Quantum Hall Geometry Preview

The integer quantum Hall effect is one of the cleanest places where Berry geometry becomes a measured transport coefficient. At the geometric level, the central statement is:

$$\sigma_{yx} = \frac{e^2}{h} \sum_{\mathrm{occupied}\ n} C_n$$

for a filled noninteracting band insulator at zero temperature. Here $e>0$, $A=i\langle u|du\rangle$, $C$ uses $dk_x\wedge dk_y$, and $j_y=\sigma_{yx}E_x$; the opposite component is $\sigma_{xy}=-\sigma_{yx}$. The integers $C_n$ are Chern numbers: quantized Berry-curvature fluxes over a closed two-dimensional parameter space.

The integer comes from the geometry of the occupied states. Connecting it to a measured Hall plateau also requires filling, a gap or localization, and a suitable transport regime.

From Hall Response to Geometry

Classically, a Hall response is a transverse current produced by an electric field in the presence of broken time-reversal symmetry. Quantum mechanically, two-dimensional gapped systems can have a transverse conductance that is quantized:

$$\sigma_{yx} = \nu\frac{e^2}{h}.$$

In the integer quantum Hall effect, $\nu$ is an integer in the simplest spin-resolved setting. The geometric interpretation identifies this integer with a Chern number of occupied quantum states.

The full material treatment is Integer Quantum Hall Effect, the compact named-effect summary is Quantum Hall Effect, and the historical card is Quantum Hall Discovery.

Band Chern Number View

For a two-dimensional isolated Bloch band, the Brillouin zone is a torus:

$$\mathrm{BZ}\simeq T^2.$$

The Berry curvature of the band is

$$F_n = dA_n, \qquad A_n = i\langle u_{n\mathbf k}|d u_{n\mathbf k}\rangle.$$

In coordinates,

$$C_n = \frac{1}{2\pi} \int_{\mathrm{BZ}} \Omega_n(\mathbf k)\,d^2k.$$

When occupied bands are separated from unoccupied bands by an energy gap, the total occupied Chern number controls the integer Hall response in the clean noninteracting band picture:

$$\sigma_{yx} = \frac{e^2}{h} C_{\mathrm{occ}}, \qquad C_{\mathrm{occ}} = \sum_{\mathrm{occupied}\ n}C_n$$

when individual occupied bands are separately isolated. If occupied bands mix among themselves, the total Chern number of the occupied subspace is the invariant object.

Landau-Level View

The original integer quantum Hall setting involves a two-dimensional electron gas in a strong magnetic field. The single-particle spectrum forms Landau Levels, each with a large degeneracy proportional to magnetic flux.

In the ideal clean spinless model, filling $\nu$ Landau levels gives

$$\sigma_{yx} = \nu\frac{e^2}{h}$$

for electrons with $B>0$ along $+\hat z$ in the response convention above. Each filled spin-resolved Landau level contributes one unit. Reversing $B$ reverses the Hall sign.

The ordinary Brillouin-zone formula is not always the natural language for continuum Landau levels. A more general geometric formulation uses boundary twists or magnetic translation parameters as the two-dimensional parameter space.

Boundary-Twist Chern Number

Place the system on a torus. For a periodic gauge background, twist the boundary conditions:

$$\Psi(\ldots,x_j+L_x,\ldots) = e^{i\theta_x} \Psi(\ldots,x_j,\ldots),$$

and similarly for $\theta_y$. With nonzero net magnetic flux, the appropriate gauge-transition factor must also multiply the right-hand side; a globally periodic vector potential is unavailable. The total flux must satisfy the consistency condition discussed under magnetic translations. The pair

$$(\theta_x,\theta_y) \in T^2$$

forms a parameter torus.

Assume the many-body ground state $\lvert\Psi_0(\theta_x,\theta_y)\rangle$ is nondegenerate and separated by a gap for all twists. Define its Berry curvature over the twist torus:

$$F_{\theta_x\theta_y} = \partial_{\theta_x}A_{\theta_y} - \partial_{\theta_y}A_{\theta_x},$$

with

$$A_{\theta_i} = i \langle\Psi_0|\partial_{\theta_i}\Psi_0\rangle.$$

The many-body Chern number is

$$C = \frac{1}{2\pi} \int_0^{2\pi} \int_0^{2\pi} F_{\theta_x\theta_y} \,d\theta_x\,d\theta_y.$$

For a finite torus, the Chern integral determines the twist-averaged Hall response. Identifying it with the response at one fixed twist uses the thermodynamic limit and appropriate locality/gap assumptions; a finite-system curvature at one twist need not be an integer. This is the formulation introduced by Niu, Thouless, and Wu (1985).

Why Quantization Is Robust

The Chern number is an integer. An integer cannot change continuously. Thus the invariant cannot change under a smooth gapped deformation. The corresponding ideal zero-temperature bulk Hall response stays quantized; finite temperature, finite size, and dissipation can affect a measured value without changing the invariant.

To change the integer, a gap must close or the assumptions behind the invariant must fail. In clean band language, Berry-curvature flux can be transferred between bands at a band touching. In a disordered quantum Hall sample, extended states and localization determine how plateaus form as magnetic field or density is varied.

The geometry explains the integer. The full plateau phenomenology also needs disorder, localization, edges, finite temperature, sample geometry, and interactions.

Edge States and Bulk Geometry

The same Chern number that quantizes bulk Hall response also predicts robust chiral edge modes in integer quantum Hall systems. This is a bulk-boundary statement: a nonzero bulk topological invariant forces protected boundary transport under appropriate physical assumptions.

At an interface, the net number and direction of chiral modes are controlled by the difference of the bulk invariants on the two sides, rather than by a local curvature peak.

Relation to Magnetic Translations

Landau levels have a large degeneracy because guiding-center coordinates and magnetic translations do not commute in the same way ordinary translations do. The magnetic translation algebra encodes flux through parallelograms:

$$T(\mathbf a)T(\mathbf b) = e^{i\varphi(\mathbf a,\mathbf b)} T(\mathbf b)T(\mathbf a).$$

This noncommuting geometry is part of the same magnetic structure that leads to Landau-level degeneracy and Hall response. See Magnetic Translations and Landau Levels Revisited.

Integer versus Fractional Hall Geometry

The integer quantum Hall effect can be understood through filled single-particle Landau levels or filled Chern bands, with Chern numbers of occupied states. The fractional quantum Hall effect is different. It is an interacting many-body phenomenon with fractional charge, anyonic statistics, topological degeneracy, and emergent gauge-theory descriptions.

Fractional Hall states also have Berry phases and topological invariants, but they are not obtained by simply assigning a noninteracting band Chern number to each electron. This page previews the integer and geometric starting point only.

Common Mistakes

• Treating the Hall conductance formula as valid without a gap or filled-band assumption.
• Ignoring charge-sign and orientation conventions in $\sigma_{yx}$.
• Thinking Berry curvature must be uniform for the Chern number to be quantized.
• Confusing a local Berry-curvature peak with the integer invariant.
• Assuming disorder is irrelevant to plateaus; topology explains robustness, but localization helps produce finite-width plateaus.
• Reducing the fractional quantum Hall effect to the integer band-Chern formula.
• Forgetting that boundary-twist Chern numbers use a parameter space, not ordinary real space.

Exercises

1. Compute the Hall conductance for occupied bands with total Chern number $C_{\mathrm{occ}}=3$ in the convention used here.

Solution

Use

$$\sigma_{yx} = \frac{e^2}{h} C_{\mathrm{occ}}.$$

For $C_{\mathrm{occ}}=3$,

$$\sigma_{yx} = 3\frac{e^2}{h}.$$

This is $j_y/E_x$. For the same axes and Berry convention, $j_x/E_y=-3e^2/h$.

2. A band has constant Berry curvature $\Omega=N/(2\pi)$ on a Brillouin-zone torus with coordinates $0\le k_x,k_y<2\pi$. Compute its Chern number.

Solution

The Chern number is

$$C = \frac{1}{2\pi} \int_{\mathrm{BZ}} \Omega\,d^2k.$$

The area of the coordinate torus is $(2\pi)^2$, so

$$\int_{\mathrm{BZ}} \Omega\,d^2k = \frac{N}{2\pi}(2\pi)^2 = 2\pi N.$$

Therefore

$$C=N.$$

3. Why can a Chern number remain fixed while the Berry curvature distribution changes?

Solution

The local Berry curvature can move around under smooth deformations of the Hamiltonian. The Chern number is the normalized integral over the whole closed parameter space. As long as the relevant gap remains open and the bundle remains defined, the integral is an integer that cannot change continuously. Therefore the local distribution may change while the total integer stays fixed.

4. What is the parameter space in the boundary-twist Chern number?

Solution

The parameter space is the torus of boundary phases:

$$(\theta_x,\theta_y) \in [0,2\pi)\times[0,2\pi) \simeq T^2.$$

The Berry curvature is computed from how the many-body ground state changes as these boundary twists are varied.

5. Why does the fractional quantum Hall effect require more than the filled-band Chern-number formula?

Solution

Fractional Hall states arise from strong interactions within a partially filled Landau level. They have correlated many-body wavefunctions, fractionalized quasiparticles, and topological order. A formula that sums Chern numbers of filled noninteracting bands does not capture those features. Many-body Berry phases and topological field theories are needed for the fractional case.

References

• R. B. Laughlin, “Quantized Hall conductivity in two dimensions,” Physical Review B 23, 5632–5633, 1981.
• Q. Niu, D. J. Thouless, and Y.-S. Wu, “Quantized Hall conductance as a topological invariant,” Physical Review B 31, 3372–3377, 1985.
• R. E. Prange and S. M. Girvin, eds., The Quantum Hall Effect, 2nd ed., Springer, 1990.
• D. J. Thouless, M. Kohmoto, M. P. Nightingale, and M. den Nijs, “Quantized Hall conductance in a two-dimensional periodic potential,” Physical Review Letters 49, 405–408, 1982.
• K. von Klitzing, G. Dorda, and M. Pepper, “New method for high-accuracy determination of the fine-structure constant based on quantized Hall resistance,” Physical Review Letters 45, 494–497, 1980.
• D. Xiao, M.-C. Chang, and Q. Niu, “Berry phase effects on electronic properties,” Reviews of Modern Physics 82, 1959–2007, 2010.