Orbital Berry Curvature in Quantum Systems

• Orbital Berry curvature is the geometric framework describing orbital hybridization in Bloch bands and its role in generating orbital magnetic moments, Hall transport, and Zeeman-like shifts.
• It quantifies how rapid changes in orbital character over momentum or real space lead to observable effects such as Berry-curvature dipoles and nonlinear responses.
• The concept underpins experimental strategies like STM, ARPES, and optical probes that reveal the underlying orbital dynamics in quantum materials.

Orbital Berry curvature is the Berry-curvature structure associated with orbital aspects of quantum states: orbital hybridization in Bloch bands, orbital magnetic moments, orbital angular-momentum transport, or local orbital texture in real or sublattice space. In the condensed-matter literature, the term is used in more than one precise sense. Most commonly it denotes the momentum-space Berry curvature of bands whose geometric response is controlled by orbital degrees of freedom; in orbital Hall theory it can also denote the Berry-curvature-like quantity entering the Kubo formula for intrinsic orbital angular-momentum transport. Related real-space, local, and dynamical generalizations extend the same geometric logic beyond static Bloch-band transport. Across these settings, the common content is that rapid orbital variation of eigenstates over parameter space generates anomalous response functions, orbital magnetization, nonlinear transport, and Zeeman-like energy shifts (Lee, 14 Aug 2025, Mercaldo et al., 2023).

1. Definitions and terminological scope

The basic Bloch-band Berry curvature is the geometric field

$$\Omega_n(\mathbf{k}) = i \left\langle \nabla_{\mathbf{k}} u_{n\mathbf{k}} \,\middle|\, \times \,\middle| \nabla_{\mathbf{k}} u_{n\mathbf{k}} \right\rangle ,$$

while the associated orbital magnetic moment is

$$\mathbf{m}_n(\mathbf{k}) = -\frac{e}{2\hbar}\,\mathrm{Im}\left\langle \nabla_{\mathbf{k}} u_{n\mathbf{k}} \middle| \times \big(H(\mathbf{k})-\varepsilon_{n\mathbf{k}}\big)\middle| \nabla_{\mathbf{k}} u_{n\mathbf{k}} \right\rangle .$$

In many orbitally active systems, these two quantities have closely parallel momentum-space structure, and the observable response is most directly orbital even when the underlying Berry curvature is computed from spin-resolved bands or multiorbital wave functions (Li et al., 2023).

The literature therefore uses “orbital Berry curvature” in a family of related but nonidentical ways.

Usage	Representative object	Typical consequence
Orbitally sourced Bloch-band curvature	$\Omega_n(\mathbf{k})$	orbital moment, Hall-like transport, Zeeman shifts
Orbital-Hall curvature	$\mathcal{O}^{ij,\sigma}_{\alpha,\mathbf{k}}$	intrinsic orbital Hall conductivity
Local or real-space orbital curvature	$B_z$, $\Omega_{n\mathbf{k}}(\mathbf{r}_\alpha)$, or $\Omega_n(\mathbf{r})$	vortex dynamics, local orbital magnetization

In a general two-band electronic system, the orbital Berry curvature governing the orbital Hall effect is defined as a band-resolved Kubo quantity $\mathcal{O}^{ij,\sigma}_{\alpha,\mathbf{k}}$, with $\alpha$ the transported orbital-angular-momentum component and $i,j$ the current and driving directions. In that framework it is not merely analogous to ordinary Berry curvature; it is exactly reducible to orbital angular momentum times Berry curvature (Lee, 14 Aug 2025). By contrast, in local theories of orbital magnetization, one resolves Berry-curvature contributions by sublattice or site, obtaining quantities that are invisible in the unit-cell-averaged modern theory (Saati et al., 13 Dec 2025). This multiplicity of usages is not a contradiction; it reflects the fact that the same geometric structure reappears in different response formalisms.

2. Relation to orbital magnetic moment, orbital magnetization, and orbital transport

A central relation in this area is that Berry curvature and orbital moment are often proportional or nearly proportional in effective two-band descriptions. In dichroic photoemission theory, the Bloch-state orbital angular momentum satisfies

$$\mathbf{m}_n(\mathbf{k}) = -\frac{e}{2\hbar}\,\mathrm{Im}\left\langle \nabla_{\mathbf{k}} u_{n\mathbf{k}} \middle| \times \big(H(\mathbf{k})-\varepsilon_{n\mathbf{k}}\big)\middle| \nabla_{\mathbf{k}} u_{n\mathbf{k}} \right\rangle .$$0

so local orbital chirality and local Berry curvature track one another directly in the valence–conduction-dominated regime (Schüler et al., 2019). In the general two-band theory of intrinsic orbital dynamics, the band-resolved orbital angular momentum is

$$\mathbf{m}_n(\mathbf{k}) = -\frac{e}{2\hbar}\,\mathrm{Im}\left\langle \nabla_{\mathbf{k}} u_{n\mathbf{k}} \middle| \times \big(H(\mathbf{k})-\varepsilon_{n\mathbf{k}}\big)\middle| \nabla_{\mathbf{k}} u_{n\mathbf{k}} \right\rangle .$$1

and the orbital Berry curvature entering orbital Hall transport is

$$\mathbf{m}_n(\mathbf{k}) = -\frac{e}{2\hbar}\,\mathrm{Im}\left\langle \nabla_{\mathbf{k}} u_{n\mathbf{k}} \middle| \times \big(H(\mathbf{k})-\varepsilon_{n\mathbf{k}}\big)\middle| \nabla_{\mathbf{k}} u_{n\mathbf{k}} \right\rangle .$$2

This implies a universal “band-splitting times Berry-curvature-squared” structure for intrinsic orbital Hall response in two-band systems (Lee, 14 Aug 2025).

The modern orbital-magnetization framework makes the same geometry explicit. In monolayer MoS$$\mathbf{m}_n(\mathbf{k}) = -\frac{e}{2\hbar}\,\mathrm{Im}\left\langle \nabla_{\mathbf{k}} u_{n\mathbf{k}} \middle| \times \big(H(\mathbf{k})-\varepsilon_{n\mathbf{k}}\big)\middle| \nabla_{\mathbf{k}} u_{n\mathbf{k}} \right\rangle .$$3, the orbital magnetization is written as

$$\mathbf{m}_n(\mathbf{k}) = -\frac{e}{2\hbar}\,\mathrm{Im}\left\langle \nabla_{\mathbf{k}} u_{n\mathbf{k}} \middle| \times \big(H(\mathbf{k})-\varepsilon_{n\mathbf{k}}\big)\middle| \nabla_{\mathbf{k}} u_{n\mathbf{k}} \right\rangle .$$4

and biased bilayer WSe$$\mathbf{m}_n(\mathbf{k}) = -\frac{e}{2\hbar}\,\mathrm{Im}\left\langle \nabla_{\mathbf{k}} u_{n\mathbf{k}} \middle| \times \big(H(\mathbf{k})-\varepsilon_{n\mathbf{k}}\big)\middle| \nabla_{\mathbf{k}} u_{n\mathbf{k}} \right\rangle .$$5 uses the corresponding decomposition into a moment term and a Berry-curvature correction (Son et al., 2019, Vargiamidis et al., 2020). The first term reflects wave-packet self-rotation; the second reflects the Berry-phase correction to phase-space structure. This decomposition is important because the same total orbital magnetization can arise from different mixtures of local orbital moment and Berry-curvature contributions.

A related misconception is that orbital Berry curvature is merely a reformulation of spin physics. The cited works do not support that reduction. In some materials the curvature is computed in spin-resolved channels, but the measured response is an orbital Zeeman coupling or an orbital magnetization. In other systems, especially multiorbital oxide and TMD platforms, the dominant curvature source is explicitly orbital hybridization rather than spin splitting (Li et al., 2023, Lesne et al., 2022).

3. Microscopic generation mechanisms

One major route is the gapping of Dirac crossings. In TbV$$\mathbf{m}_n(\mathbf{k}) = -\frac{e}{2\hbar}\,\mathrm{Im}\left\langle \nabla_{\mathbf{k}} u_{n\mathbf{k}} \middle| \times \big(H(\mathbf{k})-\varepsilon_{n\mathbf{k}}\big)\middle| \nabla_{\mathbf{k}} u_{n\mathbf{k}} \right\rangle .$$6Sn$$\mathbf{m}_n(\mathbf{k}) = -\frac{e}{2\hbar}\,\mathrm{Im}\left\langle \nabla_{\mathbf{k}} u_{n\mathbf{k}} \middle| \times \big(H(\mathbf{k})-\varepsilon_{n\mathbf{k}}\big)\middle| \nabla_{\mathbf{k}} u_{n\mathbf{k}} \right\rangle .$$7, spin-orbit coupling opens a small mass gap at a V-derived kagome Dirac crossing near the Brillouin-zone $$\mathbf{m}_n(\mathbf{k}) = -\frac{e}{2\hbar}\,\mathrm{Im}\left\langle \nabla_{\mathbf{k}} u_{n\mathbf{k}} \middle| \times \big(H(\mathbf{k})-\varepsilon_{n\mathbf{k}}\big)\middle| \nabla_{\mathbf{k}} u_{n\mathbf{k}} \right\rangle .$$8 point. The resulting massive Dirac fermion carries sharply peaked Berry curvature near the avoided crossing, which in turn generates enormous orbital magnetic moments. Spectroscopic-imaging STM and QPI resolve a field-induced splitting of the Dirac band into two branches with momentum-dependent effective $$\mathbf{m}_n(\mathbf{k}) = -\frac{e}{2\hbar}\,\mathrm{Im}\left\langle \nabla_{\mathbf{k}} u_{n\mathbf{k}} \middle| \times \big(H(\mathbf{k})-\varepsilon_{n\mathbf{k}}\big)\middle| \nabla_{\mathbf{k}} u_{n\mathbf{k}} \right\rangle .$$9 factors, reaching about $\Omega_n(\mathbf{k})$0 meV/T and corresponding to orbital moments of order $\Omega_n(\mathbf{k})$1 near the gapped Dirac point (Li et al., 2023). The same work emphasizes that the response is not conventional spin Zeeman splitting and not driven by localized Tb moments.

A distinct route is low-symmetry multiorbital hybridization. In the $\Omega_n(\mathbf{k})$2 orbital design framework, low-symmetry crystal fields and orbital Rashba couplings generate Berry-curvature hot spots and singular pinch points without requiring the usual electron–hole superposition of massive Dirac or Weyl systems. In the high-symmetry trigonal case the Berry curvature vanishes, but lowering the symmetry to $\Omega_n(\mathbf{k})$3 activates finite curvature through the coexistence of $\Omega_n(\mathbf{k})$4 and mass terms $\Omega_n(\mathbf{k})$5. The resulting hot spots, mirror-protected crossings, and pinch points produce giant Berry-curvature dipoles in time-reversal-symmetric conditions (Mercaldo et al., 2023).

A third mechanism is explicit symmetry breaking in the orbital sector. In a kagome altermagnet with compensated coplanar $\Omega_n(\mathbf{k})$6 order, noncollinear exchange alone produces altermagnetic spin splitting but not Berry curvature, because a hidden antiunitary symmetry $\Omega_n(\mathbf{k})$7 enforces $\Omega_n(\mathbf{k})$8. Finite curvature appears only when an orbital chiral flux term $\Omega_n(\mathbf{k})$9 is introduced. That term breaks the hidden symmetry, creates effective momentum-space gauge fields analogous to Haldane-type orbital flux, and produces local Berry-curvature hot spots even without spin-orbit coupling or scalar spin chirality (Tagani et al., 3 Jun 2026).

Electrostatic and Floquet control provide additional mechanisms. In biased bilayer WSe$\mathcal{O}^{ij,\sigma}_{\alpha,\mathbf{k}}$0, a perpendicular electric field breaks inversion symmetry, redistributes valley Berry curvature and orbital magnetic moment between layers, and can induce sign changes near field scales comparable to $\mathcal{O}^{ij,\sigma}_{\alpha,\mathbf{k}}$1 and $\mathcal{O}^{ij,\sigma}_{\alpha,\mathbf{k}}$2 (Vargiamidis et al., 2020). In a circularly shaken optical lattice, near-resonant $\mathcal{O}^{ij,\sigma}_{\alpha,\mathbf{k}}$3-$\mathcal{O}^{ij,\sigma}_{\alpha,\mathbf{k}}$4 Floquet hybridization and explicit chirality of the drive generate dynamical orbital magnetism and a time-dependent Berry curvature whose dominant Fourier components occur at $\mathcal{O}^{ij,\sigma}_{\alpha,\mathbf{k}}$5, while the static component vanishes (Chen et al., 2020).

4. Symmetry constraints, Berry-curvature dipoles, and local topology

Time-reversal symmetry often forces the Brillouin-zone integral of Berry curvature to vanish while leaving room for a finite first moment. In strained monolayer MoS$\mathcal{O}^{ij,\sigma}_{\alpha,\mathbf{k}}$6, the two valleys carry opposite Berry curvature, but strain lowers the in-plane symmetry and skews the distribution within each valley, producing a Berry-curvature dipole

$\mathcal{O}^{ij,\sigma}_{\alpha,\mathbf{k}}$7

Under an in-plane electric field, this dipole generates an out-of-plane valley-orbital magnetization

$\mathcal{O}^{ij,\sigma}_{\alpha,\mathbf{k}}$8

so the relevant geometric object is not the net Berry flux but the asymmetric distribution of curvature in momentum space (Son et al., 2019). Strained monolayer WSe$\mathcal{O}^{ij,\sigma}_{\alpha,\mathbf{k}}$9 realizes the same logic experimentally in a second-order nonlinear Hall setting (Qin et al., 2020).

The oxide-interface literature sharpens this distinction by separating spin-sourced and orbital-sourced Berry curvature. At the (111) LaAlO$B_z$0/SrTiO$B_z$1 interface, orbital Rashba hybridization of Ti $B_z$2 states produces an orbital Berry-curvature dipole and a zero-field nonlinear Hall response, whereas spin-sourced Berry curvature is isolated through an anomalous planar Hall effect in an in-plane field (Lesne et al., 2022). This provides a concrete example in which the internal quantum number generating the curvature matters experimentally.

A further misconception is that vanishing Chern number implies geometric triviality. Several works contradict that inference in different ways. In Floquet flat-band systems, local Berry curvature can be nonzero under detuning even when the Chern number remains zero, and orbital magnetization then depends on Berry-curvature-weighted quasienergy rather than on the global invariant alone (Dag et al., 2022). In quasi-2D parity-violating antiferromagnets, the net Berry curvature and net orbital magnetization can both vanish, yet a stacking Berry curvature dipole

$B_z$3

survives because states are weighted by layer polarization; this dipole becomes a microscopic source of longitudinal magnetoelectricity and electric-field-induced Hall response (Hu, 25 Nov 2025). At a more formal level, spin-orbit-coupled Bose–Einstein condensates admit a local topological obstruction to flattening Berry curvature even when the total Chern number vanishes, expressed through a mixed cohomology class on $B_z$4 and the bound $B_z$5 (Pigazzini et al., 22 Dec 2025). Taken together, these works distinguish global topology from locally irremovable curvature.

5. Experimental manifestations and measurement strategies

Momentum-resolved tunneling spectroscopy provides a direct probe of orbital Berry-curvature effects in band dispersion. In TbV$B_z$6Sn$B_z$7, SI-STM combined with QPI tracks a Dirac-like band near $B_z$8 and resolves its magnetic-field-induced splitting into two branches with strongly momentum-dependent effective $B_z$9 factors. The STM spectra show a peak near the band bottom that splits under field, and the QPI dispersion shows the corresponding momentum-space splitting. The effect persists to fields much higher than the saturation field of Tb magnetization, supporting its interpretation as an itinerant orbital Zeeman effect of the V kagome bands rather than localized-moment magnetism (Li et al., 2023).

Optical probes access electrically induced orbital magnetization. In strained monolayer MoS$\Omega_{n\mathbf{k}}(\mathbf{r}_\alpha)$0, Kerr rotation spectroscopy detects current-induced valley orbital magnetization: negligible Kerr signal at zero strain, approximately linear growth with strain, and sign reversal under reversal of strain or current direction (Son et al., 2019). In monolayer WSe$\Omega_{n\mathbf{k}}(\mathbf{r}_\alpha)$1, second-harmonic Hall measurements under piezoelectric strain and ionic-liquid gating reveal a strain-tunable Berry-curvature dipole, an extracted scale $\Omega_{n\mathbf{k}}(\mathbf{r}_\alpha)$2 nm, and an orbital magnetization per current density $\Omega_{n\mathbf{k}}(\mathbf{r}_\alpha)$3 in the units used in that work (Qin et al., 2020).

Transport experiments at oxide interfaces reveal both nonlinear and planar Hall signatures. At the (111) LaAlO$\Omega_{n\mathbf{k}}(\mathbf{r}_\alpha)$4/SrTiO$\Omega_{n\mathbf{k}}(\mathbf{r}_\alpha)$5 interface, a second-harmonic transverse voltage persists at zero magnetic field, consistent with an intrinsic Berry-curvature-dipole-driven nonlinear Hall effect. The extracted dipole reaches a peak near $\Omega_{n\mathbf{k}}(\mathbf{r}_\alpha)$6 nm, markedly larger than in several previously studied nonmagnetic platforms. The same system also exhibits an anomalous planar Hall effect that isolates spin-sourced Berry curvature under mirror-symmetry-breaking in-plane fields (Lesne et al., 2022).

Photoemission offers a more differential route. Circular-dichroic ARPES measures the left–right photoemission asymmetry $\Omega_{n\mathbf{k}}(\mathbf{r}_\alpha)$7, which tracks momentum-resolved orbital angular momentum and therefore local Berry curvature in the two-band regime where $\Omega_{n\mathbf{k}}(\mathbf{r}_\alpha)$8. Spin-resolved detection extends the method to spin-Chern systems, where opposite spin sectors carry opposite Berry-curvature textures that would otherwise cancel (Schüler et al., 2019). This is especially important for mapping local rather than Brillouin-zone-integrated topology.

6. Real-space, local, and dynamical generalizations

Orbital Berry curvature is not confined to static momentum-space band theory. In two-component microcavity polaritons supporting a double full Bloch beam, a real-space Berry curvature is defined from the pseudospin texture as

$\Omega_{n\mathbf{k}}(\mathbf{r}_\alpha)$9

equivalently the areal density with which the Bloch sphere is mapped onto the real plane. The velocity of observable vortex cores is then

$\Omega_n(\mathbf{r})$0

so the local curvature quantitatively controls orbital vortex kinematics (Dominici et al., 2022). This is a real-space analogue of the momentum-space statement that curvature governs anomalous dynamics.

In itinerant electron systems with spin-orbit coupling, the real-space Berry phase on a loop acquires a gauge-covariant form that generalizes scalar spin chirality. The continuum Berry curvature becomes

$\Omega_n(\mathbf{r})$1

combining the projection of the $\Omega_n(\mathbf{r})$2 field strength and a covariantized texture chirality. In this framework, collinear and coplanar magnetic textures can generate finite real-space Berry phase when spin-orbit coupling is present, thereby connecting double-exchange spin-chirality physics to Karplus–Luttinger-type orbital transport (Zhang et al., 2019).

Local formulations go still further by resolving orbital magnetization site by site. The theory of local orbital magnetization introduces a sublattice-resolved Berry curvature

$\Omega_n(\mathbf{r})$3

and, for open-boundary systems, an effective onsite Berry curvature

$\Omega_n(\mathbf{r})$4

The topological piece reduces to the usual Bloch-band curvature when summed over sublattices, whereas the geometric piece cancels only after unit-cell summation and can diagnose local orbital ferro-, antiferro-, or ferrimagnetic textures that are invisible in bulk-averaged formulations (Saati et al., 13 Dec 2025).

These extensions suggest that “orbital Berry curvature” is best understood not as a single formula but as a geometric organizing principle. In momentum space it governs orbital Hall response, orbital moments, and nonlinear transport; in real space it controls vortex and texture dynamics; in local form it resolves intra-unit-cell orbital magnetism; and in driven systems it becomes explicitly time dependent. A plausible implication is that future work will increasingly treat orbital Berry curvature as a multiscale concept linking Bloch-band geometry, local orbital texture, and nonequilibrium dynamics rather than as a property of a single transport coefficient alone.