Semiclassical framework for the calculation of transport anisotropies

Abstract: We present a procedure for finding the exact solution to the linear-response Boltzmann equation for two-dimensional anisotropic systems and demonstrate it on examples of non-crystalline anisotropic magnetoresistance in a system with spin-orbit interaction. We show that two decoupled integral equations must be solved in order to find the non-equilibrium distribution function up to linear order in the applied electric field. The examples are all based on the Rashba system with charged magnetic scatterers, a system where the non-equilibrium distribution function and anisotropic magnetoresistance can be evaluated analytically. Exact results are compared to earlier widely-used approximative approaches. We find circumstances under which approximative approaches may become unreliable even on a qualitative level.

• The paper develops an exact analytical solution to the 2D Boltzmann equation using decoupled inhomogeneous Fredholm integral equations, surpassing standard relaxation time models.
• It demonstrates how angular-dependent scattering rates and correlations critically influence the non-equilibrium distribution and anisotropic magnetoresistance in spin-orbit coupled systems.
• Applications to Rashba systems reveal resonance effects with electric-magnetic impurity admixtures, setting benchmarks for future multiband and 3D transport studies.

Semiclassical Formalism for Transport Anisotropies in 2D Systems

Introduction and Motivation

The theoretical framework for evaluating transport anisotropies—particularly the anisotropic magnetoresistance (AMR)—in solids with symmetry-breaking sources (notably spin-orbit coupling and magnetization) lacks universally accepted microscopic rigor. Historically, approaches to calculating AMR include phenomenological scattering-based models and, for sufficiently simple systems, ab initio calculations. However, discrepancies between these methods persist, especially when extending to noncrystalline (continuous, as opposed to lattice-mediated) anisotropies and in the presence of complex band structures.

This work advances the field by establishing an exact, analytic procedure for solving the semiclassical Boltzmann equation in two-dimensional, anisotropic systems. The methodology transcends prior approximations, such as the relaxation time approximation (RTA) and the approach using parallel and perpendicular relaxation times ($1/\tau^\parallel$, $1/\tau^\perp$), by reducing the determination of the non-equilibrium distribution to the solution of two decoupled inhomogeneous Fredholm integral equations. The formalism is implemented and benchmarked on Rashba-coupled systems with charged magnetic scatterers, a model capturing essential symmetry-breaking mechanisms while still allowing for analytic progress.

Limitations of Conventional Approaches

Traditional models like RTA and its generalizations are exact for isotropic scattering, where transition rates depend solely on the angle between initial and final momentum states. In anisotropic systems, however, the scattering rate acquires an explicit dependence on the absolute orientation of $\vec{k}$, yielding direction-dependent relaxation rates. The naive application of RTA or its "direction-dependent" extension thus fails, capturing only limited aspects of anisotropic transport. The popular $1/\tau$ approach, for instance, yields a non-equilibrium distribution that, while operative, does not systematically encode all relevant angular correlations.

The Schliemann-Loss "two relaxation times" method improves this by incorporating both longitudinal and transverse scattering processes, but it essentially provides an ansatz rather than an exact solution for general anisotropy. Its validity is strictly limited to cases where the distribution function contains only angular harmonics of period $2\pi$.

Exact Solution via Integral Equations

The main theoretical contribution is the demonstration that the linearized Boltzmann equation reduces, for two-dimensional continua, to a pair of uncoupled inhomogeneous Fredholm equations of the second kind for the angular dependence of the distribution function. The structure is:

$$\cos\phi = \bar{w}(\phi)a(\phi) - \int d\phi'\ w(\phi, \phi')a(\phi')\,, \ \sin\phi = \bar{w}(\phi)b(\phi) - \int d\phi'\ w(\phi,\phi')b(\phi')$$

where $w(\phi, \phi')$ is the angularly resolved transition rate, and $\bar{w}(\phi)$ its angular average over the outgoing (or incoming) direction. The functions $a(\phi)$ and $b(\phi)$ fully determine the linear (in bias) response of the distribution to the external electric field. When the scattering's angular structure is sufficiently simple (i.e., contains only lower harmonics), the equations reduce to closed systems for a finite number of Fourier coefficients; in general, however, the solution requires retaining the full Fourier expansion, involving an infinite hierarchy that can, however, be solved analytically for several physically relevant cases.

Application to Rashba Systems with Magnetic Scattering

Three Rashba-model scenarios are analyzed exhaustively:

1. Single Band, Purely Magnetic Scatterers $(\alpha=0)$: The Fermi surface is cut only through the upper Rashba band, and the potential contains only magnetic components. The exact solution yields a non-equilibrium distribution and AMR value that markedly diverges from those predicted by both $1/\tau$ and $1/\tau^\parallel, 1/\tau^\perp$ approximations. The $1/\tau$ approach underestimates AMR, while the $1/\tau^\parallel, 1/\tau^\perp$ scheme even predicts the wrong sign.
2. Single Band, Electromagnetic Scatterers: Introducing a coherent superposition of electric and magnetic impurities, parameterized by a scalar $\alpha$ characterizing the electric-magnetic admixture, the solution's algebraic structure becomes richer. The AMR is derived as $\mathrm{AMR}=1/(2-\alpha^2)$ for $|\alpha|<1$ and $1/\alpha^2$ for $|\alpha|>1$, highlighting a resonance structure when electric and magnetic contributions are balanced ($\alpha\to 1$).
3. Two Bands, Electromagnetic Scatterers and Small Rashba Coupling: When both Rashba subbands cross the Fermi level (in the limit $\lambda \to 0$), symmetry enhancements yield analytic tractability. The notable result is that the AMR computed from the exact framework coincides with the results from the $1/\tau$ and $1/\tau^\parallel, 1/\tau^\perp$ approximations, despite significant differences in the detailed structure of the non-equilibrium distribution. This equality for AMR is attributed to symmetry, with the single-band, chirality-dominated case exhibiting far greater sensitivity to approximation.

Implications for Theory and Experiment

This work demonstrates that, outside these symmetry-enhanced limits, exact calculation of anisotropic transport in low-dimensional systems cannot be reliably replaced by any local or semilocal scattering rate prescription. Only by solving the global integral equation is it possible to obtain physically correct AMR, especially when the Fermi surface and/or scattering kernel has rich angular structure. Apparent divergences in AMR (such as at $\alpha \to 1$) are artefacts of the idealized impurity model and would be regularized in real materials due to momentum-dependent scattering amplitudes.

The Rashba-plus-magnetic-scattering model provides direct qualitative connections to AMR trends in III–V semiconductor spintronics platforms such as (Ga,Mn)As, where charged, ferromagnetically ordered impurities are structurally analogous to Mn substituents. Similarly, n-type heterostructures with magnetic donors would manifest the predicted AMR signatures, which would also be tunable by varying carrier concentration and the relative electric/magnetic character of the scatters.

The theoretical framework readily generalizes to multiband and three-dimensional systems, albeit at the cost of increased mathematical complexity. The analytic tractability for 2D models provides valuable benchmarks for numerical studies of more realistic systems.

Conclusion

This work sets a new standard for the rigorous computation of transport anisotropy in anisotropic systems, explicitly demonstrating that conventional relaxation-time-based approximations can fail both quantitatively and qualitatively. The exact formalism developed, based on integral equations for the non-equilibrium distribution function, captures all essential correlations and angular structure, enabling correct predictions of transport coefficients such as AMR. The results underscore the necessity of global, rather than local, approaches to solution of the Boltzmann equation in the presence of complex symmetry-breaking interactions. The framework provides a platform for future investigation into anisotropic transport phenomena, particularly in low-dimensional materials with strong spin-orbit coupling and engineered impurity potentials (0810.5693).