Part 3, Chapter 6

Two-Fluid Theory

Separate treatment of electrons and ions, Hall MHD, and two-fluid wave modes

6.1 Motivation: Beyond Single-Fluid MHD

MHD treats plasma as a single conducting fluid, which is valid at scales much larger than the ion skin depth d_i = c/omega_pi and frequencies much below the ion cyclotron frequency. Two-fluid theory retains the separate identity of electrons and ions, capturing physics that MHD misses:

Hall effect: electrons and ions drift differently, creating a J x B/(ne) term in Ohm's law
Whistler waves: right-hand polarized electromagnetic waves above Omega_ci
Ion cyclotron waves: left-hand polarized waves near Omega_ci
Kinetic Alfven waves: dispersive corrections at k_perprho_s ~ 1
Diamagnetic drifts: species-dependent pressure-driven flows

Scale hierarchy: Two-fluid effects become important at the ion skin depth d_i = c/omega_pi (typically millimeters to centimeters in laboratory plasmas). At even smaller scales (d_e = c/omega_pe), electron inertia effects appear.

6.2 The Complete Two-Fluid Equations

We write down the full set of equations for each species s (= i for ions, e for electrons), derived from the moment equations of Chapter 1.

Continuity (each species s)

$$\frac{\partial n_s}{\partial t} + \nabla \cdot (n_s \mathbf{u}_s) = 0$$

Quasi-neutrality (n_e = Zn_i = n) constrains the two continuity equations but does not eliminate them -- we need both to track the two fluid velocities u_i and u_e independently.

Momentum (each species s)

$$m_s n_s \left(\frac{\partial \mathbf{u}_s}{\partial t} + \mathbf{u}_s \cdot \nabla \mathbf{u}_s\right) = q_s n_s (\mathbf{E} + \mathbf{u}_s \times \mathbf{B}) - \nabla p_s - \nabla\cdot\boldsymbol{\pi}_s + \mathbf{R}_s$$

where:

-- q_i = +Ze, q_e = -e are the species charges
-- p_s = n_sk_BT_s is the scalar pressure
-- pi_s is the viscous stress tensor
-- R_s is the collisional friction force (R_i = -R_e by Newton's third law)

Energy (each species s)

$$\frac{3}{2}n_s\frac{dT_s}{dt} + p_s\nabla\cdot\mathbf{u}_s + \nabla\cdot\mathbf{q}_s + \boldsymbol{\pi}_s:\nabla\mathbf{u}_s = Q_s$$

Each species has its own temperature T_s, heat flux q_s, and collisional heating Q_s. The electron and ion temperatures can differ significantly (T_e is not equal to T_i in general).

Maxwell's Equations

$$\nabla \times \mathbf{E} = -\frac{\partial \mathbf{B}}{\partial t}$$

$$\nabla \times \mathbf{B} = \mu_0 \mathbf{J} + \frac{1}{c^2}\frac{\partial \mathbf{E}}{\partial t}$$

with J = en(u_i - u_e). In the low-frequency limit, the displacement current (1/c²) dE/dt is negligible.

6.3 Derivation of the Generalized Ohm's Law

The generalized Ohm's law is obtained from the electron momentum equation. This is the key equation that distinguishes two-fluid from single-fluid MHD.

Step 1: Write the electron momentum equation

$$m_e n\left(\frac{\partial \mathbf{u}_e}{\partial t} + \mathbf{u}_e\cdot\nabla\mathbf{u}_e\right) = -en(\mathbf{E}+\mathbf{u}_e\times\mathbf{B}) - \nabla p_e + \mathbf{R}_{ei}$$

Step 2: Substitute u_e = v - J/(en)

Using the center-of-mass velocity v approximately equals u_i and u_e = v - J/(en):

$$\mathbf{u}_e\times\mathbf{B} = \left(\mathbf{v} - \frac{\mathbf{J}}{en}\right)\times\mathbf{B} = \mathbf{v}\times\mathbf{B} - \frac{\mathbf{J}\times\mathbf{B}}{en}$$

Step 3: Express the electron inertia in terms of J

The electron inertia term becomes (after some algebra):

$$m_e n\frac{d\mathbf{u}_e}{dt} \approx -\frac{m_e}{e}\frac{d\mathbf{J}}{dt} + \text{(terms involving } m_e/m_i \approx 0\text{)}$$

Step 4: Model collisional friction

$$\mathbf{R}_{ei} = m_e n\nu_{ei}(\mathbf{u}_i - \mathbf{u}_e) = \frac{m_e\nu_{ei}}{e}\mathbf{J} = \frac{\mathbf{J}}{\sigma} = \eta\mathbf{J}$$

where sigma = ne²/(m_enu_ei) is the Spitzer conductivity and eta = 1/sigma is the resistivity.

Step 5: Solve for E

Dividing the electron momentum equation by -en and rearranging:

Result: Generalized Ohm's Law

$$\boxed{\mathbf{E} + \mathbf{v}\times\mathbf{B} = \eta\mathbf{J} + \frac{1}{en}\mathbf{J}\times\mathbf{B} - \frac{1}{en}\nabla p_e + \frac{m_e}{ne^2}\frac{\partial\mathbf{J}}{\partial t}}$$

Each term on the right represents a different physical effect beyond ideal MHD:

Resistive term: eta J

Collisional drag on electrons. Enables magnetic diffusion and reconnection. Dominates at large scales.

Hall term: J x B/(en)

Decouples electron and ion motion. Important at scales ~ d_i. Introduces whistler waves and modifies reconnection.

Electron pressure: -nabla p_e/(en)

Drives diamagnetic drifts and the Biermann battery effect (generates B from crossed density and temperature gradients).

Electron inertia: (m_e/ne²) dJ/dt

Important at the electron skin depth d_e = c/omega_pe. Enables electron-scale reconnection and electromagnetic electron cyclotron waves.

6.4 Dielectric Tensor from Two-Fluid Theory

The two-fluid equations yield a dielectric tensor that describes electromagnetic wave propagation. This recovers the cold plasma dielectric tensor when thermal effects are neglected.

Step 1: Linearize the momentum equations

For each species, assume perturbations proportional to exp(-i omega t + ik dot r):

$$-i\omega m_s n_0 \mathbf{u}_{s1} = q_s n_0(\mathbf{E}_1 + \mathbf{u}_{s1}\times\mathbf{B}_0)$$

We take B₀ = B₀ z-hat and neglect pressure (cold plasma limit).

Step 2: Solve for u_s1 in terms of E₁

Writing out the components with Omega_cs = q_sB₀/m_s:

$$-i\omega u_{sx} = \frac{q_s}{m_s}E_x + \Omega_{cs}u_{sy}$$

$$-i\omega u_{sy} = \frac{q_s}{m_s}E_y - \Omega_{cs}u_{sx}$$

$$-i\omega u_{sz} = \frac{q_s}{m_s}E_z$$

Solving the coupled x-y system (substitute the first into the second):

$$u_{sx} = \frac{q_s}{m_s}\frac{-i\omega E_x + \Omega_{cs}E_y}{\omega^2 - \Omega_{cs}^2}, \quad u_{sy} = \frac{q_s}{m_s}\frac{-i\omega E_y - \Omega_{cs}E_x}{\omega^2 - \Omega_{cs}^2}$$

Step 3: Compute the current and the dielectric tensor

The current is J = sum over s of q_s n₀ u_s1. The dielectric tensor epsilon relates D = epsilon₀ epsilon dot E:

$$\boldsymbol{\epsilon} = \begin{pmatrix} S & -iD & 0 \\ iD & S & 0 \\ 0 & 0 & P \end{pmatrix}$$

with the Stix parameters:

$$S = 1 - \sum_s \frac{\omega_{ps}^2}{\omega^2 - \Omega_{cs}^2}, \quad D = \sum_s \frac{\Omega_{cs}\omega_{ps}^2}{\omega(\omega^2 - \Omega_{cs}^2)}, \quad P = 1 - \sum_s \frac{\omega_{ps}^2}{\omega^2}$$

Step 4: The dispersion relation

Wave propagation at angle theta to B₀ satisfies:

$$\det\left(\frac{c^2k^2}{\omega^2}\begin{pmatrix} \cos^2\theta & 0 & -\sin\theta\cos\theta \\ 0 & 1 & 0 \\ -\sin\theta\cos\theta & 0 & \sin^2\theta\end{pmatrix} - \boldsymbol{\epsilon}\right) = 0$$

This determinantal equation gives the full electromagnetic dispersion relation, including all the wave modes that two-fluid theory captures.

6.5 Hall MHD

Hall MHD retains only the Hall term from the generalized Ohm's law beyond ideal MHD. This is the appropriate model at scales between d_i and d_e.

Step 1: Hall Ohm's law

$$\mathbf{E} + \mathbf{v}\times\mathbf{B} = \frac{1}{en}\mathbf{J}\times\mathbf{B}$$

Dropping the resistive, electron pressure, and electron inertia terms.

Step 2: Derive the Hall induction equation

Substituting into Faraday's law and using J = (nabla x B)/mu₀:

$$\frac{\partial\mathbf{B}}{\partial t} = \nabla\times(\mathbf{v}\times\mathbf{B}) - \nabla\times\left(\frac{\mathbf{J}\times\mathbf{B}}{en}\right)$$

$$= \nabla\times(\mathbf{v}\times\mathbf{B}) - \frac{1}{en\mu_0}\nabla\times\left[(\nabla\times\mathbf{B})\times\mathbf{B}\right]$$

Result: Hall MHD Induction Equation

$$\boxed{\frac{\partial\mathbf{B}}{\partial t} = \nabla\times\left[\left(\mathbf{v} - \frac{\mathbf{J}}{en}\right)\times\mathbf{B}\right] = \nabla\times(\mathbf{u}_e\times\mathbf{B})}$$

In Hall MHD, the magnetic field is frozen to the electron fluid, not the bulk fluid. This is because the electrons, being lighter, carry the current and remain tied to the field lines at scales where ions decouple.

Consequences of the Hall term

-- Breaks the MHD symmetry between left and right circular polarization (introduces handedness)
-- Splits the Alfven wave into two branches: ion cyclotron (left-polarized) and whistler (right-polarized)
-- Modifies magnetic reconnection: the ion diffusion region is much larger than the electron one
-- Introduces dispersive corrections proportional to (kd_i)² to MHD wave modes

6.6 Two-Fluid Wave Modes

Two-fluid theory enriches the MHD wave spectrum considerably. Here we derive the key wave modes.

Whistler Waves (right-hand polarized)

For parallel propagation (k along B), the right-hand polarized mode has the dispersion relation:

$$\frac{c^2k^2}{\omega^2} = 1 - \frac{\omega_{pi}^2}{\omega(\omega - \Omega_{ci})} - \frac{\omega_{pe}^2}{\omega(\omega + |\Omega_{ce}|)}$$

In the low-frequency limit (Omega_ci much less than omega much less than |Omega_ce|):

$$\omega \approx k^2 d_i^2 \Omega_{ci} = \frac{k^2 v_A^2}{\Omega_{ci}} = k^2\frac{c^2}{\omega_{pe}^2}|\Omega_{ce}|\cos\theta$$

The quadratic dependence omega proportional to k² means whistlers are dispersive: higher frequencies travel faster. This is why lightning-generated whistlers descend in pitch.

Ion Cyclotron Waves (left-hand polarized)

The left-hand polarized mode has a resonance at omega = Omega_ci:

$$\frac{c^2k^2}{\omega^2} = 1 - \frac{\omega_{pi}^2}{\omega(\omega + \Omega_{ci})} - \frac{\omega_{pe}^2}{\omega(\omega - |\Omega_{ce}|)}$$

Near the ion cyclotron frequency, for k_perp rho_i much less than 1:

$$\omega \approx \Omega_{ci}\sqrt{1 + k_\perp^2\rho_i^2} \quad\text{(electromagnetic ion cyclotron, EMIC)}$$

These waves are used for ion cyclotron resonance heating (ICRH) in fusion devices.

Kinetic Alfven Wave (KAW)

When k_perprho_s ~ 1 (where rho_s = c_s/Omega_ci is the ion sound gyroradius), the shear Alfven wave acquires a perpendicular group velocity:

$$\omega = k_\parallel v_A\sqrt{1 + k_\perp^2\rho_s^2}$$

The KAW has a significant parallel electric field E_parallel, enabling electron heating and acceleration. It plays a crucial role in the dissipation of turbulence at sub-ion scales.

Connection to MHD modes

In the low-frequency, long-wavelength limit (omega much less than Omega_ci, kd_i much less than 1), both polarizations merge into the single MHD Alfven wave:

$$\omega = k_\parallel v_A \quad\text{(MHD limit)}$$

The two-fluid corrections appear as O(kd_i)² corrections that split the degeneracy.

6.7 Diamagnetic Drifts

Each species has a drift perpendicular to both B and nabla p, even in equilibrium:

Derivation from the momentum equation

In equilibrium, the perpendicular momentum balance for species s gives:

$$0 = q_s n_s \mathbf{u}_{s\perp}\times\mathbf{B} - \nabla_\perp p_s$$

Taking the cross product with B:

$$q_s n_s B^2 \mathbf{u}_{s\perp} = -\nabla p_s\times\mathbf{B}$$

Result: Diamagnetic drift velocity

$$\boxed{\mathbf{u}_{*s} = -\frac{\nabla p_s \times \mathbf{B}}{q_s n_s B^2}}$$

Electron diamagnetic drift

$$\mathbf{u}_{*e} = \frac{\nabla p_e \times \mathbf{B}}{en_e B^2}$$

Positive sign: drifts in the electron diamagnetic direction.

Ion diamagnetic drift

$$\mathbf{u}_{*i} = -\frac{\nabla p_i \times \mathbf{B}}{en_i B^2}$$

Negative sign: drifts opposite to electrons. The net current is the diamagnetic current.

Diamagnetic current and its role

The diamagnetic current density is:

$$\mathbf{J}_* = en(\mathbf{u}_{*i} - \mathbf{u}_{*e}) = -\frac{(\nabla p_i + \nabla p_e)\times\mathbf{B}}{B^2} = -\frac{\nabla p\times\mathbf{B}}{B^2}$$

This is exactly the current needed for pressure balance: J x B = nabla p. The diamagnetic drifts are not associated with a net guiding center drift -- they arise from the inhomogeneity of the distribution function in a pressure gradient.

Interactive Simulation

One-Fluid MHD vs Two-Fluid Wave Dispersion

Python

script.py170 lines

import numpy as np
import matplotlib
matplotlib.use('Agg')
import matplotlib.pyplot as plt

# Compare MHD vs two-fluid wave dispersion relations
# for parallel propagation along B

# Physical parameters (normalized)
mi = 1.0       # ion mass
me = 1.0/1836  # electron mass (realistic ratio)
e_charge = 1.0 # charge
n0 = 1.0       # density
B0 = 1.0       # magnetic field

# Derived quantities
Omega_ci = e_charge * B0 / mi      # ion cyclotron frequency
Omega_ce = e_charge * B0 / me      # electron cyclotron frequency
omega_pi = np.sqrt(n0 * e_charge**2 / mi)  # ion plasma frequency
omega_pe = np.sqrt(n0 * e_charge**2 / me)  # electron plasma frequency
vA = B0 / np.sqrt(n0 * mi)         # Alfven speed
di = vA / Omega_ci                  # ion skin depth = c/omega_pi in normalized units

print(f"=== Plasma Parameters ===")
print(f"Ion cyclotron frequency:  Omega_ci = {Omega_ci:.4f}")
print(f"Electron cyclotron freq:  Omega_ce = {Omega_ce:.2f}")
print(f"Alfven speed:             vA = {vA:.4f}")
print(f"Ion skin depth:           di = {di:.4f}")
print(f"Ion-electron mass ratio:  mi/me = {mi/me:.0f}")

# Wavenumber range
k = np.logspace(-2, 2, 1000)
kdi = k * di

fig, axes = plt.subplots(2, 2, figsize=(14, 10))

# ===== Plot 1: Parallel wave dispersion =====
ax = axes[0, 0]

# MHD Alfven wave: omega = k*vA (same for both polarizations)
omega_mhd = k * vA

# Two-fluid: Right-hand (whistler branch)
# omega/Omega_ci = (k*di)^2 for kdi >> 1 (whistler limit)
# omega = k*vA for kdi << 1 (Alfven limit)
# Full: omega = k*vA * sqrt(1 + (kdi)^2) approximately
# More precisely for parallel R-mode:
omega_R = k * vA * np.sqrt(1 + (k*di)**2)

# Two-fluid: Left-hand (ion cyclotron branch)
# omega -> Omega_ci as k -> infinity
# omega = k*vA for kdi << 1
omega_L = Omega_ci * k * vA / np.sqrt(Omega_ci**2 + k**2 * vA**2)

ax.loglog(kdi, omega_mhd / Omega_ci, 'w--', linewidth=1.5, label='MHD: omega = k*vA')
ax.loglog(kdi, omega_R / Omega_ci, 'cyan', linewidth=2, label='Right (whistler branch)')
ax.loglog(kdi, omega_L / Omega_ci, 'orange', linewidth=2, label='Left (ion cyclotron)')
ax.axhline(1.0, color='red', linestyle=':', alpha=0.5, label='omega = Omega_ci')
ax.axvline(1.0, color='gray', linestyle=':', alpha=0.5)
ax.set_xlabel('k * d_i', fontsize=12)
ax.set_ylabel('omega / Omega_ci', fontsize=12)
ax.set_title('Parallel Propagation: MHD vs Two-Fluid', fontsize=14, color='white')
ax.legend(fontsize=9)
ax.set_facecolor('#0a0a0a')
ax.tick_params(colors='white')
ax.set_xlim(0.01, 100)
ax.set_ylim(0.001, 100)
ax.text(0.02, 0.3, 'MHD
regime', color='gray', fontsize=9)
ax.text(5, 0.05, 'Two-fluid
regime', color='gray', fontsize=9)

# ===== Plot 2: Phase velocity =====
ax = axes[0, 1]
vph_mhd = omega_mhd / k / vA
vph_R = omega_R / k / vA
vph_L = omega_L / k / vA

ax.semilogx(kdi, vph_mhd, 'w--', linewidth=1.5, label='MHD: v_ph = vA')
ax.semilogx(kdi, vph_R, 'cyan', linewidth=2, label='Right (whistler)')
ax.semilogx(kdi, vph_L, 'orange', linewidth=2, label='Left (ion cyclotron)')
ax.axvline(1.0, color='gray', linestyle=':', alpha=0.5, label='k*d_i = 1')
ax.set_xlabel('k * d_i', fontsize=12)
ax.set_ylabel('Phase velocity / v_A', fontsize=12)
ax.set_title('Phase Velocity Comparison', fontsize=14, color='white')
ax.legend(fontsize=9)
ax.set_facecolor('#0a0a0a')
ax.tick_params(colors='white')
ax.set_xlim(0.01, 100)
ax.set_ylim(0, 5)

# ===== Plot 3: KAW dispersion for various k_perp*rho_s =====
ax = axes[1, 0]
cs = 0.5 * vA   # sound speed (beta ~ 0.25)
rho_s = cs / Omega_ci  # ion sound gyroradius

kpar = np.linspace(0.01, 2.0, 200)
kperp_rhos_values = [0, 0.5, 1.0, 2.0, 5.0]
colors_kaw = plt.cm.viridis(np.linspace(0.1, 0.9, len(kperp_rhos_values)))

for i, kpr in enumerate(kperp_rhos_values):
    omega_kaw = kpar * vA * np.sqrt(1 + kpr**2)
    ax.plot(kpar, omega_kaw / Omega_ci, color=colors_kaw[i], linewidth=2,
            label=f'k_perp*rho_s = {kpr:.1f}')

ax.set_xlabel('k_parallel * d_i', fontsize=12)
ax.set_ylabel('omega / Omega_ci', fontsize=12)
ax.set_title('Kinetic Alfven Wave Dispersion', fontsize=14, color='white')
ax.legend(fontsize=9)
ax.set_facecolor('#0a0a0a')
ax.tick_params(colors='white')

# ===== Plot 4: Scale hierarchy diagram =====
ax = axes[1, 1]
# Show the different physics at different scales
scales = np.logspace(-4, 2, 1000)
regime_colors = np.zeros((len(scales), 3))

for i, s in enumerate(scales):
    if s < me/mi:  # electron inertia scale
        regime_colors[i] = [0.8, 0.2, 0.8]  # purple
    elif s < 1:  # between d_e and d_i: Hall physics
        regime_colors[i] = [0.0, 0.8, 0.8]  # cyan
    else:  # MHD
        regime_colors[i] = [0.2, 0.8, 0.2]  # green

# Create a colorbar-style visualization
for i in range(len(scales)-1):
    ax.axvspan(scales[i], scales[i+1], color=regime_colors[i], alpha=0.3)

ax.axvline(me/mi, color='magenta', linewidth=2, linestyle='--', label=f'd_e (electron skin depth)')
ax.axvline(1.0, color='cyan', linewidth=2, linestyle='--', label=f'd_i (ion skin depth)')
ax.axvline(0.1, color='yellow', linewidth=2, linestyle=':', label=f'rho_i (ion gyroradius)')

ax.set_xscale('log')
ax.set_xlabel('Scale / d_i', fontsize=12)
ax.set_title('Scale Hierarchy & Physics Regimes', fontsize=14, color='white')
ax.set_facecolor('#0a0a0a')
ax.tick_params(colors='white')
ax.set_yticks([])
ax.legend(fontsize=9, loc='upper left')

# Add text labels
ax.text(5, 0.7, 'MHD\n(single fluid)', color='lime', fontsize=11,
        ha='center', transform=ax.get_xaxis_transform())
ax.text(0.1, 0.4, 'Hall\nMHD', color='cyan', fontsize=11,
        ha='center', transform=ax.get_xaxis_transform())
ax.text(0.001, 0.4, 'Electron\ninertia', color='magenta', fontsize=10,
        ha='center', transform=ax.get_xaxis_transform())

for ax_item in axes.flat:
    ax_item.spines['bottom'].set_color('white')
    ax_item.spines['left'].set_color('white')

fig.patch.set_facecolor('#111111')
plt.tight_layout()
plt.savefig('output.png', dpi=150, bbox_inches='tight', facecolor='#111111')

print(f"\n=== Key Results ===")
print(f"At k*d_i = 0.1 (MHD regime):")
print(f"  Both modes: omega/Omega_ci ~ {0.1*vA/Omega_ci:.3f} (Alfven wave)")
print(f"At k*d_i = 1 (Hall regime):")
print(f"  Right mode: omega/Omega_ci ~ {1*vA*np.sqrt(2)/Omega_ci:.3f}")
print(f"  Left mode:  omega/Omega_ci ~ {Omega_ci*1*vA/np.sqrt(Omega_ci**2+1*vA**2)/Omega_ci:.3f}")
print(f"At k*d_i = 10 (deep Hall):")
print(f"  Right (whistler): omega/Omega_ci ~ {10*vA*np.sqrt(101)/Omega_ci:.1f}")
print(f"  Left (EMIC): omega/Omega_ci -> {Omega_ci*10*vA/np.sqrt(Omega_ci**2+100*vA**2)/Omega_ci:.3f} (approaching 1)")

print(f"\nPlot saved to output.png")

Click Run to execute the Python code

Code will be executed with Python 3 on the server

Key Takeaways

-- Two-fluid theory treats electrons and ions as separate fluids with their own continuity, momentum, and energy equations
-- The generalized Ohm's law (from electron momentum) contains Hall, electron pressure, resistive, and electron inertia terms beyond ideal MHD
-- The cold plasma dielectric tensor with Stix parameters S, D, P follows directly from linearized two-fluid equations
-- Hall MHD: magnetic field is frozen to the electron fluid, not the bulk plasma; introduces dispersive corrections at scale d_i
-- The MHD Alfven wave splits into whistler (right, omega proportional to k²) and ion cyclotron (left, omega approaches Omega_ci) branches
-- Diamagnetic drifts arise from pressure gradients and are responsible for equilibrium current in confined plasmas

← MHD Instabilities Next: Closure Problem →