Skip to content

Optical coherence of diamond nitrogen-vacancy centers formed by ion implantation and annealing

MetadataDetails
Publication Date2019-04-19
JournalPhysical review. B./Physical review. B
AuthorsSuzanne van Dam, Michael Walsh, Maarten Degen, Eric Bersin, Sara Mouradian
InstitutionsQuTech, Delft University of Technology
Citations140
AnalysisFull AI Review Included

Optical Coherence of Diamond Nitrogen-Vacancy Centers Formed by Ion Implantation and Annealing

Section titled “Optical Coherence of Diamond Nitrogen-Vacancy Centers Formed by Ion Implantation and Annealing”
  • Research Focus: Investigates the optical coherence (linewidth/stability) of Nitrogen-Vacancy (NV) centers in diamond when created via ion implantation versus natural growth defects.
  • Critical Finding: NV centers formed from implanted 15N atoms exhibit significantly broader optical transitions (> 1 GHz median linewidth) and higher local strain compared to naturally occurring 14NVs (< 500 MHz linewidth).
  • Root Cause Analysis: The data explicitly shows that poor optical coherence in implanted NVs is not an intrinsic isotope effect, but a direct result of residual lattice damage caused by ion bombardment.
  • Strain Correlation: 15NVs show 3-4× greater axial strain spread (up to 60 GHz) and higher transverse strain than native NVs, linking implantation damage to degraded optical properties.
  • Methodology: Used confocal fluorescence mapping, Optically Detected Magnetic Resonance (ODMR) for isotope identification, and Photoluminescence Excitation (PLE) spectroscopy at 4 K.
  • Engineering Implication: Standard nitrogen ion implantation + annealing recipes are insufficient for high-yield creation of coherent, narrow-linewidth NV centers required for quantum networking and photonic integration.
  • Path Forward: The paper suggests exploring 2D nitrogen-doped diamond layers, electron irradiation for vacancy creation, or laser-writing strategies to achieve precise NV positioning with minimal lattice disruption.
ParameterValueUnitContext
Diamond TypeBulk ⟨100⟩ CVD-Element 6, electronic grade
Membrane Thickness≈ 14µmSample B (Delft)
Implantation Ion15N+-Isotope tracing
Implantation Energy (Sample A)85keVMIT sample
Implantation Energy (Sample B)400keVDelft sample
Fluence (Sample A)109N/cm2Low dose
Fluence (Sample B)108N/cm2Low dose
Annealing Temperature (A)1200°CMax temp
Annealing Temperature (B)1100°CMax temp
NV ZPL Wavelength≈ 637nmZero-phonon line
ODMR Zero-Field Splitting2.88GHzNV identification
14NV Hyperfine (A)2.2MHzIsotope fingerprint
15NV Hyperfine (A)3.1MHzIsotope fingerprint
14NV Median Linewidth (A)0.33GHzCoherent NVs
15NV Median Linewidth (A)3.1GHzBroadened NVs
14NV Median Linewidth (B)0.14GHzCoherent NVs
15NV Median Linewidth (B)4.1GHzBroadened NVs
Narrowest 14NV FWHM64 ± 4MHzLow-strain native NV
15NV FWHM860 ± 236MHzImplanted NV
Axial Strain Spread (14NVs, A)13GHzStandard deviation
Axial Strain Spread (15NVs, A)44GHzStandard deviation
Transverse Strain Median (15NVs, A)5.4GHzEx/Ey splitting
Measurement Temperature≈ 4KCryogenic PLE
External Magnetic Field5-10GODMR bias
  • Substrate Selection: Bulk ⟨100⟩ CVD diamond (Element 6) prepared as bulk (Sample A) or thinned to ≈ 14 µm membrane (Sample B).
  • Ion Implantation: 15N+ ions implanted at 85 keV (109 N/cm2) or 400 keV (108 N/cm2) to track isotope-specific NV formation.
  • High-Temperature Annealing: Samples annealed at 1100 °C or 1200 °C in vacuum to mobilize vacancies (> 600 °C) and repair lattice damage.
  • Confocal Fluorescence Mapping: Automated spot recognition at the implantation depth (4 K, 532 nm excitation) to localize NV candidates.
  • Isotope Identification (ODMR): Continuous-wave Optically Detected Magnetic Resonance at 2.88 GHz distinguishes 14NVs (triplet, A = 2.2 MHz) from 15NVs (doublet, A = 3.1 MHz).
  • Photoluminescence Excitation (PLE): Resonant laser scanning near 637 nm to measure zero-phonon line (ZPL) linewidth with interleaved 532 nm repump pulses.
  • Spectral Diffusion Analysis: Repetitive ZPL scans with single off-resonant repump to separate intrinsic linewidth from laser-induced spectral diffusion.
  • Strain Characterization: ZPL absolute frequency maps axial strain; Ex/Ey splitting maps transverse strain.

Replicating or extending this quantum optics research requires ultra-low-nitrogen, high-purity diamond substrates to isolate the effects of implantation damage:

  • Optical Grade Single Crystal Diamond (SCD): The paper relies on Element 6 ⟨100⟩ CVD bulk diamond. 6CCVD provides equivalent and superior optical-grade SCD substrates with custom crystal orientations, thicknesses from 0.1 µm to 500 µm, and bulk substrates up to 10 mm.
  • Electronic Grade SCD Membranes: Sample B utilized a ≈ 14 µm thinned membrane for photonic integration. 6CCVD specializes in custom-thickness SCD membranes with surface roughness Ra < 1 nm, ideal for coupling to optical microcavities and nanophotonic structures.
  • Ultra-Low Nitrogen SCD: The baseline native 14N density in the diamond directly impacts the ratio of 14NV to 15NV centers observed. 6CCVD can provide SCD with controlled, ultra-low nitrogen concentrations to serve as the clean background substrate required for unambiguous isotope studies.

This research demanded precise implantation layers at specific depths and post-annealing steps. 6CCVD supports the full experimental workflow:

  • Custom Geometries: Whether the experiment requires bulk plates, thin membranes (≤ 14 µm), or specific die dimensions, 6CCVD delivers custom-cut SCD and PCD wafers up to 125 mm.
  • Laser Cutting Services: To prepare membrane samples or isolate specific implantation zones, 6CCVD offers precision laser cutting to customer-specified geometries without introducing mechanical damage.
  • Surface Preparation: For the cryogenic PLE measurements and ODMR mapping, surface quality is critical. 6CCVD provides super-polishing (Ra < 1 nm for SCD) to minimize surface-induced spectral diffusion.
  • Custom Metalization: For follow-up experiments integrating NVs with superconducting resonators, microwave striplines, or photonic waveguides, 6CCVD offers in-house deposition of Ti, W, Au, Pt, Pd, and Cu metal layers.

The paper highlights a fundamental bottleneck: ion-implanted NVs suffer from lattice damage that degrades optical coherence, limiting scalability for quantum networks. Addressing this challenge requires material engineering at the substrate level.

  • Material Selection Consultation: 6CCVD’s in-house PhD engineering team can advise on the optimal SCD substrate purity, crystal orientation, and post-processing for NV quantum optics experiments, quantum networking, and quantum sensing.
  • Alternative NV Creation Strategies: The authors suggest investigating 2D nitrogen-doped diamond layers and electron irradiation. 6CCVD can develop custom CVD-grown nitrogen-doped layers or supply Boron-Doped (BDD) PCD variants for novel defect-engineering approaches.
  • Scalable Quantum Platforms: For researchers moving from single-NV proof-of-concept to scalable arrays, 6CCVD provides consistent, high-quality MPCVD diamond tailored for photonic integration, cavity coupling, and ion-implantation protocols.

For custom specifications, high-purity SCD substrates, or material consultation for quantum optics and NV center research, visit 6ccvd.com or contact our engineering team directly.

View Original Abstract

Nitrogen-vacancy (NV) centers in diamond with coherent optical transitions are a promising candidate for quantum science and technology protocols. A widely used strategy for their creation is nitrogen-ion implantation and high-temperature annealing. The data presented in this work show that while low-strain, narrow-optical-linewidth NV centers are formed from naturally occurring ${}^{14}$N atoms, NV centers formed from the implanted ${}^{15}$N atoms show broadened optical transitions and experience higher strain. These results have immediate implications for the positioning accuracy of current NV center creation protocols.