Inhomogeneous XY Spin Chains

• Inhomogeneous XY spin chains are one-dimensional spin-½ models with spatially varying couplings and anisotropies that remain exactly solvable via Jordan–Wigner fermionization.
• These systems exhibit rich phenomena including quantum phase transitions, dimerization, and topological boundary modes, making them vital for studying criticality and entanglement.
• Analytical techniques such as Bogoliubov–de Gennes transformations and orthogonal-polynomial methods provide precise spectral engineering and insights into transport and non-equilibrium dynamics.

Inhomogeneous XY spin chains are one-dimensional spin-$\tfrac12$ systems in which the transverse exchange couplings, anisotropies, magnetic fields, or boundary data vary with position rather than remaining uniform. In their most common form they are written as

$$H_{XY}=\sum_n\Bigl(J_n^x\,\sigma_n^x\sigma_{n+1}^x+J_n^y\,\sigma_n^y\sigma_{n+1}^y\Bigr)+\sum_n h_n\,\sigma_n^z,$$

or, equivalently,

$$H_{XY}=\sum_n J_n\left[(1+\gamma_n)S_n^xS_{n+1}^x+(1-\gamma_n)S_n^yS_{n+1}^y\right]+\sum_n h_n S_n^z.$$

Because these models remain quadratic after Jordan–Wigner fermionization in the XX limit and in the general anisotropic case become Bogoliubov–de Gennes free-fermion problems, they form a central class of exactly solvable or partially solvable quantum many-body systems for studying spectral engineering, quantum phase transitions, transport, entanglement, topological boundary modes, and boundary-driven non-equilibrium steady states (Crampé et al., 27 Aug 2025, Bernard et al., 8 Jul 2025).

1. Canonical definitions and limiting cases

A widely used inhomogeneous realization is the period-two anisotropic XY chain in a transverse field, with two sites $a,b$ per unit cell and alternating couplings and anisotropies,

$$\begin{aligned} \mathscr{H} = -\sum_{l=1}^{N} \Big[ & J_{a}(1+\gamma_{a}) S_{l,a}^{x}S_{l,b}^{x} + J_{a}(1-\gamma_{a}) S_{l,a}^{y}S_{l,b}^{y} \ & + J_{b}(1+\gamma_{b}) S_{l,b}^{x}S_{l+1,a}^{x} + J_{b}(1-\gamma_{b}) S_{l,b}^{y}S_{l+1,a}^{y} \ & + h (S_{l,a}^{z} + S_{l,b}^{z}) \Big]. \end{aligned}$$

Here the inhomogeneity is encoded by the alternating bond data $(J_a,\gamma_a)$ and $(J_b,\gamma_b)$, so that intracell and intercell links are inequivalent (Ma et al., 2012).

Several standard homogeneous chains appear as special cases. Setting $\gamma_a=\gamma_b=1$ and $J_a=J_b\equiv J$ gives the uniform quantum Ising chain in a transverse field. Setting $\gamma_a=\gamma_b=0$ and $$H_{XY}=\sum_n\Bigl(J_n^x\,\sigma_n^x\sigma_{n+1}^x+J_n^y\,\sigma_n^y\sigma_{n+1}^y\Bigr)+\sum_n h_n\,\sigma_n^z,$$0 gives the uniform XX chain. Setting $$H_{XY}=\sum_n\Bigl(J_n^x\,\sigma_n^x\sigma_{n+1}^x+J_n^y\,\sigma_n^y\sigma_{n+1}^y\Bigr)+\sum_n h_n\,\sigma_n^z,$$1 and $$H_{XY}=\sum_n\Bigl(J_n^x\,\sigma_n^x\sigma_{n+1}^x+J_n^y\,\sigma_n^y\sigma_{n+1}^y\Bigr)+\sum_n h_n\,\sigma_n^z,$$2 gives the standard homogeneous anisotropic XY model in a transverse field. More generally, $$H_{XY}=\sum_n\Bigl(J_n^x\,\sigma_n^x\sigma_{n+1}^x+J_n^y\,\sigma_n^y\sigma_{n+1}^y\Bigr)+\sum_n h_n\,\sigma_n^z,$$3 with $$H_{XY}=\sum_n\Bigl(J_n^x\,\sigma_n^x\sigma_{n+1}^x+J_n^y\,\sigma_n^y\sigma_{n+1}^y\Bigr)+\sum_n h_n\,\sigma_n^z,$$4 gives an alternating-coupling or dimerized XY chain, while $$H_{XY}=\sum_n\Bigl(J_n^x\,\sigma_n^x\sigma_{n+1}^x+J_n^y\,\sigma_n^y\sigma_{n+1}^y\Bigr)+\sum_n h_n\,\sigma_n^z,$$5 with $$H_{XY}=\sum_n\Bigl(J_n^x\,\sigma_n^x\sigma_{n+1}^x+J_n^y\,\sigma_n^y\sigma_{n+1}^y\Bigr)+\sum_n h_n\,\sigma_n^z,$$6 gives alternating anisotropy (Ma et al., 2012).

Open-chain formulations are equally important. One exactly solvable inhomogeneous XY chain is defined on $$H_{XY}=\sum_n\Bigl(J_n^x\,\sigma_n^x\sigma_{n+1}^x+J_n^y\,\sigma_n^y\sigma_{n+1}^y\Bigr)+\sum_n h_n\,\sigma_n^z,$$7 sites by

$$H_{XY}=\sum_n\Bigl(J_n^x\,\sigma_n^x\sigma_{n+1}^x+J_n^y\,\sigma_n^y\sigma_{n+1}^y\Bigr)+\sum_n h_n\,\sigma_n^z,$$8

with arbitrary real sequences $$H_{XY}=\sum_n\Bigl(J_n^x\,\sigma_n^x\sigma_{n+1}^x+J_n^y\,\sigma_n^y\sigma_{n+1}^y\Bigr)+\sum_n h_n\,\sigma_n^z,$$9. In this form $$H_{XY}=\sum_n J_n\left[(1+\gamma_n)S_n^xS_{n+1}^x+(1-\gamma_n)S_n^yS_{n+1}^y\right]+\sum_n h_n S_n^z.$$0 controls the average bond exchange, $$H_{XY}=\sum_n J_n\left[(1+\gamma_n)S_n^xS_{n+1}^x+(1-\gamma_n)S_n^yS_{n+1}^y\right]+\sum_n h_n S_n^z.$$1 the bond anisotropy, and $$H_{XY}=\sum_n J_n\left[(1+\gamma_n)S_n^xS_{n+1}^x+(1-\gamma_n)S_n^yS_{n+1}^y\right]+\sum_n h_n S_n^z.$$2 the local longitudinal field (Bernard et al., 8 Jul 2025).

“Inhomogeneous” is therefore broader than random disorder. The literature includes periodic alternation, exponentially deformed “rainbow” couplings, Harper-type superlattice modulations, boundary inhomogeneities introduced at the transfer-matrix level, and long-range-correlated field disorder (Buruaga et al., 2018, Lado et al., 2019, Nepomechie et al., 2021, Almeida et al., 2017). A common misconception is that inhomogeneity necessarily means loss of analytic control; in this class, many inhomogeneous patterns remain exactly solvable or reducible to controlled finite-dimensional spectral problems.

2. Jordan–Wigner fermionization and exact solution methods

The structural reason for the prominence of XY chains is that they map to quadratic fermionic models. In the two-sublattice period-two chain, the Jordan–Wigner map introduces spinless fermions $$H_{XY}=\sum_n J_n\left[(1+\gamma_n)S_n^xS_{n+1}^x+(1-\gamma_n)S_n^yS_{n+1}^y\right]+\sum_n h_n S_n^z.$$3 with

$$H_{XY}=\sum_n J_n\left[(1+\gamma_n)S_n^xS_{n+1}^x+(1-\gamma_n)S_n^yS_{n+1}^y\right]+\sum_n h_n S_n^z.$$4

and the Hamiltonian becomes quadratic, containing hopping, pairing, and on-site terms. After Fourier transform, one obtains a $$H_{XY}=\sum_n J_n\left[(1+\gamma_n)S_n^xS_{n+1}^x+(1-\gamma_n)S_n^yS_{n+1}^y\right]+\sum_n h_n S_n^z.$$5 Hermitian matrix $$H_{XY}=\sum_n J_n\left[(1+\gamma_n)S_n^xS_{n+1}^x+(1-\gamma_n)S_n^yS_{n+1}^y\right]+\sum_n h_n S_n^z.$$6 acting on

$$H_{XY}=\sum_n J_n\left[(1+\gamma_n)S_n^xS_{n+1}^x+(1-\gamma_n)S_n^yS_{n+1}^y\right]+\sum_n h_n S_n^z.$$7

so the anisotropic inhomogeneous problem becomes a coupled $$H_{XY}=\sum_n J_n\left[(1+\gamma_n)S_n^xS_{n+1}^x+(1-\gamma_n)S_n^yS_{n+1}^y\right]+\sum_n h_n S_n^z.$$8-space Bogoliubov problem. The period-two structure appears through the $$H_{XY}=\sum_n J_n\left[(1+\gamma_n)S_n^xS_{n+1}^x+(1-\gamma_n)S_n^yS_{n+1}^y\right]+\sum_n h_n S_n^z.$$9-dependent hopping amplitude $a,b$0 and pairing amplitude $a,b$1, producing a two-band fermionic spectrum (Ma et al., 2012).

For general open inhomogeneous XY chains, the quadratic fermionic Hamiltonian can be written in Nambu form,

$a,b$2

where $a,b$3 is real symmetric tridiagonal and $a,b$4 is real antisymmetric tridiagonal. The eigenvalues of $a,b$5 come in pairs $a,b$6, and the many-body spectrum is generated by Bogoliubov quasiparticles with energies $a,b$7 (Bernard et al., 8 Jul 2025).

In the XX limit, pairing vanishes and the problem collapses to a single real symmetric tridiagonal matrix,

$a,b$8

or equivalently to a Jacobi recurrence. This is the setting in which orthogonal-polynomial methods become especially powerful (Bernard et al., 2024, Crampé et al., 27 Aug 2025).

A distinctive recent development is the construction of exactly solvable inhomogeneous XY chains through $a,b$9-contiguity relations of $$\begin{aligned} \mathscr{H} = -\sum_{l=1}^{N} \Big[ & J_{a}(1+\gamma_{a}) S_{l,a}^{x}S_{l,b}^{x} + J_{a}(1-\gamma_{a}) S_{l,a}^{y}S_{l,b}^{y} \ & + J_{b}(1+\gamma_{b}) S_{l,b}^{x}S_{l+1,a}^{x} + J_{b}(1-\gamma_{b}) S_{l,b}^{y}S_{l+1,a}^{y} \ & + h (S_{l,a}^{z} + S_{l,b}^{z}) \Big]. \end{aligned}$$0-Racah polynomials. There the coupled recurrences for Bogoliubov amplitudes are matched to contiguity relations, yielding explicit formulas for $$\begin{aligned} \mathscr{H} = -\sum_{l=1}^{N} \Big[ & J_{a}(1+\gamma_{a}) S_{l,a}^{x}S_{l,b}^{x} + J_{a}(1-\gamma_{a}) S_{l,a}^{y}S_{l,b}^{y} \ & + J_{b}(1+\gamma_{b}) S_{l,b}^{x}S_{l+1,a}^{x} + J_{b}(1-\gamma_{b}) S_{l,b}^{y}S_{l+1,a}^{y} \ & + h (S_{l,a}^{z} + S_{l,b}^{z}) \Big]. \end{aligned}$$1 and closed forms for the single-particle energies $$\begin{aligned} \mathscr{H} = -\sum_{l=1}^{N} \Big[ & J_{a}(1+\gamma_{a}) S_{l,a}^{x}S_{l,b}^{x} + J_{a}(1-\gamma_{a}) S_{l,a}^{y}S_{l,b}^{y} \ & + J_{b}(1+\gamma_{b}) S_{l,b}^{x}S_{l+1,a}^{x} + J_{b}(1-\gamma_{b}) S_{l,b}^{y}S_{l+1,a}^{y} \ & + h (S_{l,a}^{z} + S_{l,b}^{z}) \Big]. \end{aligned}$$2 (Bernard et al., 8 Jul 2025). In the XX case, a complementary route maps inhomogeneous chains to quasi-exactly solvable Schrödinger problems and weakly orthogonal polynomials, with the chain couplings read off from the three-term recurrence coefficients (Finkel et al., 2020).

This divide between XX and anisotropic XY is methodologically important. Inhomogeneous XX chains are governed by scalar Jacobi matrices; anisotropic XY chains require Nambu doubling and block-tridiagonal structure. This suggests why exact algebraic symmetries that are transparent in XX do not automatically extend to generic anisotropic XY chains.

3. Structured inhomogeneity: dimerization, rainbow deformations, superlattices, and disorder

Alternating or dimerized chains are the simplest structured inhomogeneous XY systems. The period-two chain already shows that alternating bonds and anisotropies generate richer critical behavior than the homogeneous model and can produce more than one quantum phase-transition point in some parameter regions (Ma et al., 2012).

A different deformation is the “rainbow” XX chain, where couplings decay exponentially away from the center. In spin language,

$$\begin{aligned} \mathscr{H} = -\sum_{l=1}^{N} \Big[ & J_{a}(1+\gamma_{a}) S_{l,a}^{x}S_{l,b}^{x} + J_{a}(1-\gamma_{a}) S_{l,a}^{y}S_{l,b}^{y} \ & + J_{b}(1+\gamma_{b}) S_{l,b}^{x}S_{l+1,a}^{x} + J_{b}(1-\gamma_{b}) S_{l,b}^{y}S_{l+1,a}^{y} \ & + h (S_{l,a}^{z} + S_{l,b}^{z}) \Big]. \end{aligned}$$3

with either bond-centered or site-centered inversion symmetry. In the unfolded chain, these couplings produce long-range entanglement and volume-law entanglement entropy; after folding, the long-range entanglement becomes short-range. Bond-centered foldings lead to trivial phases, whereas site-centered foldings lead to nontrivial symmetry-protected phases: in the $$\begin{aligned} \mathscr{H} = -\sum_{l=1}^{N} \Big[ & J_{a}(1+\gamma_{a}) S_{l,a}^{x}S_{l,b}^{x} + J_{a}(1-\gamma_{a}) S_{l,a}^{y}S_{l,b}^{y} \ & + J_{b}(1+\gamma_{b}) S_{l,b}^{x}S_{l+1,a}^{x} + J_{b}(1-\gamma_{b}) S_{l,b}^{y}S_{l+1,a}^{y} \ & + h (S_{l,a}^{z} + S_{l,b}^{z}) \Big]. \end{aligned}$$4-symmetric folded spin-$$\begin{aligned} \mathscr{H} = -\sum_{l=1}^{N} \Big[ & J_{a}(1+\gamma_{a}) S_{l,a}^{x}S_{l,b}^{x} + J_{a}(1-\gamma_{a}) S_{l,a}^{y}S_{l,b}^{y} \ & + J_{b}(1+\gamma_{b}) S_{l,b}^{x}S_{l+1,a}^{x} + J_{b}(1-\gamma_{b}) S_{l,b}^{y}S_{l+1,a}^{y} \ & + h (S_{l,a}^{z} + S_{l,b}^{z}) \Big]. \end{aligned}$$5 chain the phase belongs to the Su–Schrieffer–Heeger class, and in the $$\begin{aligned} \mathscr{H} = -\sum_{l=1}^{N} \Big[ & J_{a}(1+\gamma_{a}) S_{l,a}^{x}S_{l,b}^{x} + J_{a}(1-\gamma_{a}) S_{l,a}^{y}S_{l,b}^{y} \ & + J_{b}(1+\gamma_{b}) S_{l,b}^{x}S_{l+1,a}^{x} + J_{b}(1-\gamma_{b}) S_{l,b}^{y}S_{l+1,a}^{y} \ & + h (S_{l,a}^{z} + S_{l,b}^{z}) \Big]. \end{aligned}$$6-symmetric folded chain it is in the Haldane phase (Buruaga et al., 2018).

Harper-type superlattice modulation is another major pattern. The modulated XY limit

$$\begin{aligned} \mathscr{H} = -\sum_{l=1}^{N} \Big[ & J_{a}(1+\gamma_{a}) S_{l,a}^{x}S_{l,b}^{x} + J_{a}(1-\gamma_{a}) S_{l,a}^{y}S_{l,b}^{y} \ & + J_{b}(1+\gamma_{b}) S_{l,b}^{x}S_{l+1,a}^{x} + J_{b}(1-\gamma_{b}) S_{l,b}^{y}S_{l+1,a}^{y} \ & + h (S_{l,a}^{z} + S_{l,b}^{z}) \Big]. \end{aligned}$$7

maps, for spin $$\begin{aligned} \mathscr{H} = -\sum_{l=1}^{N} \Big[ & J_{a}(1+\gamma_{a}) S_{l,a}^{x}S_{l,b}^{x} + J_{a}(1-\gamma_{a}) S_{l,a}^{y}S_{l,b}^{y} \ & + J_{b}(1+\gamma_{b}) S_{l,b}^{x}S_{l+1,a}^{x} + J_{b}(1-\gamma_{b}) S_{l,b}^{y}S_{l+1,a}^{y} \ & + h (S_{l,a}^{z} + S_{l,b}^{z}) \Big]. \end{aligned}$$8, to the off-diagonal Harper model

$$\begin{aligned} \mathscr{H} = -\sum_{l=1}^{N} \Big[ & J_{a}(1+\gamma_{a}) S_{l,a}^{x}S_{l,b}^{x} + J_{a}(1-\gamma_{a}) S_{l,a}^{y}S_{l,b}^{y} \ & + J_{b}(1+\gamma_{b}) S_{l,b}^{x}S_{l+1,a}^{x} + J_{b}(1-\gamma_{b}) S_{l,b}^{y}S_{l+1,a}^{y} \ & + h (S_{l,a}^{z} + S_{l,b}^{z}) \Big]. \end{aligned}$$9

which is the one-dimensional representative of a two-dimensional Harper–Hofstadter problem. Here $(J_a,\gamma_a)$0 plays the role of flux and $(J_a,\gamma_a)$1 the role of a pump parameter (Lado et al., 2019).

A fourth pattern is long-range-correlated disorder in the local fields. In the isotropic XX chain with Hamiltonian

$(J_a,\gamma_a)$2

the fields $(J_a,\gamma_a)$3 are chosen from a long-range-correlated sequence with power spectrum $(J_a,\gamma_a)$4. For $(J_a,\gamma_a)$5, a finite band of extended states emerges around the band center, while localized states survive near the band edges; the resulting interplay between localization and delocalization changes the spatial pattern of entanglement distribution (Almeida et al., 2017).

These examples show that inhomogeneity is not a single perturbative motif but a design variable. Period-two modulation, exponentially deformed couplings, quasiperiodic superlattices, and correlated disorder generate qualitatively different spectra and correlations even though all remain within the XY/XX framework.

4. Criticality, Berry geometry, and topological structure

The geometric approach to quantum criticality has a natural realization in inhomogeneous XY chains. For the period-two anisotropic XY model, one introduces a one-parameter family

$(J_a,\gamma_a)$6

which is $(J_a,\gamma_a)$7-periodic in $(J_a,\gamma_a)$8 and leaves the spectrum unchanged. The geometric phase of the ground state then detects the quantum phase transitions: in some parameter regions there may exist more than one transition point, and these points correspond to divergence or extremum properties of the Berry curvature (Ma et al., 2012).

Topological interpretations are especially transparent in superlattice-modulated chains. In the Harper–XY chain, the Jordan–Wigner image is an off-diagonal Harper model, and the one-dimensional family labeled by $(J_a,\gamma_a)$9 can be embedded into a two-dimensional Hofstadter problem. The excitation gaps therefore inherit Chern numbers, and open chains exhibit boundary-localized in-gap states whose energies cross the gaps as $(J_b,\gamma_b)$0 is varied. In spin language these appear as boundary spin excitations in the local dynamical structure factor, and the same topological boundary modes remain visible when one turns on the $(J_b,\gamma_b)$1 interaction and moves to the isotropic Heisenberg limit (Lado et al., 2019).

Site-centered rainbow inhomogeneity provides a different topological mechanism. After folding, the $(J_b,\gamma_b)$2-symmetric spin-$(J_b,\gamma_b)$3 chain falls into the SSH class, its entanglement spectrum is twofold degenerate for any $(J_b,\gamma_b)$4, and the topological structure is encoded in a robust zero mode of the single-particle entanglement spectrum. Bond-centered folding, by contrast, remains topologically trivial (Buruaga et al., 2018).

A useful corrective to oversimplified topological language is that topology in inhomogeneous XY chains is model-dependent. Harper modulation, folded rainbow deformations, and Majorana-supporting anisotropic chains each realize different mechanisms. The data support no universal claim that generic inhomogeneity automatically produces protected boundary modes; rather, specific symmetry classes, spatial patterns, and spectral gaps are decisive.

5. Exact states, emergent symmetry, and boundary inhomogeneity

One striking exact result concerns the isotropic XX limit. For a generic inhomogeneous open XX chain in a uniform transverse field,

$(J_b,\gamma_b)$5

a suitable uniform shift of the transverse field can force the one-particle hopping matrix to have an exact zero-energy mode. From this zero mode one constructs non-local operators $(J_b,\gamma_b)$6 that commute with $(J_b,\gamma_b)$7, satisfy the $(J_b,\gamma_b)$8 commutation relations, and obey $(J_b,\gamma_b)$9. As a consequence every many-body eigenvalue is at least two-fold degenerate. The generators are “inhomogeneous” because they carry site-dependent coefficients $\gamma_a=\gamma_b=1$0 and Jordan–Wigner strings (Crampé et al., 27 Aug 2025).

This mechanism is special to the XX limit. In fermionic language it relies on the fact that the one-particle problem is governed by a single real symmetric tridiagonal matrix and that the exact zero mode is a single annihilation/creation operator. In a generic anisotropic XY chain, the free-fermion problem lives in Nambu space, zero modes are Bogoliubov or Majorana modes, and they do not generically organize into an exact global $\gamma_a=\gamma_b=1$1 algebra commuting with the Hamiltonian (Crampé et al., 27 Aug 2025).

Anisotropic XY chains nevertheless support their own exact inhomogeneous states. For the anisotropic spin-$\gamma_a=\gamma_b=1$2 XY model obtained at $\gamma_a=\gamma_b=1$3, the spin-helix states

$\gamma_a=\gamma_b=1$4

are exact eigenstates provided each linear system size satisfies $\gamma_a=\gamma_b=1$5. In this XY limit they are exact zero-energy eigenstates. For special $\gamma_a=\gamma_b=1$6, the local spinors reduce to $\gamma_a=\gamma_b=1$7 and $\gamma_a=\gamma_b=1$8, producing a four-site cycle in the $\gamma_a=\gamma_b=1$9-plane (Zheng et al., 21 May 2025).

Integrable boundary inhomogeneity leads to yet another exact structure. In open trigonometric chains one can modify the transfer matrix by shifting the spectral parameter in one reflected monodromy matrix,

$J_a=J_b\equiv J$0

For a special choice $J_a=J_b\equiv J$1 together with staggered bulk inhomogeneities, the logarithmic derivative at $J_a=J_b\equiv J$2 yields a local Hamiltonian, and the Bethe ansatz is profoundly modified: the crossing relation is shifted, the $J_a=J_b\equiv J$3-functions take the form

$J_a=J_b\equiv J$4

and the boundary factors enter nontrivially into the Bethe equations (Nepomechie et al., 2021). Although developed for higher-rank trigonometric chains, this construction identifies boundary inhomogeneity as a distinct integrable mechanism rather than a mere local boundary field.

6. Transport, non-equilibrium dynamics, and quantum-information applications

Inhomogeneous XX chains coupled to bosonic heat baths admit an exact non-equilibrium steady state. With baths attached at the two ends, the reduced density matrix satisfies a Lindblad equation, and the NESS is diagonal in the fermionic normal-mode occupations. The stationary spin and heat currents are mode sums weighted by the edge amplitudes $J_a=J_b\equiv J$5 and $J_a=J_b\equiv J$6. For mirror-symmetric chains, the thermal conductivity in linear response is

$J_a=J_b\equiv J$7

At high temperature this implies $J_a=J_b\equiv J$8, so transport is ballistic for any mirror-symmetric inhomogeneous XX chain. Breaking mirror symmetry can drastically reduce both heat and spin conductivities, and in examples with linear or random field perturbations the relevant coefficient $J_a=J_b\equiv J$9 decreases strongly with system size, indicating subdiffusive transport (Bernard et al., 2024).

The same review literature ties mirror symmetry to perfect state transfer. For an open XX chain, PST from one end to the other requires mirror symmetry of the couplings and a spectral spacing condition

$\gamma_a=\gamma_b=0$0

The Krawtchouk chain satisfies this criterion and exhibits PST, but exact open-system calculations show that PST in the closed chain does not automatically imply superior low-temperature heat transport in the corresponding driven open system (Bernard et al., 2024, Bernard et al., 2024). This corrects a common overidentification of state-transfer optimality with transport optimality.

Time-dependent impurities can also be treated rigorously. For inhomogeneous XY chains with time-dependent bond and field data, the Heisenberg dynamics of the Jordan–Wigner fermions is governed by Bogoliubov–Valatin transformations satisfying linear ODE or integro-differential systems. Under boundedness, summability, and piecewise continuity assumptions, one has local existence of the dynamics, and under additional assumptions global existence follows. In the XX limit with field impurities, the resulting site-variable equations contain Bessel-function memory kernels, making the non-Markovian character explicit (Genovese, 2014).

Entanglement transport provides another operational application. In the XX chain with long-range-correlated diagonal disorder $\gamma_a=\gamma_b=0$1, the concurrence between two sites in the single-excitation sector is $\gamma_a=\gamma_b=0$2. As $\gamma_a=\gamma_b=0$3 increases, extended modes appear in the middle of the band, and the entanglement wave can spread to distant sites; correlated disorder also improves the transmission of an initially maximally entangled state when compared with the uncorrelated case (Almeida et al., 2017).

Taken together, these results show that inhomogeneous XY spin chains are not merely perturbed homogeneous models. They are a broad analytical and phenomenological category in which spatial variation serves as a control parameter for spectra, criticality, edge structure, transport, and entanglement. The XX limit remains the most algebraically transparent sector, but the anisotropic XY case extends the same themes into BdG spectral theory, exact spin-helix states, and boundary-sensitive integrable constructions.