Energy Gap in Weakly Disordered Fractional Quantum Hall Liquids: Quantitative Comparison to GaAs Quantum Well Experiments at ν = 1 / 3 \nu=1/3
Yi-Han Zhou, Zi-Ang Wang, Xin Wan, Zhao Liu · 2026
Read Energy Gap in Weakly Disordered Fractional Quantum Hall Liquids: Quantitative Comparison to GaAs Quantum Well Experiments at ν = 1 / 3 \nu=1/3 by Yi-Han Zhou, Zi-Ang Wang, Xin Wan, Zhao Liu (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-2606-29305.txt.
Introduction
Zhou and three colleagues compute the energy gap of a fractional quantum Hall liquid and compare it with experiment. At filling factor one third, electrons in a strong magnetic field form a collective state with a gap: a minimum energy to excite it. A recent experiment on high-quality gallium arsenide quantum wells measured that gap, but theory often ignores layer thickness and disorder.
The team model the real devices. Solving coupled Poisson-Schrodinger equations gives the electron wavefunctions across the layer, and they take the disorder energy from the experiment itself. For narrow wells, both their charge gap and mobility gap match the measured activation gap well. For wide wells, the agreement points to a missing ingredient: higher subbands may be needed before theory can predict the gap accurately there.
Catalog record
| CREATOR | Yi-Han Zhou, Zi-Ang Wang, Xin Wan, Zhao Liu |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 40 KB |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2606.29305
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Energy Gap in Weakly Disordered Fractional Quantum Hall Liquids: Quantitative Comparison to GaAs Quantum Well Experiments at ν = 1 / 3 \nu=1/3 Yi-Han Zhou, Zi-Ang Wang, Xin Wan, Zhao Liu Abstract Based on a recent experiment in high-quality GaAs quantum wells [Phys. Rev. Lett. 127, 056801 (2021)], we present a microscopic study of the energy gap in two-dimensional electron gases at filling factor ν = 1 / 3 \nu=1/3 , explicitly incorporating both finite layer thickness and disorder effects. The finite layer thickness is modeled by solving the Poisson-Schrödinger equations for the experimental devices, yielding the electron wave functions in the perpendicular direction. Using these and the disorder energy extracted from the experiment, we estimate the charge gap and the mobility gap at ν = 1 / 3 \nu=1/3 in the weakly disordered lowest Landau level. Remarkably, both gaps show good quantitative agreement with the activation gap measured from the experiment in narrow quantum wells. Our results also indicate the potential need of incorporating higher subbands to make accurate theoretical predictions of the energy gap in wide quantum wells. Introduction. Rich many-body phenomena emerge in two-dimensional electron gases (2DEGs) subjected to strong perpendicular magnetic fields and cooled to ultra-low temperatures. Among these, the fractional quantum Hall effect (FQHE) stands out as one of the most celebrated examples, having attracted tremendous attention since its discovery over 40 years ago 30 . The intrinsic topological order of FQHE states strikingly distinguishes them from traditional phases characterized by symmetry breaking. This unique feature renders certain FQHE states promising platforms for topological quantum computation 20 . Electrons form an incompressible liquid when the FQHE occurs. This incompressibility is characterized by a nonzero charge gap \Delta_{c} , which measures the energy cost to create a far-separated quasiparticle-quasihole pair in the system. In experiments at very low temperatures, the charge gap manifests itself in the thermally activated longitudinal resistance R_{xx}\propto e^{-\frac{\Delta_{a}}{2k_{B}T}} , where k_{B} is the Boltzmann constant and T is the temperature. The activation gap \Delta_{a} depends on many details of the system, including the finite layer thickness of the 2DEG, the disorder strength, Landau level (LL) mixing, etc. For a long time, there has been interest in quantitatively comparing the theoretically calculated and experimentally measured energy gap, particularly for the strongest and first observed FQHE at filling factor \nu=1/3 12 ; 11 ; 2 ; 34 ; 6 ; 17 ; 24 ; 19 ; 33 ; 23 ; 31 . On the theoretical side, numerical simulations in the lowest Landau level (LLL) for an ideal 2DEG without disorder and layer thickness yield \Delta_{c}\approx 0.1e^{2}/(4\pi\epsilon l_{B}) 12 ; 11 , where \epsilon is the dielectric constant and l_{B}=\sqrt{\hbar/(eB)} is the magnetic length. Efforts have also been made to incorporate realistic factors, such as finite layer thickness 37 ; 17 ; 24 ; 19 ; 38 and LL mixing (LLM) 17 ; 35 ; 5 ; 38 . Although numerical calculations successfully capture the suppression of the gap with increasing layer thickness and LLM, a recent experiment that systematically extracted the activation gap at \nu=1/3 in extremely high-quality 2DEGs reported appreciably lower \Delta_{a} than theoretical \Delta_{c} obtained in clean systems, even after the reduction of energy gap caused by disorder has been empirically compensated 31 . In this work, we aim to resolve the discrepancy between the experimental gap data in Ref. 31 and theoretical predictions. To accurately account for the finite thickness of realistic 2DEG samples, we consider a typical modulation-doped GaAs quantum well structure that approximates the experimental setup of Ref. 31 . Instead of using trial wave functions 8 ; 37 ; 33 ; 19 , we numerically solve the self-consistent Poisson-Schrödinger equations 18 ; 32 ; 26 to extract the electron subband wave function in the perpendicular direction. On the other hand, while most previous theoretical works focused on clean systems, we quantitatively model the disorder in experimental devices instead of empirically compensating the disorder effect. In the presence of random impurities, excitations localized by the disorder cannot contribute to the longitudinal resistance. Therefore, the mobility gap \Delta_{m} , the energy required to excite delocalized quasiparticles and quasiholes, is a proper estimate of the experimentally measured \Delta_{a} 27 ; 33 . By contrast, the charge gap \Delta_{c} may underestimate \Delta_{a} due to the disorder induced localized excitations. However, we expect this difference to be sufficiently small in ultra-high quality samples with very weak disorder. To corroborate this point, we keep the lowest subband and compute \Delta_{c} and \Delta_{m} using disorder strengths extracted from the experimental data. Remarkably, they are indeed close to each other, and both show good quantitative agreement with experimentally measured \Delta_{a} for various narrow quantum wells in Ref. 31 . Since the computation of \Delta_{c} only requires the lowest energies under periodic boundary conditions, our work provides a more efficient method for the quantitative comparison between theoretical and experimental data of the energy gap in weakly disordered samples. In wide quantum wells, our theoretical prediction of the energy gap shows discrepancy from the experimental data, which may result from the effects of higher subbands. Model. We consider N_{e} spin-polarized electrons of charge -e on an L\times L=2\pi N_{\phi}l_{B}^{2} square torus with a perpendicular uniform magnetic field, where N_{\phi} is the number of magnetic flux quanta penetrating the torus. The LL filling factor is defined as \nu=N_{e}/N_{\phi} . The Coulomb energy E_{C}=e^{2}/(4\pi\epsilon l_{B}) is used as the energy unit in this section. For numerical efficiency, we neglect the LLM which is suppressed at strong magnetic fields. The many-body Hamiltonian projected to the LLL, consisting of the interaction and disorder potential, takes the form \displaystyle H \displaystyle= \displaystyle\sum_{m_{1},m_{2},m_{3},m_{4}=0}^{N_{\phi}-1}V_{m_{1},m_{2},m_{3},m_{4}}c^{\dagger}_{m_{1}}c^{\dagger}_{m_{2}}c_{m_{3}}c_{m_{4}} (1) \displaystyle+ \displaystyle\sum_{m_{1},m_{2}=0}^{N_{\phi}-1}U_{m_{1},m_{2}}c^{\dagger}_{m_{1}}c_{m_{2}}, where c^{\dagger}_{m} and c_{m} are the operators creating and annihilating an electron in the LLL orbital m , respectively. We impose the twisted boundary condition \mathcal{T}_{\mu}\psi_{m}({\bf r})=e^{i\theta_{\mu}}\psi_{m}({\bf r}) on the LLL orbitals \psi_{m}({\bf r}) , where \mathcal{T}_{\mu} is the magnetic translation operator in the \mu=x,y direction and \theta_{\mu} is the boundary phase. Under the Landau gauge \bm{A}=(0,-Bx) , the interaction matrix element V_{\{m_{i}\}} and disorder matrix element U_{m_{1},m_{2}} are \displaystyle V_{\{m_{i}\}}=\frac{1}{2}\delta^{\text{mod}\ N_{\phi}}_{m_{1}+m_{2},m_{3}+m_{4}}\sum_{s,t=-\infty}^{+\infty}\delta^{\text{mod}\ N_{\phi}}_{t,m_{1}-m_{4}}V_{\bm{q}} \displaystyle\times e^{-\frac{1}{2}q^{2}l_{B}^{2}}e^{i\frac{2\pi s}{N_{\phi}}(m_{1}-m_{3})}e^{i\frac{\theta_{x}}{N_{\phi}}(m_{1}+m_{2}-m_{3}-m_{4})} (2) and \displaystyle U_{m_{1},m_{2}}=\sum_{s,t=-\infty}^{+\infty}\delta_{t,m_{1}-m_{2}}^{\text{mod}\ N_{\phi}}U_{\bm{q}}e^{-\frac{1}{4}q^{2}l_{B}^{2}} \displaystyle\times e^{i\frac{\pi s}{N_{\phi}}(2m_{1}-t+\theta_{y}/\pi)}e^{i\frac{\theta_{x}}{N_{\phi}}(m_{1}-m_{2}-t)}, (3) respectively, where \delta_{i,j}^{\text{mod}\ N_{\phi}} is the periodic Kronecker delta function with period N_{\phi} , \bm{q}=(q_{x},q_{y})=(\frac{2\pi s}{L},\frac{2\pi t}{L}) with integers s and t , q\equiv|{\bm{q}}| , and V_{\bm{q}} and U_{\bm{q}} are the Fourier transforms of the interaction potential and the impurity potential, respectively. We use the Coulomb interaction with V_{\bm{q}}=\frac{1}{N_{\phi}}\frac{1}{ql_{B}}F(q) , where the factor F(q) accounts for the finite layer thickness and depends on the electron distribution along the z -axis (the form will be given later). The s=t=0 must be excluded from the sum in Eq. ( 2 ) to remove the q=0 singularity of V_{\bm{q}} . Physically this can be achieved by involving the interaction of electrons with their images and a neutralizing background charge 36 . We use the Gaussian white noise to model the disorder potential U({\bm{r}}) , which satisfies \langle U({\bm{r}})\rangle=0,\langle U({\bm{r}})U({\bm{r}}^{\prime})\rangle=W^{2}l_{B}^{2}\delta({\bm{r}}-{\bm{r}}^{\prime}) . The disorder strength is measured by W , and \langle\cdots\rangle represents the average over disorder configurations. Accordingly, we have \langle U_{\bm{q}}\rangle=0,\ \langle U_{\bm{q}}U_{\bm{q}^{\prime}}\rangle=\frac{W^{2}}{2\pi N_{\phi}}\delta_{\bm{q},-\bm{q}^{\prime}} (4) in the momentum space. In each disorder configuration, the real U_{{\bm{q}}={\bm{0}}}