Spin currents in crystals with spin-orbit coupling: multi-band effects in an effective Hamiltonian formalism

This paper demonstrates that calculating spin currents in crystals with spin-orbit coupling requires a modified operator derived from iterative elimination of remote bands, as the standard definition based on group velocity fails to capture dominant interband mixing effects that can lead to qualitatively incorrect results.

The Gist

Here is an explanation of the paper using simple language and creative analogies.

The Big Picture: The "Spin Current" Mystery

Imagine a crystal lattice (a solid material) as a giant, busy city. Inside this city, electrons are the citizens. Usually, we think of electricity as the flow of these citizens (charge current). But electrons also have a little internal spinning motion, like a top. When these "spinning tops" flow in a specific direction, we call it a spin current.

Scientists want to use these spin currents to build faster, cooler computers (spintronics). To do that, they need to calculate exactly how much spin current is flowing.

The Problem:
When scientists try to calculate this, they often use a "simplified map" of the city. They only look at the main streets (the "essential bands" where electrons actually live) and ignore the tiny alleyways and basements (the "remote bands").

The authors of this paper, Kirill Samokhin, Manfred Sigrist, and Mark Fischer, discovered a major flaw in this approach. They found that ignoring the alleyways changes the result completely. If you use the standard simplified map, you get the wrong answer. In fact, the "real" spin current is much stronger and behaves differently than the old formulas predicted.

---

The Analogy: The "Shadow" Effect

To understand why the simplified map fails, imagine you are trying to measure the wind speed on a rooftop (the essential band).

1. The Old Way (Standard Definition): You stand on the roof and measure the wind directly. You assume the air above and below the roof doesn't matter. You calculate the wind speed based only on what you see right there.
2. The New Way (This Paper): The authors realized that the wind on the roof is actually being pushed and pulled by massive air currents swirling in the basement and the attic (the remote bands). Even though you can't see the basement, the air pressure from down there is pushing the wind on the roof sideways.

If you ignore the basement, you think the wind is blowing straight. But in reality, the "shadow" of the basement air is creating a huge, swirling vortex on the roof that you missed.

In physics terms:

• The Roof: The "Essential Bands" (the energy levels we care about).
• The Basement/Attic: The "Remote Bands" (energy levels far away that we usually ignore).
• The Interaction: Spin-Orbit Coupling. This is a quantum force where the electron's movement (orbit) affects its spin. In crystals without a center of symmetry (like a tilted building), the "basement" and "roof" talk to each other.

The Core Discovery

The paper argues that when you "integrate out" (mathematically remove) the remote bands to create a simple model, you don't just get a simpler version of the same physics. You accidentally throw away the most important part of the story.

The "Hidden" Terms:
When you do the math correctly, you find that the formula for spin current needs extra terms.

• The Standard Formula: Says Spin Current = (Speed of Electron) × (Spin).
• The New Formula: Says Spin Current = (Speed of Electron) × (Spin) + [Huge Correction from Remote Bands].

The authors show that this "Huge Correction" is actually the dominant part of the spin current in many materials. It's like realizing that the reason the wind is blowing so hard isn't because the fan on the roof is strong, but because a giant jet engine in the basement is pushing it.

Why Does This Matter?

1. Magnitude: The new calculation predicts that the equilibrium spin current (the spin flow that exists naturally without any external power) is much larger than previously thought.
2. Dependence: The old formula said the spin current didn't depend on how many electrons were in the material (chemical potential). The new formula says it does depend on the electron count.
3. Accuracy: If engineers use the old, simplified formulas to design new spintronic devices, their designs might be off by a huge margin. They might think a device will generate a weak signal when it could actually generate a massive one (or vice versa).

The "Rashba" Example

The paper uses a specific model called the Rashba model (named after physicist Rashba) to prove their point.

• Imagine a 2D material (like a very thin sheet of metal).
• In the old view, the spin current in this sheet was thought to be tiny and cubic (related to the cube of the interaction strength).
• In the new view, by accounting for the mixing with distant bands, the spin current becomes linear (directly proportional) and much stronger.

Summary in One Sentence

You cannot accurately calculate the flow of spinning electrons in a crystal by only looking at the electrons' immediate neighborhood; you must account for the invisible "ghosts" of distant energy bands, because those ghosts are actually the ones pushing the spin current around.

The Takeaway for Everyone

This paper is a warning to physicists and engineers: "Don't oversimplify." When dealing with the quantum world, the things you ignore (the distant bands) often have the biggest impact on the things you are trying to measure. By fixing the math, we might be able to build much better quantum computers and sensors than we thought possible.

Technical Summary

Here is a detailed technical summary of the paper "Spin currents in crystals with spin-orbit coupling: multi-band effects in an effective Hamiltonian formalism" by Samokhin, Sigrist, and Fischer.

1. Problem Statement

The paper addresses a fundamental issue in the theoretical calculation of spin currents in solids, particularly in noncentrosymmetric crystals with spin-orbit coupling (SOC).

• The Context: Spintronics relies on understanding spin currents, which can be extrinsic (driven by electric fields) or intrinsic (present in equilibrium). Calculating these often involves using effective Hamiltonians (e.g., Rashba models) that focus only on a few "essential" bands near the Fermi level, having "integrated out" distant bands.
• The Flaw: Standard practice defines the spin current operator in an effective model simply as the anticommutator of the group velocity and the spin operator: $\hat{J} = \frac{1}{2}\{\hat{v}, \hat{\sigma}\}$.
• The Core Question: The authors argue that this standard definition is microscopically unjustified when derived from a full microscopic theory that includes inter-band mixing. Ignoring the contributions of remote bands (which are integrated out to form the effective Hamiltonian) leads to qualitatively and quantitatively incorrect results for the spin current, particularly the equilibrium spin current.

2. Methodology

The authors develop a rigorous formalism to derive the correct spin current operator within an effective band description by systematically eliminating inter-band couplings.

• Microscopic Starting Point: They begin with the exact microscopic Hamiltonian in the Luttinger-Kohn (LK) basis, which includes all orbital states and the full electron-lattice SOC. The spin current operator is defined microscopically using non-Abelian gauge invariance.
• Canonical Transformation (Schrieffer-Wolff): To reduce the infinite-dimensional microscopic problem to a manageable effective model, they apply a unitary canonical transformation (Luttinger-Kohn or Schrieffer-Wolff transformation). This transformation decouples the "essential" orbital subspace (near the chemical potential) from "distant" orbitals.
• Iterative Elimination: They perform this elimination perturbatively (iteratively) to second order in the inter-orbital couplings.
  • This procedure generates an effective Hamiltonian ($\hat{H}_{eff}$) for the essential bands.
  • Crucially, the same transformation is applied to the spin current operator ($\hat{J}$), the spin torque, and the spin density.
• Derivation of the Effective Operator: By expanding the transformed operator in powers of the mixing parameter, they derive an expression for the effective spin current operator that includes correction terms arising from the eliminated distant bands.

3. Key Contributions

The paper makes several theoretical breakthroughs:

1. Refutation of the Standard Definition: They prove that the standard definition $\hat{J} = \frac{1}{2}\{\hat{v}, \hat{\sigma}\}$ is invalid for effective Hamiltonians derived from multi-band systems. The effective spin current operator contains additional terms that cannot be expressed solely in terms of the effective Hamiltonian and the spin operator.
2. Identification of Dominant Terms: The derived effective operator includes a "trivial" term (the standard definition) and "correction" terms ($\delta \hat{J}$) arising from inter-band mixing. The authors show that in equilibrium, these correction terms dominate the spin current.
3. Quantitative Discrepancy: They demonstrate that the magnitude of the equilibrium spin current calculated using the correct multi-band formalism is significantly larger (potentially by orders of magnitude) than that predicted by the standard single-band Rashba model.
4. Dependence on Chemical Potential: Unlike the standard model prediction (where equilibrium spin current is independent of chemical potential), the corrected theory predicts a strong dependence on the chemical potential (electron concentration).

4. Results

The authors apply their general theory to a specific model: a 2D noncentrosymmetric crystal with $C_{4v}$ symmetry, coupling a 1D orbital ($\Gamma_1$) and a 2D orbital ($\Gamma_5$).

• Effective Hamiltonian: The inter-orbital mixing generates an effective Rashba SOC ($\alpha_R$) in the essential band, proportional to the product of inter-orbital coupling and energy gap.
• Corrected Spin Current Operator: The effective spin current operator takes the form:
  $$\hat{J}_{\mu,i} = \frac{1}{2}\left\{ \frac{\partial \hat{H}_{eff}}{\partial k_i}, \hat{\sigma}_\mu \right\} + \delta \hat{J}_{\mu,i}$$
  where $\delta \hat{J}_{\mu,i}$ contains terms proportional to the inter-orbital couplings and the energy gap.
• Equilibrium Spin Current Calculation:
  • Standard Model Prediction: Using the standard definition on the effective Hamiltonian yields $J_s \propto \alpha_R^3$. This result is independent of the chemical potential ($\mu$).
  • Corrected Model Prediction: Using the full derived operator yields a result where the leading term is linear in $\alpha_R$ and quadratic in the chemical potential:
    $$J_s(T=0) \propto \left(\frac{\mu}{E_g}\right)^2 \alpha_R$$
    (where $E_g$ is the energy gap between orbitals).
• Magnitude Comparison: The ratio of the corrected current to the standard prediction scales as $\mu / E_R$ (where $E_R$ is the Rashba splitting). Since $\mu \gg E_R$ in typical metals, the corrected equilibrium spin current is much larger than previously thought.

5. Significance

• Theoretical Correction: The paper corrects a long-standing assumption in condensed matter physics regarding the calculation of spin currents in effective models. It establishes that "integrating out" bands is not a passive procedure; it fundamentally alters the form of transport operators.
• Experimental Implications: The results suggest that equilibrium spin currents in noncentrosymmetric materials (such as oxide interfaces, chalcogenide surfaces, or noncentrosymmetric superconductors) may be much stronger than previously estimated. This has direct implications for:
  • The interpretation of mechanical torque measurements on samples.
  • The design of spintronic devices where equilibrium spin accumulation or spin-momentum locking is utilized.
  • The understanding of spin superfluidity and spin transport in superconductors.
• General Framework: The methodology provides a robust framework for calculating not just spin currents, but any observable (like spin torque or conductivity) in effective Hamiltonian formalisms, ensuring consistency with the underlying microscopic physics.

In summary, the paper argues that multi-band effects are not a minor correction but the dominant mechanism for equilibrium spin currents in noncentrosymmetric crystals, necessitating a complete reformulation of the spin current operator in effective theories.