Skip to content

Optimal control of a nitrogen-vacancy spin ensemble in diamond for sensing in the pulsed domain

MetadataDetails
Publication Date2022-07-05
JournalPhysical review. B./Physical review. B
AuthorsAndreas F. L. Poulsen, Joshua D. Clement, James L. Webb, Rasmus Høy Jensen, Luca Troise
InstitutionsTechnical University of Denmark
Citations20
AnalysisFull AI Review Included
  • Core Achievement: Demonstrated robust, coherent optimal control over a large, macroscopic ensemble of up to 4 x 109 Nitrogen-Vacancy (NV) centers in off-the-shelf diamond, overcoming challenges posed by spatial inhomogeneity.
  • Sensitivity Gain: Shaped microwave control pulses, designed using Floquet theory and optimal control, achieved an 11% enhancement in the Optically Detected Magnetic Resonance (ODMR) slope (C’) compared to the best conventional three-frequency flat pulse scheme.
  • Performance Benchmark: This enhancement translates to a 78% improvement in sensitivity compared to standard single-frequency flat (π-)pulses commonly used in coherent control literature.
  • Methodology Advancement: The optimal control algorithm explicitly included the effects of the 14N hyperfine interaction and ensemble inhomogeneity (detuning and Rabi frequency variation) to ensure simultaneous high-fidelity state transfer across all relevant spin transitions.
  • Readout Optimization: Dynamical modeling showed that reliable, hysteresis-free contrast measurements can be achieved using short laser reinitialization pulses (3 ms), effectively controlling the high-intensity core (≈ 25% or 1 billion NV centers) of the ensemble, thereby maximizing measurement bandwidth.
  • Practical Application: The successful operation at low Rabi frequencies (1.4 MHz) proves the viability of this technique for compact, low-power portable quantum sensor devices.
ParameterValueUnitContext
Diamond GradeOptical-grade (Element 6)N/AOff-the-shelf material used.
NV- Concentration~0.5ppbBulk concentration.
Diamond Dimensions5 x 5 x 1.2mm3Sample size.
T1 Coherence Time7.1msEnsemble average measurement.
T2 Coherence Time7.0ÂľsEnsemble average measurement.
T2* Coherence Time0.44ÂľsEnsemble average measurement.
Maximum Ensemble Rabi Frequency (Rmax)3.2MHzMaximum achievable experimental frequency.
Optimized Shaped Pulse Duration (tp)1.85ÂľsDuration of the best-performing pulse.
Microwave Carrier Frequency (ωD)≈ 2.8GHzNV ground state splitting (with bias field).
14N Hyperfine Splitting (δI)2.16MHzSeparation between hyperfine resonances.
Laser Wavelength532nmDPSS pump laser.
Maximum Laser Power500mWDelivered to the diamond.
Estimated Total Ensemble Size≈ 4 x 109NV centersBased on fluorescence emission volume.
Contributing Ensemble Size (Readout Core)≈ 1billionNV centers
Sensitivity Improvement (vs. 3-Freq Flat)11%Increase in ODMR slope C’.
Sensitivity Improvement (vs. Single-Freq Flat)78%Increase in ODMR slope C’.
Estimated Shot-Noise Sensitivity (η)≈ 10nT/sqrt(Hz)Sensitivity estimate for the current setup.
Minimum Reinitialization Time (tI)3msRequired for hysteresis-free contrast.
  1. Optimal Control Pulse Design: Shaped microwave pulses (defined by In-phase I(t) and Quadrature Q(t) components) were generated using smooth optimal control theory, maximizing the state transfer fidelity (Fst) from the bright state (|0>) to the dark state (|-1>).
  2. Floquet Theory Integration: The time-periodic Hamiltonian describing the system was solved using Floquet theory, allowing the explicit inclusion of the 14N hyperfine splitting (three levels) into the optimization algorithm.
  3. Ensemble Inhomogeneity Modeling: The optimization maximized the weighted average fidelity across a representative ensemble (12x12 defects) that modeled Gaussian distributions of frequency detuning (Δ) and relative control amplitude (â) inhomogeneity.
  4. Pulse Generation and Delivery: Control pulses were created by an Arbitrary Waveform Generator (AWG) performing IQ modulation of a 2.8 GHz RF signal. The signal was amplified and delivered to the diamond via a near-field square split-ring resonator antenna.
  5. Pulsed ODMR Sequence: A repeating sequence was used: laser initialization (tI), microwave control pulse (tp), and laser readout (tR). The minimum tI was set to 3 ms to ensure reliable reinitialization of the high-intensity core of the ensemble.
  6. Noise Reduction Readout: Fluorescence was collected by an Avalanche Photodiode (APD). Contrast (C) was measured using software lock-in detection against a reference photodetector (Vref) to reject common-mode DC and high-frequency laser technical noise.
  7. Performance Quantification: The sensitivity of the control pulse was quantified by measuring the maximum slope (C’) of the resulting ODMR spectrum, which is directly proportional to the sensor’s response strength.
  • DC and Low-Frequency Quantum Sensing: The demonstrated robust control is ideal for applications requiring high sensitivity in the DC to low-frequency range, such as measuring slowly varying magnetic fields or temperature changes where long coherence times are beneficial.
  • Biosensing and Medical Diagnostics: Diamond’s robustness and biocompatibility, combined with enhanced sensitivity, enable high-precision magnetic imaging of biological samples, including single-neuron action potential detection and real-time thermometry in living cells.
  • Nanoscale Magnetic Resonance: The technique provides improved coherent control for large ensembles, enhancing the signal-to-noise ratio for bulk nanoscale Nuclear Magnetic Resonance (NMR) and Electron Spin Resonance (ESR) experiments.
  • Portable Sensor Devices: The ability to achieve high performance using low Rabi frequencies (low microwave power) makes this methodology suitable for integration into compact, battery-operated quantum magnetometer systems.
  • Solid-State Quantum Computing/Control: The methodology for designing robust control pulses that compensate for inhomogeneous broadening is transferable to other solid-state quantum systems (e.g., defects in 2D materials) where coherent manipulation of large, imperfect ensembles is required.
View Original Abstract

Defects in solid state materials provide an ideal, robust platform for quantum sensing. To deliver maximum sensitivity, a large ensemble of non-interacting defects hosting coherent quantum states are required. Control of such an ensemble is challenging due to the spatial variation in both the defect energy levels and in any control field across a macroscopic sample. In this work we experimentally demonstrate that we can overcome these challenges using Floquet theory and optimal control optimization methods to efficiently and coherently control a large defect ensemble, suitable for sensing. We apply our methods experimentally to a spin ensemble of up to 4 $\times$ 10$^9$ nitrogen vacancy (NV) centers in diamond. By considering the physics of the system and explicitly including the hyperfine interaction in the optimization, we design shaped microwave control pulses that can outperform conventional ($\pi$-) pulses when applied to sensing of temperature or magnetic field, with a potential sensitivity improvement between 11 and 78%. Through dynamical modelling of the behaviour of the ensemble, we shed light on the physical behaviour of the ensemble system and propose new routes for further improvement.

  1. 2016 - Quantum Optics