A Zoology of Quantum Turing Patterns

AI summary

Overview

Research area: Open quantum systems and quantum pattern formation — specifically Turing pattern formation in driven-dissipative bosonic lattices governed by a Lindblad master equation, and how the morphology of those patterns relates to Gaussian nonclassicality witnesses.

Technical level: Advanced. The paper assumes familiarity with Lindblad generators, Gaussian covariance matrices, symplectic eigenvalues, partial transposition, and Glauber–Sudarshan P-function witnesses.

Scope (one sentence): The paper maps a "zoology" of Turing morphologies produced by a single Lindblad generator and shows quantitatively that the noise level at which visible stripe order disappears is parametrically separated from the noise levels at which two Gaussian nonclassicality witnesses on the same k* mode vanish.

What This Paper Is About

Turing mechanisms explain how local reaction dynamics plus differential transport select a finite wavelength, and a single selected wavelength can support stripes, spots, holes, labyrinths, or defects depending on nearby parameters. The open question the paper addresses is whether the pattern morphology and the quantum-nonclassicality witnesses associated with that same ordering mode fail at the same physical noise level. Using one completely positive Lindblad generator that carries both the drift (first moments) and the diffusion (second moments), the author finds that they do not: resolved stripes can survive after both tested Gaussian witness margins cross zero.

Key Contributions

1. A morphology atlas from one generator. At λ = 0.38 the same Lindblad equation is mapped across a parameter grid into stripe, spot-rich, hole-rich, labyrinthine, and defect-rich fields, with a principal stripe branch (P) and an orthogonal stripe branch (H) marked as stable solutions. All atlas cells start from the same deterministic preparation.

2. Two distinct noise scalings derived from one generator. The first moments depend on Θ = (2n_th + 1)/N (with Θ ≥ N⁻¹), while the canonical covariance depends on n_th, so the two Gaussian witness thresholds on the exact k* shell fall as N⁻¹ whereas the stripe nematic threshold tends to a nonzero constant Θ_M^lim > 0 at fixed lattice size, time window, and morphology criterion.

3. A quantitative separation result. The ratio Θ_M^c(N)/Θ_G,k*^c(N) grows proportionally to N; at N = 10⁴ the morphology threshold is reported as about 54 times the NPT threshold.

4. Covariance structure beyond pairwise reductions. A scanned Bloch sector retains nonzero full-sector negativity after the pairwise shell witnesses cross zero, and broken translation symmetry produces nonzero covariance asymmetry even at zero bath occupation.

Main Findings

• Witness thresholds on the exact shell: n_NPT,k*^c ≃ 1.02 and n_P,k*^c ≃ 1.04, where NPT is the negative-partial-transpose witness and P is the second-moment Glauber–Sudarshan witness. These values are unchanged to numerical precision across commensurate lattices L = 60, 120, 180, 240, and 300.
• Threshold formula: Θ_G,k*^c(N) = (2n_G,k*^{c,(0)} + 1)/N + O(N⁻²), so witness crossings sit at bath occupations of order unity and their thresholds in physical noise fall as N⁻¹.
• Morphology threshold: The long-time trajectory ensemble gives Θ_M^c ≃ 1.64 × 10⁻², while direct integration of the first-moment equation at fixed Θ gives Θ_M^lim ≃ 1.60 × 10⁻².
• Stripes outlive the witnesses: Visible, resolved Turing stripes can remain after both Gaussian witness margins on the same k* mode cross zero, and the interval in which this happens widens with N.
• Multimode sector persists: In the scanned L = 60 sector with |n_y| = 29 (|k_y| = 29π/30) and n_x mod 5 = 0, containing 24 traveling modes, the full-sector negativity remains nonzero beyond the pairwise shell crossings; the shaded quantity is the full-sector result minus the sum of the tested opposite-momentum reductions.
• Translation asymmetry: At zero bath occupation, after removing the k_y = 0 sector containing the nearly neutral translation mode, U_SB is 0.34 for the principal stripe and 0.47 for the horizontal stripe. The k_y = 0 blocks of the principal stripe account for about 96% of Tr V₀, and the uniform comparison states are translation invariant.
• Robustness checks: Including covariance feedback shifts the exact-shell crossings by less than 0.3%; the stripe crossing stays below all uniform comparison states whose linearized drift eigenvalues have negative real parts. Results hold for longer trajectories, the orthogonal stripe branch, and morphology cutoffs from 0.68 to 0.76.
• Weak pinning control: At the one percent boundary max{ε_pin, |δA₁|, |δA₂|} = 0.01, giving h_pin^{1%} ≃ 2.3 × 10⁻⁴; a weak period-12 field fixes the registry without changing the sign of the localized 2k* response.
• Atlas robustness to preparation: Eight preparations are compared; subsampling the same arrays changes the extracted contour by less than two percent and fewer than one percent of resolved cells.

Methodology in Plain English

The author builds a periodic square lattice of bosonic modes governed by a Lindblad equation with one-photon gain and loss, two-photon loss, and parametric/dissipative bonds. Transport is made unequal for the two quadratures (D_qμ = K_μ − G_μ, D_pμ = K_μ + G_μ), nonlinear loss saturates the finite-wavelength instability, and a cubic Hamiltonian breaks parity.

The calculation is split into two layers:

• First moments (the pattern): The mean field is treated by a deterministic drift equation with a scaled amplitude ⟨a_r⟩ = √N (q_r + i p_r)/√2. The parameter N controls local inverse noise and is independent of the number of lattice sites. A linear stability analysis produces a finite-wavenumber wedge in (λ, ρ_D), and ρ_D = 3 + 2√3 selects k* = π/6 with period 12.
• Second moments (the nonclassicality): Fluctuations about a stationary stripe obey V̇ = JV + VJᵀ + D₀ + n_th D_T, solved at fixed registry. Two witnesses are computed on opposite-momentum pairs: w_NPT = 1/2 − ν̃₋^PT (positive certifies Gaussian entanglement for the tested bipartition) and w_P = −λ_min(V_pair − I₄/2) (positive detects second-moment nonclassicality in the Glauber–Sudarshan sense).

Morphology is quantified with the stripe nematic moment Φ₂^(S) on a shell defined by ||k| − k*| ≤ 0.14, together with shell concentration C_shell, real-space contrast σ_R, signed area imbalance, normalized shell entropy, Bragg peak width, and Bragg concentration. Atlas cells are displayed only when σ_R ≥ 0.015 and C_shell ≥ 0.10; the morphology crossover is located at ⟨|Φ₂^(S)|⟩ = 0.72.

To compare momentum-space structure against a null model, the author constructs uniform comparison states from 64 amplitude-matched phases, of which 38 have Hurwitz-stable Jacobians; the stripe crossing is standardized against this family (Z_ref), displayed on a linear scale from −3 to 0.25 with zero mapped to white. Thermal trajectories use a first-order implicit-explicit Euler–Maruyama scheme with Δt = 0.1; the L = 96, N = 10⁴, T = 640 ensemble averages over 320 < t ≤ 640 at Θ = 0.014, 0.015, 0.016, 0.018, 0.020 with six paired counter-based Philox streams, while the limiting first-moment equation uses L = 48, T = 320, 160 < t ≤ 320, and nine noise values from 0.014 to 0.018.

Why This Matters

Impact on research: The paper separates two things that are often conflated — macroscopic spatial order and Gaussian nonclassicality. It shows that a witness crossing zero on the ordering mode does not mean the order has vanished, and that the separation widens with N. It also reports nonclassicality lying outside pairwise reductions (a full Bloch sector retaining negativity after pair witnesses vanish) and covariance asymmetry from broken translations at zero bath occupation, which affects how entanglement is quantified from momentum-resolved data. Practically, the paper argues that imaging or Bragg measurements and covariance reconstruction near k* probe these sectors separately.

Real-world applications (as framed by the paper):

• Driven photonic arrays, where the separation between first-moment order and second-moment nonclassicality could guide which observable is used for characterization.
• Polariton fluids, where Turing order has been observed in a coherent polariton fluid.
• Cold exciton systems, where Turing order has been predicted.
• Other bosonic lattices whose first moments carry order while second moments carry the tested nonclassicality.

Industry relevance: The results are relevant to platforms that engineer lattices of nonlinear resonators and read them out optically or via covariance tomography — photonic and polaritonic hardware, and quantum simulation of pattern-forming dynamics. The paper reports no device-level benchmarks, fidelities, or hardware demonstrations, so industrial applicability is currently at the level of measurement-protocol guidance rather than a demonstrated technology.

Future Directions

• Experimental test of the separation: The paper proposes that imaging or Bragg measurements combined with momentum-resolved covariance reconstruction near k* can distinguish macroscopic order from the Gaussian correlations studied. Whether this separation is observable in a real photonic array or polariton fluid is untested here.
• Extending quantitative control beyond stripes: The stable stripe species is the only case described as quantitatively controlled; spots, holes, labyrinths, and defects appear in the atlas but the threshold separation is established for stripes.
• Generality of the scaling laws: Whether Θ_G ∝ N⁻¹ versus Θ_M → Θ_M^lim > 0 holds for other Lindblad generators, other lattice geometries, and other morphology criteria is left open, as is the behavior when lattice size and time window are not held fixed.
• Role of the multimode sector: The full-sector negativity surviving beyond pairwise crossings raises the question of how multimode nonclassicality should be quantified and used, and how translation-symmetry breaking at zero bath occupation affects such measures.

Target Audience

Researchers in open quantum systems, quantum optics, and driven-dissipative many-body physics; theorists working on pattern formation and Turing instabilities outside the classical reaction–diffusion setting; and experimentalists in photonic, polaritonic, or excitonic lattices who need to know which measurement — real-space imaging, Bragg scattering, or covariance reconstruction — probes which property. The paper is not beginner-friendly: readers should be comfortable with Lindblad master equations, Gaussian covariance formalism, and symplectic eigenvalue criteria for entanglement.

Authors’ abstract

We explore quantum Turing pattern zoology, where the same Lindblad equation supports a morphology atlas of stripes, spots, holes, labyrinths, and defects. The stable stripe species provides a quantitatively controlled case in which morphology and Gaussian witness loss separate parametrically. In particular, visible Turing stripes can remain after two Gaussian witness margins associated with the same $k_*$ mode cross zero in a completely positive Lindblad lattice. The witness thresholds on the exact shell fall as $\mathcal{N}^{-1}$. The stripe nematic threshold tends to a nonzero value at fixed lattice size, time window, and morphology criterion. The ratio of the morphology threshold to either witness threshold therefore grows with $\mathcal{N}$. Imaging and momentum-resolved covariance measurements probe these sectors separately.