Driven-Dissipative Condensate, Cavity, and Circuit Matter

Driven quantum matter reaches states set by coherent forcing, interactions, loss, noise, and measurement rather than by minimizing a free energy. A sharp phase claim therefore needs an open-system limit, a steady-state or trajectory observable, and a scaling variable that distinguishes a true Liouvillian transition from long-lived switching in a finite device.

Required background. Lindblad field dynamics defines the generator and complete-positivity assumptions; quenches and relaxation supplies finite-time tests. Helpful background. Driven-dissipative steady states and criticality develops the general field-theory classification.

A driven Kerr resonator

In a frame rotating at the coherent drive frequency, a single nonlinear mode can be modeled by

$$H=-\Delta a^\dagger a+\frac{U}{2}a^{\dagger2}a^2+F(a+a^\dagger),$$

with one-photon loss

$$\dot\rho=\mathcal L\rho =-i[H,\rho]+\kappa\mathcal D[a]\rho, \qquad \mathcal D[a]\rho=a\rho a^\dagger-\frac12\{a^\dagger a,\rho\}.$$

Here $\Delta=\omega_d-\omega_c$ is the drive frequency minus the bare cavity frequency in this sign convention, $F$ is the calibrated drive amplitude, and $\kappa$ is an energy-decay rate. Factor-of-two mistakes arise when an amplitude linewidth is substituted for an energy linewidth.

Mean-field factorization with $\alpha=\langle a\rangle$ gives

$$\dot\alpha =\left[i(\Delta-U\lvert\alpha\rvert^2)-\frac\kappa2\right]\alpha-iF.$$

For a stationary occupation $n=\lvert\alpha\rvert^2$,

$$F^2=n\left[(\Delta-Un)^2+\left(\frac\kappa2\right)^2\right].$$

This cubic can have three solutions, two dynamically stable at mean-field level. It predicts an optical-bistability region, but a finite quantum resonator with a primitive Lindbladian commonly has a unique stationary density matrix. Its apparently bistable trajectories switch between metastable bright and dim states. Calling the finite-device histogram a pair of thermodynamic phases skips the essential limit.

The closing-gap and metastable-switching structure of the Kerr problem is analyzed by Casteels, Fazio, and Ciuti 2017.

Liouvillian spectrum and metastability

Let $\mathcal L R_j=\lambda_jR_j$, with $\lambda_0=0$ for a stationary state and $\operatorname{Re}\lambda_j\le0$. Define the asymptotic relaxation gap

$$\Delta_{\mathcal L} =-\max_{j\ne0}\operatorname{Re}\lambda_j.$$

A small gap produces long switching or relaxation times. A dissipative phase transition requires the gap to close in a controlled thermodynamic limit, together with nonanalytic steady-state behavior or diverging correlations. For a single Kerr mode, one useful large-occupation limit scales $U\to U/N$ and $F\to\sqrt N F$ while $N\to\infty$. For a cavity array, the spatial volume can instead supply the limit. These are different physical extrapolations and should not be mixed.

At finite $N$, a bimodal Wigner or photon-number distribution and hysteresis under a finite-rate sweep can be consequences of metastability. Slow the sweep, lengthen the dwell time, and compare with independently measured switching rates. True steady-state hysteresis vanishes in a finite ergodic system when the sweep is made asymptotically slow; a shrinking but very long-lived loop is dynamical evidence for a nearby first-order transition, not two stationary density matrices.

From internal fields to measured signals

Circuit and optical platforms usually measure an output field. For one monitored port,

$$a_{\mathrm{out}}(t)=a_{\mathrm{in}}(t)-\sqrt{\kappa_{\mathrm{ext}}}\,a(t),$$

so transmission includes interference between the incident coherent tone and the emitted cavity field. Inferring $n=\langle a^\dagger a\rangle$ from power alone requires the external coupling, gain chain, phase response, and added noise. The second-order correlation

$$g^{(2)}(0)= \frac{\langle a^{\dagger2}a^2\rangle}{\langle a^\dagger a\rangle^2}$$

can distinguish coherent, bunched, and antibunched regimes, but detector bandwidth and background subtraction must be applied to the theoretical correlator before comparison.

For condensates and cavity arrays, the same discipline extends to spatial correlations, structure factors, and coherence lengths. Mean-field criticality can be changed by fluctuations, dimensionality, conservation laws, and whether noise is Markovian. A finite correlation-length peak near a closed-system critical parameter is not automatically an open-system phase transition.

The corresponding driven-open Keldysh classification and its dimensional limitations are reviewed by Sieberer, Buchhold, and Diehl 2016.

Unconditional, conditional, and synchronized dynamics

The Lindblad density matrix averages over unobserved quantum jumps. A monitored trajectory follows a stochastic conditional state and can display switching, purification, or oscillations absent from the unconditional average. Postselection can amplify rare records. Every claim must identify which object is used:

Object	Evolution	Appropriate claim
Unconditional $\rho$	Trace-preserving master equation	Steady state, Liouvillian gap, averaged correlations
Monitored trajectory	Stochastic update conditioned on a record	Switching statistics, trajectory transition, feedback response
No-jump state	Non-Hermitian conditional evolution	Survival-conditioned dynamics only

Synchronization similarly requires more than a spectral peak. One should measure phase locking across subsystems, its robustness interval, noise-induced phase diffusion, and scaling with the number of modes. A dark state requires verification that the proposed jump operators annihilate it and that Hamiltonian terms do not leak it into bright sectors.

Experiments in circuit-QED arrays, open atom–cavity matter, and photonic fluids have observed metastability, switching, hysteresis, and critical slowing compatible with dissipative transitions Fitzpatrick et al. 2017, Klinder et al. 2015, Li et al. 2022. Their interpretation is strongest when a device-size or occupation scaling is paired with calibrated output correlations and when mean-field multistability is not presented as the exact finite-system steady-state spectrum.

Evidence status

The platform evidence and methods summarized here were checked through 10 August 2026. Driven-dissipative critical phenomena are established across several architectures, but the thermodynamic variable, Markov approximation, and trajectory-versus-ensemble meaning remain platform specific. Dated calibrations, null tests, and superseding results belong in the Quantum Matter and Emergence Research dossier.

Exercises

1. Mean-field response. Derive the cubic stationary equation for the Kerr resonator from the equation for $\alpha$.

Solution

Setting $\dot\alpha=0$ gives $iF=[i(\Delta-Un)-\kappa/2]\alpha$. Taking the absolute square gives $F^2=[(\Delta-Un)^2+(\kappa/2)^2]n$. Multiple positive roots are possible, but their dynamical stability must be checked by linearizing the complex equation for $\alpha$ and $\alpha^*$.

2. Switching and the gap. A two-state metastable model switches bright-to-dim at rate $r_{bd}$ and dim-to-bright at rate $r_{db}$. Find its nonzero relaxation eigenvalue.

Solution

The probability generator is $M=\begin{pmatrix}-r_{bd}&r_{db}\\r_{bd}&-r_{db}\end{pmatrix}$. Its eigenvalues are $0$ and $-(r_{bd}+r_{db})$. Thus the slow Liouvillian scale inferred from the reduced dynamics is $\Delta_{\mathcal L}=r_{bd}+r_{db}$, and long dwell times correspond to a small gap.

References

• Casteels, Wim, Rosario Fazio, and Cristiano Ciuti. “Critical Dynamical Properties of a First-Order Dissipative Phase Transition.” Physical Review A 95, 012128 (2017). DOI.
• Fitzpatrick, Mattias, Neereja M. Sundaresan, Andy C. Y. Li, Jens Koch, and Andrew A. Houck. “Observation of a Dissipative Phase Transition in a One-Dimensional Circuit QED Lattice.” Physical Review X 7, 011016 (2017). DOI.
• Klinder, Jens, Hans Keßler, Matthias Wolke, Ludwig Mathey, and Andreas Hemmerich. “Dynamical Phase Transition in the Open Dicke Model.” Proceedings of the National Academy of Sciences 112, 3290–3295 (2015). DOI.
• Li, Zejian, Ferdinand Claude, Thomas Boulier, Elisabeth Giacobino, Quentin Glorieux, Alberto Bramati, and Cristiano Ciuti. “Dissipative Phase Transition with Driving-Controlled Spatial Dimension and Diffusive Boundary Conditions.” Physical Review Letters 128, 093601 (2022). DOI.
• Sieberer, Lukas M., Michael Buchhold, and Sebastian Diehl. “Keldysh Field Theory for Driven Open Quantum Systems.” Reports on Progress in Physics 79, 096001 (2016). DOI.