Symmetry-enforced Fermi surfaces
Minho Luke Kim, Salvatore D. Pace, Shu-Heng Shao · 2025
Read Symmetry-enforced Fermi surfaces by Minho Luke Kim, Salvatore D. Pace, Shu-Heng Shao (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-2512-04150.txt.
Introduction
A Fermi surface, the boundary in momentum space between filled and empty electron states, is usually something a material happens to have. Minho Luke Kim, Salvatore D. Pace, and Shu-Heng Shao prove some systems cannot avoid having one, in a 2025 preprint. They construct a symmetry forcing every model that respects it to be gapless in exactly this way.
The symmetry mixes ordinary fermion number conservation with a non-local Majorana translation, generating a noncompact Lie group related to the Onsager algebra. For any Fermi surface this symmetry enforces, the authors show the symmetry group always contains the even part of the ersatz Fermi liquid symmetry that such surfaces carry. They also prove these enforced surfaces generically have at least two noncontractible components, like open orbits wrapping momentum space.
Catalog record
| CREATOR | Minho Luke Kim, Salvatore D. Pace, Shu-Heng Shao |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 67 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2512-04150
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Symmetry-enforced Fermi surfaces Minho Luke Kim, Salvatore D. Pace, Shu-Heng Shao Abstract We identify a symmetry that enforces every symmetric model to have a Fermi surface. These symmetry-enforced Fermi surfaces are realizations of a powerful form of symmetry-enforced gaplessness. The symmetry we construct exists in quantum lattice fermion models on a d d -dimensional Bravais lattice, and is generated by the onsite U(1) fermion number symmetry and non-onsite Majorana translation symmetry. The resulting symmetry group is a non-compact Lie group closely related to the Onsager algebra. For a symmetry-enforced Fermi surface ℱ {\cal F} , we show that this UV symmetry group always includes the subgroup of the ersatz Fermi liquid L ℱ U(1) symmetry group formed by even functions f ( 𝐤 ) ∈ U ( 1 ) {f(\mathbf{k})\in\mathrm{U}(1)} with 𝐤 ∈ ℱ {\mathbf{k}\in{\cal F}} . Furthermore, we comment on the topology of these symmetry-enforced Fermi surfaces, proving they generically exhibit at least two non‑contractible components (i.e., open orbits). MIT-CTP/5947 I Introduction Quantum phases of matter come in two distinct flavors: gapped phases and gapless phases. A gapped phase is a phase whose energy spectrum in the thermodynamic limit has a finite gap separating the ground state energy from the first excited state energy. The landscape of gapped phases has been heavily explored and is remarkably rich. It includes, for example, discrete symmetry breaking phases, symmetry-protected topological and topologically ordered phases W161003911 , and fracton phases NH180311196 . A gapless phase is a phase whose energy spectrum in the thermodynamic limit does not have a finite gap separating the ground state energy from the first excited state energy. The landscape of gapless phases is under much less control compared to that of gapped phases. One clear thing, however, is that not all gapless phases are equally gapless. Indeed, gapless phases can be organized by, roughly speaking, the number of gapless excitations—the number of gapless fields in the infrared (IR) effective field theory description of the phase. Some gapless phases have a finite number of gapless excitation types. This includes, for example, conventional ordered phases with Goldstone bosons and critical phases whose IR limit is described by conformal field theories. There are also gapless phases with an infinite number of gapless excitations. A prototypical example of such a gapless phase is a metal, which has an entire Fermi surface worth of gapless excitations. A fruitful approach to studying gapless phases is to identify ultraviolet (UV) constraints that forbid gapped phases, thereby enforcing gaplessness. This is most commonly realized by imposing anomalous UV symmetries that enforce gaplessness WS14011142 ; WS160406807 ; SKW160908616 ; WNM170302426 ; GHO171004218 ; CO191004962 ; CO191213069 ; ACL221214605 ; CPS240912220 ; PCS241218606 ; PKC250504684 . We distinguish two notions of this symmetry-enforced gaplessness (SEG): weak and strong SEG. 1 1 1 What we call weak SEG is often referred to as just SEG in the literature. Weak SEG arises when an anomalous symmetry is incompatible with topological/fracton order, and enforces gaplessness whenever its discrete symmetries are not spontaneously broken. Weak SEG permits gapped, discrete symmetry-broken phases. A trivial example of weak SEG is from anomalous discrete symmetries in {(1+1)} D because there is no topological/fracton order in {(1+1)} D. One of the first nontrivial examples of weak SEG WS14011142 is from a {(\mathrm{SU}(2)\times\mathbb{Z}_{4}^{T})/\mathbb{Z}_{2}} symmetry in {(2+1)} D whose ’t Hooft anomaly enforces either gaplessness or a gapped phase with spontaneously broken \mathbb{Z}_{4}^{T} time-reversal symmetry. Strong SEG, on the other hand, occurs when an anomalous symmetry is incompatible with all gapped phases. Strong SEG enforces gaplessness regardless of whether its symmetry is spontaneously broken. For example, perturbative anomalies of continuous global symmetries cause strong SEG because they are encoded in the local operator product expansion and, therefore, cannot be matched by any gapped phase. A less trivial example of strong SEG is Witten’s SU(2) anomaly in {(3+1)} D GHO171004218 . All known examples of SEG, both of the weak and strong form, enforce gaplessness that is compatible with a finite number of gapless excitations described by a conventional continuum field theory. It is interesting to wonder, however, if there can be an SEG that requires an infinite number of gapless excitations. In this paper, we answer the following question in the affirmative: Is there a symmetry that can enforce a Fermi surface? In particular, we show that quantum lattice models with a single, spinless fermion per unit cell must have a Fermi surface if they (1) conserve the total fermion number and (2) commute with Majorana lattice translations. This is a powerful form of strong SEG. II Enforcing a Fermi surface A Fermi surface is a codimension-1 locus in momentum space where fermionic gapless excitations reside. 2 2 2 Our definition of Fermi surface relies on momentum space and, therefore, requires translation symmetry. Accordingly, we assume lattice translation symmetry throughout so that crystal momentum is a good quantum number. We will comment on both {d=1} and {d>1} spatial dimensions throughout the paper, but primarily focus on {d>1} and only use the term Fermi surface in that case. If a symmetry enforces a Fermi surface, then every microscopic model with that symmetry must have a Fermi surface. Importantly, this symmetry is only a sufficient condition for a Fermi surface, and it does not mean that every model with a Fermi surface must have this UV symmetry. Given such a symmetry and a symmetric model, the only way to destroy the model’s symmetry-enforced Fermi surface would be to explicitly break its UV symmetry. We investigate the possibility of symmetry-enforced Fermi surfaces within a familiar class of UV models, described by quantum Hamiltonian lattice models. We assume the d -dimensional spatial lattice \Lambda has one lattice site per unit cell and denote by \mathbf{a}_{i} , with {i=1,2,\cdots,d} , a choice of its primitive lattice vectors. For instance, in {d=2} , \Lambda can be a square lattice but not a honeycomb lattice. The lattice vectors {\mathbf{r}=\sum_{i=1}^{d}n_{i}\mathbf{a}_{i}} satisfy the periodic boundary conditions {\mathbf{r}\sim\mathbf{r}+L_{i}\mathbf{a}_{i}} for each i , where L_{i} is a positive integer. We assume the total number of lattice sites {|\Lambda|=\prod_{i=1}^{d}L_{i}} is even. The Hilbert space is defined by having a single complex fermion reside at each site {\mathbf{r}} . The local Hilbert space on each site is a two-level system, acted on by the complex fermion operator c_{\mathbf{r}} . These fermion operators satisfy the standard anticommutation relations {\{c_{\mathbf{r}},c_{\mathbf{r}^{\prime}}^{\dagger}\}=\delta_{\mathbf{r},\mathbf{r}^{\prime}}} and {\{c_{\mathbf{r}},c_{\mathbf{r}^{\prime}}\}=0} . The Hamiltonian H is assumed to be a local Hamiltonian and admits the decomposition {H=\sum_{\mathbf{r}\in\Lambda}H_{\mathbf{r}}} , where H_{\mathbf{r}} is a bosonic, hermitian operator acting nontrivially only within a finite range about \mathbf{r} . Lastly, we consider UV symmetries described by a collection of unitary operators {\{U_{g}\mid g\in G\}} acting as representations of a group G for the total Hilbert space and commuting with H : {[H,U_{g}]=0} . A symmetry that enforces a Fermi surface rules out a wide class of terms in the Hamiltonian H . For example, the symmetry must forbid the chemical potential term {H_{\mu}=-\mu\sum_{\mathbf{r}}c^{\dagger}_{\mathbf{r}}c_{\mathbf{r}}} . That is, there must be at least one {g\in G} for which {[H_{\mu},U_{g}]\neq 0} . If this were not the case, then H_{\mu} by itself would be an allowed Hamiltonian, but H_{\mu} is a gapped Hamiltonian and does not have