Skip to content

Electric Noise Spectra of a Near-Surface Nitrogen-Vacancy Center in Diamond with a Protective Layer

MetadataDetails
Publication Date2018-12-21
JournalPhysical Review Applied
AuthorsPhilip Chrostoski, H. R. Sadeghpour, D. H. Santamore
InstitutionsDelaware State University, Center for Astrophysics Harvard & Smithsonian
Citations22
AnalysisFull AI Review Included
  • Core Challenge: Surface electric field fluctuations in shallow nitrogen-vacancy (NV) diamond devices degrade sensitivity, shorten T2 coherence times, and broaden fluorescent lines at room temperature.
  • Methodology: Application of the quantum fluctuation-dissipation theorem to evaluate charge noise spectra for six different dielectric protective coatings modeled as parallel-plate capacitors in series.
  • Key Finding #1: Four critical parameters govern noise performance: effective relaxation time (τeff), effective loss tangent (tan φeff(ω)), power law exponent (a), and protective layer thickness (d).
  • Key Finding #2: Liquid overlayers outperform solids in the 103-107 Hz band due to their picosecond-to-nanosecond relaxation times, while solids dominate above 107 Hz owing to steeper power law decay (a ≈ −2).
  • Key Finding #3: An optimal coating would be a thin (<10 nm) liquid layer with low real permittivity, a sub-nanosecond relaxation time, and minimal lossy behavior across the target operational frequency.
  • Material Suitability: Propylene carbonate (PC) and dimethyl sulfoxide (DMSO) yield lowest noise floors at low frequencies because their loss tangents fall below diamond’s at sub-105 Hz regimes.
  • Strategic Value: Material choice must be matched to operational frequency band; no single universal coating satisfies all sensing regimes.

ParameterValueUnitContext
Diamond permittivity (static) εs7.99—Bare substrate dielectric constant
Diamond permittivity (infinite freq.) ε∞5.7—High-frequency dielectric constant
Diamond Young’s modulus Y443GPaMechanical stiffness
Diamond relaxation time τdia2480nsCharacteristic dipole reorientation
Diamond power law exponents (γ, β)0.97 / 0.89—Havriliak-Negami shape factors
Glycerol εs / ε∞ / Y / τ51.8 / 3.9 / 4.80 / 19.9— / — / GPa / nsLiquid overcoat parameter set
Propylene Carbonate (PC) εs / ε∞ / Y / τ64.9 / 4.7 / 8.09 / 0.0433— / — / GPa / nsBest low-frequency candidate
Dimethyl Sulfoxide (DMSO) εs / ε∞ / Y / τ45.9 / 6.36 / 4.50 / 11.0— / — / GPa / nsShort relaxation, low noise floor
PMMA εs / ε∞ / Y / τ3.60 / 2.6 / 5.00 / 500— / — / GPa / nsSolid polymer overcoat
PVDF εs / ε∞ / Y / τ7.50 / 5.0 / 1.10 / 100,000— / — / GPa / nsSolid polymer overcoat
PFPE εs / ε∞ / Y / τ104 / 5.0 / 34.5 / 7.00— / — / GPa / nsLiquid fluorinated overcoat
Temperature of operation298 (room temp)KThermal equilibrium condition
Frequency range modeled103 to 107HzWhite noise to power-law transition
Optimal layer thickness dL< 10nmBelow this threshold, noise reduces
High-frequency crossover> 107HzSolid coatings dominate above this point
Effective capacitance modelCeff = series composite—Diamond + cover layer treated as capacitors in series
Critical noise equationS(ω) ∝ kBT tan φeff(ω) / [ω Ceff(ω)]—Fluctuation-dissipation result
Coating Materiala (d = 5 nm)a (d = 1 µm)
PC−1.6−0.91
Glycerol−0.5−0.15
DMSO−1.9−1.1
PFPE−1.6−0.2
PMMA−2.2−2.1
PVDF−2.0−1.94

  1. Theoretical Framework: Applied the quantum fluctuation-dissipation theorem treating noise as thermally-activated dipole fluctuations at room temperature.
  2. Capacitor Model: Treated diamond and surface coating as two parallel-plate capacitors in series, with the NV center dipole acting at the interface (z-axis normal).
  3. Dielectric Function: Implemented the Havriliak-Negami relaxation model with complex permittivity ε*(ω) = ε∞ + (εs − ε∞) / [1 + (iωτ)1−β]γ.
  4. Effective Loss Tangent: Combined diamond and coating loss tangents using series-weighting governed by acoustic impedance (Young’s modulus ratio) and layer thickness.
  5. Protection Layer Sweep: Evaluated six candidate materials (glycerol, PC, DMSO, PMMA, PVDF, PFPE) at two thicknesses (5 nm and 1 µm).
  6. Spectral Analysis: Calculated noise spectra S(ω) across 103 to 107 Hz, identifying the transition frequency ft = 2π/τeff between white noise and 1/fa regime.
  7. Sensitivity Mapping: Cross-referenced the four governing parameters (τeff, tan φeff(ω), a, d) against operational frequency band to determine best-fit coating.

To replicate or extend this research, the following 6CCVD materials are most directly applicable:

Application RequirementRecommended 6CCVD Material
Shallow NV center sensing layerOptical Grade SCD (polished Ra < 1 nm)
Solid-state spin register prototypingHigh-Purity SCD with controlled nitrogen delta-doping
Cost-effective ensemble sensorIsotopically-Enriched PCD
Charge-noise characterization across broad frequenciesBDD thin films as reference/control dielectric

For the sub-10 nm dielectric top-layer requirements described in the paper, 6CCVD’s single crystal diamond (SCD) substrates — available from 0.1 µm to 500 µm thickness — provide the bulk dielectric foundation whose permittivity (εs = 7.99) was used throughout the calculations. Our SCD plates can be supplied at thicknesses matching the 5 nm top-diamond layer modeled in the series-capacitor geometry, giving researchers a precisely defined εdia baseline.

The protective overlayers evaluated in this study (PMMA, PVDF, PFPE, glycerol, PC, DMSO) require either deposition atop the diamond surface or integration into a device stack. When your research progresses toward a packaged sensor:

  • Custom Metallization: 6CCVD offers Ti, Pt, Au, Pd, W, and Cu metallization for bonding pads, microwave striplines, or ground planes used in NV ODMR (Optically Detected Magnetic Resonance) experiments.
  • Laser Cutting & Dicing: Custom-shaped SCD/PCD chips up to 125 mm (PCD) — ideal when integrating diamond dies into coplanar waveguide resonators or scanning probe tips.
  • Thickness Engineering: Need an NV-center depth of exactly 5 nm below the surface to match the modeled series capacitance? Our SCD plates from 0.1 µm upward can be engineered with sub-micron depth precision.
  • Surface Finishing: Ra < 1 nm polish on SCD ensures the dielectric interface modeled in Eq. 8 is physically realized, eliminating surface roughness as a noise contributor the model does not account for.

Reproducing noise-spectrum results requires a diamond substrate whose bulk dielectric properties (εs, ε∞, Y, τ) closely match literature values used in theoretical modeling. 6CCVD’s in-house PhD engineering team can:

  • Material Selection: Recommend the optimal SCD, PCD, or BDD grade for your specific NV magnetometry, NMR, or thermometry project.
  • Dimensional Matching: Confirm that your chosen cover-layer thickness is compatible with our supply range (e.g., 5 nm top layer achievable only via etching of thicker stock).
  • Noise Floor Baseline: Provide measurement-ready surfaces with documented Ra and roughness data, ensuring that experimental deviation from theoretical predictions is attributable to your coating — not substrate variability.

Practical Insights from the Paper Relevant to Sourcing

Section titled “Practical Insights from the Paper Relevant to Sourcing”
  • Frequency-Dependent Material Choice: The paper shows that no single coating is universally optimal. For below 105 Hz sensing (e.g., DC magnetometry), liquid overlayers (PC, DMSO) on SCD outperform solids; for above 107 Hz (e.g., high-frequency ESR/NMR), PMMA or PVDF on SCD are preferred.
  • Thickness Constraint: Substrates thinner than 10 nm are not practical for handling, so the modeled “5 nm diamond top layer” is more reasonably implemented as a 5 nm implanted NV depth in a bulk SCD wafer — a configuration where 6CCVD’s nitrogen-implanted SCD plates become directly applicable.
  • Optical Transparency Requirement: All coatings analyzed are optically transparent at the NV zero-phonon line (637 nm). 6CCVD’s optical-grade SCD ensures that the underlying substrate’s optical clarity preserves this critical readout path.

Call to Action: For custom specifications, material consultation, or to discuss how 6CCVD’s MPCVD diamond can support your NV center noise-reduction research, visit 6ccvd.com or contact our engineering team directly.

View Original Abstract

Nitrogen-vacancy center diamonds are important for quantum sensors, due to their ultrasensitivity to a magnetic field and manipulability at room temperature. A major hindrance of device sensitivity is electric field noise. This work theoretically investigates the effect of a surface covering on N-$V$-center diamonds on reducing such noise. Six covering layers currently used in experiments are studied. The results show that a covering layer is useful for decreasing surface noise, but the most suitable material depends on the device’s operational frequency range.