Continuous transition from Fermi liquid to A fractional Chern insulator

Ya-Hui Zhang · 2025

Read Continuous transition from Fermi liquid to A fractional Chern insulator by Ya-Hui Zhang (2025) 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-2507-22130.txt.

Introduction

Ya-Hui Zhang writes a solo 2025 physics paper on a smooth shift between two states of matter. Recent tests on moiré materials, made by stacking thin sheets with a twist, found a rare state called a fractional Chern insulator with no magnet applied. Zhang asks how that state can melt into a plain Fermi liquid, the state of electrons in most metals, as forces or band width change.

The theory is built at a filling of two thirds. On one side sits a state with the same charge signals as the rare state but with an added neutral part. On the other side, hot behavior looks like a gas of odd charge carriers. Only at very low heat does the plain metal state come back. Twisted MoTe2 may show this shift, Zhang notes.

Catalog record

CREATORYa-Hui Zhang
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE47 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2507-22130

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Continuous transition from Fermi liquid to A fractional Chern insulator Ya-Hui Zhang Abstract Recent experiments in moiré materials have observed fractional Chern insulators (FCI) at zero magnetic field, providing an opportunity to study the transition from FCI to the more conventional phases such as Fermi liquid (FL) and superconductor (SC) by tuning the interaction strength or bandwidth. In this work, we formulate a critical theory for a continuous transition at the filling ν = 2 3 \nu=\frac{2}{3} between the FL and a FCI* phase that hosts an additional neutral sector, but has the same transport signatures as the usual FCI. In our framework, this corresponds to a transition from a composite Fermi liquid (CFL) to a superfluid* phase of the composite bosons. The Fermi liquid close to the transition has an additional factor | z i − z j | 6 |z_{i}-z_{j}|^{6} in its wavefunction. Also, the transport behavior at high temperature on the FL side is actually like an ‘anyon gas’ phase on top of the FCI, with ρ x ​ y \rho_{xy} close to 3 2 ​ h e 2 \frac{3}{2}\frac{h}{e^{2}} . FL behavior with ρ x ​ y ≈ 0 \rho_{xy}\approx 0 is recovered only at very low temperature. We also briefly discuss the possibility of a chiral superconductor as the descendant of this strongly correlated FL and the potential relevance to the twisted MoTe 2 system. I Introduction Fractional Chern insulators Sun et al. 2011 ; Sheng et al. 2011 ; Neupert et al. 2011a ; Wang et al. 2011 ; Tang et al. 2011 ; Regnault and Bernevig 2011 ; Bergholtz and Liu 2013 ; Parameswaran et al. 2013 ; Neupert et al. 2011b at fractional filling of the topological bands of moiré systems Zhang et al. 2019 ; Wu et al. 2019 have attracted lots of attentions after the experimental realizations in twisted MoTe 2 homobilayer Cai et al. 2023 ; Zeng et al. 2023 ; Park et al. 2023 ; Xu et al. 2023 and in rhombohedrally stacked multilayer graphene proximated to an aligned hexagon boron nitride (hBN) Lu et al. 2024 . There are already many theoretical studies Yu et al. 2020 ; Devakul et al. 2021 ; Li et al. 2021 ; Crépel and Fu 2023 ; Wang et al. 2023 ; Reddy et al. 2023 ; Xu et al. 2023 ; Yu et al. 2023 ; Goldman et al. 2023 ; Dong et al. 2023a ; Morales-Durán et al. 2023 ; Song et al. 2024 ; Dong et al. 2023b ; Zhou et al. 2023 ; Dong et al. 2023c ; Guo et al. 2023 ; Kwan et al. 2023 trying to understand the ideal condition to optimize FCIs and their competing states. From a purely conceptual perspective, the currently realized FCI is likely in the same phase as the fractional quantum Hall (FQH) state at high magnetic field Stormer et al. 1999 ; Stormer 1999 , which has been well understood. One natural next question is: can we realize completely new physics unprecedented in the Landau level systems at high magnetic field? One unique opportunity of the moiré systems is that it is easy to tune the ratio between the bandwidth and the interaction. Therefore it is possible to study the evolution from the FCI to a conventional state such as a Fermi liquid or a superconductor. Theories already exist for continuous transition between the bosonic Laughlin state and superfluid at filling \nu=\frac{1}{2} Barkeshli and McGreevy 2014 ; Song and Zhang 2023 ; Lu et al. 2025 ; Wang et al. 2025 ; Lotrič and Simon 2025 and between a composite Fermi liquid (CFL) and Fermi liquid for electron at \nu=\frac{1}{2} Barkeshli and McGreevy 2012 ; Song et al. 2024 . In this work we turn our attention to electron at the filling \nu=\frac{1}{3} . A continuous transition between a FCI and a charge density wave (CDW) insulator has already been pointed out Song et al. 2024 . However, the most robust state at larger bandwidth is the Fermi liquid (FL) and it is not clear whether the FL can directly transits into a FCI insulator. In this work, we provide a theory of a possible continuous transition from the FL to an intermeidate FCI phase (dubbed as FCI*) at \nu=\frac{2}{3} by decreasing the bandwidth. The FCI* phase has exactly the same resistivity tensor \rho as the familiar FCI, but it hosts an additional neutral sector which only contributes to thermodynamic measurement (such as specific heat) and thermal transport. Unfortunately we are still not able to find a direct transition from FL to the conventional FCI without the additional neutral degrees of freedom. We use the composite boson theory, which can be constructed by the usual slave boson parton decomposition: c(\mathbf{x})=b(\mathbf{x})f(\mathbf{x}) . Then we put f in the FCI phase. Then the composite boson b feels an effective flux of \frac{2\pi}{3} per unit cell. If b is in a composite Fermi liquid (CFL) phase, electron is in a conventional FL phase, but with an additional factor \prod_{i<j}|z_{i}-z_{j}|^{6} in its wavefunction. On the other hand, if b forms a superfluid (SF), electron is in the usual FCI phase. In this language, the FL-FCI transition corresponds to the CFL-SF transition of the composite boson b . The later is still a quite challenging problem. Here we make a compromise and consider a simpler transition for the boson. Starting from the CFL of b , we simply condense the boson operator b and reach a SF* phase with superfluid of the boson coexisting with a neutral sector. For the electron c , this is the FCI* phase. Therefore the critical theory of the FL-FCI* transition is from the condensation of a boson field. In the FCI* phase, the composite boson b is in a mixture of condensation part and a normal part formed by a composite fermion \psi . We expect the FCI* phase to occupy at most a very narrow region of the phase diagram and the density of \psi may quickly vanish at a second critical point between the FCI* phase and the usual FCI phase by further decreasing the bandwidth. The existence of an intermediate FCI* phase may depend on model details and the critical point may be replaced with another competing state such as a CDW insulator. On the other hand, some of our predictions may only rely on a correlated FL phase proximate to the FCI, as long as it hosts a small gap to the composte boson excitations. We find that the resistivity tensor has a strong deviation from typical FL phase when increasing the temperature above the gap of composite boson. At finite T , there is a finite density of thermally excited composite bosons, which depletes the density associated with the Fermi surface. The high temperature regime has a \rho_{xy} close to \frac{3}{2}\frac{h}{e^{2}} and a small \rho_{xx} , similar to an ‘anyon gas’ phase on top of a FCI. At lower temperature the FL behavior is recovered with \rho_{xy}\approx 0 . In principle, this unusual temperature dependence does not rely on a critical point, and may be associated with a strongly correlated Fermi liquid with cheap composite boson excitations. We propose to search for

Read the complete text