Analytical solution of boundary time crystals via the superspin basis

Boundary time crystals (BTCs) in dissipative collective spin systems have been extensively studied using numerical, mean-field, and perturbative approaches. However, an explicit Liouvillian description governing the long-time dynamics deep within the time crystal phase has remained elusive. Here, we derive an effective Liouvillian that analytically captures the extreme BTC regime, where dissipation is parametrically weak and oscillatory order is maximally robust. By introducing a superspin representation of Liouville space, we obtain closed-form expressions for the Liouvillian eigenvalues to first order in the dissipation strength, providing direct access to decay rates, oscillation frequencies, and their thermodynamic scaling. Applying this framework to the canonical BTC model we analytically recover spontaneous breaking of continuous time-translation symmetry and persistent oscillations in the thermodynamic limit. In contrast, we show that other dissipative spin models exhibit only single-frequency oscillatory dynamics and therefore do not support genuine BTC phases. Our results establish a controlled analytical framework for the long-time dynamics in the extreme BTC regime.

💡 Research Summary

The paper tackles the long‑standing problem of obtaining an explicit Liouvillian description for the deep‑time‑crystal regime of boundary time crystals (BTCs) in dissipative collective‑spin systems. While previous works have relied on numerical simulations, mean‑field theory, or weak‑coupling perturbation that required numerical diagonalisation of tridiagonal matrices, the authors introduce a superspin representation of Liouville space that diagonalises the Liouvillian analytically to first order in the dissipation strength Γ.

Starting from the canonical BTC model – a collective spin of size N/2 with Hamiltonian Hₛ = −NΩₓJₓ and a single Lindblad jump operator J₋ at rate NΓ – the authors rewrite the Lindblad superoperator L = L₀ + L_D in a doubled Hilbert space. They define the superspin operators S_α = J_α⊗𝟙 − 𝟙⊗J_αᵀ (α = x,y,z) and note that the coherent part L₀ acts only through Sₓ, giving eigenvalues 2iΩₓsₓ where sₓ = mₓ − mₓ′. The superspin magnitude S² has eigenvalues (2/N)² s(s+1) with s ranging from 0 to N, mirroring ordinary angular‑momentum addition.

Crucially, the dissipative part L_D simplifies in this superspin basis: after discarding terms that change sₓ, it reduces to −(NΓ/4)(Sₓ² + S²). Thus L_D is already diagonal in each degenerate subspace labelled by (s, sₓ). The first‑order correction to the eigenvalues is therefore

λ_{s,sₓ} = 2iΩₓ sₓ − (Γ/N)(sₓ² + s(s+1)).

These closed‑form expressions reproduce the qualitative features identified in earlier perturbative studies (purely imaginary eigenvalues in the thermodynamic limit, closing of the Liouvillian gap) but now provide quantitative access to decay rates, oscillation frequencies, and their scaling with N. In the N → ∞ limit, sectors with small s/N become densely packed near the imaginary axis; the density of sectors diverges as 1/(2Γ \bar{s}), leading to a macroscopic accumulation of modes with vanishing real parts. Consequently, a highly degenerate nonequilibrium steady‑state subspace emerges, signalling spontaneous breaking of continuous time‑translation symmetry and the appearance of persistent, multifrequency oscillations – the hallmark of a BTC.

The authors also explore how increasing Γ drives the spectrum toward the real axis. Complex‑conjugate pairs coalesce at exceptional points, first affecting the least‑damped modes. This explains the fragility of the BTC phase against stronger dissipation and indicates that higher‑order (second‑order) corrections are needed to capture the shift of imaginary parts for larger Γ.

To test the generality of the superspin framework, the paper applies the same analysis to alternative collective‑spin models. Model A (Hₛ = −NΩ_zJ_z with a J₊ jump) cannot be expressed purely in superspin form; its spectrum, obtained in the uncoupled basis, shows only a single oscillation frequency. Models B and C, previously claimed to host BTCs, also exhibit only single‑frequency dynamics despite possessing gapless Liouvillian spectra. The authors therefore argue that the presence of purely imaginary eigenvalues alone does not guarantee a genuine BTC; a multifrequency structure arising from a dense set of superspin sectors is essential.

Numerical benchmarks using QuTiP confirm that the first‑order analytical eigenvalues match exact diagonalisation for small Γ, validating the perturbative approach. The paper concludes that the superspin representation offers a controlled, analytically tractable method for describing the extreme BTC regime, opening avenues for studying more complex open quantum systems, including those with multiple jump operators or nonlinear dissipation, where traditional methods become intractable.