Fractional topology in open systems
Xi Wu, Xiang Zhang, Fuxiang Li · 2026
Read Fractional topology in open systems by Xi Wu, Xiang Zhang, Fuxiang Li (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-2603-03854.txt.
Introduction
Xi Wu and others study fractional topological invariants in an open quantum system: a periodic Su-Schrieffer-Heeger chain subject to gain and loss, described by Lindblad master equations. They first set out the symmetry settings under which integer topological invariants exist, then study how those invariants turn fractional in steady states. The change can come from tuning parameters in the jump operators or arise as a dynamic shift during time evolution.
The authors show that the fractional invariants lose the usual quantized topology, but stretching the Brillouin zone over many cycles restores integer quantization of the total winding. They also show how the effects could be observed in long-range hopping photonic lattices with fractional fillings, using Bloch state tomography. That opens a path to studying fractional topology in open systems.
Catalog record
| CREATOR | Xi Wu, Xiang Zhang, Fuxiang Li |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 30999 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2603.03854
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Fractional topology in open systems Xi Wu, Xiang Zhang, Fuxiang Li Abstract We investigate the emergence of fractional topological invariants in a periodic Su–Schrieffer–Heeger chain subject to gain and loss, governed by the Gorini–Kossakowski–Sudarshan–Lindblad master equations. After preparing the symmetry condition for integer topological invariants, we investigate their transition to fractional ones in steady states, which can happen either by tuning parameters in jump operators or as a dynamical transition during time evolution. Moreover, we show that these fractional topological invariants no longer possess quantized topology in the conventional sense. However, by extending the Brillouin zone to cover multiple cycles, the total winding regains integer quantization. Finally, we show how such effects can be observed in long-range hopping photonic lattices with fractional fillings, via Bloch state tomography. Our results open a new pathway to understand fractional topology in open quantum systems. 1 1 footnotetext: These authors contributed equally to this work. I Introduction In recent decades, people have gradually drawn attention to the study of open systems 7 ; 9 ; 8 ; 33 ; 43 ; 44 ; 21 ; 48 ; 10 ; 42 ; 36 ; 25 ; 51 ; 53 ; 45 ; 49 ; 37 ; 31 ; 30 ; 40 ; 41 ; 35 ; 26 ; 24 ; 17 ; 28 described by Gorini–Kossakowski–Sudarshan–Lindblad master equation(Lindblad master equation in short) 32 ; 18 ; 22 that governs the time evolution of density matrix \frac{d\rho}{dt}=-i[H,\rho]+\sum_{\mu}(2L_{\mu}\rho L_{\mu}^{\dagger}-\{L_{\mu}^{\dagger}L_{\mu},\rho\})\equiv\mathcal{L}\rho, (1) where H=\sum_{ij}c_{i}^{\dagger}h_{ij}c_{j} is quadratic in terms of the fermionic creation (annihilation) operators c_{i}^{\dagger} ( c_{i} ). Here, h_{ij} are the elements of matrix H , which denotes the Hamiltonian in the single-particle basis and L_{\mu} ’s are the Lindblad dissipators describing quantum jumps due to coupling to the environment. \mathcal{L} is called the Liouvillian superoperator. Formal solutions were established 22 ; 43 ; 44 , through the third quantization procedure, which makes the study in the cases without interactions straightforward. From Eq. ( 1 ) one can see that non-Hermiticity naturally arises in the Liouvillian superoperator. Rapid growth was seen in the field of topology of non-Hermitian matrices in the past decade 27 ; 52 ; 12 ; 46 ; 55 ; 54 ; 39 ; 56 ; 23 ; 47 ; 6 ; 13 ; 57 . Exceptional points(EP) are considered as a key signal for fractional topological invariant. In open systems described by Lindblad equation, the dynamics is related with the damping matrix X , which can be regarded as a non-Hermitian Hamiltonian. Note that this is different from the post-selected Hamiltonian that was considered as the effective Hamiltonian 11 . A natural intuition is that EP also can be used as a mechanism for fractional topological invariant in Lindblad open systems. However, there are several limitations of this EP approach: First, from a practical point of view, it is very difficult to engineer eigenstates of X in steady states through gain matrix. 15 shows that EPs of X may not reach steady states. Therefore, a fractional winding number does not appear in steady states even if EPs occur in X. Second, to generate this construction to include mixed states is challenging. For generic mixed states, the density matrix should be used to define the topological invariant and it is only expanded by right eigenstates of X , not the left eigenstates. The famous construction of fractional winding number, for non-Hermitian Su–Schrieffer–Heeger(NHSSH) model, through right states is done by Tony Lee, who considers the trajectory of momentum in \left\langle\sigma_{x}\right\rangle,\left\langle\sigma_{z}\right\rangle space 27 , excluding \left\langle\sigma_{y}\right\rangle for no specific reason. The orbit of these expectation values in the Bloch sphere is not symmetric, making the total Berry phase non-topological. Can we cook up a simple setup for fractional winding number in terms of Berry phase in open systems, which is also topological? EPs themselves are not the necessary condition for the fractional topological invariant. What is important is the multi-valueness of the physical states, which is also the essence of EP approach. Moreover, as we shall see the text below, at least in the NHSSH model, the parameter that induces EP in the eigenstates breaks the inversion symmetry, which keeps the Berry phase topological otherwise. The above concerns motivate us to take an alternative approach: we introduce the multi-valueness directly in the gain/loss matrix by a fractional momentum and keep the inversion symmetry in the Hamiltonian, gain/loss matrices and initial states. As a result, the above drawbacks can be overcome, and fractional winding numbers appear in the steady states. Based on our definition, our findings are as follows: Under the symmetries required for quantization, single-valued gain and damping matrices guarantee that a completely filled lower band yields an integer winding number in the steady state; if the initial state is also single-valued, the winding number remains integer throughout the evolution. For fractional fillings, tuning the gain matrix drives the winding number away from integrality. It can vary continuously and undergo abrupt jumps from integer to fractional values. These discontinuities occur when the purity gap closes, mirroring the mechanism of topological transitions in integer cases. In contrast to pure states in closed or non-Hermitian systems—where the winding numbers are rational—the values here depend sensitively on the structure of the states. Yet the topology survives in a re-quantized form: when the states exhibit n branches, the winding number defined over an n-period is strictly integer. We call this multi-period re-quantization , indicating that n Brillouin zones behave collectively as a single topological unit, insensitive to local details. Finally, the proposed model is experimentally accessible. In a 1D optical superlattice of ultracold fermions, the Berry phase of both steady and transient states can be directly probed. II Our model We will consider single particle loss and gain with loss dissipators L_{\mu}^{l}=\sum_{i}D_{\mu i}^{l}c_{i} and gain dissipators L_{\mu}^{g}=\sum_{i}D_{\mu i}^{g}c_{i}^{\dagger} , where D_{\mu i} are the complex coefficients. With the single-particle correlation 47 ; 20 \Delta_{ij}(t)=\mathrm{Tr}[c_{i}^{\dagger}c_{j}\rho] , Eq. ( 1 ) can be recast as \frac{d\Delta(t)}{dt}=X\Delta(t)+\Delta(t)X^{\dagger}+2M_{g} (2) where the damping matrix \displaystyle X=ih^{T}-M_{l}^{T}-M_{g}, (3) with (M_{g})_{ij}=\sum_{\mu}D_{\mu i}^{g*}D_{\mu j}^{g} and (M_{l})_{ij}=\sum_{\mu}D_{\mu i}^{l*}D_{\mu j}^{l} . Note that M_{l} and M_{g} are Hermitian matrices. The steady state correlation \Delta_{s}=\Delta(\infty) , to which long time evolution of any initial state converges, is determined by d\Delta_{s}/dt=0 or X\Delta_{s}+\Delta_{s}X^{\dagger}+2M_{g}=0 . This equation is solved by: \Delta_{s}=2\int_{0}^{\infty}e^{Xt}M_{g}e^{X^{\dagger}t}dt. (4) It can be seen that the steady state is determined by the gain matrix M_{g} and the damping matrix X . We present a simple model of open quantum system: a standard Su-Schrieffer-Heeger(SSH) model with single particle gain and loss with fractional momentum: H=\sum_{ij}h_{ij}c^{\dagger}_{i}c_{j} and L^{g}_{\mu}=L^{g}_{\mu i}c^{\dagger}_{i} and L^{l}_{\mu}=L^{l}_{\mu i}c_{i} , In momentum space, L^{g}_{1}(k)=\sqrt{\gamma_{1}}(c^{\dagger}_{A}+c^{\dagger}_{B}) , L^{g}_{2}(k)=\sqrt{\gamma_{2}}(c^{\dagger}_{A}e^{-ik/n}+c^{\dagger}_{B}) , L^{g}_{3}(k)=\sqrt{\gamma_{0}}c^{\dagger}_{A} , L^{g}_{4}(k)=\sqrt{\gamma_{0}}c^{\dagger}_{B} , L^{l}_{1}(k)=\sqrt{\gamma_{1}}(c_{A}-c_{B}) , L^{l}_{2}(k)=\sqrt{\gamma_{2}}(c_{A}e^{-ik/n}-c_{B}) , L^{l}_{3}(k)=\sqrt{\gamma_{0}}(c_{A}+ic_{B}) . For concreteness we choose n=3 without loss of generality. Then the damping matrix X and the gain matrix M_{g}(k) read \displaystyle X(k) \displaystyle=i \displaystyle\left[(t_{1}+t_{2}\cos k)\sigma_{x}+t_{2}\sin k\sigma_{y}\right]-\Gamma\sigma_{0} \displaystyle M_{l} \displaystyle= \displaystyle(\gamma_{1}+\gamma_{2})\sigma_{0}-(\gamma+\cos\frac{k}{3})\sigma_{x}-\sin\frac{k}{3}\sigma_{y} (6) \displaystyle M_{g} \displaystyle= \displaystyle(\gamma_{1}+\gamma_{2})\sigma_{0}+(\gamma+\cos\frac{k}{3})\sigma_{x}+\sin\frac{k}{3}\sigma_{y}\,. (7) where \Gamma=\gamma_{1}+\gamma_{2} which is also called the Liouvillian gap, governing the relaxation of the open quantum system. For simplicity, we set \gamma_{1}=\gamma_{2}=0.5 as follows. In general, the Liouvillian gap \Gamma is less than 0, ensuring the existence of the steady state. It can also be seen that the