Non-Reciprocal Open Quantum Spin Chains

• Non-Reciprocal Open Quantum Spin Chains are one-dimensional systems combining coherent Hamiltonian dynamics with asymmetric Lindblad dissipation to yield directed transport and unique non-equilibrium states.
• The time-dependent Generalized Gibbs Ensemble (t-GGE) framework analytically captures the system's evolution by approximating its state as a Gaussian density matrix reflecting conserved mode occupations.
• Nonlinear rate equations demonstrate anomalous power-law decay in observables like magnetization, emphasizing the critical role of non-reciprocal dissipation in shaping relaxation dynamics.

Non-reciprocal open quantum spin chains are one-dimensional quantum lattice systems wherein spin degrees of freedom are subject to both coherent Hamiltonian evolution and engineered, non-reciprocal dissipative processes. Such systems realize fundamentally non-equilibrium steady states and transient dynamics that go beyond any description in terms of closed integrable models or standard (reciprocal) dissipation. Non-reciprocality, often implemented via Lindblad jump operators that lack left-right symmetry, induces directed (chiral) transport phenomena and non-trivial late-time decay exponents, and is analytically tractable within a time-dependent generalized Gibbs ensemble (t-GGE) formalism in the weak dissipation regime (Marché et al., 13 Jan 2026). This framework connects nonequilibrium quantum statistical mechanics, integrability, and open quantum system theory.

1. Model Definition and Non-reciprocal Dissipation

The canonical model is the open XX spin chain under a Lindblad master equation,

$$\frac{d\rho}{dt} = -i[H, \rho] + \kappa \sum_j \left( L_j\rho L_j^\dagger - \frac{1}{2} \{ L_j^\dagger L_j, \rho \} \right)$$

with the XX Hamiltonian,

$$H = -\frac{J}{2}\sum_j (S^+_{j+1} S^-_j + S^-_{j+1} S^+_j)\,,$$

and non-reciprocal two-site lowering jumps,

$$L_j = S^-_j + e^{i\phi} S^-_{j+1},\quad \phi\in(-\pi, \pi].$$

The presence of the phase $\phi$ in $L_j$ breaks left-right symmetry, leading to fundamentally asymmetric dissipative dynamics. Under the Jordan–Wigner transformation, the system maps to a free-fermion chain with string-modified jump operators. This non-reciprocity is not gauge-removable and gives rise to directed transport at both transient and stationary levels (Marché et al., 13 Jan 2026).

2. Time-dependent Generalized Gibbs Ensemble (t-GGE) Approach

For weak dissipation ($\kappa\ll J$), the system remains locally quasi-equilibrated with respect to the integrable charges of the closed XX model. At all times the state is approximated by a Gaussian density matrix constructed as a t-GGE: $$\rho_{\mathrm{GGE}}(t) = Z^{-1}(t)\exp\left[-\sum_m \beta_m(t) Q_m\right]$$ or, equivalently, in terms of momentum-mode occupations,

$$\rho_{\mathrm{GGE}}(t) = \tilde{Z}^{-1}(t)\exp\left[-\int_0^{2\pi} \lambda_k(t) c^\dagger(k)c(k) dk\right].$$

The Lagrange multipliers $\lambda_k(t)$ parametrize the instantaneous expectation values of the conserved charges and are related to the $\beta_m(t)$ via discrete Fourier transform. Physical observables are then linear functionals of the rapidity distribution $$\rho(k,t) = \mathrm{Tr}[c^\dagger(k)c(k)\rho_{GGE}(t)]$$ (Marché et al., 13 Jan 2026).

3. Nonlinear Rate Equation and Rapidities

The evolution of the rapidity distribution is governed by a closed, nonlinear, integro-differential rate equation derived by projecting the Lindblad dynamics onto the t-GGE ansatz: $$-\frac{1}{2\kappa}\frac{d\rho(k)}{dt} = \rho(k)[1-\rho(k)][1+\cos(k+\phi)] + 2\left[PV\int \frac{dq}{2\pi} \rho(q)\frac{\cos[(q+\phi)/2]}{\sin[(k-q)/2]}\right]^2 + \int \frac{dq}{2\pi} \rho(q)[1+\cos(q+\phi)]\, PV\int \frac{dp}{2\pi} \frac{\rho(k)-\rho(p)}{\sin^2[(k-p)/2]}.$$ This equation encapsulates the nonlinear mixing of occupations at different rapidities, crucial to the breakdown of simple exponential or mean-field algebraic decay and the emergence of anomalous scaling (Marché et al., 13 Jan 2026). Numerically efficient forms using circular Hilbert transforms are provided for practical computation.

4. Observables: Magnetization and Current

Key observables accessible in the t-GGE formalism are simple integrals over the rapidity distribution:

• Magnetization density: $n(t)=\int_0^{2\pi}\rho(k,t)\,dk/(2\pi)$
• Spin current density: $$\mathcal{J}(t)=J\int_0^{2\pi}\sin k\, \rho(k,t)\,dk/(2\pi)$$

At late times, $\rho(k)$ sharpens around the slowest decaying mode $k^*=\pi-\phi$, leading to the relation

$\frac{\mathcal{J}}{n} \rightarrow J\sin\phi\,,$

and for the energy density, $\epsilon/n \rightarrow J\cos\phi$. Subleading corrections are captured by fits assuming a Gaussian profile around $k^*$ (Marché et al., 13 Jan 2026).

5. Anomalous Power-law Relaxation and Nonlinear Effects

Numerical solutions of the rate equation reveal that the magnetization decays as a nontrivial algebraic law, $n(t)\sim t^{-\chi}$, with exponents depending on initial state and the non-reciprocity parameter $\phi$. Specifically, for the all-up state ($\theta=0$), $\chi\approx0.58$, while for $\theta=\pi/4$, $\chi\approx0.515$ (Marché et al., 13 Jan 2026). In contrast, the free-fermion analogue yields analytic exponents $\chi=1/2$ or $3/2$. Logarithmic derivatives ($D_1(t) = -d\ln n/d\ln t$) exhibit slow drifts and higher-order corrections, suggesting either very slow crossovers or logarithmic, rather than pure, power-law scaling. This anomalous kinetics is a direct manifestation of the nonlinear rapidity mixing induced by non-reciprocal dissipation, which is absent in mean-field or naive non-interacting models.

6. Integration with Broader t-GGE and GHD Frameworks

The t-GGE strategy as developed for non-reciprocal spin chains is structurally similar to that for weakly open integrable bosonic chains subject to loss/gain processes (Rossini et al., 2020, Lumia et al., 2024), as well as classical models (Pandey et al., 2024). In all such cases, the combination of local quasi-equilibration and slow drift due to dissipation enables a closed-form evolution for overlap densities (mode occupations/rapidities). The t-GGE is thus a unifying formalism for describing the nonequilibrium hydrodynamic evolution and non-stationary relaxation in a range of open integrable systems. However, the nonlinearities and exponents in the non-reciprocal spin case are more intricate due to the specific structure of the Lindblad operators and Jordan–Wigner strings (Marché et al., 13 Jan 2026).

7. Open Problems and Future Directions

Determining the precise asymptotic form of relaxation—specifically, whether the observed exponents cross over to the simple $1/2$ power law with subleading logarithmic or other corrections at very late times—remains unresolved. Further, the extension of the approach to strong dissipation or strongly non-integrable settings is nontrivial and likely requires moving beyond Gaussian/t-GGE ansätze (Marché et al., 13 Jan 2026). The methods are generalizable to weakly open integrable chains provided a full set of conserved charges is available, suggesting future studies may systematically explore dissipative engineering of nonequilibrium steady states and transport in more complex quantum media.