Effect of Electron Correlation on the Integer Quantum Hall Effect
Daniel Staros 1, Christopher Lane 1, Roxanne Tutchton 1, Jian-Xin Zhu 1,† · 2026
Read Effect of Electron Correlation on the Integer Quantum Hall Effect by Daniel Staros 1, Christopher Lane 1, Roxanne Tutchton 1, Jian-Xin Zhu 1,† (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2601-16453.txt.
Introduction
Daniel Staros and three coauthors test how grain links bend the integer quantum Hall act. Their 2026 preprint deals with the Hall act in a square grid. In that act, flow steps sit at whole count marks. Links twixt grains can shift or blur those steps. The team varies link strength by a push term in the maths.
Runs show how steps move as the push term grows. Clear shift maps help read Hall tests in real stuff. Real stuff always holds some grain links in its grid. A map of link acts firms up the pure Hall tale. The work is a code run study of linked grids.
Catalog record
| CREATOR | Daniel Staros 1, Christopher Lane 1, Roxanne Tutchton 1, Jian-Xin Zhu 1,† |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 30 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2601-16453
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Effect of Electron Correlation on the Integer Quantum Hall Effect Daniel Staros 1, Christopher Lane 1, Roxanne Tutchton 1, Jian-Xin Zhu 1,† Abstract We numerically investigate the effect of electron correlation on the integer quantum Hall effect in a square lattice. Increasing the correlation strength via the effective onsite repulsion parameter U U degrades the quantization of ν = 1 \nu=1 transverse conductance due to the interplay of correlation and the external magnetic field, which together induce periodic modulations in renormalized hopping parameters and site energies. Overall, this work demonstrates that the strength of electron correlation can significantly impact conductivity in the integer quantum Hall regime. Introduction. —The integer quantum Hall effect (IQHE) is a cornerstone of modern condensed matter theory 13 ; 10 ; 7 ; 1 . Fundamentally, it serves as a first example of topological quantum phase. The robust integer quantization of transverse conductivity, which defines the IQHE has led to multiple technological innovations including metrological standards for electrical resistance and IQHE-based interferometers for probing anyonic quasiparticle statistics 4 ; 2 ; 14 . The robustness of the IQHE stems from the topological protection afforded by highly degenerate cyclotron orbits, or Landau levels (LLs), against external perturbations. One type of perturbation is site disorder, an increase in which was previously shown to degrade the integer conductance in a noninteracting lattice model 2 ; 12 . Although its effects could also be considered a perturbation, electron correlation is presumed to play a minor role in the IQHE when the disorder effect is dominant, in contrast to the fractional quantum Hall effect 11 where correlation plays a major role. Accordingly, there has been little motivation to investigate the effect of correlation on microscopic models of the IQHE. Very recently, experimental evidence of an interaction-driven breakdown of the IQHE in GaAs/AlGaAs quantum wells was reported 5 . Specifically, a transport instability for the \nu=1 LL was identified, and a correlation-induced asymmetry in the dissipation current for filling just above and below \nu=1 provided evidence that the instability is dominated by electron-electron interactions. This is a striking demonstration that correlation can have an important impact on the IQHE. However, a fundamental understanding of the origin of such phenomena remains missing. In fact, the effect of correlation on the IQHE has been virtually unexplored in general, leaving a gap in understanding that could help bridge the single-particle-like nature of the IQHE with the many-body nature of the fractional quantum Hall effect 5 . In this Letter, we present a simplified study of the effect of electron correlation on the IQHE. Within the Gutzwiller approximation, we elucidate how electron correlation renormalizes the bands of electrons on a square lattice subject to a strong, perpendicular magnetic field. Our band renormalization accounts for the fact that the translation invariance of the square lattice is modified by the external magnetic field 1 . Remarkably, we find a key similarity to the recent experiment of Ref. 5 , where correlation degrades the quantization of our \nu=1 LL. We identify the correlation-induced closing of the mobility gap between the \nu=1 and \nu=2 LLs as the explanation for this effect, and show that band bending, ultimately due to spatial modulations of the local electronic structure, underlies this gap closing. Our work thus establishes that correlation alone can degrade the IQHE in spatially unrestricted Gutzwiller-renormalized lattice models, and outlines a clear electronic mechanism. More broadly, it provides a conceptual link between noncorrelated (integer) and correlated (integer-degraded) quantum Hall effects from a real-space perspective, which allows for the spatial modulation. Figure 1: Schematic of the magnetized square lattice model illustrating nearest neighbor hopping parameterized by t_{a} and t_{b} , lattice dimensions and perpendicular magnetic field B . Theoretical model and methods. —We consider a Hubbard model defined on a two-dimensional (2D) square lattice with a strong magnetic field applied perpendicular the lattice plane, as schematically shown in Fig. 1 . Within a Landau gauge, the model Hamiltonian can be written in real space as \displaystyle\hat{H}=\sum_{m,n,s}\biggl\{\epsilon_{m,n,s}c_{m,n,s}^{\dagger}c_{m,n,s}-\biggl(te^{-i2\pi\phi n}c_{m+1,n,s}^{\dagger}c_{m,n,s} \displaystyle+tc_{m,n+1,s}^{\dagger}c_{m,n,s}+\textnormal{H.c.}\biggr)\biggr\}+\sum_{m,n}Un_{m,n,\uparrow}n_{m,n,\downarrow}\;. (1) Here the operators c_{m,n,s}^{\dagger} ( c_{m,n,s} ) create (annihilate) an electron at site \textbf{r}_{m,n}=ma\hat{x}+na\hat{y} for integers m and n such that 0\leq m<N_{x} and 0\leq n<N_{y} (we choose a=1 ) with periodic boundary conditions. Local onsite energies are denoted by \epsilon_{m,n,s} , the physical quantity t is the hopping integral along \hat{x} and \hat{y} , and s\in\{\uparrow,\downarrow\} indexes the spin. The perpendicular magnetic field \mathbf{B}=(0,0,B_{z}) has vector potential \mathbf{A}=(-yB_{z},0,0) where \phi=1/q defines the corresponding magnetic flux per plaquette and q is commensurate with N_{y} . Figure 2: Optimal parameters as a function of site index and U ( q=20 ); (top left) hopping renormalization \alpha_{j,s} , (top right) double occupancy d_{j} , (bottom left) site renormalization energies \lambda_{j,s} , and (bottom right) quasiparticle occupations \bar{n}_{j,s}^{qp} . Under the Gutzwiller wavefunction approximation 3 ; 8 ; 15 ; 9 , the Hamiltonian of Eq. ( 1 ) can be partitioned into q sectors along k_{y} due to the magnetic field, \displaystyle H=\sum_{s}\sum_{k_{x}=-\pi}^{\pi}\sum_{k_{y}=-\pi/q}^{\pi/q}\hat{h}_{s}(k_{x},k_{y})+N_{k_{x}}N_{k_{y}}\sum_{j=0}^{q-1}Ud_{j} (2) where N_{k_{x}}=N_{x} and N_{k_{y}}=N_{y}/q of the corresponding real-space lattice of size N_{x}\times N_{y} , and \displaystyle\hat{h}_{s}(k_{x},k_{y})=\sum_{j=0}^{q-1}\left[(\epsilon_{j,s}+\lambda_{j,s})-2t\alpha_{j,s}\cos(k_{x}+2\pi\phi j)\right]\times \displaystyle c_{k_{x},k_{y},j,s}^{\dagger}c_{k_{x},k_{y},j,s}-t\sqrt{\alpha_{j,s}}\left(e^{-ik_{y}}\sqrt{\alpha_{j+1,s}}c_{k_{x},k_{y},j+1,s}^{\dagger}\right. (3) \displaystyle\left.+e^{ik_{y}}\sqrt{\alpha_{j-1,s}}c_{k_{x},k_{y},j-1,s}^{\dagger}\right)c_{k_{x},k_{y},j,s}\;. Here, k_{x} and k_{y} are the x - and y -components of the reciprocal lattice vector describing periodicity of the magnetized unit cell, and j indexes the j^{th} plaquette within a unit cell at a given reciprocal lattice point. The local electronic structure of our solutions is governed by: (1) the Gutzwiller renormalization parameters \alpha_{j,s} , (2) the double occupancy parameters d_{j} , and (3) the site renormalization energies \lambda_{j,s} . Hopping parameters are scaled by Gutzwiller renormalization factors \sqrt{\alpha_{j,s}} defined by \displaystyle\sqrt{\alpha_{j,s}}=\left[\frac{\left(\bar{n}_{j,s}-d_{j}\right)\left(1-\sum_{s}\bar{n}_{j,s}+d_{j}\right)}{\bar{n}_{j,s}(1-\bar{n}_{j,s})}\right]^{1/2} \displaystyle+\left[\frac{d_{j}\left(\bar{n}_{j,\bar{s}}-d_{j}\right)}{\bar{n}_{j,s}(1-\bar{n}_{j,s})}\right]^{1/2}, (4) where stronger correlation (larger values of U ) corresponds to smaller values of \sqrt{\alpha} , thereby suppressing hopping between sites. We self-consistently solve for the values of \{\alpha_{j,s}\} , \{d_{j}\} and \{\lambda_{j,s}\} that minimize the expectation value of the Hamiltonian in Eq. ( 2 ) subject to the self-consistency conditions on \lambda_{j,s} and d_{j} that are provided in the SI. Electronic structure. —To investigate the effect of correlation on the electronic structure of the magnetized square lattice, we test our model over correlation strengths ranging from U=0.1t to U=9t . The real-space and k -space grids underlying these calculations correspond to N_{x}=N_{y}=20 , which are sufficiently converged to resolve quantization of the first three Landau levels in the noninteracting limit. We choose the hopping integral t=1 to ensure that any anisotropy in the optimized Gutzwiller parameters is solely due to the interplay between the correlation and magnetic orbital effects, and consider the smallest value of the magnetic field which is consistent with the given lattice dimensions, B=2\pi/q with q=20 . All Gutzwiller optimizations are performed at half filling. As a rule of thumb, solutions of a Gutzwiller-renormalized lattice model are best behaved over a range of U values corresponding to twice the total bandwidth. Our model converges up to U=8t , converges with self-consistent mixing at U=9t , but is significantly more challenging to converge for greater values of U . This is consistent with our bandwidth of \sim 8t . Figure 3: Plots of transverse conductivity in units of e^{2}/h vs. chemical potential \mu (left) and corresponding k_{x} ( k_{y}=0 ) band dispersion (right)