Entanglement negativity in free fermions: twisted characteristic polynomial, universal bounds, and area laws

Ryota Matsuda, Zongping Gong · 2025

Read Entanglement negativity in free fermions: twisted characteristic polynomial, universal bounds, and area laws by Ryota Matsuda, Zongping Gong (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-2507-20947.txt.

Introduction

Ryota Matsuda and Zongping Gong give a plain sums line for a knot score in free charge crowds and draw wide caps from it. The score, entanglement negativity, counts how tied a mixed set stands. Their 2025 line runs through a bent shape sum of the set's own marks. Caps on the score and on its shift pace in play fall out of the line.

One cap ties the score straight to how fast ties fade with span, the root of the area law for tied size. Best terms for that law in long-tie mixed crowds are pinned. A like cap binds tie growth in open sets. Charge-set tie craft gains a set of hard walls.

Catalog record

CREATORRyota Matsuda, Zongping Gong
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE55 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2507-20947

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Entanglement negativity in free fermions: twisted characteristic polynomial, universal bounds, and area laws Ryota Matsuda, Zongping Gong Abstract We present a general and simple formula for computing the entanglement negativity in free fermions. Our formula allows for deriving several universal bounds on negativity and its rate of change in dynamics. The bound on negativity directly relates the clustering property of correlations in free-fermion states to the entanglement area law, and provides the optimal condition for the area law in mixed free fermion states with long-range correlations. In addition, we establish an area-law bound on entanglement generation in open systems, analogous to previously known results for entanglement entropy in unitary dynamics. Our work provides new analytical insights into fermionic mixed-state entanglement. Introduction.— Understanding quantum entanglement in mixed states is a central challenge in modern physics and quantum information science. This is essential for deciphering the quantum features in realistic systems at finite temperatures or coupled to an environment Horodecki et al. 2009 ; Horodecki et al. 1998 ; Wootters 1998 ; Peres 1996 . Entanglement negativity has emerged as an indispensable tool for this challenge, providing a computable measure that successfully captures quantum correlations beyond the scope of entanglement entropy for bipartite pure states Vidal and Werner 2002 . Its application to both (multipartite) closed and open quantum many-body systems has yielded fruitful insights across various contexts, including critical phenomena Calabrese et al. 2012 ; Calabrese et al. 2015 ; Sherman et al. 2016 , nonequilibrium dynamics Wichterich et al. 2009 ; Eisler and Zimborás 2014 , and topological phases Lee and Vidal 2013 ; Castelnovo 2013 . In particular, its computability has served as a powerful analytical tool for deriving rigorous results about mixed-state entanglement in bosonic and spin systems Eisert and Prosen 2010 ; Cramer et al. 2006 ; Cramer and Eisert 2006 ; Eisler and Zimborás 2014 ; Sherman et al. 2016 ; Audenaert et al. 2002 . However, extending the notion of negativity to fermionic systems turns out to be highly nontrivial, as the entanglement structure differs fundamentally from that of bosons due to the anticommuting nature of fermionic operators and the intrinsic superselection rules Wick et al. 1952 ; Bañuls et al. 2007 ; Spee et al. 2018 ; Verstraete and Cirac 2003 ; Schuch et al. 2004a ; Schuch et al. 2004b . These properties necessitate a different definition of negativity to correctly capture fermionic entanglement. A breakthrough was made in Ref. Shapourian et al. 2017 , which reinterprets the partial transpose as a partial time-reversal transformation. This allows for a quantum-information-theoretically consistent extension to fermionic systems Shapourian et al. 2017 ; Shiozaki et al. 2018 . The transposed density matrix can be restored to be Hermitian by the fermion parity operator Shapourian et al. 2019 . Nevertheless, both analytical and practical computations of negativity turn out to be extremely involved even for free fermions. It thus remains elusive to rigorously analyze universal properties of fermionic mixed-state entanglement. In this Letter, we demonstrate that the negativity in free fermions is simply determined by the zeros of a “twisted” characteristic polynomial of the covariance matrix. As shown in Eq. ( 6 ), the twist is realized by replacing the variable in one diagonal block with its minus inverse. Utilizing our formula and the monotonicity of negativity under local operations, we establish upper and lower bounds on both the negativity and its rate of change in dissipative dynamics 1 1 1 Prior work exists Eisert et al. 2018 deriving rigorous upper and lower bounds on negativity using Gaussian channels, but these bounds concern the naive bosonic negativity, which cannot fully capture fermionic entanglement such as that of Majorana bonds Shapourian et al. 2017 . . By combining these results with locality, we discuss entanglement area laws in static and dynamical settings. Regarding the static case, we unveil how entanglement is related to the clustering property of the system. Notably, application of this argument to the Gibbs states answers an open problem in Ref. Kim et al. 2025 concerning long-range systems. In addition, the dynamical area law represents the first attempt to extend previous rigorous results about unitary dynamics Van Acoleyen et al. 2013 ; Bravyi et al. 2006 ; Bravyi 2007 ; Gong et al. 2017 to nonunitary settings. Explicit formula of the negativity for free fermions.— We consider a general fermionic system with N modes and define the fermionic creation and annihilation operators \hat{f}_{j}^{\dagger},\hat{f}_{j} for each mode. From these operators we can construct 2N Majorana operators \hat{c}_{j} ( j=1,2,...,2N ) by \hat{c}_{2j-1}=\hat{f}_{j}+\hat{f}_{j}^{\dagger}\;,\;\hat{c}_{2j}=-i(\hat{f}_{j}-\hat{f}_{j}^{\dagger}) satisfying \{\hat{c}_{j},\hat{c}_{j^{\prime}}\}=2\delta_{jj^{\prime}} . Any free-fermion state \hat{\rho} is given by Bravyi 2004 \displaystyle\hat{\rho}=Z^{-1}e^{\frac{1}{4}\sum_{j,j^{\prime}}\Omega_{j,j^{\prime}}\hat{c}_{j}\hat{c}_{j^{\prime}}}, (1) where \Omega is a 2N\times 2N purely imaginary antisymmetric matrix and Z is a normalization constant such that \Tr\hat{\rho}=1 . This is a Gaussian state fully characterized by its covariance matrix \Gamma_{jj^{\prime}}=\frac{1}{2}\Tr[[\hat{c}_{j},\hat{c}_{j^{\prime}}]\hat{\rho}]. (2) Since \Gamma is related to \Omega via \Gamma=\tanh\frac{\Omega}{2} , its spectrum consists of pairs \pm\nu_{j}\in[-1,1] for j=1,2,\ldots,N . To study bipartite entanglement, we partition the covariance matrix into blocks corresponding to subsystems A with N_{A} modes and B with N_{B}=N-N_{A} modes: \displaystyle\Gamma=\begin{pmatrix}\Gamma_{A}&\Gamma_{AB}\\ \Gamma_{BA}&\Gamma_{B}\end{pmatrix}. (3) The logarithmic entanglement negativity with respect to such a bipartition is given by Vidal and Werner 2002 \mathcal{E}=\ln\|\hat{\rho}^{{\rm T}_{A}}\|_{1}, (4) where {\rm T}_{A} denotes the partial time-reversal transformation of subsystem A and \|O\|_{1}={\rm tr}\sqrt{O^{\dagger}O} denotes the trace norm (i.e., the sum of singular values of O ). Since \hat{\rho}^{{\rm T}_{A}} is no longer Hermitian, we add a parity twist to make it Hermitian without changing the trace norm Shapourian et al. 2019 : \displaystyle\hat{\rho}^{\tilde{{\rm T}}_{A}}=\hat{\rho}^{{\rm T}_{A}}\hat{P}_{A}, (5) where \hat{P}_{A}=\prod^{N_{A}}_{j=1}(i\hat{c}_{2j-1}\hat{c}_{2j}) is the subsystem fermion parity operator Fidkowski and Kitaev 2011 . The negativity ( 4 ) is then computable from the (real) spectrum of \hat{\rho}^{\tilde{\rm T}_{A}} . Concerning the effect of this twisted partial transpose ( 5 ) on the level of covariance matrix \Gamma , we define the twisted characteristic polynomial as \displaystyle P(\lambda) \displaystyle=\lambda^{2N_{A}}\det\begin{pmatrix}\Gamma_{A}+\lambda^{-1}\mathbb{I}_{A}&\Gamma_{AB}\\ \Gamma_{BA}&\Gamma_{B}-\lambda\mathbb{I}_{B}\end{pmatrix}. (6) Note that P(\lambda) is different from the ordinary characteristic polynomial by the transformation \lambda\to-\lambda^{-1} in the subsystem A . From P(\lambda) , we define P_{>}(\lambda) as its factor with zeros \lambda_{j} having absolute value larger than 1: \displaystyle P_{>}(\lambda)=\frac{P(\lambda)}{\prod_{j:|\lambda_{j}|\leq 1}(\lambda-\lambda_{j})}. (7) We find the logarithmic negativity ( 4 ) simply reads \displaystyle\mathcal{E}=\frac{1}{2}\ln|P_{>}(0)|. (8) We note that our formula is consistent with the results in Ref. Shapourian et al. 2019 provided the invertibility of \Gamma_{A} Sup . Nonetheless, our formula is applicable regardless of the invertibility, and this is necessary to derive the following universal bounds. Bounds on the negativity.— We show that our formula ( 8 ) is useful for deriving upper and lower bounds on negativity when combined with the monotonicity of negativity under local operations Shapourian and Ryu 2019a . A key observation is that if \Gamma_{A} or \Gamma_{B} vanishes, Eq. ( 8 ) reduces to a remarkably simple formula \displaystyle\mathcal{E}=\frac{1}{2}\tr[\ln(\matidentity_B+\Gamma_{BA}\Gamma_{AB})]. (9) This means that if one applies local operations to transform some \Gamma with vanishing \Gamma_{A} or \Gamma_{B} into / from the target state, negativity can be bounded using the off-diagonal blocks of \Gamma , which is responsible for the inter-subsystem correlation. Concretely, a general Gaussian operation on a fermionic Gaussian state with covariance matrix \Gamma_{\rm in} is given by Bravyi 2004 \displaystyle\Gamma_{\text{out}}=K(\Gamma_{\text{in}}^{-1}+D)^{-1}K^{\dagger}+C,

Read the complete text