Entanglement Across Fields

Entanglement is one mathematical structure with several scientific roles. In quantum information it is a resource constrained by local operations. In many-body physics it records correlations across spatial cuts and controls efficient representations. In chemistry it helps diagnose correlation among orbitals and spins. In optics it is prepared and measured through field modes. In foundations it separates quantum inseparability from stronger notions of nonclassicality. In quantum field theory it meets locality, ultraviolet structure, and operator-algebra subtleties.

The formal definition does not change from field to field. What changes is the physically meaningful subsystem decomposition, the accessible operations, the useful diagnostic, and the claims that diagnostic can support.

Chapter Map

Field	Canonical bridge	Central question
quantum information	Entanglement in Quantum Information	what tasks become possible under local-operation constraints?
many-body physics	Entanglement in Many-Body Physics	what does a spatial or mode cut reveal about phases and dynamics?
quantum chemistry	Entanglement in Quantum Chemistry	how do antisymmetry, orbital choice, and electronic correlation differ?
quantum optics	Entanglement in Quantum Optics	how are polarization, path, frequency, and quadrature modes prepared and measured?
foundations	Entanglement in Foundations	how does entanglement relate to steering, Bell nonlocality, and measurement?
quantum field theory	Entanglement in QFT Preview	what survives when subsystems are modes or spatial operator algebras?

These are bridge pages. Protocol proofs, phase classifications, electronic-structure methods, optical hardware, Bell-test analysis, and relativistic field theory retain their own canonical homes.

The Common Backbone

For a bipartite tensor product,

$$\mathcal H = \mathcal H_A\otimes\mathcal H_B,$$

a state is separable when it can be written

$$\begin{aligned} \rho_{AB} &= \sum_k p_k \rho_A^{(k)}\otimes\rho_B^{(k)}, \\ p_k&\ge0, \qquad \sum_kp_k=1. \end{aligned}$$

If no such decomposition exists, the state is entangled across the stated split. Local predictions are controlled by

$$\rho_A = \operatorname{Tr}_B\rho_{AB}, \qquad \rho_B = \operatorname{Tr}_A\rho_{AB}.$$

For a pure state, the nonzero spectra of $\rho_A$ and $\rho_B$ coincide. The von Neumann entropy

$$S(A) = -\operatorname{Tr} (\rho_A\ln\rho_A)$$

is then an entanglement measure. For mixed states, $S(A)$ also includes classical and thermal uncertainty, so one must use a mixed-state diagnostic matched to the question.

Across every application, four decisions come first:

1. What physically defines $A$ and $B$?
2. Is the state pure, mixed, Gaussian, fixed-number, or field-theoretic?
3. Which operations and measurements are available locally?
4. Which conclusion is the chosen quantity actually licensed to support?

Quantum Information: Entanglement as a Resource

The Bell state

$$\lvert\Phi^+\rangle = \frac{\lvert00\rangle+\lvert11\rangle}{\sqrt2}$$

contains one ebit of pure bipartite entanglement. Each qubit is maximally mixed:

$$\rho_A = \rho_B = \frac{I_2}{2},$$

and therefore

$$S(A)=S(B)=\ln2.$$

The operational language is usually LOCC: local quantum operations supplemented by classical communication. Entanglement cannot increase on average under LOCC when measured by a valid entanglement monotone. This turns entanglement into a resource for teleportation, superdense coding, distributed computation, networking, and some error-correction tasks.

The resource statement must remain qualified:

• a noisy entangled state need not be directly useful for every protocol;
• different tasks order mixed states differently;
• distillability, bound entanglement, fidelity, and channel noise are separate properties;
• a protocol requires classical communication, measurements, and calibrated local controls in addition to a shared state.

Entanglement in Quantum Information owns the bridge. Full resource theories and communication protocols belong in the quantum-information volume.

Many-Body Physics: Entanglement Across a Cut

For a lattice ground state, choose a region $A$ and trace out its complement. In many gapped local systems, the leading entropy scales with the boundary rather than the volume:

$$S(A) \sim \alpha\lvert\partial A\rvert+cdots.$$

The coefficient $\alpha$ is generally nonuniversal and cutoff dependent. Subleading terms can carry universal information in suitable settings. In a one-dimensional critical system described by a conformal field theory, an interval of length $\ell$ in an infinite system has the characteristic form

$$S(\ell) = \frac{c}{3} \ln\left(\frac{\ell}{a}\right) + s_0,$$

where $c$ is the central charge, $a$ is a short-distance cutoff, and boundary conditions or finite geometry can change the coefficient and scaling function.

The reduced state may be written formally as

$$\rho_A = \frac{e^{-H_E}}{Z_E}, \qquad Z_E = \operatorname{Tr}(e^{-H_E}),$$

which defines an entanglement Hamiltonian $H_E$. Its spectrum can reveal structure not visible in the entropy alone. Area laws also help explain why matrix-product and tensor-network descriptions can be efficient for restricted classes of low-entanglement states.

Entanglement growth after a quench, volume-law eigenstates, localization, and topological corrections require model-specific assumptions. Entanglement Entropy in Many-Body Systems develops the scaling laws and their qualifications; Entanglement in Many-Body Physics provides the cross-volume reading map.

Quantum Chemistry: Correlation Beyond Antisymmetry

An $N$-electron state lies in the antisymmetric sector of the one-electron Hilbert space. A single Slater determinant,

$$\lvert\Phi\rangle = c_{i_1}^\dagger \cdots c_{i_N}^\dagger \lvert0\rangle,$$

already contains the exchange structure required for fermions. That antisymmetry alone is not what quantum chemistry calls correlation beyond Hartree–Fock.

In a spin-orbital basis, the electronic Hamiltonian has the standard form

$$\begin{aligned} H &= \sum_{p,q}h_{pq} c_p^\dagger c_q \\ &\quad+ \frac12 \sum_{p,q,r,s} V_{pq;rs} c_p^\dagger c_q^\dagger c_s c_r. \end{aligned}$$

Configuration interaction expands the state in multiple determinants. Orbital reduced states, mutual information, and entanglement diagnostics can help identify active spaces and strongly correlated orbital groups, but their numerical values depend on orbital choice and partition.

Useful distinctions are:

• exchange structure: imposed by fermionic antisymmetry;
• single-reference correlation: often captured efficiently around one dominant determinant;
• multireference correlation: requires several important configurations;
• orbital entanglement: defined relative to a chosen mode decomposition;
• spin entanglement: defined relative to a specified spin partition and observable access.

Entanglement in Quantum Chemistry owns this translation. Detailed Hartree–Fock, coupled-cluster, configuration-interaction, and density-matrix-renormalization methods belong in electronic structure and many-body theory.

Quantum Optics: Entanglement of Accessible Modes

Optical subsystems are commonly polarization, path, frequency, time-bin, or spatial modes. A polarization-entangled pair may be written

$$\lvert\Psi^-\rangle = \frac{ \lvert H\rangle_A\lvert V\rangle_B - \lvert V\rangle_A\lvert H\rangle_B }{\sqrt2}.$$

A single excitation delocalized across paths,

$$\lvert\psi\rangle = \alpha\lvert1,0\rangle + \beta\lvert0,1\rangle,$$

is entangled across the path-mode split when both amplitudes are nonzero. Continuous-variable experiments often use two-mode squeezing,

$$\lvert\mathrm{TMSV}\rangle = \sqrt{1-\lambda^2} \sum_{n=0}^{\infty} \lambda^n\lvert n,n\rangle,$$

whose quadrature correlations can be accessed with homodyne detection.

Source claims must be separated from state claims. Parametric down-conversion is a preparation mechanism, not a guarantee that every emitted or postselected pair realizes the target pure state. Loss, distinguishability, multipair emission, detector response, mode mismatch, and phase references all affect the operational state.

Entanglement in Quantum Optics owns the mode-level bridge. Continuous Variables and Modes supplies the quadrature and Gaussian-state orientation.

Foundations: A Hierarchy of Nonclassical Correlations

Entanglement, steering, and Bell nonlocality are not synonyms. Denoting the corresponding state classes by $\mathcal C_{\mathrm{ent}}$, $\mathcal C_{\mathrm{steer}}$, and $\mathcal C_{\mathrm{Bell}}$, standard bipartite definitions give strict inclusions:

$$\mathcal C_{\mathrm{Bell}} \subsetneq \mathcal C_{\mathrm{steer}} \subsetneq \mathcal C_{\mathrm{ent}}.$$

The precise boundaries depend on the allowed measurements, number of settings, direction of steering, and whether hidden states or hidden variables are being modeled. A state can therefore be entangled without violating the tested Bell inequality.

Entanglement also respects no-signaling. If a trace-preserving local channel $\mathcal E_B$ acts on $B$, then

$$\operatorname{Tr}_B \left[ (I_A\otimes\mathcal E_B)(\rho_{AB}) \right] = \rho_A.$$

Local statistics at $A$ do not reveal which trace-preserving operation was chosen at spacelike-separated $B$. Correlations become visible only when records are compared.

Measurement interactions can create entanglement between a system and apparatus, but that fact alone does not select an interpretation or solve the measurement problem. Decoherence explains suppression of interference in suitable reduced descriptions; it does not by itself choose a unique outcome.

Entanglement in Foundations owns this boundary. EPR arguments, Bell inequalities, no-signaling proofs, interpretations, and decoherence machinery remain in their dedicated volumes.

Quantum Field Theory: Locality Changes the Technical Setting

Finite collections of regulated modes can be treated with ordinary tensor products and Fock-space methods. A continuum quantum field adds sharper questions:

• local operator algebras need not factorize like finite-dimensional matrix algebras;
• vacuum correlations extend across spatial boundaries;
• sharp-region entanglement entropy is typically ultraviolet divergent;
• gauge constraints can obstruct naive spatial factorization;
• particle and mode decompositions may depend on observer or background.

For a regulated region $A$, one often finds an area-divergent leading term,

$$S(A) \sim \kappa \frac{\operatorname{Area}(\partial A)} {\epsilon^{d-2}} + \cdots,$$

in $d$ spacetime dimensions, with cutoff $\epsilon$ and theory-dependent coefficient $\kappa$. The precise power changes in low dimensions, and universal logarithmic or finite terms require a specified theory and geometry.

Mutual information,

$$I(A:B) = S(A)+S(B)-S(AB),$$

often cancels leading local boundary divergences for separated regions. Relative entropy,

$$D(\rho\Vert\sigma) = \operatorname{Tr} \left[ \rho(\ln\rho-\ln\sigma) \right],$$

is another robust comparison quantity with an operator-algebraic extension.

Entanglement in QFT Preview is deliberately bounded. It prepares the concepts but does not replace relativistic QFT, algebraic QFT, gauge-theory factorization, renormalization, or holography.

Comparison Matrix

Application	Natural subsystem	Useful first diagnostic	Essential caveat
communication protocol	remote laboratories or registers	fidelity and task-specific entanglement monotone	entangled does not mean useful for every task
lattice ground state	spatial region or orbital block	entropy, spectrum, or mutual information	scaling depends on dimension, gap, and geometry
electronic structure	spin-orbitals or active-space blocks	orbital entropy and mutual information	values depend on orbital basis and particle constraints
optical experiment	calibrated field modes	tomography, witnesses, or covariance criteria	loss and mode mismatch change the inferred state
foundations test	separated parties and measurement settings	steering or Bell inequality matched to the claim	entanglement alone is not Bell nonlocality
continuum field theory	regulated modes or local algebras	mutual information or relative entropy	sharp-region entropy is cutoff sensitive

The diagnostic should follow from the question, not the other way around. Reporting a scalar without the partition, state class, and operational assumptions makes comparisons unreliable.

A Cross-Field Workflow

1. Name the factors or algebras

Specify laboratories, spins, lattice regions, orbitals, optical modes, field modes, or local operator algebras.

2. State preparation and state class

Distinguish a target pure state from the measured mixed state. Record symmetry sectors, conserved charges, Gaussianity, and regulator choices.

3. Identify accessible local operations

LOCC, particle-number constraints, spatial locality, detector bandwidth, and gauge constraints can change which entanglement is operationally available.

4. Compute the appropriate reduced object

Use a reduced density matrix, covariance matrix, correlation matrix, or restricted state on an observable algebra as the setting requires.

5. Choose a claim-sized diagnostic

Use entropy for pure-state bipartite entanglement, mixed-state measures for resource questions, scaling data for many-body structure, witnesses for experiments, and Bell inequalities only for Bell-nonlocality claims.

6. Test robustness

Vary basis, cutoff, subsystem size, noise model, postselection rule, and measurement set where those choices are not physical invariants.

7. State the boundary

Say whether the result demonstrates inseparability, resource usefulness, a phase signature, electronic correlation, squeezing, steering, Bell violation, or a regulated field-theory quantity.

Common Mistakes

• Treating entanglement as one universal scalar. Different state classes and tasks require different diagnostics.
• Omitting the subsystem decomposition. Entanglement is always relative to a specified split or observable algebra.
• Using pure-state entropy for an arbitrary mixed state. Local entropy can include classical and thermal uncertainty.
• Calling fermionic antisymmetry correlation beyond Hartree–Fock. Exchange structure and many-determinant correlation are distinct.
• Treating an optical source label as a state certificate. Loss, distinguishability, and postselection must be included.
• Equating entanglement, steering, and Bell nonlocality. They form distinct operational classes.
• Claiming no-signaling means no correlation. It forbids controllable local signaling, not joint correlations.
• Using an area law as a universal phase identifier. Many states share leading scaling while differing in subleading structure.
• Reporting orbital entanglement without the orbital basis. Mode transformations can change the partition and values.
• Removing a QFT cutoff without tracking divergences. Regulator dependence is part of the result.
• Presenting holographic or topological previews as generic quantum-mechanical theorems. Their assumptions belong to specialized theories.

Reading Paths

Information and foundations: Entanglement in Quantum Information → Entanglement in Foundations → Entanglement Witnesses.

Matter and chemistry: Entanglement in Many-Body Physics → Entanglement in Quantum Chemistry → Many-Particle Hamiltonians.

Optics and continuous variables: Entanglement in Quantum Optics → Continuous Variables and Modes → Squeezed States as Entangled Modes.

Field-theory preparation: Creation, Annihilation, and Second Quantization → Entanglement in QFT Preview → Mutual Information.

Diagnostic and practice bridge: Reference, Problems, and Notebooks → Entanglement Diagnostic Table → Computational Notebooks.

References

• M. A. Nielsen and I. L. Chuang, Quantum Computation and Quantum Information, Cambridge University Press, 2010.
• R. Horodecki et al., “Quantum entanglement,” Reviews of Modern Physics 81, 865–942, 2009.
• L. Amico, R. Fazio, A. Osterloh, and V. Vedral, “Entanglement in many-body systems,” Reviews of Modern Physics 80, 517–576, 2008.
• J. Eisert, M. Cramer, and M. B. Plenio, “Area laws for the entanglement entropy,” Reviews of Modern Physics 82, 277–306, 2010.
• U. Schollwöck, “The density-matrix renormalization group in the age of matrix product states,” Annals of Physics 326, 96–192, 2011.
• A. Szabo and N. S. Ostlund, Modern Quantum Chemistry, Dover, 1996.
• G. K.-L. Chan and S. Sharma, “The density matrix renormalization group in quantum chemistry,” Annual Review of Physical Chemistry 62, 465–481, 2011.
• L. Mandel and E. Wolf, Optical Coherence and Quantum Optics, Cambridge University Press, 1995.
• D. F. Walls and G. J. Milburn, Quantum Optics, 2nd ed., Springer, 2008.
• H. M. Wiseman, S. J. Jones, and A. C. Doherty, “Steering, entanglement, nonlocality, and the Einstein–Podolsky–Rosen paradox,” Physical Review Letters 98, 140402, 2007.
• H. Casini and M. Huerta, “Entanglement entropy in free quantum field theory,” Journal of Physics A 42, 504007, 2009.
• P. Calabrese and J. Cardy, “Entanglement entropy and quantum field theory,” Journal of Statistical Mechanics P06002, 2004.

Exercises

Exercise 1: Local unitaries preserve pure-state entanglement

Let $\lvert\psi'\rangle=(U_A\otimes U_B)\lvert\psi\rangle$. Show that the reduced density operators before and after the transformation have the same eigenvalues.

Solution

The transformed density operator is

$$\rho_{AB}' = (U_A\otimes U_B) \rho_{AB} (U_A^\dagger\otimes U_B^\dagger).$$

Taking the partial trace over $B$ and using its invariance under unitary conjugation on the traced subsystem gives

$$\rho_A' = U_A\rho_AU_A^\dagger.$$

Unitary conjugation preserves eigenvalues. Hence the Schmidt coefficients and every pure-state entanglement measure depending only on them are unchanged.

Exercise 2: No signaling from a local measurement

Suppose a measurement on $B$ has Kraus operators $M_b$ satisfying $\sum_bM_b^\dagger M_b=I_B$. Show that the unconditioned state of $A$ is unchanged.

Solution

When the outcome is not communicated, the joint state becomes

$$\rho_{AB}' = \sum_b (I_A\otimes M_b) \rho_{AB} (I_A\otimes M_b^\dagger).$$

For any observable $X_A$,

$$\begin{aligned} Y &= X_A\otimes \sum_bM_b^\dagger M_b \\ &= X_A\otimes I_B, \\ \operatorname{Tr}(X_A\rho_A') &= \operatorname{Tr}(Y\rho_{AB}) \\ &= \operatorname{Tr}(X_A\rho_A). \end{aligned}$$

Thus all local expectation values agree, so $\rho_A'=\rho_A$.

Exercise 3: Bell-pair mutual information

Compute $I(A:B)$ for $\lvert\Phi^+\rangle$ using natural logarithms.

Solution

The joint state is pure, so $S(AB)=0$. Each reduced qubit is maximally mixed, giving $S(A)=S(B)=\ln2$. Therefore

$$\begin{aligned} I(A:B) &= S(A)+S(B)-S(AB) \\ &= 2\ln2. \end{aligned}$$

For a pure bipartite state, the mutual information is twice the entanglement entropy.

Exercise 4: What an area law does not determine

Two one-dimensional gapped ground states both have block entropy approaching a constant for large blocks. Does this prove they are in the same phase? Explain.

Solution

No. Saturation is a broad leading-scaling property shared by many short-range-entangled gapped states. It does not determine symmetry realization, edge structure, topological or symmetry-protected indices, degeneracy, or the detailed entanglement spectrum.

One must add the relevant symmetries, gap assumptions, boundary conditions, and phase diagnostics. The area law explains why low-entanglement representations may work; it is not by itself a complete phase classifier.

Exercise 5: Exchange structure versus orbital choice

Explain why the fermionic anticommutation relations are invariant under a unitary spin-orbital rotation while orbital entanglement values can change.

Solution

For $d_\alpha=\sum_pU_{\alpha p}^*c_p$, unitarity gives

$$\{d_\alpha,d_\beta^\dagger\} = \delta_{\alpha\beta}.$$

Thus the antisymmetric fermionic state space and exclusion rule are basis independent. Orbital entanglement, however, is defined by treating selected modes or mode blocks as subsystems. A unitary transformation that mixes those blocks changes the tensor-factor identification and therefore can change reduced orbital states and their entropies.

The invariant physics is recovered only after the partition and accessible observables are specified operationally.