Shakura–Sunyaev Accretion Disk

The Shakura–Sunyaev Accretion Disk (often termed the standard α-disk model) is the foundational analytical model describing radiatively efficient, geometrically thin, optically thick gas disks accreting onto black holes, neutron stars, and white dwarfs. Formulated by the Soviet astrophysicists Nikolay Ivanovich Shakura and Rashid Alievich Sunyaev in 1973, the theory resolved a central obstacle in high-energy astrophysics: ordinary microscopic molecular viscosity in ionized astrophysical plasma is roughly ten orders of magnitude too weak to explain the rapid inward mass accretion rates observed in X-ray binaries and active galactic nuclei (AGN). By replacing unknown magnetohydrodynamic and hydrodynamic turbulence with a single dimensionless stress prescription (τ_rφ = αP), Shakura and Sunyaev derived closed-form radial profiles for disk temperature, thickness, density, and multi-temperature blackbody radiation spectra.

Consider gas with surface mass density Σ(r) = 2ρ(r)H(r) orbiting a central compact object of mass M in cylindrical coordinates (r, φ, z). In the standard Shakura–Sunyaev regime, the accretion rate Ṁ is sub-Eddington (L ≲ 0.3 L_Edd), allowing radiative cooling to balance local viscous dissipation on a timescale much shorter than the radial inflow time. Consequently, the vertical pressure scale height H(r) = c_s / Ω_K remains small compared to the cylindrical radius (H / r = c_s / v_K ≪ 1, typically 10⁻³ to 10⁻²), where c_s = √(P / ρ) is the isothermal sound speed and Ω_K(r) = √(GM / r³) is the Keplerian angular velocity.

Because radial pressure gradients are suppressed by (H / r)² ≪ 1, the azimuthal orbital velocity is closely Keplerian, v_φ(r) ≈ rΩ_K(r) = √(GM / r), while the radial inflow velocity v_r ≪ c_s ≪ v_φ is governed by outward angular momentum transport.

Dimensional Scaling of Turbulent Shear Stress

For turbulent eddies or tangled magnetic fields embedded within a shearing disk, the maximum characteristic length scale of a coherent turbulent cell cannot exceed the vertical disk half-thickness H, and the turbulent velocity fluctuations v_turb cannot comfortably exceed the local sound speed c_s without shocking and dissipating into heat. Combining hydrodynamics with magnetic Maxwell stress B_rB_φ / (4π), Shakura and Sunyaev parameterized the vertically averaged rφ-component of the shear stress tensor τ_rφ as directly proportional to the total local pressure P = P_gas + P_rad:

τ_rφ = −ρν r (dΩ_K / dr) = (3 / 2)*ρνΩ*_K = αP

Equivalently, the effective kinematic turbulent viscosity ν is written as:

ν = (2 / 3)α c_s H = (2 / 3)α (c_s² / Ω_K)

Here α is a dimensionless constant constrained physically to 0 < α ≤ 1 (empirically α ≈ 0.01–0.3 in ionized outbursts). In 1991, Steven Balbus and John Hawley proved that the physical mechanism generating this α-viscosity in weakly magnetized Keplerian disks is the magnetorotational instability (MRI).

Integrating the steady-state angular momentum conservation equation from the innermost stable circular orbit r_in = r_ISCO (where shear stress vanishes under the zero-torque boundary condition) to radius r yields the vertically integrated viscous heating rate Q⁺(r) per unit face area of the disk:

Q⁺(r) = (3*GMṀ* / 8*πr*³) × [1 − √(r_in / r)]

Notice that Q⁺(r) is independent of the viscosity parameter α: in steady state, α sets the surface density Σ(r) and radial drift speed v_r(r) = −3ν / (2r), whereas energy conservation fixes the radiative flux F(r) = Q⁺(r) = σ_SBT_eff⁴(r).

Radial Temperature Profile

Equating viscous heating Q⁺(r) to blackbody radiative cooling Q⁻(r) = σ_SBT_eff⁴(r) gives the characteristic radial effective temperature profile:

T_eff(r) = [(3*GMṀ* / 8*πσ*_SBr³) × (1 − √(r_in / r))]^1/4 ∝ M^−1/4 (Ṁ / Ṁ_Edd)^1/4 (r / R_S)^−3/4

Because T_eff(r) scales inversely with black hole mass as M^−1/4 at fixed Eddington ratio Ṁ / Ṁ_Edd, inner disks around stellar-mass black holes (M ≈ 10 M_☉) reach T_eff ≈ 10⁷ K and peak in soft X-rays (~1 keV), whereas disks around supermassive black holes (M ≈ 10⁸ M_☉) in quasars reach T_eff ≈ 10⁵ K, producing the optical–ultraviolet "Big Blue Bump." Summing concentric blackbody annuli over radius produces a power-law spectral energy distribution F_ν ∝ ν^1/3 at intermediate frequencies k_BT_out ≪ hν ≪ k_BT_in.

Shakura and Sunyaev divided the radial structure of the disk into three distinct regimes according to the dominant pressure source (radiation pressure P_rad = aT_c⁴ / 3 versus ideal gas pressure P_gas = ρk_BT_c / μm_p) and the dominant radiative opacity (electron-scattering Thomson opacity κ_es = 0.40 cm² g⁻¹ versus free–free and bound–free Kramers opacity κ_ff ∝ ρT_c^−7/2):

Lightman–Eardley and Thermal Instabilities

In Zone A (where radiation pressure dominates), the disk scale height H becomes independent of radius and proportional to the local accretion rate Ṁ, while surface density scales inversely with accretion rate (∂Σ / ∂Ṁ < 0). As Alan Lightman and Douglas Eardley demonstrated in 1974, this negative slope triggers a viscous clumping instability in which the inner disk breaks into concentric rings, accompanied by a runaway thermal instability (Q⁺ ∝ T_c⁸ growing faster than radiative cooling Q⁻ ∝ T_c⁴). Similarly, in the cool outer regions of Zone C (T ≈ 6,500 K), partial ionization of hydrogen causes opacity to vary steeply with temperature, producing the S-curve thermal–viscous limit-cycle instability that drives dwarf nova outbursts and soft X-ray transient flares.

In 1973, Igor Novikov and Kip Thorne generalized the Newtonian Shakura–Sunyaev model to full Kerr black hole spacetime, showing that the radiative efficiency η = L / (Ṁc²) = 1 − E_ISCO increases from η ≈ 0.0572 for a non-rotating Schwarzschild black hole (r_ISCO = 6*GM* / c²) to η ≈ 0.423 for a maximally spinning prograde Kerr black hole (a_* → 1, r_ISCO → GM / c²). When Ṁ falls far below Ṁ_Edd (producing optically thin, geometrically thick Advection-Dominated Accretion Flows, or ADAFs) or exceeds Ṁ_Edd (producing radiation-trapped Polish doughnuts and slim disks), the thin-disk approximation H / r ≪ 1 breaks down and radial heat advection dominates.

Why can ordinary molecular viscosity not drive astrophysical accretion disks?

In a fully ionized plasma orbiting a stellar-mass compact object at r ~ 10⁹ cm with number density n ~ 10¹⁶ cm⁻³ and temperature T ~ 10⁶ K, the Coulomb mean free path is small and the Reynolds number Re = r v_K / ν_mol exceeds 10¹⁴. The corresponding viscous inflow timescale t_visc ~ r² / ν_mol is longer than the age of the universe, requiring turbulent magnetohydrodynamic stresses parameterized by α ~ 0.01–0.1 to match observed outburst durations of days to weeks.

Does the value of alpha change the total luminosity of a steady-state accretion disk?

No. In a steady-state, radiatively efficient Shakura–Sunyaev disk with a fixed mass accretion rate Ṁ, the total bolometric luminosity L = GMṀ / (2r_in) and the surface temperature profile T_eff(r) are determined purely by virial energy conservation and the conservation of angular momentumflux, independent of α. Instead, α controls how much mass Σ(r) must reside in the disk at any instant to sustain that accretion rate Ṁ (larger α yields lower surface density and faster radial inflow).

How is the Shakura–Sunyaev parameter α measured observationally?

Because α governs the viscous diffusion time t_visc = r² / ν, astronomers measure α empirically by timing the exponential decay and rise phases of dwarf nova and X-ray binary outbursts. During the hot ionized outburst state, observed decay timescales yield α_hot ≈ 0.1–0.3, whereas during the cool, weakly ionized quiescent state, α_cold drops to ≈ 0.01–0.03.

1. Shakura, N. I., & Sunyaev, R. A. (1973). Black holes in binary systems. Observational appearance. Astronomy and Astrophysics, 24, 337–355. https://ui.adsabs.harvard.edu/abs/1973A%26A....24..337S
2. Novikov, I. D., & Thorne, K. S. (1973). Astrophysics of black holes. In Black Holes (Les Astres Occlus) (pp. 343–450). Gordon and Breach. https://ui.adsabs.harvard.edu/abs/1973blho.conf..343N
3. Lightman, A. P., & Eardley, D. M. (1974). Black holes in binary systems: Instability of disk accretion. The Astrophysical Journal Letters, 187, L1–L3. https://doi.org/10.1086/181377
4. Balbus, S. A., & Hawley, J. F. (1991). A powerful local shear instability in weakly magnetized disks. I. Linear analysis. The Astrophysical Journal, 376, 214–222. https://doi.org/10.1086/170270