Beam-tracing and profile evolution for localised beams in inhomogeneous plasmas

We derive the beam tracing and profile evolution for the propagation of any localised beam with arbitrary profile through an inhomogeneous cold plasma. We recover standard Gaussian beam-tracing, with an additional PDE describing the evolution of the beam’s profile as it propagates through the plasma. We then solve for generic families of solutions to the PDE using ladder operators, which can be chosen to reduce to Gauss-Hermite beams in homogeneous media. We importantly obtain an exact expression for the resulting beam profile, demonstrating that Hermite modes will generally evolve into a superposition of different modes during propagation through inhomogeneous plasmas, contrary to prior work on the subject. Importantly, this approach facilitates future analysis of the diagnostic signal received from arbitrary beams.

💡 Research Summary

This paper presents a comprehensive theory for the propagation of arbitrary localized electromagnetic beams through an inhomogeneous cold plasma. Starting from the Helmholtz equation with the cold‑plasma dielectric tensor, the authors introduce a beam ansatz in which the electric field is written as a slowly varying amplitude multiplied by a rapidly varying phase. The amplitude is split into a basic envelope (A⁽⁰⁾ ê) and a profile function P(τ, wₓ, w_y), while the phase contains a linear term in the transverse coordinates and a quadratic term governed by a complex tensor Ψ. The small‑parameter ordering λ/W ≪ 1, W/L ≪ 1 (λ: wavelength, W: beam width, L: plasma inhomogeneity scale) allows a systematic expansion up to second order.

At zeroth order the dispersion relation D·ê = 0 yields a Hamiltonian H(K, r) whose characteristics give the usual ray‑tracing equations ∇ₖH = dq/dτ and ∇H = −dK/dτ. The first‑order equations reproduce standard Gaussian beam tracing. The second‑order expansion produces two coupled evolution equations: (i) an envelope‑tensor equation for Ψ that includes curvature, medium gradients, and the dielectric response, and (ii) a partial differential equation for the profile P:

∂P/∂τ − (i/2) ∇ₖ∇ₖH : ∇_w∇_w P + w·T·∇_w P = 0,

where T combines ray curvature, the Ψ‑tensor, and second derivatives of the Hamiltonian. This PDE resembles an Ornstein‑Uhlenbeck equation and captures diffraction, focusing, and the effect of plasma inhomogeneity on the transverse shape of the beam.

To solve the profile equation analytically, the authors construct ladder operators L₊ and L₋ that commute with the differential operator D̂ defined by the left‑hand side of the PDE. By choosing the coefficients of the ladder operators appropriately, they generate an infinite bi‑orthogonal set of eigenfunctions. In a homogeneous plasma the ladder operators reduce to the familiar creation and annihilation operators of quantum harmonic oscillators, and the eigenfunctions become the standard Gauss‑Hermite modes, confirming consistency with established Gaussian beam theory. In an inhomogeneous plasma, however, the coefficients become τ‑dependent, causing the ladder operators to mix modes. Consequently, an initial Hermite mode evolves into a superposition of several Hermite modes—a result that directly contradicts earlier work (e.g., Pereversev et al.) which assumed mode preservation. The paper validates this claim through both analytical arguments and numerical examples.

The theory also incorporates weak dissipative effects by adding a small imaginary part to Ψ, leading to modified amplitude and phase evolution equations that account for power absorption and phase delay—essential for realistic microwave heating simulations.

Practically, the derived ray‑tracing and envelope equations are already implemented in existing codes such as TORBEAM, WKBeam, and SCOTTY. The new profile‑evolution module can be added as a lightweight post‑processor, enabling real‑time prediction of beam shape changes for arbitrary antenna patterns, non‑Gaussian imperfections, or complex plasma profiles.

In conclusion, the paper delivers (1) a rigorous derivation of beam‑tracing equations extended to include full transverse profile dynamics, (2) an elegant ladder‑operator solution that unifies Gaussian and non‑Gaussian beam evolution, and (3) a clear pathway for integrating these results into current plasma‑diagnostic and heating simulation tools. Future directions suggested include extensions to multi‑frequency beams, strongly non‑uniform plasmas (temperature or current gradients), and nonlinear effects such as relativistic self‑focusing.