Rényi entropy of single–character CFTs on the torus
and Rahul Poddar · 2024
Read Rényi entropy of single–character CFTs on the torus by and Rahul Poddar (2024) 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-2412-00192.txt.
Introduction
Luis Alberto León Andonayre and Rahul Poddar compute an entanglement measure for a special class of quantum field theories, in a 2024 paper. The theories, called single-character or meromorphic, are the simplest conformal field theories and include a model based on the E8 symmetry group. Their method builds differential equations for the theory's building blocks on a torus, following a classic Wrońskian approach. From those blocks they extract the second Rényi entropy of a single interval.
The answer repeats with period two along both cycles of the torus. It diverges logarithmically in the expected universal way when the torus grows large, and again when the interval nears the full cycle size. A side construction shows how folding such a theory produces multi-character theories, in one case realizing the toric code. The authors present the method as a nonperturbative handle on entanglement in these tidy theories.
Catalog record
| CREATOR | and Rahul Poddar |
|---|---|
| YEAR | 2024 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 87 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2412-00192
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Rényi entropy of single–character CFTs on the torus and Rahul Poddar Abstract We introduce a non–perturbative approach to calculate the Rényi entropy of a single interval on the torus for single–character (meromorphic) conformal field theories. Our prescription uses the Wrońskian method of Mathur, Mukhi and Sen [ 1 ] , in which we construct differential equations for torus conformal blocks of the twist two–point function. As an illustrative example, we provide a detailed calculation of the second Rényi entropy for the E 8 , 1 \rm E_{8,1} WZW model. We find that the ℤ 2 \mathbb{Z}_{2} cyclic orbifold of a meromorphic CFT results in a four–character CFT which realizes the toric code modular tensor category. The ℤ 2 \mathbb{Z}_{2} cyclic orbifold of the E 8 , 1 \rm E_{8,1} WZW model, however, yields a three–character CFT since two of the characters coincide. On computing the twist two–point function, the second Rényi entropy for a single interval of the E 8 , 1 \rm E_{8,1} WZW model has the universal logarithmic divergent behaviour in the decompactification limit of the torus, as expected. Furthermore, we see that the q q –expansion is UV finite, apart from the leading universal logarithmic divergence. 1 Introduction Quantum entanglement is a fundamental property of quantum systems [ 2 ] , with significant relevance across various areas of physics, including condensed matter physics [ 3 , 4 , 5 , 6 , 7 ] , quantum information theory [ 8 ] , black holes physics [ 9 , 10 , 11 , 12 ] , quantum field theory (QFT) [ 13 , 14 , 15 , 16 ] , and holography [ 17 ] . Over the past 20 years, entanglement has become a crucial area of study, particularly through quantum informational measures such as entanglement entropy [ 18 , 13 , 17 , 15 ] . Since the discovery that the entropy of a black hole is proportional to their area [ 19 ] , understanding how entropy and entanglement arise in systems has been crucial. This shows that black hole microstates are associated with the event horizon and should help us to gain some insight into the Hilbert space of quantum gravity [ 20 ] . However, it is still a challenge to compute measures of entropy and entanglement in interacting systems and systems on higher–genus Riemann surfaces. For QFT, some measures of quantum entanglement are not well–defined, as there is an infinite amount of quantum entanglement between any two given subregions. This can be quantified by noting that the algebra of local operators for a quantum field theory without a UV cut–off has a von Neumann factor of type III 1 [ 21 , 22 , 23 ] . In other words, without imposing a UV cut–off, objects like reduced density matrices and traces over subregions cannot be defined in a quantum field theory, and the Hilbert space of a QFT can not be factorized into separate Hilbert spaces for each spatial subregion. For two–dimensional conformal field theories (CFTs) on the plane, it is well known that the entanglement entropy for single intervals satisfies a universal UV divergent behaviour [ 24 ] . Computing measures like entanglement entropy becomes much harder on higher genus surfaces and with a larger number of intervals for interacting systems. Entanglement entropy, or von Neumann entropy for a subregion A is defined by, S_{\rm vN}(A)=-\Tr(\rho_A \ln|\rho_A|), (1.1) where \rho_{A} is the reduced density matrix for the subregion A . In practice, this quantity is difficult to compute. A computationally useful measure of quantum information is the Rényi entropy [ 25 ] . To compute this, we employ the replica trick, which involves replicating the field theory and sewing together the surfaces by identifying the ends of the interval of interest. This can be achieved by introducing a branch cut between the ends of the intervals, where the replicated theory is defined on the corresponding Riemann surface. The N^{\rm th} Rényi entropy, S_{N}(A)=\frac{1}{1-N}\ln\left|\Tr\rho_{A}^{N}\right|=\frac{1}{1-N}\ln\left|\frac{Z(A,N)}{Z(0,N)}\right|, (1.2) is in terms of the reduced density matrix \rho_{A} for the subregion A , when possible to define (i.e. with a UV cut–off or for operator algebras of type I or II), or in terms of the partition function on the Riemann surface of the replicated theory Z(A,N) . The normalization Z(0,N) is the replicated surface without sewing together the replicated copies. On the plane, Z(0,N)=Z^{N} , where Z is the partition function of the unreplicated theory. The entanglement entropy, or von Neumann entropy, can be recovered with an analytical continuation in the limit, \lim_{N\to 1}S_{N}(A)=S_{\rm vN}(A). (1.3) To work with the replicated theory, one must compute the \mathbb{Z}_{N} cyclic orbifold of the original theory. For a CFT, this introduces new primary fields, corresponding to the new twisted sectors, or heighest weight representations of the new \mathbb{Z}_{N} symmetry. These are known as twist operators \sigma [ 26 , 27 ] . Inserting a twist operator introduces a monodromy e^{2\pi i/N} at the insertion point. Therefore, when inserting a twist–anti-twist pair, this introduces a branch cut between the insertion points. One can then work out correlation functions of these twist operators, to compute the partition function of the replicated CFT on the Riemann surface created by introducing the branch cut [ 13 ] , Z(A,N)=\expectationvalue{\sigma(z_1,\bar z_1) \overline\sigma(z_2, \bar z_2)}. (1.4) Here, the subregion A has end points (z_{1},\bar{z}_{1}) and (z_{2},\bar{z}_{2}) on the unreplicated Riemann surface. On the plane, we can use conformal invariance to compute this two–point function trivially, which allows one to make a universal statement about the Rényi and von Neumann entropies for conformal field theories on the plane. This universal behaviour is no longer true for theories on higher–genus surfaces. Rényi entropies on the torus have only been calculated exactly for free bosons and free fermions, and have been perturbatively calculated for other interacting systems. Holographic techniques have also been used to calculate the entropy of CFTs at large central charge [ 18 , 28 , 29 , 30 , 31 ] . Torus correlators in the holographic or semi–classical limit have also been computed [ 32 , 33 , 34 , 35 , 36 , 37 , 38 , 39 ] . To compute the entanglement entropy of disjoint intervals on a torus for free bosonic and fermionic theories, strategies involving explicit computations of propagators and Green’s functions have been employed. For a CFT of free bosons, it is possible to explicitly compute Green’s functions by means of cut–differentials and Ward identities on the torus. The problem is broken down into computations of the classical and quantum part, respectively, and has been tackled accordingly [ 26 , 30 , 40 ] . Corrections to the entanglement entropy on the torus for d+1 dimensional generalized quantum Lifshitz models have been calculated in [ 41 ] . Similarly, the task of computing the entanglement entropy of free fermions on the torus has been done by considering the different boundary conditions of the twisted sectors and computing propagators in these twisted sectors. Alternative approaches have employed resolvent analysis, enabling the determination of modular 1 1 1 The term “modular” here refers to the modular operators of Tomita–Takesaki theory [ 42 ] . data associated with a