Goldreich-Julian Charge Density

The Goldreich-Julian charge density (denoted ρ_GJ, with corresponding number density n_GJ = |ρ_GJ| / e) is the fundamental space-charge density that must populate the corotating magnetosphere of a spinning, magnetized neutron star (pulsar) or magnetized black hole in order to short out the electric field component parallel to the magnetic field (E · B = 0). Derived by American astrophysicists Peter Goldreich and William H. Julian in a landmark 1969 Astrophysical Journal paper published just one year after the discovery of radio pulsars, the Goldreich-Julian theorem proves that a rotating magnetized neutron star cannot be surrounded by an empty vacuum. Because rotation of the highly conducting stellar interior through its own magnetic field B acts as a unipolar inductor, an external vacuum would possess a parallel surface electric field strong enough to exert an electrostatic pull on crustal electrons and ions 10⁹ to 10¹² times stronger than surface gravity, ripping charges out of the neutron star crust to fill a force-free corotating magnetosphere Bounded by the light cylinder (R_LC = c / Ω). Within relativistic astronomy, ρ_GJ sets the baseline electrodynamic scale for pulsar polar-cap pair cascades, relativistic pulsar winds, magnetar magnetospheres, and energy extraction from spinning black holes via the Blandford-Znajek process.

Prior to 1969, early electrodynamic models of radio pulsars (such as Pacini 1967 and Ostriker & Gunn 1969) treated a spinning neutron star of radius R ≈ 10 km, angular velocity vector Ω (period P = 2π / Ω), and surface polar magnetic field B_p ≈ 10¹² Gauss (10⁸ Tesla) as a magnetic dipole rotating in an empty vacuum, radiating low-frequency electromagnetic waves at the rotation frequency Ω.

Goldreich and Julian (1969) analyzed the electrodynamics of an aligned rotator (Ω parallel to the magnetic dipole moment μ) whose superconducting/highly conducting interior satisfies the ideal magnetohydrodynamic (MHD) frozen-in condition:

E_int + ( v_rot / c ) × B = E_int + [ (Ω × r) / c ] × B = 0 (Gaussian CGS units; in SI, omit 1/c)

Inside the neutron star (r ≤ R), free charges redistribute themselves to maintain a radial polarization electric field E_int = −[(Ω × r) / c] × B, which generates a quadrupolar electrostatic surface potential Φ(R, θ) = −(Ω B_p R² / 6c) (3 cos²θ − 1) between the magnetic poles (θ = 0, π) and the equator (θ = π/2). For a typical pulsar (P = 0.1 s, B_p = 10¹² G, R = 10⁶ cm), this unipolar-induction voltage drop across the stellar surface exceeds 10¹⁵ to 10¹⁶ Volts.

If the exterior (r > R) were a charge-free vacuum (∇ · E_vac = 0), matching the tangential electric field across the stellar surface (r = R) yields an external quadrupole electric field possessing a huge component parallel to the magnetic field lines (E_vac · B ≠ 0):

E_∥,vac(R, θ) = ( E_vac · B ) / |B| ≈ −( Ω R B_p / c ) cos³θ

Comparing the upward electrostatic force |e E_∥,vac| on a surface electron or iron ion to the downward Newtonian gravitational force G M m / R² on a 1.4 M_☉ neutron star reveals an overwhelming force ratio:

• For Surface Electrons (m = m_e): |e E_∥,vac| / (G M m_e / R²) ≈ 10¹² (for P = 1 s, B_p = 10¹² G)
• For Surface Protons or Iron Nuclei (m ≈ m_p): |e E_∥,vac| / (G M m_p / R²) ≈ 10⁹

Because the electrostatic pull exceeds gravity by nine to twelve orders of magnitude (and easily overcomes the ~keV cohesive work function of the condensed iron crust), the vacuum state is electrodynamically unstable: electric fields strip electrons and ions from the stellar surface and accelerate them out along the magnetic field lines until their collective space charge screens E_∥ to zero (E · B = 0), establishing a charge-separated corotating magnetosphere.

Within the corotating magnetosphere where plasma is tied to the magnetic field lines and rotates rigidly with the star at velocity v = Ω × r (valid inside the light cylinder where |Ω × r| < c), screening of the parallel electric field (E_∥ = 0) requires that the magnetospheric plasma obey the same ideal corotation electric field as the stellar interior:

E(r) = −[ (Ω × r) / c ] × B(r)

Taking the divergence of E via Gauss's law (∇ · E = 4π ρ_GJ in CGS units, or ρ_GJ / ε₀ in SI units) and applying the vector identity ∇ · (A × B) = B · (∇ × A) − A · (∇ × B) with A = Ω × r (for which ∇ × (Ω × r) = 2Ω) gives:

ρ_GJ = ( ∇ · E ) / (4π) = −( 1 / 4πc ) ∇ · [ (Ω × r) × B ] = −( Ω · B ) / ( 2π c ) + [ (Ω × r) · (∇ × B) ] / ( 4π c )

Substituting Ampère's law ∇ × B = (4π / c)J + (1 / c)(∂E/∂t) yields the exact relativistic Goldreich-Julian charge density (in Gaussian CGS units; in SI units, replace 1/(2πc) with −2ε₀ (Ω · B) / [1 − |Ω × r|²/c²]):

ρ_GJ(r) = −( Ω · B ) / ( 2π c ) × 1 / [ 1 − ( r sin θ / R_LC )² ] ≈ −( Ω · B ) / ( 2π c ) (for r ≪ R_LC)

Dividing |ρ_GJ| by the elementary charge e = 4.803 × 10⁻¹⁰ esu gives the Goldreich-Julian number density n_GJ:

n_GJ = |ρ_GJ| / e = |Ω · B| / ( 2π e c ) ≈ 6.94 × 10⁻² ( B_z / 1 Gauss ) ( 1 s / P ) cm⁻³

For a canonical radio pulsar at the stellar surface (B_p = 10¹² G, P = 1 s), n_GJ ≈ 7 × 10¹⁰ charges cm⁻³; for the young Crab Pulsar (P = 0.033 s, B_p ≈ 3.8 × 10¹² G), n_GJ ≈ 8 × 10¹² cm⁻³ at the polar cap.

Charge-Separated Zones and the Holloway Null Surface (Ω · B = 0)

Because ρ_GJ ∝ −(Ω · B), the sign of the Goldreich-Julian space charge depends on the local projection of the magnetic field B onto the spin axis Ω. For an aligned dipole rotator (Ω · μ > 0):

• Polar Regions (|Ω** · B| > 0, θ < 54.7°):** Where dipole field lines point upward/downward with B_z > 0, the required space charge is negative (ρ_GJ < 0, populated by electrons).
• Equatorial Torus (B_z < 0, 54.7° < θ < 125.3°): Where closed dipole field lines curve downward in the equatorial belt, B_z reverses sign and the required space charge is positive (ρ_GJ > 0, populated by positive ions or positrons).
• The Goldreich-Julian / Holloway Null Surface (Ω** · B = 0):** The conical surface where B_z = 0 (at colatitude cos²θ_null = 1/3, or θ_null ≈ 54.7° for a pure aligned dipole) has ρ_GJ = 0, separating the negative polar dome from the positive equatorial torus and serving as the site of outer-gap particle acceleration (Cheng, Ho & Ruderman, 1986; Holloway, 1973).

The Light Cylinder (R_LC = c / Ω) and Polar Cap Pair Multiplicity (κ = n / n_GJ)

Rigid corotation at speed v_φ = Ω r sin θ cannot extend beyond the cylindrical radius where the corotation velocity reaches the speed of light c—the light cylinder:

R_LC = c / Ω = ( c P ) / ( 2π ) ≈ 4.77 × 10⁴ ( P / 1 s ) km

The light cylinder divides the pulsar magnetosphere into two topological regions:

1. Closed Field-Line Zone: Magnetic field lines whose equatorial apex lies inside the light cylinder (r_max < R_LC) form closed loops that corotate with the neutron star, trapping a charge-separated Goldreich-Julian plasma.
2. Open Field-Line Bundle and Polar Caps: Field lines that cross the light cylinder (r_max ≥ R_LC) cannot close and are swept back into an open, relativistic pulsar wind carrying a Poynting flux spin-down luminosity L_sd ≈ (μ² Ω⁴) / c³ (Spitkovsky, 2006). These open field lines root at the neutron star surface within two circular polar caps of radius r_pc = R (R / R_LC)^1/2 ≈ 145 m (P / 1 s)^−1/2.

Because charges continuously escape along the open field lines across the light cylinder, unscreened vacuum-like accelerate gaps (the Ruderman-Sutherland 1975 polar cap gap, the Arons-Scharlemann slot gap, the outer gap, and the equatorial current sheet beyond the Y-point at R_LC) develop where E_∥ ≠ 0. Across these gaps, primary electrons or positrons are accelerated to Lorentz factors γ ~ 10⁶–10⁷ and emit curvature gamma-ray photons (E_γ ≫ 2 m_ec² = 1.022 MeV) that convert into electron–positron (e⁻e⁺) pairs via single-photon magnetic pair production (γ + B → e⁻ + e⁺). This quantum electrodynamic cascade floods the open magnetosphere with a neutral pair plasma whose total number density n_pair exceeds the baseline Goldreich-Julian density by the pair multiplicity factor κ = n_pair / n_GJ ≈ 10³ to 10⁵, until the secondary pairs reach ρ_GJ and screen out E_∥, terminating the pulsar when spin-down drops the polar cap voltage below the pair-creation threshold (the pulsar "death line").

In 1977, Roger D. Blandford and Roman L. Znajek extended the Goldreich-Julian electrodynamic theorem to a spinning Kerr black hole of mass M and horizon angular velocity Ω_H threaded by an external magnetic field B_⊥ supported by a surrounding accretion disk (Blandford-Znajek process). Because relativistic frame dragging inside the ergosphere forces magnetic field lines threading the event horizon to rotate at a field-line angular velocity Ω_F ≈ (1/2) Ω_H, space-time curvature itself induces a general-relativistic Goldreich-Julian charge density:

ρ_GJ,BH ≈ −( Ω_F − ω_LT ) · B / ( 2π c α_lapse )

Unlike a neutron star, a black hole possesses no solid crust from which electrons or ions can be extracted. Consequently, where inflows of disk plasma and outflows of relativistic jet plasma diverge along a stagnation surface near the horizon, a charge-starved black hole magnetospheric gap opens (n < n_GJ, E · B ≠ 0), accelerating leptons that trigger inverse-Compton and photon–photon (γ + γ_disk → e⁻ + e⁺) pair-production cascades to supply the Goldreich-Julian charge density required for force-free relativistic jet launching in active galactic nuclei such as Messier 87.

Why can a rotating magnetized neutron star not exist in an empty vacuum?

Inside the highly conducting neutron star crust, free electrons move to cancel out the Lorentz force of rotation through the star's magnetic field (E + (v × B)/c = 0). Matching this internal electric potential to an external vacuum generates a surface electric field parallel to the magnetic field lines whose electrostatic pull on crustal electrons and ions is 10⁹ to 10¹² times stronger than the star's surface gravity. This electric field immediately rips charges out of the surface until they fill the magnetosphere at the Goldreich-Julian charge density ρ_GJ, shorting out the parallel electric field.

What is the pulsar light cylinder and why must field lines open beyond it?

The light cylinder (R_LC = c / Ω) is the imaginary cylinder coaxial with the pulsar's rotation axis at which a point corotating with the star at angular speed Ω would have to travel at the speed of light c. Because special relativity forbids plasma particles from moving at or faster than c, magnetic field lines that extend beyond R_LC cannot corotate as closed loops; instead, plasma inertia sweeps them backward into an open Archimedes spiral, launching a magnetized relativistic pulsar wind that carries away rotational energy and slows the pulsar down.

What causes a pulsar to cross the "death line" and turn off as a radio source?

Coherent radio emission in pulsars requires continuous electron–positron pair production cascades (n_pair = κ n_GJ with κ ~ 10³–10⁵) inside unscreened polar-cap or outer-gap regions where ρ deviates from ρ_GJ. Because the maximum polar-cap voltage scales as ΔΦ_max ∝ B_p / P², as a pulsar ages and its spin period P lengthens, ΔΦ_max eventually drops below the ~10¹² Volt threshold needed to ignite γ + B → e⁻ + e⁺ pair cascades. At that point, the magnetosphere settles into a quiescent, pair-free Goldreich-Julian electrosphere ("dead pulsar") and radio pulsing ceases.

1. Goldreich, P., & Julian, W. H. (1969). "Pulsar electrodynamics." The Astrophysical Journal, 157, 869–880. https://doi.org/10.1086/150119
2. Ruderman, M. A., & Sutherland, P. G. (1975). "Theory of pulsars: Polar gaps, sparks, and coherent microwave radiation." The Astrophysical Journal, 196, 51–72. https://doi.org/10.1086/153393
3. Blandford, R. D., & Znajek, R. L. (1977). "Electromagnetic extraction of energy from Kerr black holes." Monthly Notices of the Royal Astronomical Society, 179(3), 433–456. https://doi.org/10.1093/mnras/179.3.433
4. Spitkovsky, A. (2006). "Time-dependent force-free pulsar magnetospheres: Axisymmetric and oblique rotators." The Astrophysical Journal, 648(1), L51–L54. https://doi.org/10.1086/507518
5. Philippov, A., & Kramer, M. (2022). "Pulsar magnetospheres and their radiation." Annual Review of Astronomy and Astrophysics, 60, 495–548. https://doi.org/10.1146/annurev-astro-052920-112338