Plasma Physics: Kinetics, Fields and Waves: /20 Summary

Context and scope

the supplied physics reference, Applications of Classical Physics — Chapters 20–23 Particle Kinetics · Cold-Plasma Waves · Kinetic Theory of Warm Plasmas · Nonlinear Dynamics

/20 Summary

• Plasma = quasineutral ionized gas exhibiting collective behavior at scales larger than the Debye length.
• Three defining length / time scales:
  • Debye length $\lambda_D = \sqrt{\epsilon_0 k_B T/(n_e e^2)}$ — screening of charges.
  • Plasma frequency $\omega_p = \sqrt{n_e e^2/(\epsilon_0 m_e)}$ — fundamental oscillation.
  • Plasma parameter $\Lambda = n_e\lambda_D^3 \gg 1$ — collective regime, weak coupling.
• Three descriptions, increasing detail:
  1. MHD (Part V, Ch. 19): treats plasma as a single conducting fluid.
  2. Two-fluid (Ch. 21): separate electron and ion fluids — captures plasma waves.
  3. Kinetic / Vlasov (Ch. 22): full distribution function $f(\mathbf{x}, \mathbf{v}, t)$ — captures wave-particle resonances, Landau damping.
• Single-particle motion decomposes into rapid gyration + slow guiding-center drift when $r_L \ll L$. Adiabatic invariants ($\mu$, $J$, $\Phi$) preserved.
• Landau damping — collisionless dissipation: energy transferred from waves to resonant particles ($v \approx \omega/k$) without entropy generation in the Vlasov equation itself (it's a phase-mixing effect).
• Nonlinear plasma physics = trapping, quasilinear diffusion, mode coupling, solitons, anomalous resistivity. Where MHD fails and the field stays alive.
• Reach: fusion devices, ionosphere, magnetosphere, solar wind, stellar coronae, AGN jets, ICM, ISM, accretion disks, lab discharges, semiconductors at extreme conditions.

Master Map

mindmap
  root((Plasma))
    Ch.20 Particle Kinetics
      Debye λ_D
      Plasma freq ω_p
      Plasma param Λ
      Quasineutrality
      Gyration
        Ω_c = qB/m
        Larmor r_L
      Drifts
        E×B
        grad-B
        curvature
        polarization
        gravitational
      Adiabatic invariants
        μ magnetic moment
        J longitudinal
        Φ flux
      Magnetic mirrors
      Loss cone
      Collisions
        Coulomb log
        Spitzer resistivity
    Ch.21 Cold Plasma Waves
      Two-fluid eqs
      Langmuir wave
      EM wave ω² = ω_p² + c²k²
      Cutoffs / resonances
      O / X modes
      Whistler
      Alfvén / MS recap
      Faraday rotation
      Dielectric tensor
    Ch.22 Warm/Kinetic
      Vlasov eq
      BBGKY
      Plasma dispersion Z(ζ)
      Landau damping
      Bump on tail
      Two-stream
      Penrose criterion
      Echoes
      Cyclotron resonance
    Ch.23 Nonlinear
      Trapping
      Quasilinear diffusion
      Wave-wave coupling
      Ion-acoustic soliton (KdV)
      Double layers
      Fermi acceleration
      Anomalous resistivity
      Plasma turbulence

Particle Kinetics of Plasma

What is a Plasma?

Definition (operational): ionized gas where collective effects dominate, i.e., all three criteria hold:

1. $L \gg \lambda_D$ (system larger than screening length)
2. $\Lambda = n_e\lambda_D^3 \gg 1$ (many particles in Debye sphere)
3. $\omega\tau \gg 1$ where $\tau$ = collision time (plasma oscillations faster than collisions)

Below any of these: weakly ionized gas, or strongly coupled (dusty / liquid) plasma, but not standard collective plasma.

Debye Shielding

Insert test charge $Q$ at origin in plasma at temperature $T$. Boltzmann distribution + Poisson:

$$\nabla^2\phi - \phi/\lambda_D^2 = -Q\delta^3(\mathbf{r})/\epsilon_0$$

Solution: screened Coulomb (Yukawa) potential:

$$\phi(r) = \frac{Q}{4\pi\epsilon_0 r}\,e^{-r/\lambda_D}$$

Debye length: $$\boxed{\lambda_D = \sqrt{\frac{\epsilon_0 k_B T_e}{n_e e^2}}}$$

(For unequal $T_e, T_i$: include both species, $\lambda_D^{-2} = \sum_s \lambda_{D,s}^{-2}$.)

Plasma parameter: $$\Lambda \equiv n_e\,\lambda_D^3 = \frac{1}{4\pi}\frac{(\epsilon_0 k_BT_e)^{3/2}}{n_e^{1/2} e^3}$$

A "weakly coupled" plasma has $\Lambda \gg 1$ — Coulomb collisions are rare per gyration / oscillation period.

Typical $(n, T, \lambda_D, \omega_p, \Lambda)$ values

Plasma Frequency

Displace all electrons by $\xi$; restoring force from the resulting space charge. Solving Poisson + Newton:

$$\boxed{\omega_p = \sqrt{\frac{n_e e^2}{\epsilon_0 m_e}} = 5.64\times 10^4\,\sqrt{n_e/\text{m}^{-3}}\,\text{rad/s}}$$

Ions oscillate too but at lower frequency $\omega_{p,i} = \sqrt{m_e/m_i}\,\omega_p \ll \omega_p$.

Connection: $\lambda_D \omega_p = v_{T,e} = \sqrt{k_BT_e/m_e}$ (electron thermal speed) — Debye length is one thermal-speed step per plasma period.

Single-Particle Motion in EM Fields

Uniform $\mathbf{B}$: Gyration

Equation: $m\dot{\mathbf{v}} = q\,\mathbf{v}\times\mathbf{B}$.

Solution: helical motion with axis $\parallel \mathbf{B}$.

Magnetic moment (the famous adiabatic invariant):

$$\boxed{\mu = \frac{m v_\perp^2}{2 B}}$$

— invariant if $B$ varies slowly on the gyration timescale ($|\dot B/B| \ll \Omega_c$).

Uniform $\mathbf{E} + \mathbf{B}$: $\mathbf{E}\times\mathbf{B}$ Drift

Average velocity of guiding center:

$$\boxed{\mathbf{v}_E = \frac{\mathbf{E}\times\mathbf{B}}{B^2}}$$

Properties:

• Independent of charge and mass — all particles drift the same way.
• Perpendicular to both $\mathbf{E}$ and $\mathbf{B}$.
• No net current ⇒ no Joule heating.

Other Drifts (Guiding-Center Approximation)

General formula for slowly-varying force $\mathbf{F}$: $$\mathbf{v}_d = \frac{\mathbf{F}\times\mathbf{B}}{q B^2}$$

The charge-dependent drifts produce currents and feed back on the field (e.g., ring current in Earth's magnetosphere, ion-acoustic instabilities).

Adiabatic Invariants

When motion has $N$ separable periodic degrees of freedom, $N$ action integrals $J_i = \oint p_i\,dq_i$ are adiabatic invariants — conserved if external parameters change slowly compared to oscillation period.

For magnetized particles:

Magnetic Mirrors

In a field with $B$ varying along its lines (e.g., terminating at "high-$B$ throats"), $\mu$-conservation gives:

$$v_\perp^2/B = \text{const}, \quad \tfrac{1}{2}m(v_\perp^2 + v_\|^2) = \text{const (energy)}$$

⇒ as particle moves to higher $B$, $v_\perp$ increases, $v_\|$ decreases, until $v_\| = 0$ → particle reflects.

Mirror ratio: $R = B_{\max}/B_{\min}$.

Loss cone (half-angle from $\mathbf{B}$ at $B_{\min}$): $$\sin^2\theta_{\text{lc}} = 1/R$$

Particles with $\sin^2\theta_0 < 1/R$ escape; others are trapped. Foundation of magnetic confinement (mirror machines, tokamaks, magnetosphere).

Diffusion into loss cone drives auroral precipitation and tokamak ion losses.

Collisions in a Plasma

Coulomb collisions are small-angle (long-range $\propto 1/r$). The cumulative effect of many small kicks dominates over rare large-angle scatterings.

Coulomb logarithm: $\ln\Lambda \approx \ln(\lambda_D/b_{90})$, where $b_{90}$ = impact parameter for 90° deflection.

Typical $\ln\Lambda \approx 10$–$20$ for laboratory and astrophysical plasmas. Weak dependence on parameters — often treated as constant.

Collision Frequencies

Electron-electron collision frequency (90° scatter equivalent): $$\nu_{ee} \approx \frac{n_e e^4 \ln\Lambda}{4\pi\epsilon_0^2 m_e^2 v_{T,e}^3} = \frac{\omega_p\ln\Lambda}{4\pi\Lambda}$$

Key inequality: $\nu_{ee}/\omega_p \sim \ln\Lambda/\Lambda \ll 1$ in weakly coupled plasmas ⇒ collective effects faster than collisions.

Spitzer resistivity (electrons scattering off ions): $$\boxed{\eta_{\text{Sp}} = \frac{m_e\nu_{ei}}{n_e e^2} = \frac{e^2 m_e^{1/2}\ln\Lambda}{(4\pi\epsilon_0)^2(k_B T_e)^{3/2}}\times\text{const}}$$

$\eta \propto T_e^{-3/2}$ — hot plasmas are very good conductors. This is why fusion plasmas don't need explicit resistivity in models (they're nearly ideal), but reconnection still happens because of anomalous (turbulent) resistivity.

Cold-Plasma Waves (Two-Fluid Formalism)

Two-Fluid Equations

Treat electrons and ions as separate fluids:

$$\partial_t n_s + \nabla\cdot(n_s\mathbf{u}_s) = 0$$

$$m_s n_s(\partial_t + \mathbf{u}_s\cdot\nabla)\mathbf{u}_s = n_s q_s(\mathbf{E} + \mathbf{u}_s\times\mathbf{B}) - \nabla P_s + \mathbf{R}_s$$

with $\mathbf{R}_s$ = collisional momentum transfer. Plus Maxwell's equations with sources $\rho = \sum n_s q_s$, $\mathbf{J} = \sum n_s q_s\mathbf{u}_s$.

Cold plasma $\Rightarrow$ drop $\nabla P_s$ (set $T \to 0$). Captures dispersive wave physics but not Landau damping (need Ch. 22).

Langmuir (Plasma) Oscillation

Linearize about uniform background; ions immobile. Continuity + momentum + Poisson give:

$$\omega^2 = \omega_p^2 + 3 v_{T,e}^2 k^2 \quad \text{(Bohm-Gross)}$$

— "cold" $\to \omega = \omega_p$ for all $k$ (longitudinal, no group velocity). Warm gives slight dispersion.

Properties:

• Longitudinal: $\mathbf{E}\parallel\mathbf{k}$, $\mathbf{B}_1 = 0$ (electrostatic).
• Sometimes called "plasma oscillations" or "Langmuir waves" (Tonks-Langmuir 1929).

EM Waves in Unmagnetized Plasma

Linearize and seek transverse $\mathbf{E}_1, \mathbf{B}_1 \perp \mathbf{k}$. Get:

$$\boxed{\omega^2 = \omega_p^2 + c^2 k^2}$$

Properties:

• Cutoff at $\omega = \omega_p$: waves with $\omega < \omega_p$ cannot propagate (purely evanescent).
• High-frequency: approaches vacuum $\omega = ck$ for $\omega \gg \omega_p$.
• Phase velocity $v_\phi = c/\sqrt{1 - \omega_p^2/\omega^2} > c$.
• Group velocity $v_g = c\sqrt{1 - \omega_p^2/\omega^2} < c$.
• Refractive index $n = c k/\omega = \sqrt{1 - \omega_p^2/\omega^2}$.

Ionospheric reflection: $\omega_p$ in ionosphere is a few MHz; shortwave radio bounces off the underside. Above $\sim 30$ MHz (limit of $f_p$): transparent, used for satellite communication.

Magnetized Plasma: O and X Modes

In presence of $\mathbf{B}_0$, plasma response is anisotropic; the dielectric tensor $\boldsymbol\epsilon(\omega)$ has multiple components (Stix tensor).

For propagation perpendicular to $\mathbf{B}_0$:

X-mode cutoffs: $\omega_R = \tfrac{1}{2}[\Omega_e + \sqrt{\Omega_e^2 + 4\omega_p^2}]$ (right-hand cutoff), and $\omega_L = \tfrac{1}{2}[-\Omega_e + \sqrt{\Omega_e^2 + 4\omega_p^2}]$ (left-hand).

Upper hybrid resonance: $\omega_{UH}^2 = \omega_p^2 + \Omega_e^2$.

Parallel Propagation: R/L Waves, Whistlers

For $\mathbf{k}\parallel\mathbf{B}_0$, dispersion factorizes into right-hand circularly polarized (R) and left-hand (L):

$$n_{R,L}^2 = 1 - \frac{\omega_p^2/\omega}{\omega \mp \Omega_e}$$

• R-mode ($\omega < \Omega_e$): whistler waves, frequency rises from below — used for ionospheric/magnetospheric remote sensing. Famous "whistler" sound (Eckersley, Storey) from lightning.
• L-mode ($\omega < \Omega_i$): ion cyclotron waves.
• Faraday rotation: different phase speeds for R and L cause rotation of linearly polarized light through magnetized plasma — diagnostic for $B_\|$ in ISM, galaxy clusters.

CMA Diagram & Stix Tensor

The Clemmow-Mullaly-Allis (CMA) diagram plots wave-propagation regimes in $(\omega_p^2/\omega^2, \Omega_e^2/\omega^2)$ space: shows cutoffs, resonances, and allowed modes vs orientation. Master tool for radio-frequency heating, propagation in magnetosphere.

Stix dielectric tensor (in $\mathbf{B}_0 = B_0\hat z$ frame):

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

with $S = \tfrac{1}{2}(R+L)$, $D = \tfrac{1}{2}(R-L)$, $P = 1 - \omega_p^2/\omega^2$. The dispersion equation factors elegantly.

Recap: MHD Waves

From Ch. 19: Alfvén $\omega = v_A k_\|$, fast/slow magnetosonic. These are the low-frequency limit ($\omega \ll \Omega_i$) of the full plasma-wave menu.

flowchart TD
    A[Plasma wave taxonomy] --> B[Low ω, MHD]
    A --> C[ω ~ Ω_i: ion-cyclotron]
    A --> D[ω ~ ω_p: Langmuir]
    A --> E[ω ~ Ω_e: whistlers, ECR]
    A --> F[ω >> ω_p: transparent]
    B --> G[Alfvén, slow MS, fast MS]
    C --> H[L-mode, ion-Bernstein]
    D --> I[Langmuir, EM]
    E --> J[R-mode, X-mode]

Kinetic Theory of Warm Plasmas

The Vlasov Equation

The single most important equation of plasma kinetic theory.

Distribution function $f_s(\mathbf{x}, \mathbf{v}, t)$ for species $s$. Collisionless evolution:

$$\boxed{\frac{\partial f_s}{\partial t} + \mathbf{v}\cdot\nabla_{\mathbf{x}} f_s + \frac{q_s}{m_s}(\mathbf{E} + \mathbf{v}\times\mathbf{B})\cdot\nabla_{\mathbf{v}} f_s = 0}}

— Boltzmann eq. with the collision integral removed and self-consistent mean fields $\mathbf{E}$, $\mathbf{B}$ obeying Maxwell's eqs. with sources from $f_s$:

$$\rho = \sum_s q_s\int f_s\,d^3v, \quad \mathbf{J} = \sum_s q_s\int\mathbf{v}\,f_s\,d^3v$$

Vlasov-Poisson (electrostatic): drop $\mathbf{B}$, replace Ampère with Poisson $\nabla\cdot\mathbf{E} = \rho/\epsilon_0$.

Properties of Vlasov:

• Phase-space density is conserved along characteristics ("Vlasov-Liouville theorem").
• $f \log f$ is conserved (no entropy production!).
• Infinite number of conserved "Casimirs" $\int g(f)\,d^3x\,d^3v$.
• Time-reversible.

Yet damping occurs. Resolution: phase-mixing / Landau damping — see §22.4.

BBGKY Hierarchy → Vlasov

Bogoliubov-Born-Green-Kirkwood-Yvon hierarchy: equations for $N$-particle correlations.

• $f_1$ (single-particle distribution) couples to $f_2$ (pair correlations).
• $f_2$ couples to $f_3$, etc.
• Truncation: for $\Lambda \gg 1$, pair correlations $\sim 1/\Lambda$. Neglect $\Rightarrow$ Vlasov.
• Keep $f_2$ at leading order $\Rightarrow$ Landau or Lenard-Balescu collision operator (small-angle Coulomb dynamics).

Linearized Vlasov & Plasma Dispersion Function

Linearize $f_s = f_{s,0}(\mathbf{v}) + f_{s,1}$, $\mathbf{E} = \mathbf{E}_1$. Fourier transform: $e^{i(\mathbf{k}\cdot\mathbf{x} - \omega t)}$.

For 1-D electrostatic, ions immobile, plasma:

$$1 + \frac{\omega_p^2}{k^2}\int\frac{\partial f_0/\partial v}{v - \omega/k}\,dv = 0$$

— a singular integral when $v = \omega/k$. Landau's prescription (causality, $t\to+\infty$): integrate along contour deformed below the singularity:

$$\int\frac{\partial f_0/\partial v}{v - \omega/k}\,dv = \text{P.V.}\int + i\pi\,\frac{\partial f_0}{\partial v}\bigg|_{v = \omega/k}$$

Plasma dispersion function (Fried-Conte):

$$Z(\zeta) = \frac{1}{\sqrt\pi}\int_{-\infty}^\infty \frac{e^{-x^2}}{x - \zeta}\,dx, \quad \text{Im}\,\zeta > 0$$

with analytic continuation. Workhorse for Maxwellian plasma linear theory.

Landau Damping

For a Maxwellian electron distribution and $\omega/k \gg v_T$:

$$\omega^2 \approx \omega_p^2(1 + 3 k^2 \lambda_D^2)$$

with damping rate (real part of $\omega$ → $\omega_r + i\gamma$):

$$\boxed{\gamma_L = -\sqrt{\tfrac{\pi}{8}}\,\frac{\omega_p}{(k\lambda_D)^3}\,\exp\!\left[-\frac{1}{2 k^2\lambda_D^2} - \tfrac{3}{2}\right]}$$

Negative ⇒ wave amplitude decays as $e^{\gamma_L t}$.

Crucial physical interpretation:

flowchart LR
    A[Wave at phase velocity v_p = ω/k] --> B[Particles with v ≈ v_p see slowly varying E]
    B --> C{Slope of f₀ at v_p?}
    C -->|∂f/∂v < 0| D[More slow than fast particles]
    D --> E[Net energy from wave to particles]
    E --> F[Wave damps]
    C -->|∂f/∂v > 0| G[More fast than slow]
    G --> H[Net energy from particles to wave]
    H --> I[Instability!]

Key facts about Landau damping:

• Reversible at single-particle level (Vlasov is reversible).
• "Damping" is phase mixing — energy redistributes from coherent wave to incoherent particle motion.
• Demonstrated experimentally (Malmberg & Wharton 1964, after 19 years of theory).
• Mathematically: time-irreversible behavior emerging from reversible equation via initial conditions + spectral analysis.
• Sometimes "plasma echo" (Gould-O'Neil-Malmberg 1967): two waves at $t_1, t_2$ produce coherent response at $2t_2 - t_1$ — phase memory recovered. Direct evidence of non-dissipative damping.

Linear Instabilities

Landau formula $\gamma_L \propto \partial f_0/\partial v|_{v = \omega/k}$ ⇒ instability if there's positive slope in the distribution.

Two-Stream Instability

Two cold beams ($\pm v_0$) ⇒ rich double-peaked distribution. Cold-plasma analysis gives:

$$\gamma_{\max} \sim \omega_p/\sqrt{8}\quad \text{at}\quad k v_0 = \omega_p/\sqrt{2}$$

— growth rate of order $\omega_p$, exponentially fast. Prototype for beam-plasma interactions, particle accelerator wakefields, type-III solar radio bursts.

Bump-on-Tail Instability

Maxwellian + small high-velocity bump (e.g., suprathermal beam). $\partial f/\partial v > 0$ on inner edge of bump ⇒ Langmuir waves grow.

Saturation: quasilinear plateau formation (Ch. 23).

Penrose Criterion

Necessary + sufficient for instability of 1-D electrostatic mode: $$\int \frac{\partial f_0/\partial v}{v - v_0}\,dv > 0 \text{ for some } v_0 \text{ at which } f_0' = 0$$

⇒ a "dip" in distribution between two maxima ⇒ instability.

Other Instabilities

Magnetized Kinetic Theory

In presence of $\mathbf{B}_0$, particle motion is no longer free-streaming — it's helical. Vlasov equation in cylindrical $(\mathbf{v}_\perp, v_\|)$:

Resonance condition for wave at $(\omega, \mathbf{k})$:

$$\boxed{\omega - k_\| v_\| - n\Omega_c = 0}$$

— $n = 0$: Landau (parallel transit). $n = \pm 1, \pm 2,\ldots$: cyclotron harmonics.

Cyclotron damping ($n = 1$): wave at frequency $\omega$ damps on particles for which $v_\| = (\omega - \Omega_c)/k_\|$. Foundation of ICRH (ion-cyclotron resonance heating) in fusion devices.

Nonlinear Dynamics of Plasmas

Particle Trapping in a Wave

Linear theory assumes infinitesimal $\mathbf{E}_1$. Beyond threshold, particles with $v\approx v_p$ are trapped in the wave potential, executing closed orbits in the wave frame.

Trapping width (velocity range trapped): $$\Delta v_{\text{trap}} = 2\sqrt{e\phi_1/m_e}$$

Bounce frequency (oscillation in trap): $$\omega_b = \sqrt{e\phi_1 k^2/m_e} = k\sqrt{e\phi_1/m_e}$$

Trapping breaks Landau linear analysis at amplitudes $\phi_1 \gtrsim$ thermal energy / $e$.

O'Neil saturation: when $\omega_b \sim \gamma_L$, linear damping/growth halts and oscillates at $\omega_b$.

Quasilinear Theory

When wave amplitudes are weak but the spectrum is broad: particles diffuse in velocity space due to scattering off many uncorrelated waves.

QL diffusion equation:

$$\frac{\partial f_0}{\partial t} = \frac{\partial}{\partial v_\|}\!\left[D_{QL}(v_\|)\,\frac{\partial f_0}{\partial v_\|}\right]$$

with $D_{QL}(v_\|) = (\pi e^2/m_e^2)\int|E_k|^2\,\delta(\omega_k - k v_\|)\,dk$.

Result: flattening of $\partial f/\partial v$ at resonant velocities — plateau formation. Saturates the bump-on-tail instability.

QL theory bridges deterministic Vlasov and statistical fluid descriptions. Foundation of weak plasma turbulence.

Wave-Wave Interactions (Mode Coupling)

Three-wave resonance: $\omega_1 + \omega_2 = \omega_3$, $\mathbf{k}_1 + \mathbf{k}_2 = \mathbf{k}_3$ (energy + momentum).

Examples in plasmas:

• Decay instability: large-amplitude Langmuir wave decays into Langmuir + ion-acoustic.
• Stimulated Raman scattering: EM wave + Langmuir → EM (lower frequency). Limits laser-plasma interaction efficiency.
• Stimulated Brillouin scattering: EM + ion-acoustic → EM.

Four-wave coupling at higher order; relevant to Langmuir collapse (Zakharov), strong plasma turbulence.

Solitons in Plasmas

Ion-Acoustic Solitons (KdV)

In a weakly nonlinear, weakly dispersive ion-acoustic wave:

$$\frac{\partial \phi}{\partial t} + (1 + \alpha\phi)\frac{\partial\phi}{\partial x} + \beta\frac{\partial^3\phi}{\partial x^3} = 0 \quad \text{(KdV)}$$

Soliton solution:

$$\phi(x,t) = \phi_0\,\mathrm{sech}^2\!\left[\sqrt{\alpha\phi_0/(12\beta)}\,(x - ct)\right]$$

— amplitude-dependent speed $c \propto \phi_0$. Solitons collide elastically (preserve identity).

Observed in low-temperature plasma columns, planetary magnetospheres.

Langmuir Solitons & Zakharov Equations

Coupled NLSE-like system describing Langmuir wave envelope + ion density. Predicts Langmuir collapse — wave field intensifies in shrinking region until kinetic effects (trapping) intervene.

Double Layers and Beam-Plasma Phenomena

Double layer: localized potential jump of order $T_e/e$ over a few Debye lengths, separating regions of different potential. Drives field-aligned acceleration (auroral electrons!). Sustained by current.

Plasma sheath: boundary layer at conductor immersed in plasma; thickness ~ $\lambda_D$.

Particle Acceleration

Fermi Acceleration

Particles bouncing between converging magnetic "mirrors" gain energy on each encounter.

• First-order: systematic compression → power-law spectrum $f(p)\propto p^{-q}$ with $q = 3 - 2/(r-1)$ for shock compression ratio $r$. For strong shock ($r=4$): $q = 4$ ⇒ $dN/dE \propto E^{-2}$.

• Second-order: stochastic gains/losses on moving inhomogeneities. Slower; energy ∝ $\langle v_{\text{scattering}}^2\rangle$.

Diffusive shock acceleration (DSA): the standard model for galactic cosmic-ray origin. Strong supernova-remnant shocks produce $E^{-2}$ source spectra, modified by propagation losses to observed $E^{-2.7}$.

Stochastic Heating

For finite-amplitude waves, phase-space dynamics becomes chaotic above a threshold (Chirikov criterion: overlap of resonance islands). Above threshold, particles random-walk in energy → effective heating.

Foundation of: lower-hybrid heating in fusion plasmas, ion cyclotron heating, auroral kilometric radiation.

Plasma Turbulence

Weak Turbulence

QL theory + wave-wave interactions. Energy spectrum from cascade among normal modes. Successful for: ionospheric F-region, ICRH-heated tokamak plasmas, laser-driven coronae.

Strong Turbulence

Beyond weak: large-amplitude coherent structures (cavitons, filaments, Langmuir collapse). Resists analytical treatment.

Kolmogorov-like cascades exist in MHD turbulence:

• Iroshnikov-Kraichnan (1965): $E(k)\propto k^{-3/2}$ — wave-packet interactions in MHD.
• Goldreich-Sridhar (1995): anisotropic cascade — $k_\perp$ grows faster than $k_\|$. Predicts $E(k_\perp)\propto k_\perp^{-5/3}$, $k_\| \propto k_\perp^{2/3}$.

Solar wind, ICM, ISM all show MHD-turbulence spectra consistent with G-S.

Anomalous Resistivity & Reconnection

When current density $J$ drives velocity drift $v_d > v_T$: current instability (Buneman, ion-acoustic) generates turbulent fluctuations. Wave-particle scattering → effective collision frequency $\nu_{\text{eff}} \gg \nu_{\text{Spitzer}}$:

$$\eta_{\text{anom}} \sim \frac{m_e\,\omega_p\,\delta n/n}{n_e e^2}$$

— orders of magnitude above classical. Resolves the Sweet-Parker reconnection paradox (Part V Ch. 19): real reconnection in solar flares, magnetospheric substorms is fast because anomalous resistivity (or collisionless electron physics: Hall, electron inertia) takes over in thin current sheets.

Workflow / Process

flowchart TD
    A[Plasma problem] --> B{Length scale L vs λ_D?}
    B -->|L >> λ_D and Λ >> 1| C[Collective plasma]
    B -->|L < λ_D| D[Single-particle physics]
    C --> E{Collisional or collisionless?}
    E -->|ω τ_coll < 1| F[Fluid/MHD]
    E -->|ω τ_coll > 1| G[Need kinetic]
    F --> H{One- or two-fluid?}
    H -->|Low ω, large scale| I[MHD]
    H -->|High ω, plasma waves| J[Two-fluid cold/warm]
    G --> K[Vlasov-Maxwell]
    K --> L{Linear?}
    L -->|Yes| M[Solve dispersion D(ω,k) = 0]
    M --> N[Landau prescription if singular]
    N --> O[Damping or instability]
    L -->|No, large amplitude| P[Trapping, QL, mode coupling]
    A --> Q{Magnetized?}
    Q -->|Yes, ω_c >> ω| R[Guiding-center: drifts, adiabatic invariants]
    Q -->|Yes, ω ~ ω_c| S[Full kinetic with cyclotron resonance]
    Q -->|No| T[Isotropic plasma]

Comparison Tables

Plasma Length/Time Scales

Three Descriptions Compared

Wave Modes Summary

Plasma Drifts

Instabilities: Quick Reference

Common Mistakes

• ❌ Calling a partly ionized gas a "plasma" without checking $\Lambda\gg 1$ and $L\gg\lambda_D$.
• ❌ Using cold-plasma dispersion for kinetic phenomena (e.g., Landau damping). Must use Vlasov.
• ❌ Forgetting the Landau contour deformation. Causality (initial value) forces below-pole prescription; PV alone gives wrong damping sign.
• ❌ Treating ω = ω_p as the only Langmuir wave. Warm correction $\omega^2 = \omega_p^2 + 3 v_T^2 k^2$ matters for short wavelengths.
• ❌ Mixing SI and Gaussian units in formulas. $\omega_p^2 = n_e e^2/(\epsilon_0 m_e)$ in SI, $4\pi n_e e^2/m_e$ in Gaussian.
• ❌ Using guiding-center theory at high $\omega \sim \Omega_c$. Validity is $\omega\ll\Omega_c$, $r_L\ll L$.
• ❌ Confusing $\mu$ (magnetic moment) with $\mu_0$ (permeability of vacuum). Notation collision!
• ❌ Treating Landau damping as dissipation in entropy sense. Vlasov conserves entropy; damping is phase-mixing.
• ❌ Applying Landau formula outside its validity ($\omega \gg k v_T$). For $\omega \sim k v_T$ (e.g., ion-acoustic in $T_i \sim T_e$), full $Z$-function calculation needed.
• ❌ Forgetting the species sum in dielectric. Both electrons and ions contribute; ions matter for low-frequency waves.
• ❌ Using Spitzer resistivity in fast-reconnection contexts. Anomalous / Hall / electron-inertia effects dominate at small scales.
• ❌ Stating "all" plasmas are quasineutral. Sheaths and double layers violate it locally; very thin transitions over $\lambda_D$.
• ❌ Conflating two-stream and bump-on-tail. Two-stream needs comparable beams; bump-on-tail is a small perturbation on Maxwellian. Different growth rates.
• ❌ Treating reconnection as purely fluid (MHD). Inside the diffusion region, kinetic effects (Hall, EMHD, two-fluid) essential at PIC-resolved scales.
• ❌ Confusing whistlers with Alfvén waves. Whistlers are dispersive (R-mode), Alfvén non-dispersive in cold plasma.
• ❌ Using Penrose criterion in multi-dimensional cases without care. Criterion is for 1-D electrostatic; full kinetic theory needed for 2D/3D.
• ❌ Applying ideal MHD frozen-in flux at scales of $\rho_i$ (ion Larmor) or smaller. Hall MHD takes over.

Expert Insights

A plasma is more than an ionized gas — it's an ionized gas with collective behavior. The three criteria ($L\gg\lambda_D$, $\Lambda\gg 1$, $\omega\tau\gg 1$) are non-negotiable.

The plasma frequency is the system's heartbeat. All "fast" phenomena live near $\omega_p$; "slow" ones (MHD) at $\omega\ll\omega_p$.

Drifts decompose the motion of a magnetized particle. Rapid gyration averages out; only the guiding-center drift remains observable on long timescales.

The magnetic moment $\mu$ is an adiabatic invariant, not an exact one. Slow violations are crucial: scattering by waves, sudden field changes, all break $\mu$ and feed particles into the loss cone.

Landau damping is the most counterintuitive result of plasma theory. A dissipation-free (Vlasov) equation produces decaying waves because phase mixing redistributes information into finer velocity-space structure.

The plasma dispersion function $Z(\zeta)$ is the workhorse. Tabulate / compute it once; every Maxwellian linear theory calculation reduces to it.

Penrose's criterion is the most powerful tool for plasma stability: $f_0(v)$ stable iff it has no dip between maxima.

The two-stream / bump-on-tail instabilities are the basis of beam-plasma physics: type-III solar radio bursts, particle-accelerator wakefields, free-electron lasers.

Quasilinear theory is the workhorse of plasma turbulence applications: ICRH heating, lower-hybrid current drive, cosmic-ray scattering, radial transport in tokamaks.

Reconnection is the great non-MHD problem. Sweet-Parker fails (too slow); modern resolutions: plasmoid instability (still MHD), Hall MHD (kinetic at $\rho_i$), or full PIC simulation showing electron-scale physics matters.

Fermi acceleration explains cosmic rays. Diffusive shock acceleration produces $E^{-2}$ source spectra, modified to $E^{-2.7}$ observed by propagation.

Anomalous resistivity is real but unintuitive. Wave-particle scattering provides enormous effective collision rates in current sheets and shocks — much larger than Spitzer.

The Vlasov equation has infinitely many conserved quantities (Casimirs $\int g(f)$). This is why plasma equilibria are highly non-unique, and why heating is "anomalous" — entropy creation requires breaking Vlasov.

Solar wind is a magnetized turbulent plasma with extremes: $\beta\sim 1$, $\text{Re}\sim 10^{11}$, $\text{Rm}\sim 10^{15}$. It's the best-instrumented plasma laboratory anywhere.

Tokamak confinement is fundamentally about microinstabilities. Drift waves driven by pressure gradients cause "anomalous transport" — far above neoclassical (collisional) levels. Decades of research on gyrokinetic theory (subset of Vlasov suited to magnetized plasmas).

Magnetic confinement and inertial confinement face the same enemy: plasma instabilities. ITER battles drift modes; NIF battles Rayleigh-Taylor and Brillouin/Raman parametric instabilities.

The CMA diagram is genuinely useful for navigating wave propagation in magnetized plasmas — once you learn to read it, decades of empirical knowledge come at a glance.

Strong plasma turbulence remains an open problem — between weak (perturbative) and full kinetic. Best progress: Iroshnikov-Kraichnan and Goldreich-Sridhar MHD-turbulence cascades, validated in solar wind.

Plasma physics has more elements that look like other physics than other physics combined: sound waves, EM waves, magnetic waves, Landau damping (phase mixing), trapping (Hamiltonian chaos), turbulence cascades, dispersion, dispersion relations, reconnection (topological change), wave-particle resonance (Cherenkov). It's a survey of all of classical physics in one medium.