Note on Generalized Symmetries, Gapless Excitations, Generalized Symmetry Protected Topological states, and Anomaly

Chao-Ming Jian,Cenke Xu
DOI: https://doi.org/10.1088/1742-5468/abe411
2021-01-20
Abstract:We consider quantum many body systems with generalized symmetries, such as the higher form symmetries introduced recently, and the "tensor symmetry". We consider a general form of lattice Hamiltonians which allow a certain level of nonlocality. Based on the assumption of dual generalized symmetries, we explicitly construct low energy excited states. We also derive the 't Hooft anomaly for the general Hamiltonians after "gauging" the dual generalized symmetries. A 3d system with dual anomalous 1-form symmetries can be viewed as the boundary of a 4d generalized symmetry protected topological (SPT) state with 1-form symmetries. We also present a prototype example of 4d SPT state with mixed 1-form and 0-form symmetry topological response theory as well as its physical construction. The boundary of this SPT state can be a 3d anomalous QED state, or an anomalous 1-form symmetry enriched topological order. Insights are gained by dimensional compatification/reduction. After dimensional compactification, the 3d system with N pairs of dual 1-form symmetries reduces to a 1d system with 2N pairs of dual U(1) global symmetries, which is the boundary of an ordinary 2d SPT state; while the 3d system with the tensor symmetry reduces to a 1d Lifshitz theory, which is protected by the center of mass conservation of the system.
Strongly Correlated Electrons
What problem does this paper attempt to address?
The problems that this paper attempts to solve mainly focus on the generalized symmetries in quantum many - body systems and their related gapless excitations, generalized symmetry - protected topological states (SPT), and anomalies. Specifically, the paper explores quantum many - body systems with generalized symmetries, such as higher - order form symmetries and "tensor symmetries", and considers lattice Hamiltonians in general forms, allowing a certain degree of non - locality. Based on the hypothesized dual generalized symmetries, the paper constructs low - energy excited states and derives the 't Hooft anomaly after "gauging" the dual generalized symmetries. ### Main research questions: 1. **Existence and properties of generalized symmetries**: - The paper assumes that the system has a \( \text{U}(1)_g \) generalized symmetry on long scales, such as higher - order form symmetries or "tensor symmetries". What are the impacts of the existence of these symmetries on the physical properties of the system? 2. **Construction of low - energy excited states**: - Based on the assumption of dual generalized symmetries, the paper explicitly constructs low - energy excited states and discusses the properties of these excited states. 3. **Derivation of 't Hooft anomaly**: - The paper derives the 't Hooft anomaly after "gauging" the dual generalized symmetries and explores how this anomaly affects the physical behavior of the system. 4. **Relationship between three - dimensional systems and four - dimensional SPT states**: - A three - dimensional system with dual anomalous 1 - form symmetry can be regarded as the boundary of a four - dimensional generalized symmetry - protected topological state (SPT). The paper explores this relationship in detail and gives specific examples. 5. **SPT states with mixed 0 - form and 1 - form symmetries**: - The paper proposes a prototypical example of a four - dimensional SPT state with mixed 0 - form and 1 - form symmetries and discusses its physical construction. The boundary state can be an anomalously 1 - form symmetry - enriched topological order. 6. **Insights into dimension compactification/reduction**: - Through dimension compactification, the paper shows how a three - dimensional system can be simplified into a one - dimensional system and discusses the relationships between these simplified systems and ordinary two - dimensional SPT states. ### Formula summary: - **1 - form charge density**: \[ Q_A=\int_A d\vec{S}\cdot\vec{\rho} \] \[ \nabla\cdot\vec{\rho}=\text{const.} \] - **Maxwell equations**: \[ \partial_\mu J^{(e)}_\mu=\frac{\partial e_i}{\partial t}-\partial_j\epsilon_{ijk}b_k = 0 \] \[ \partial_\mu J^{(m)}_\mu=\frac{\partial b_i}{\partial t}+\partial_j\epsilon_{ijk}e_k = 0 \] - **Currents**: \[ J^{(e)}=(\rho^{(e)}_i,J^{(e)}_{ij})=(e_i,\epsilon_{ijk}b_k) \] \[ J^{(m)}=(\rho^{(m)}_i,J^{(m)}_{ij})=(b_i,-\epsilon_{ijk}e_k) \] - **1 - form currents**: \[ J^{(e)}_{ij}(x)=\epsilon_{ijk}\frac{\partial H}{\partial\hat{b}_k(x)} \] \[ J^{(m)}_{ij}(x)=-\epsilon_{ijk}\frac{\partial H}{\partial\hat{e}_k(x)} \] - **'t Hooft anomaly**: \[ S_{CS}=\int d\tau d^4x\frac{