Emulating non-Markovian system-bath dynamics with parametrically driven cavities

AI summary

Overview

Research area: Quantum simulation of open quantum systems — specifically, engineering controllable bosonic baths with tunable spectral density using parametrically driven electromagnetic cavities (including fiber-loop cavities operating with classical light).

Technical level: Intermediate. The paper combines cavity QED / quantum optics, tight-binding lattice models in synthetic dimensions, and open-system theory (spin-boson and Wigner-Weisskopf models). The core idea is intuitive, but the derivations use interaction-picture transformations, Bloch modes, and spectral-density arguments.

Scope (1 sentence): The paper proposes and characterizes a scheme in which the periodic waveform of a parametric cavity drive directly programs the spectral density of an emulated bosonic bath, and analyzes — with a fiber-loop implementation in mind — whether the resulting device can reproduce the sharp sub-ohmic-to-super-ohmic "localization" transition of the spin-boson / Wigner-Weisskopf model.

What This Paper Is About

Most treatments of open quantum systems assume a memoryless (Markovian) environment, but real environments often retain memory, producing revivals, altered thermodynamics, and hard-to-correct errors. The authors ask whether a single parametrically driven cavity can act as a modular, highly tunable "bath module" whose spectral density $J(\omega)\propto\omega^{s}$ can be continuously tuned in the exponent $s$, and whether such a device can reproduce a genuine non-Markovian effect — a sharp transition from partial to complete decay of a two-level system — that is only exact in the thermodynamic limit and at weak system-bath coupling.

Key Contributions

1. A waveform-to-spectral-density mapping. The authors show that a periodically modulated cavity realizes an effective one-dimensional long-range tight-binding model in a synthetic dimension (the undriven cavity modes), where the hopping amplitudes are the Fourier components of the drive. The time-averaged dispersion $\varepsilon_k$ is fixed by the drive waveform, and since $J(\omega)=\pi D(\omega)$ with $D(\varepsilon)\propto|d\varepsilon(k)/dk|^{-1}$, the bath spectral density is programmed directly by the shape of $v(\tau)$.

2. An explicit design for a power-law bath. For a target $J(\omega)\propto\omega^{s}$ with cutoff $\omega_c$, they derive the required dispersion $\varepsilon_s(k)=\omega_c(|k|/\pi)^{1/(s+1)}$ for $-\pi\le k\le\pi$, give the hopping amplitudes $t_{sl}$ as a Fourier integral of that dispersion, and analyze the "filtered" dispersion obtained by truncating to $N_F$ Fourier coefficients (illustrated for $N_F=100$).

3. A concrete experimental blueprint and readout protocol. They characterize a fiber-loop cavity implementation and show that the density of states can be verified directly through a simple transmission experiment: the time-averaged output intensity $\bar I/I_0$ maps onto the density of states, reducing to $\bar I/I_0\simeq 1-\pi\kappa D(\Delta\omega)$ when $D$ varies slowly on the scale of the cavity linewidth $\gamma$.

4. An analysis of the limits of the emulation. Section V is dedicated to quantifying systematic error from a finite number of cavity modes $N$ and finite drive bandwidth $N_F$ when approximating an infinite continuum bath, and the paper highlights a competition between the weak-coupling limit and the thermodynamic limit that governs how closely the localization transition can be approximated.

Main Findings

• The bath spectral density is set by the drive waveform. In the simplest case where the drive amplitude is independent of the frequency difference between modes, $v_l(\tau)=v(\tau)$, the dispersion relation is exactly the drive waveform evaluated at $\tau=k/\Omega$: $\varepsilon(k)=v(\tau=k/\Omega)$. Long-range hopping in the synthetic dimension, unlike short-range tight-binding models (which give $D(\varepsilon)\propto\varepsilon^{d/2-1}$ for $d\in{1,2,3}$), admits a near-arbitrary dispersion and therefore a continuously tunable low-energy exponent $s$.

• A genuine sharp transition is targeted. For $J(\omega)\propto\omega^s$, the Wigner-Weisskopf atom decays completely for ohmic or sub-ohmic baths ($s\le 1$) and only partially for super-ohmic baths ($s>1$). This discontinuity in the long-time excited-state probability is the signature the scheme is designed to reproduce.

• Uniform coupling follows from the Bloch-mode structure. Because the system couples only to the central mode and the bath modes are plane waves, the effective coupling is $g_k=1/\sqrt{N}$, independent of $k$, so that $J(\omega)=(\pi/N)\sum_k\delta(\omega-\varepsilon_k)\equiv\pi D(\omega)$.

• Non-stationary corrections are controllable. The driven bath produces a two-time correlation function $c(\tau,\tau')$ with an extra phase $\phi_k(\tau)$; this phase vanishes stroboscopically at $\tau=nT$, and under the rotating-wave condition $|\varepsilon_k|\ll\Omega$ one has $|\phi_k(\tau)-\phi_k(\tau')|\ll 1$, justifying neglect of the non-stationary phase and recovery of a time-local $c(\tau)=\frac{1}{N}\sum_k e^{-i\varepsilon_k\tau}$.

• Large mode and coefficient counts are available. Truncating to $N=2M+1$ modes, the authors note $N>10^{4}$ is achievable for optical fiber loop cavities and $N_F\gg 1$ is achievable with a high-quality waveform generator, which they argue gives an advantage over platforms limited to a few bosonic modes.

• Positioning against prior work. Trapped-ion realizations have achieved $s=\pm 1$ (proposal with $\sim 50$ ions), spin-boson dynamics with up to 20 ions but without fine control of $s$, and $s=0.5, 1.0, 2.0$ with only three motional modes (limited frequency range). Earlier cold-atom work focused on a Markovian-to-non-Markovian crossover via an energy offset to a continuum band, and a prior driven-resonator proposal targeted giant-atom physics rather than spectral-density engineering.

• Numerical results for the fiber-loop parameters: The provided text is truncated within Section IV.2, so the specific simulated decay curves, error magnitudes, and the quantitative bounds on systematic error advertised in the abstract are not reported in the available content.

Methodology in Plain English

The authors start from the standard system-bath Hamiltonian — an arbitrary system coupled with strength $\delta$ to a collective bath operator $B=\sum_n g_n b_n$ — and then build an artificial version of that bath out of the modes of one driven cavity.

The trick is that periodic modulation of the cavity at the free spectral range $\Omega$ couples modes that differ in frequency by multiples of $\Omega$. Viewed in a rotating frame, those modes become sites on a one-dimensional lattice (a "synthetic dimension"), and the modulation strengths become hopping amplitudes between sites. Crucially, the modulation can connect distant sites, so the hopping is long-range: $t_l = v_l^l$, the Fourier coefficient of the drive at the $l$-th harmonic. Averaging over one drive period gives a hopping model whose band structure $\varepsilon_k=\sum_l t_l e^{-ilk}$ is whatever the experimenter chooses.

Because the system couples equally to all these modes, the bath spectral density is just proportional to the density of states, which in turn is inversely proportional to the slope of the dispersion. So by sculpting the band structure — equivalently, by sculpting the drive waveform — one sculpts the bath. To get a power law $J\propto\omega^s$, one needs the dispersion $\varepsilon_s(k)=\omega_c(|k|/\pi)^{1/(s+1)}$, whose Fourier transform gives the required drive harmonics; truncating at $N_F$ harmonics gives an approximate ("filtered") version.

To check the device, they propose shining a laser through a weakly coupled transmission line and recording the time-resolved output. The transmission dip traces out the dispersion, and the period-averaged intensity traces out the density of states — a direct, in-situ calibration that is possible precisely because the bath is not traced out as it would be in a real open quantum system.

Finally, they argue that the classical light amplitudes in a fiber-loop cavity obey the same linear equations as the quantum probability amplitudes of a two-level atom coupled to a bath, which is what allows a classical optical experiment to emulate Wigner-Weisskopf decay.

Why This Matters

Non-Markovian open-system dynamics are notoriously hard to compute: exact numerics become intractable as the bath grows, yet truncating the bath destroys the very long-time behavior one wants to study. A hardware platform whose bath spectral density is a continuously tunable dial and whose bath is directly observable offers a way around that bottleneck, and it does so with a modular "bath module" that can be attached to qubits, anharmonic modes, or other cavities.

Potential applications (as identified in the paper):

• Quantum thermodynamics, where history-dependent (non-Markovian) dynamics change the rules.
• Thermalization of many-body systems.
• Resource-efficient quantum simulations of open quantum systems.
• Quantum memory and enhanced qubit readout, both of which exploit non-Markovian memory effects.
• Understanding non-Markovian error mechanisms relevant to fault-tolerant quantum computing.

Industry relevance: The scheme is compatible with existing photonic hardware (fiber loops, waveform generators, high-bandwidth detectors), making it attractive for photonic quantum-computing and quantum-simulation platforms that need calibrated, reproducible noise and bath models. The authors explicitly frame the near-term value as "quantum mimetics" — borrowing useful quantum effects into a classical analogue system to inspire new functionality — rather than as universal quantum simulation.

Future Directions

• Quantitative error budgets. The abstract promises bounds on systematic error for arbitrary system observables; extending these from bounds to practical design rules for choosing $N$ and $N_F$ for a given target observable is a natural next step (the details lie in Section V, beyond the provided excerpt).
• Experimental realization. The fiber-loop proposal is argued to be feasible with "currently achievable parameters," so demonstrating the transmission-based calibration and then the decay transition experimentally is the obvious follow-up.
• Coupling to richer systems. The bath module is described as modular; attaching it to qubits, anharmonic oscillators, or multiple cavities would test how far the analogy extends beyond the single two-level emitter case.
• Accessing classically intractable regimes. The authors suggest that, if the criteria in the paper are met, the device could probe non-Markovian dynamics that current numerical methods cannot reach — a claim that requires benchmarking against exact results at the boundary of tractability.

Target Audience

Quantum-optics and superconducting-circuit experimentalists interested in bath engineering; theorists working on non-Markovian open quantum systems and the spin-boson / Caldeira-Leggett / Wigner-Weisskopf models; photonic quantum-simulation researchers using fiber loops or synthetic dimensions; and graduate students who want a worked example of how a Hamiltonian's band structure translates into a physical spectral density.

Authors’ abstract

We introduce and characterize a scheme for emulating non-Markovian quantum system-bath dynamics using the discrete electromagnetic field modes of a parametrically driven cavity. In this scheme, the character of a bosonic bath (the form of the bath spectral density) is tied directly to the shape of the real-time waveform describing periodic parametric cavity modulation. As an explicit example, we demonstrate the feasibility of this scheme for a fiber-loop experiment with currently achievable parameters, supported by numerical simulations and analytical estimates. In particular, we show that a localization transition of the spin-boson model can be accurately emulated in this fiber-loop cavity system. This localization effect is characterized by a sharp transition from partial decay to complete decay for a two-level system coupled to a bosonic bath as the bath spectral density $J(ω)\propto ω^{s}$ is continously tuned from the sub-ohmic ($s<1$) to the super-ohmic ($s>1$) regime. The transition is only exactly realized in the thermodynamic limit (for an infinite number of bath modes) and for sufficiently weak system-bath coupling. We highlight a competition between the weak-coupling and thermodynamic limits in this problem and show that challenges in approximately realizing the transition can nevertheless be overcome. Finally, we provide bounds on the systematic error introduced when emulating non-Markovian dynamics for arbitrary system observables. The scheme presented here can enable the realization of a modular platform for engineering custom non-Markovian baths, with potential applications in quantum thermodynamics, thermalization of many-body systems, and resource-efficient quantum simulations of open quantum systems.