Quantum States and Spectra of Small Cylindrical and Toroidal Lattices

AI summary

Overview

Research area: Quantum physics of tight-binding lattices in synthetic gauge fields, specifically the spectral and eigenstate structure of small two-dimensional lattices wrapped into cylinders and tori (the finite-size, boundary-sensitive counterpart of Hofstadter/Harper physics).

Technical level: Advanced. The paper assumes familiarity with tight-binding Hamiltonians, Landau and symmetric gauges, the Hofstadter butterfly, the Harper equation, and quantum Hall edge states.

Scope (one sentence): A comparative, ground-up numerical study of how gauge choice, boundary conditions, coupling anisotropy, and an added axial flux shape the spectra and eigenstates of small cylindrical and toroidal lattices of a few sites.

What This Paper Is About

In real electronic systems the magnetic field is fundamental and the vector potential is a free choice, so different gauges are interchangeable. In lattices with synthetic gauge fields (ultracold atoms, where a position-dependent phase is engineered directly on the inter-site coupling) the potential is the fundamental object, and wrapping the lattice into a cylinder or torus with periodic boundary conditions can break the gauge equivalence normally assumed for infinite or box-boundary lattices. The authors ask what happens to the spectrum and eigenstates when that equivalence is lost, and they deliberately shrink the lattice to a few sites so that boundary and topology effects are not washed out.

Key Contributions

1. Identifies when gauge equivalence is restored on a cylinder. For rational α = p/q, the Landau x and Landau y gauges give identical spectra when the number of sites around the periodic direction is commensurate with the magnetic period, M × α = k for natural number k. This is generalized to arbitrary uniform-field gauges of the form A = (∓(r/s)By, ±((s−r)/s)Bx, 0), where the condition becomes M × α × f(1) = k, valid for any coupling ratio J_y/J_x.

2. Maps how the Hofstadter (butterfly) shape emerges. By interpolating the gauge factor as f(n) = −n^r with r ∈ [0, 1], the paper shows the butterfly is absent when the phase on one hopping direction does not depend on the transverse lattice index, and grows a "skeletal" butterfly only as that dependence is turned on, with an exact closed-form spectrum for the 2 × 2 cylinder.

3. Classifies spectral intersections and their eigenstate consequences. Intersections in the decoupled-ring limit fall into three types — same state on different rings (becomes an avoided crossing), different states on the same ring (slides with α), and different states on different rings (persistent degeneracy at α M (N+1) = k) — and each type has a distinct effect on how eigenstates blend or stay localized.

4. Documents edge-state gaps unique to small cylinders and extends the analysis to tori. When the number of rings N is commensurate with α (N α = k), avoided crossings open between the edge-state spectral lines that span the bulk gap; these shrink with system size and are absent for incommensurate N. The toroidal case reproduces most cylindrical features but adds a localization-to-delocalization transition set by the relation between the field parameter and the site number.

Main Findings

• Gauge equivalence survives only at commensurate flux. For M = 4 cylinders with N = 3, 4, and 5 rings, the two Landau gauges coincide at α = 1/4, 2/4, 3/4, 4/4, and the coincidence persists when an axial phase φ is added via J_x → J_x e^{iφ} (the Laughlin-type flux insertion).

• The symmetric gauge never joins them. Even at those commensurate α values, the symmetric gauge (f(n) = ∓½n, g(m) = ±½m) produces a different spectrum.

• Transverse phase dependence is required for the butterfly. With a constant phase (f(n) = −1) the α dependence in a cylinder merely spreads the degenerate lines apart uniformly and no butterfly appears; the butterfly requires the phase on J_x to depend on the y index and vice versa. For the 2 × 2 cylinder, E = ±√(J_x² + J_y² ± 2J_xJ_y cos[(−1 + 2^r)απ]), giving a period 2/(−1 + 2^r) that runs from no period at r = 0 to period 2 at r = 1.

• Three intersection classes, three behaviors. In the Landau x gauge, starting from decoupled rings: same state m on different rings opens a gap and becomes an avoided crossing as J_y increases; different states on the same ring slide past each other to different α; different states on different rings remain degenerate at exactly the same α for any J_y/J_x, at α M (N+1) = k.

• Ring-to-ring phase difference resists mixing. With f(n) = −1, even a tiny inter-ring coupling instantly blends all states at any α; with f(n) = −n, blending is gradual and concentrated near avoided crossings (one shown at α = 0.0715). Larger J_y is needed to gap intersections between rings that are farther apart.

• The Landau y gauge has no true avoided crossings. Because α dependence vanishes in both limits J_y = 0 and J_x = 0, the gaps that look like avoided crossings are "faux": at J_y = 0.001 the states are already fully blended across all α, unlike genuine avoided crossings where blending fades away from closest approach. A phase that varies between strips (g(m) = m versus g(m) = 1) only produces visible modulation once J_y reaches 1.

• Coupling-dependent degeneracies appear in the y gauge. At α = 0.345 a pair of spectral lines becomes degenerate at J_y = 2.316 with J_x = 1 fixed, splitting above and below that ratio; the states above and below the gap switch identity across the degeneracy point. This also occurs with box boundary conditions, so it is not caused by periodicity.

• Small cylinders host gaps in the edge spectrum. With M = 4, α = 1/4, J_x = J_y = 1, and M α = 1, edge states are the N-th and (N+1)-th states and avoided crossings appear for N = 4 but not N = 5, and for N = 44 but not N = 45. The crossing width shrinks progressively with ring number and is visible only on zooming in; it can be widened by increasing J_y/J_x, but that pushes the system toward weakly coupled strips where edge-state language loses meaning. Lower α requires proportionally more rings, so larger magnetic fields favor observation.

• Edge-line counting generalizes. For M α > 1 and commensurate, more spectral lines appear in the gap; avoided crossings then occur at lines numbered N M α and N M α + 1, while the crossing width is unaffected by the number of azimuthal sites.

• Toroidal lattices add a localization transition. Per the abstract, tori reproduce most cylindrical features but exhibit a transition from localization to delocalization governed by the relation between the field parameter and the number of lattice sites.

• Finite size blocks the usual topological diagnostic. The paper states that for the small finite periodic lattices studied, Chern numbers cannot be computed accurately and are not necessarily quantized, so sharp topological transitions characteristic of the thermodynamic limit may become gradual.

Methodology in Plain English

The authors write down a standard tight-binding Hamiltonian on an M × N grid where hopping in x carries a phase 2πα f(n) and hopping in y carries a phase 2πα g(m), with α proportional to the flux per plaquette. Different choices of f and g are different gauges of the same field. They then close the lattice into a cylinder by identifying one direction periodically (or into a torus by closing both) and diagonalize the resulting finite matrix numerically.

The strategy is incremental: start with decoupled rings (J_y = 0), where each spectral line can be labeled unambiguously by a ring index and a state index, then switch on the inter-ring coupling and watch how each intersection of lines evolves. Eigenstates are visualized by mapping the 2D lattice onto a 1D sequence using M × (n − 1) + m, so that localization on particular rings can be read off directly. An extra flux is threaded through the cylinder as a phase φ on the hopping around the ring, which is the lattice analogue of the Laughlin flux-insertion argument, and the spectra and states are tracked as a function of φ to expose edge states traversing the bulk gap. No experimental data are reported; the work is analytical plus numerical.

Why This Matters

The paper fills a gap the authors identify explicitly: most work on Hofstadter physics assumes infinite lattices or ignores boundary conditions, while the small number of studies on cylindrical and toroidal real-space lattices are focused on quantum Hall physics. This work maps the basic spectral and eigenstate structure of such lattices, which is the prerequisite for interpreting dynamics in synthetic-gauge-field experiments.

Applications and connections (drawn from the physics discussed in the paper, not from reported device results):

• Ultracold-atom quantum simulators using synthetic dimensions, where periodic boundary conditions are implemented by cyclic coupling of typically three or four internal states — a regime that maps naturally onto the few-site lattices studied here.
• Ring-trap and azimuthal-lattice experiments, where atoms are confined in ring geometries topologically equivalent to a cylinder and azimuthal lattice structure has been demonstrated.
• Topological-insulator characterization in small devices, since the appearance of gaps in edge-state lines signals a possible change in topological character where Chern numbers are unreliable.
• Quantum Hall edge-state physics, where the small-system edge gaps the authors find are a finite-size counterpart of the gapless edge channels of macroscopic samples.

Industry relevance: the results are directly relevant to groups building neutral-atom quantum simulators and atomtronic sensors based on ring geometries, and they provide benchmark spectra that hardware-facing simulation efforts can use to validate few-site lattice models before scaling up.

Future Directions

• Extend the edge-gap analysis to M α > 1, where multiple spectral lines occupy the band gap and crossings become more frequent and can coincide at the same φ, complicating the identification of edge states.
• Find a finite-size topological diagnostic to replace Chern numbers, given the authors' statement that the latter are not accurately computable or necessarily quantized in these small periodic lattices.
• Push the toroidal analysis further, in particular the localization-to-delocalization transition and how it depends quantitatively on the field parameter relative to site number (the supplied text is truncated before Section VII, so those details are not available here).
• Connect to experiment: identify which of the predicted features — gauge-dependent avoided crossings, persistent degeneracies at α M (N+1) = k, edge-state gaps at N α = k — are observable with existing imaging and stroboscopic correlation techniques, and how disorder or interactions would smear them.

Target Audience

Advanced graduate students and researchers in ultracold-atom quantum simulation, synthetic gauge fields, and topological band theory, particularly those working with few-site or ring-geometry lattices where boundary conditions dominate. It is also useful for theorists who model finite Hofstadter systems and need a reference for how gauge choice changes the answer, and less suitable for readers without prior exposure to tight-binding lattice models or gauge arguments.

Authors’ abstract

We examine the spectrum and quantum states of small lattices with cylindrical and toroidal topology subject to a scalar gauge potential that introduces a position dependent phase in the inter-site coupling. Equivalency of gauges assumed in infinite lattices is generally lost due to the periodic boundary conditions, and conditions that restore it are identified. We trace the impact of various system parameters including gauge choice, boundary conditions and inter-site coupling strengths, and an additional axial field. We find gauge dependent appearance of avoided crossings and persistent degeneracies, and we show their impact on the associated eigenstates. Smaller lattices develop prominent gaps in spectral lines associated with edge states, which are suppressed in the thermodynamic limit. Toroidal lattices have counterparts of most of the features observed in cylindrical lattices, but notably they display a transition from localization to delocalization determined by the relation between the field parameter and the number of lattice sites.