's rotational spin angular momentum (J_spin ≈ 5.86 × 10³³ kg m² s⁻¹), and because the Moon orbits outside synchronous altitude (Ω_⊕ > n), tidal torques transfer angular momentum outward from Earth's spin to the lunar orbit (J_orb increasing). If the Earth–Moon system survived long enough (~50 billion years) to reach full mutual synchronous lock at a ~47-day period, J_orb would exceed 3J_spin by more than an order of magnitude, placing it on the stable outer synchronous branch.

1. Darwin, G. H. (1879). On the precession of a viscous spheroid, and on the remote history of the Earth. Philosophical Transactions of the Royal Society of London, 170, 447–538. https://doi.org/10.1098/rstl.1879.0073
2. Counselman, C. C., III. (1973). Outcomes of tidal evolution. The Astrophysical Journal, 180, 307–316. https://doi.org/10.1086/151964
3. Hut, P. (1980). Stability of tidal equilibrium. Astronomy and Astrophysics, 92(1–2), 167–170. https://ui.adsabs.harvard.edu/abs/1980A%26A....92..167H
4. Tylenda, R., Hajduk, M., Kamiński, T., Udalski, A., Soszyński, I., Szymański, M. K., ... &Ulaczyk, K. (2011). V1309 Scorpii: Merger of a contact binary. Astronomy & Astrophysics, 528, A114. https://doi.org/10.1051/0004-6361/201016221