Skip to content

Witnessing nonstationary and non-Markovian environments with a quantum sensor

MetadataDetails
Publication Date2026-01-02
JournalPhysical Review Applied
AuthorsJohn W. Rosenberg, Martín Kuffer, Inbar Zohar, Rainer Stöhr, Andrej Denisenko
InstitutionsWeizmann Institute of Science, Balseiro Institute
Citations1
AnalysisFull AI Review Included

Witnessing Non-Stationary and Non-Markovian Environments with a Quantum Sensor — Technical Brief

Section titled “Witnessing Non-Stationary and Non-Markovian Environments with a Quantum Sensor — Technical Brief”
  • Platform: Single nitrogen-vacancy (NV) centers in chemical-vapor-deposition (CVD) diamond membranes, hosted on nanopillar arrays for enhanced photon collection.
  • Static Bias Field: B0 ≈ 298 G aligned with the N→V axis to maximize ODMR splitting (0↔−1 transition near 2.87 GHz − γ·B0).
  • Detection Scheme: Ramsey interferometry with 100 noise realizations per time point × 4·105 repetitions to sample the full decoherence distribution.
  • Modeling Framework: Langevin-type stochastic differential equation mn″ + Γn′ + κn = η(t), with an effective inertia term m governing Markovian (m = 0, Ornstein-Uhlenbeck) versus non-Markovian (m ≠ 0, underdamped oscillator) regimes.
  • Key Discriminators: Distinct short-time scaling exponents χ(t) — t2 (stationary Markovian), t3 (quenched Markovian), and t5 (quenched non-Markovian) — plus collapse-and-revival signatures at t = 2πm/Ω in stationary non-Markovian baths.
  • Validated Capabilities: Differentiation of stationary vs. non-stationary noise, Markovian vs. non-Markovian dynamics, quench timing detection (fit error < ±100 ns), and identification of abrupt correlation-time switching.
  • Practical Implication: Combining stationary + quenched measurements resolves parameter degeneracies inherent in single-regime spectroscopy.
ParameterValueUnitContext
Diamond typeSingle NV in nanopillar—CVD-grown diamond membrane
Static field B0298GAligned to N→V axis
ZFS2.87GHz|0⟩↔|−1⟩ reference
Rabi contrast34.1%Single NV characterization
π-pulse duration344nsRabi driving at 1.45 MHz
Native T2* (no injection)1.65µsDrift-corrected Ramsey
Native T2* (detuned)3.52µs1 MHz detuning
Coupling check oscillation3.6MHz0.45 V DC injection
Decay rate (control)0.608µs−1NV inherent decoherence
Laser wavelength520nmInitiation + readout
Laser power (init)2 (typ.)mWPre-objective
Read window220-400µsAfter green pulse
Clock resolution Δt1-4nsOPX pulser
Noise std. dev. Δ0.10VEquilibrium Markovian set
Correlation times tc1.25, 2.5, 5, 10µsMarkovian swept set
Restoring freq. ω05, 6, 7MHzNon-Markovian sweep
Damping 2tc−10.025-0.20MHzNon-Markovian sweep
Driving strength A/m20.054V2/µs3Non-Markovian fixed
Effective freq. Ω√(ω02 − tc−2)MHzRevival spacing
Iteration loops (per point)4·105shotsInnermost Ramsey loop
Realizations per point100—Outermost noise loop
  • Sample Preparation: Single NV center in a CVD diamond membrane fabricated with an array of nanopillars for fluorescence enhancement.
  • Optical Initialization & Readout: 520 nm green laser (3 µs init / 1.6 µs read) through a confocal microscope; photon counting via dichroic mirror + beam-splitter to two avalanche photodiodes (300 ps resolution).
  • Static Field Alignment: Permanent magnet positioned to maximize ODMR splitting, confirming axial alignment with the NV center.
  • Microwave Control: MW pulses generated by mixing a local oscillator with an AWG signal, delivered via a copper wire above the diamond membrane.
  • Synthetic Noise Injection: Arbitrary waveforms (Lorentzian-approximated white noise) generated by a Quantum Machines OPX and applied through a spiral antenna beneath the diamond, with a central hole for optical access.
  • Ramsey Sequence:
    • π/2 pulse → evolution time τ with injected noise → −π/2 pulse → fluorescence readout.
    • Normalized signal computed from dual-phase (±π/2) readouts.
  • Drift Correction: Every 30 minutes, a Ramsey measurement with 1 MHz detuning and zero injection is performed; the MW frequency is adjusted to compensate drift.
  • Loop Hierarchy: 4·105 shots × 10 noise realizations × 100 time points × 10 outer-loop repetitions = 100 realizations per time point.
  • Quench Protocols:
    • Pure Quench: Initial conditions n(0) = 0 (Markovian) or n(0) = n′(0) = 0 (non-Markovian).
    • Delayed Quench: Noise sequence extended by td before discarding initial samples.
    • Correlation-Time Switch: Abrupt change in tc from ta → tb at switching time ts.
  • Fitting Procedure: Analytical expressions for χ(t) (Appendix B) used to extract Δ, tc, ω0, with td or ts as free parameters where applicable.

This protocol hinges on a diamond host whose spin-bath contribution is minimal and well-characterized — a regime where 6CCVD’s engineered MPCVD material delivers a decisive advantage.

  • Applicable Materials:
    • Optical-Grade Single-Crystal Diamond (SCD) membranes with ultra-low nitrogen concentration (< 5 ppb) directly mirror the high-T2* substrate used here (T2* = 1.65 µs baseline, 3.52 µs detuned). Such material preserves the long intrinsic coherence required to isolate injected statistical signatures from native bath noise.
    • Custom Polycrystalline Diamond (PCD) wafers (up to 125 mm) are ideal for scalable nanopillar array fabrication, supporting high-density NV ensemble or single-NV device batches.
    • Engineered P1 Spin-Bath Diamond can be provided when intentional, controlled decoherence channels are needed to benchmark the noise-spectroscopy framework under realistic bath conditions.
  • Customization Potential:
    • Dimensional Control: Diamond membranes or pillars can be cut to user-defined geometries by laser micromachining, accommodating spiral-antenna housings, microwave striplines, or confocal access windows.
    • Metalization: Ti / Pt / Au, W, or Cu contact stacks can be deposited in-house for integration with microwave delivery structures and spiral antennas.
    • Surface Engineering: Sub-nanometer polishing (Ra < 1 nm for SCD; < 5 nm for inch-size PCD) supports shallow NV implantation protocols for high-sensitivity near-surface sensing.
    • Doping Tailoring: Both nitrogen-vacancy-friendly SCD and controlled P1 bath densities are available for systematic study of stationary vs. non-stationary decoherence regimes.
  • Engineering Support:
    • 6CCVD’s in-house PhD team can assist with material selection — matching diamond grade, defect density, and geometry to specific quantum-sensing protocols (Ramsey, Hahn echo, dynamical decoupling).
    • Consultation available for customers seeking to replicate the noise-witnessing framework demonstrated here, including recommendations on T2* benchmarks, NV yield optimization, and membrane thinning for nanopillar processing.
    • For applications requiring combined Ramsey + dynamical decoupling sequences or advanced pulse protocols (e.g., Uhrig, concatenated DD), 6CCVD can tailor substrate parameters to maximize protocol efficacy.
  • Call to Action: For custom specifications, material consultation, or to discuss how 6CCVD diamond can accelerate your quantum-sensing research, visit 6ccvd.com or contact our engineering team directly.
View Original Abstract

Quantum sensors offer exceptional sensitivity to nanoscale magnetic fluctuations, where non-stationary effects — such as spin diffusion — and non-Markovian dynamics arising from coupling to few environmental degrees of freedom play critical roles. Because fully reconstructing the microscopic structure of realistic spin baths is often infeasible, a practical challenge is to identify the dynamical features that are actually encoded in the sensor’s decoherence signal. Here, we demonstrate how quantum sensors can operationally characterize the statistical nature of environmental noise, distinguishing between stationary and non-stationary behaviors, as well as Markovian and non-Markovian dynamics. Using nitrogen-vacancy (NV) centers in diamond as a platform, we develop a physical noise model that captures the essential dynamical features of realistic environments relevant to sensor observables — independently of the microscopic bath details — and provides analytical predictions for Ramsey decay across different regimes. These predictions are experimentally validated through controlled noise injection with tunable correlation properties. Our results showcase the capability of quantum sensors to isolate and identify key dynamical properties of complex environments, without requiring full microscopic bath reconstruction. This work clarifies the operational signatures of non-stationarity and non-Markovian behavior at the nanoscale and lays the foundation for strategies that mitigates decoherence while exploiting environmental dynamics for enhanced quantum sensing.