Momentum-Dependent Spin Splitting

Updated 23 April 2026

Momentum-dependent spin splitting is a phenomenon where electronic spin states split as a function of crystal momentum, driven by relativistic, symmetry, and exchange mechanisms.
It manifests across diverse platforms such as semiconductors, antiferromagnets, and twisted 2D materials, with its magnitude tunable by strain, electric fields, and ligand engineering.
This phenomenon underpins advanced spintronic applications, offering insights into ultrafast dynamics, topological states, and controllable spin-transport in next-generation devices.

Momentum-dependent spin splitting refers to the lifting of spin degeneracy in the electronic, magnonic, or relativistic particle spectra, whereby the energy difference ΔE(k) = E↑(k) − E↓(k) acquires explicit dependence on the crystal momentum k, beyond a trivial Zeeman non-dispersive term. It encompasses a variety of mechanisms—relativistic, symmetry-driven, exchange-driven, and topological—that manifest in metals, semiconductors, antiferromagnets, chiral systems, and even bosonic excitations. Unlike conventional Zeeman splitting, which is momentum-independent and requires a net magnetization or external field, momentum-dependent spin splitting can arise in systems with zero net M, and often without the need for atomic spin–orbit coupling (SOC). The classification, magnitude, and applications of these effects are determined by the interplay of local symmetry, exchange interactions, relativistic corrections, and, in certain platforms, the presence of ultrafast nonequilibrium drive.

1. Fundamental Mechanisms and Symmetry Classification

Relativistic and SOC-Mediated Spin Splittings

The canonical mechanisms of momentum-dependent spin splitting are rooted in relativistic effects—specifically, the interplay of Dirac electron motion in an external field and atomic SOC. In III–V semiconductors, two archetypal contributions are identified:

Dresselhaus (bulk inversion asymmetry, BIA) terms: Cubic and linear-in-k terms of the form γ [kₓ(kᵧ²−k_z²)σₓ + …] arise in noncentrosymmetric zincblende structures and produce momentum-dependent splitting that vanishes at Γ but grows for finite k, controlled by the BIA symmetry (Norman et al., 2010).
Rashba (structural inversion asymmetry, SIA) terms: Manifesting as α(k_y σ_x − k_x σ_y), the Rashba effect produces a linear-in-k splitting whose direction and magnitude are tunable by electric fields or interface asymmetry. The Rashba parameter α can reach values on the order of several eV·Å in engineered heavy-element interfaces or noncentrosymmetric bulk systems (Feng et al., 2019, Hong et al., 2017).

Chiral (enantiomorphic) 1D systems like InSeI exhibit a longitudinal analog of the Rashba effect, with linear-in-k splitting ΔE(k) = 2|α k| and collinear spin-momentum locking. Notably, chirality reverses the sign of spin polarization at each k point, locking the spin quantization axis along the chain direction (Zhao et al., 2023).

Exchange-Driven, SOC-Independent Splittings (Altermagnetism)

Recent advances have revealed that collinear and noncollinear antiferromagnets with certain magnetic space groups can host intrinsic, momentum-dependent spin splitting even in the absence of atomic SOC—termed "altermagnetism" or "AFM-induced spin splitting" (Yuan et al., 2019, Zeng et al., 2024). Key symmetry criteria are:

Breaking PT: The combined space inversion (P) and time reversal (T) symmetry must be broken. In monolayer or twisted bilayer systems, AB-stacking, layer twist, or ligand displacement can achieve this condition (Pathak et al., 23 Feb 2026, Yuan et al., 2021).
Absence of antiunitary flipping symmetries (UT): The magnetic space group must preclude antiunitary elements that guarantee Kramers degeneracy at each k; only types I and III in the Shubnikov classification permit pure exchange-induced splitting (Yuan et al., 2019, Yuan et al., 2020).

Microscopically, exchange-driven splittings are governed by the local motif symmetry (site multipoles—monopole, dipole, quadrupole) and their transformation under space-group operations. Quadrupole moments (e.g. Q_{xy}) connected by axial symmetries lead to quadratic-in-k splitting (e.g. ΔE(k) ∝ k_x k_y), while dipole differences yield linear-in-k splitting (Acosta, 28 May 2025). Representative H_eff(k) forms are tabulated below.

Mechanism	ΔE(k) functional form	Physical Symmetry Condition
Zeeman (monopole, l=0)	const. (k-independent)	two inequivalent, charge-disproportionated sites, no P or T
Rashba/Dresselhaus (SOC)	linear (αk), cubic	inversion symmetry breaking for Rashba; noncentrosymmetric point group for Dresselhaus
Altermagnetic (quadrupole, l=2)	ΔE(k) ∝ k_x k_y, etc.	PT-breaking, motif quadrupole differences
Chiral 1D (Rashba-type)	ΔE(k) ∝ αk	chiral space group (e.g. no mirror, P)
AFM-induced (twistronic, 2D)	ΔE(k) ∝ α^{(1)} k_{⊥}	breaking of combined [C_2

2. Model Hamiltonians and Spin Splitting Magnitudes

The Hamiltonians capturing momentum-dependent spin splitting admit a range of forms, dictated by spatial, spin, and time-reversal symmetries. Paradigmatic examples include:

Linear/k-dependent:
- Rashba: H = α_R (k_x σ_y − k_y σ_x) → ΔE(k) = 2 α_R |k|
- Chiral 1D: H_eff(k) = ε₀ + ħ²k²/(2m*) + α k σ_z → ΔE(k) = 2|αk| (Zhao et al., 2023)
Quadratic/k-dependent (altermagnetic):
- H(k) = h_Γ Q_Γ(k) σ, with Q_Γ(k) = sin(k_x a) sin(k_y a) ≈ k_x k_y for k ≈ 0. ΔE(k) = 2|h_Γ Q_Γ(k)| (Hayami et al., 2019, Yuan et al., 2019)
Twisted 2D systems:
- H_eff(q) = (ħ² q_x'^2)/(2m_{x'}) + (ħ² q_y'^2)/(2m_{y'}) + [α^{(1)} q_{y'} + α^{(3)} q_{y'}^3] σ_z (Pathak et al., 23 Feb 2026)

Numerical first-principles and tight-binding calculations yield giant exchange-driven splittings: in MnF₂, ΔE(k) approaches 300 meV (quadratic in k along certain directions) (Yuan et al., 2019); in CrSb, ARPES and DFT confirm ΔE(k) up to 0.8–1.1 eV at generic k (Zeng et al., 2024). In NiO-based prototypes, "DFT model Hamiltonian" studies show tunable ΔE(k) from 0 to 1 eV by ligand displacement of only ~0.1 Å (Yuan et al., 2021).

In contrast, classic SOC-induced Rashba splittings in PtBi₂ measured as α_R ≈ 4.36 eV·Å (ΔE(k) = 2 α_R |k|), already among the largest, remain smaller than the exchange-driven splittings in low-Z antiferromagnetic compounds (Feng et al., 2019).

3. Experimental Realizations and Tuning

Bulk and 2D Antiferromagnets

CrSb: ARPES and DFT resolve anisotropic ΔE(k), vanishing along high-symmetry mirror planes and peaking at 0.8+ eV away from symmetry-enforced nodes (Zeng et al., 2024).
CrO monolayer: 2D antiferromagnetic Weyl semimetal with two pairs of spin-polarized Weyl points at E_F. Strain tunes the separation, enabling transitions between Weyl, half-metal, and AFM semiconductor phases, all with spin-momentum–locked transport channels (Chen et al., 2021).
NiO “model system”: O-ligand displacement by ~0.04 Å in SST-4 geometry produces 400–950 meV spin splitting even without SOC, with negligible energy cost, highlighting the strong ligand-lattice–exchange coupling (Yuan et al., 2021).
Twisted-bilayer altermagnets: Twist angle between bilayers of 2D AFMs (e.g. CoCl₂, FeS, MnTe₂) breaks [C_2||P] and point-group rotations, yielding linear-in-k NRSS with α^{(1)} up to ~1100 meV·Å, exceeding typical SOC splittings by an order of magnitude (Pathak et al., 23 Feb 2026).

Ultrafast Nonequilibrium Dynamics

Ultrafast light pulses provide a route to dynamically induce momentum-dependent altermagnetic splitting in centrosymmetric insulating AFMs, e.g., KNiF₃. Linearly polarized excitation of zone-center Eg phonons (symmetry selection Sym²(e)) breaks effective time-reversal and reveals transient d- or g-wave splittings in ΔE(k), detectable by time-resolved spectroscopies (Chen et al., 3 Apr 2026). The effect is tunable in amplitude and symmetry via laser parameters and is fundamentally nonrelativistic.

Functional Control

Strain: Biaxial and uniaxial strain linearly tunes NRSS magnitude and can drive symmetry transitions (g/i → d wave)—crucially, activating finite spin conductivity in systems previously forbidden by rotational symmetries (Pathak et al., 23 Feb 2026).
Electric field: An out-of-plane field superposes a Zeeman-type, k-independent ΔE(k), up to ~110 meV at 10 MV/cm in certain 2D bilayers (Pathak et al., 23 Feb 2026).
Ligand engineering: Sub-ångström ligand displacements switch between symmetry-distinct SSTs at constant magnetic sublattice chemistry, enabling phase diagrams of nonrelativistic, relativistic, and trivial splitting (Yuan et al., 2021).

4. Advanced Manifestations: Magnons, Topology, Spin Transport

Magnon Spin-Momentum Locking

Momentum-dependent spin splitting generalizes to bosonic magnon spectra in noncollinear antiferromagnets. The spin expectation of magnon modes S_{k,α} acquires k-dependence if the ground-state order spontaneously breaks enough spin-rotation symmetry. Kagome 120° magnets display winding numbers Q = ±2 in the vortex spin textures of magnon bands, and magnon Dirac cones with Q = +1; the Poincaré–Hopf theorem enforces that the sum of winding indices equals zero on the BZ torus (Okuma, 2017).

Topologically Protected Splittings

In 2D AFM Weyl semimetals (e.g., CrO), spin-split pairs of Weyl points at E_F result in k-dependent, symmetry-protected splitting up to ~0.5 eV, controlled by AFM lattice symmetry. Charge currents inject spin-polarized states with distinct velocities; strain steers the number and character of topological transport channels (Chen et al., 2021).

Spin–Transport and Spin-Splitter Response

Altermagnets with momentum-dependent spin splitting host unconventional spin–splitter currents. Under applied fields, differing Fermi-surface shifts for opposite spins produce pure spin currents, governed by both intrinsic band geometry and extrinsic side-jump/skew-scattering contributions. FeSb₂ displays spin Hall angles up to ≳0.5 with the effect maximized in d-wave NRSS systems (Sarkar et al., 26 Feb 2026).

5. Theoretical Frameworks: Magnetoelectric Unification and Multipole Analysis

A unified relativistic framework identifies an intrinsic magnetoelectric correction to the Hamiltonian, derived from the Foldy–Wouthuysen expansion of the Dirac equation. Local electric multipoles (monopole, dipole, quadrupole) couple via σ·m, producing the full taxonomy of momentum-dependent splitting:

Monopole: ΔE(k) = 2 μ_B η₀ λ₀ Q₀(σ·m), k-independent (Γ-point);
Dipole: ΔE(k) ∝ (d·k)(σ·m), linear-in-k;
Quadrupole: ΔE(k) ∝ k_i k_j (σ·m), quadratic-in-k (altermagnetic) (Acosta, 28 May 2025).

Motif connectivity under space group operations U determines which multipoles survive inter-sublattice cancellation. This formalism predicts, from symmetry and site analysis, the allowed k-dependence and offers a roadmap for material discovery and design.

6. Materials Design and Applications in Spintronics

Strategic symmetry and chemical engineering allow tuning and exploitation of momentum-dependent spin splitting:

AFM materials: Identify magnetic space groups lacking PT and UT, screen for strong local moments and enabled motif multipoles (Yuan et al., 2019, Yuan et al., 2020).
Twistronic 2D platforms: Exploit layer twist and strain to realize and modulate NRSS, driving transitions in spin conductivity and enabling nonrelativistic, field-free spin filtering and generation (Pathak et al., 23 Feb 2026).
Ultrafast photonics: Use laser-driven symmetry breaking for THz-frequency, reversible switching of k-space spin textures (Chen et al., 3 Apr 2026).
Low-Z, stable compounds: Advances allow "giant" ΔE(k) in materials free of heavy elements, minimizing disorder, and maximizing scalability for memory and spin–charge conversion technologies (Zeng et al., 2024, Yuan et al., 2021).

Device concepts include AFM spin-FETs, spin-splitter tunneling junctions with 100% k-selective polarization, and low-loss, high-speed spin–charge interconversion.

In summary, momentum-dependent spin splitting is an organizing principle unifying diverse forms of symmetry-driven, exchange-mediated, and relativistic band structure phenomena, with a well-defined mechanistic, symmetry, and materials design underpinning. Both fundamental and applied ramifications continue to drive exploration of new classes of magnets, semiconductors, and 2D materials for advanced spintronic technology.