Entanglement Negativity in Quantum Systems

• Entanglement Negativity is a quantitative measure that evaluates entanglement in mixed bipartite quantum systems using the partial transpose method.
• It provides an operational framework via semidefinite programming to certify minimal entangled dimensions and certify quantum correlations.
• Scaling laws demonstrate universal features, with area laws in bosonic systems and logarithmic corrections in fermionic and conformal field theory settings.

Entanglement negativity is a quantitative measure of quantum correlations—particularly, quantum entanglement—in bipartite quantum systems, designed to capture and quantify entanglement in general mixed states beyond the scope of pure-state measures such as the von Neumann or Rényi entropies. It serves not only as a practical entanglement monotone with an operational interpretation in a diversity of quantum information and many-body contexts, but also as a technical device for studying entanglement scaling laws, multipartite structure, and universal features in both quantum field theoretic and condensed matter systems.

1. Formal Definition and Fundamental Properties

Given a density matrix $\rho_{AB}$ over a finite-dimensional bipartite Hilbert space $\mathcal{H}_A \otimes \mathcal{H}_B$, the partial transpose with respect to subsystem $B$ (denoted $T_B$) is

$$\langle a, b|\,\rho_{AB}^{T_B}\,|a', b'\rangle = \langle a, b'|\,\rho_{AB}\,|a', b\rangle$$

in any orthonormal basis. The trace norm is $\|O\|_1 = \mathrm{Tr}\sqrt{O^\dagger O}$. The (Vidal–Werner) negativity is

$$N(\rho_{AB}) = \frac{1}{2}\left(\|\rho_{AB}^{T_B}\|_1 - 1\right) = \sum_{i:\lambda_i<0} |\lambda_i|$$

where $\{\lambda_i\}$ are the eigenvalues of $\rho_{AB}^{T_B}$. The logarithmic negativity is

$$\mathcal{E}(\rho_{AB}) = \log \|\rho_{AB}^{T_B}\|_1 = \log(2 N + 1)$$

$\mathcal{H}_A \otimes \mathcal{H}_B$0 if and only if $\mathcal{H}_A \otimes \mathcal{H}_B$1 is positive (the PPT criterion), that is, for separable states. Negativity is an entanglement monotone under LOCC and provides an upper bound on distillable entanglement. For pure states, the logarithmic negativity coincides with the Rényi-$\mathcal{H}_A \otimes \mathcal{H}_B$2 entropy of the reduced density matrix (Eltschka et al., 2013, Eisler et al., 2015, Ruggiero et al., 2016).

2. Operational Meaning and Dimensional Quantification

Negativity has an immediate operational interpretation: for pure states of Schmidt rank $\mathcal{H}_A \otimes \mathcal{H}_B$3,

$\mathcal{H}_A \otimes \mathcal{H}_B$4

$\mathcal{H}_A \otimes \mathcal{H}_B$5

where $\mathcal{H}_A \otimes \mathcal{H}_B$6 [as in (Eltschka et al., 2013)]. For arbitrary mixed states, $\mathcal{H}_A \otimes \mathcal{H}_B$7, so $\mathcal{H}_A \otimes \mathcal{H}_B$8 is a lower bound on the number of entangled dimensions.

Device-independent lower bounds are accessible via the moment matrix constructed from measurement outcomes; by semidefinite programming one obtains a certified minimum value of the negativity, which in turn certifies a minimum underlying Hilbert space dimension even when the system implementation or measurement operators are not fully specified (Eltschka et al., 2013).

3. Scaling Laws and Spectrum in Quantum Lattice Systems

In free lattice models, the scaling of entanglement negativity reflects universal features:

• In 2D bosonic (harmonic) lattices, the logarithmic negativity obeys a strict area law: $\mathcal{H}_A \otimes \mathcal{H}_B$9, with $B$0 non-critical. There are no logarithmic violations (Eisler et al., 2015).
• In 2D free fermionic systems, the negativity exhibits a multiplicative log correction: $B$1, where $B$2 depends on geometric integrals over the Fermi surface and subsystem boundary.
• In 1D CFT, the negativity between two adjacent intervals scales logarithmically with subsystem length $B$3, with universal central-charge dependence (Ruggiero et al., 2016, Chung et al., 2013): $B$4 The negativity spectrum, i.e., the eigenvalue distribution of $B$5, exhibits universal scaling functions depending on the central charge and block geometry, with bulk sign symmetry and Bessel-function scaling in the tails (Ruggiero et al., 2016).

For random-singlet phases in strongly disordered spin chains, the disorder-averaged negativity is proportional to the number of singlets shared by the intervals, leading to logarithmic scaling but with a reduced effective central charge $B$6 (Ruggiero et al., 2016).

4. Monogamy, Disentangling, and Multipartite Structure

Negativity is strictly monogamous: $B$7 for any pure $B$8 state. Equality is achieved only in trivial cases forced by LOCC monotonicity. The tight boundary for the set of achievable $B$9 tuples in the three-qubit case is given by a non-linear polynomial equation; in high-dimensional systems, attainable distributions of negativity asymptotically become linear in local dimension (He et al., 2014, Allen et al., 2015).

The disentangling theorem gives necessary and sufficient conditions for when $T_B$0 can be factorized from $T_B$1 given $T_B$2. Physically, this allows negativity to diagnose factorization structures and topological boundaries in many-body states, and the absence or presence of residual entanglement between separated regions (He et al., 2014, Lim et al., 2021).

5. Holographic and Field-Theoretic Perspectives

In holographic AdS/CFT duality, negativity is computed by a specific algebraic sum of minimal surfaces: $T_B$3 leading to

$T_B$4

for pure AdS and

$T_B$5

for AdS-Schwarzschild (finite $T_B$6), isolating distillable quantum correlations via the subtraction of thermal entropy. In 1+1d CFT, the same structure emerges via the replica-twist formalism, matching the universal forms at large central charge (Chaturvedi et al., 2016, Chaturvedi et al., 2016).

In random tensor networks and spin network states for quantum gravity, the typical logarithmic negativity between two boundary regions is an area law, matching the Ryu–Takayanagi prescription, and is robust in the large bond/spin regime (Chirco et al., 2022). Nonlocal intertwiner entanglement in the bulk induces corrections, modifying the area-law coefficient.

6. Dynamics and Measurement-Induced Transitions

Negativity tracks the growth and propagation of quantum correlations in out-of-equilibrium systems. Following a global quantum quench in 1D CFT, negativity exhibits linear spreading in time, plateau, and eventual decay, mirroring entanglement entropy but with distinct lattice-specific effects such as the "late birth" and "sudden death" of entanglement on finite lattices (Coser et al., 2014).

In hybrid quantum circuits undergoing measurement-induced transitions, the scaling of "mutual negativity"—the negativity between two disjoint subregions—diagnoses the emergence of bipartite vs multipartite entanglement, and reveals a robust universal scaling exponent for negativity distinct from that governing mutual information (Sang et al., 2020).

7. Perturbative and Spectral Analyses

The negativity, being a non-analytic function of the density matrix, admits a systematic perturbative expansion with well-defined matrix calculus (i.e., derivatives/Jacobians and Hessians involving the spectral decomposition of $T_B$7). The first and second derivatives furnish the linear and quadratic responses of negativity under state deformations, applicable to the characterization of entanglement growth, response to perturbations, and entanglement phase transitions (Cresswell et al., 2018).

The full negativity spectrum in random ensembles and tensor networks (especially holographic or large-$T_B$8 limits) can be analyzed via replica moments and combinatorial methods (e.g., sums over noncrossing permutations, hypergeometric resummations, and max-flow diagrams), elucidating universal spectral laws (e.g., semicircle, Marchenko–Pastur, Motzkin), Page-like transitions, and the structure of Hawking-radiation negativity in island calculations (Kudler-Flam et al., 2021, McBride et al., 2023).

---

Entanglement negativity, thus, serves as a cornerstone entanglement measure for both fundamental and practical investigations across quantum information, condensed matter, and quantum gravity, tightly linking operational meaning, scaling universality, and the multipartite structure of quantum correlations. For explicit analytic results, lattice and field-theoretic scaling forms, device-independent certification protocols, holographic prescriptions, and algorithmic recipes for perturbative computation, see the cited works (Eltschka et al., 2013, Eisler et al., 2015, Ruggiero et al., 2016, He et al., 2014, Allen et al., 2015, Chaturvedi et al., 2016, Sang et al., 2020, Kudler-Flam et al., 2021, Ruggiero et al., 2016, Chirco et al., 2022, Chaturvedi et al., 2016, Coser et al., 2014).