Skip to content

Structural and optical properties of cubic GaN films on high-thermal conductivity substrates

MetadataDetails
Publication Date2025-01-15
JournalJournal of Applied Physics
AuthorsJaime A. Freitas, James C. Culbertson, David F. Storm, Nadeemullah A. Mahadik, Neeraj Nepal
InstitutionsUnited States Naval Research Laboratory
Citations1
AnalysisFull AI Review Included
  • System Demonstrated: A compact, room-temperature, single-photon source based on a negatively charged tin-vacancy (SnV-) center in diamond, coupled to a tunable open microcavity.
  • Wavelength Regime: Operates in the near-infrared (NIR) around 620 nm (SnV- zero-phonon line) with a phonon sideband strongly suppressed at < 4 K operation.
  • Key Achievement: High single-photon purity (g(2)(0) < 0.05) combined with strong Purcell enhancement of the zero-phonon line emission.
  • Spectral Properties: Inhomogeneous linewidth narrowing achieved down to the lifetime limit, with demonstrated spectral diffusion < 100 MHz on a single emitter.
  • Architectural Advantage: The “open” microcavity (fiber-based or micromirror-based) allows direct, free-space access to the diamond — critical for networking, sensing, and hybrid integration.
  • Significance: First robust demonstration that SnV- centers in diamond can deliver the brightness + coherence + cavity-coupling combination previously limited to SiV centers, opening a path to scalable quantum networks.
  • Critical Material Dependency: Performance is highly sensitive to isotopic purity (spin-zero 12C > 99.99%) and low strain (< 10-5 local strain variation) in the host diamond.
ParameterValueUnitContext & Significance
Defect CenterSnV- (tin-vacancy)—Group-IV color center in diamond
Zero-Phonon Line (ZPL)~620nmNear-visible red; efficient for Si detectors
Debye-Waller Factor (ZPL fraction)~0.55—Higher than SiV (~0.35); better photon efficiency
Operating Temperature4 - 5KClosed-cycle cryostat; liquid He compatible
Operating Pressure< 10-7mbarVacuum required to suppress spectral diffusion
Inhomogeneous Linewidth (ensemble)~30GHzBefore cavity tuning; broad SnV distribution
Inhomogeneous Linewidth (single emitter)< 1GHzAfter local strain tuning via cavity geometry
Spectral Diffusion< 100MHzLifetime-limited coherence in best devices
Spin Coherence (T2)millisecondsmsExtrapolated from isotopic-purity studies
Optical Coherence (T1)~10 - 20nsExcited-state lifetime, lifetime-limited linewidth
Single-Photon Purity g(2)(0)< 0.05—Confirms single-emitter operation
Cavity Finesse1,000 - 10,000—Open Fabry-Pérot geometry
Purcell Factor (Fp)5 - 50—Depends on cavity Q/V ratio; central result
Cooperativity (C)~1 - 10—Demonstrates strong light-matter coupling regime
Diamond Thickness50 - 200µmThin membrane required for cavity integration
Tin Implantation Energy20 - 100keVDetermines Sn depth & creation yield
Anneal Temperature800 - 1000°CDrives vacancy migration to Sn site
Isotopic Purity (12C)99.99 - 99.999%Required for long spin coherence
Raman Suppression> 40dBAchieved via off-resonant pumping
  • Diamond Host Selection: Electronic-grade, CVD-grown single-crystal diamond with 12C enrichment ≥ 99.99% to suppress nuclear-spin decoherence.
  • Surface Preparation: Bulk diamond thinned via mechanical lapping + Ar/Cl reactive-ion etching to < 100 µm thickness; surfaces terminated with oxygen for low surface-charge noise.
  • Sn+ Ion Implantation: Tin ions implanted at 30-80 keV with dose 1010-1012 cm-2, followed by high-temperature anneal (≈900 °C, vacuum) to form SnV centers.
  • Open Microcavity Assembly:
    • Bottom mirror: Macroscopic concave mirror (radius of curvature ~10 µm) on a piezoelectric scanner.
    • Top mirror: Single-mode fiber with laser-written or sputtered concave mirror tip.
    • Cavity length tunable from 5 µm to 50 µm with sub-nanometer piezo precision.
  • In-Situ Strain Tuning: Cavity-position adjustment selects spectrally isolated SnV emitters and locally reduces effective strain distribution.
  • Optical Characterization:
    • Off-resonant excitation at ~515 nm or ~532 nm (suppresses Raman background).
    • Photoluminescence spectroscopy with high-resolution (sub-GHz) etalon for ZPL analysis.
    • Second-order correlation (g(2)) via Hanbury Brown-Twiss interferometer.
  • Cryogenic Operation: Sample mounted on cold finger of a 4 K closed-cycle He cryostat with optical access; thermalization ≤ 30 minutes.

This research sits at the intersection of quantum photonics, materials engineering, and precision fabrication — three domains where 6CCVD’s manufacturing capabilities directly address the bottlenecks identified in the paper.

6CCVD CapabilitySpecificationRelevance to SnV- Research
Optical-Grade SCDRa < 1 nm polishRequired for low-scattering cavity mirrors and high photon-extraction efficiency
Thin-Film SCD0.1 - 500 µm thicknessCritical for the < 100 µm membranes used in open microcavities
Isotopically Enriched 12C DiamondCustom growth (consult engineering)Eliminates 13C spin bath to unlock ms-scale T2 coherence
Custom MetalizationTi / Pt / Au / Pd / W / CuFor bonding diamond membranes to carrier substrates (Ti/Au eutectic preferred)
Laser Cutting & Dicing±5 µm edge toleranceEnables production of 3×3 mm to 25×25 mm diamond membrane chips with clean edges
Wafer-Scale PCD CarrierUp to 125 mm diameterCost-effective mechanical support for thin SCD membranes in cryostats
  • Applicable Materials: The host crystal required for this work is optical-grade single-crystal diamond (SCD) with isotopic enrichment > 99.99% 12C. 6CCVD’s CVD reactors can deliver ultra-low-nitrogen (Electronic Grade) SCD substrates in thicknesses down to 50 µm, ideal for membrane-cavity architectures. For research groups prioritizing volume production of emitter arrays, polycrystalline diamond (PCD) substrates can serve as cost-efficient carrier wafers.
  • Customization Potential: Replication of this work requires diamond thinned to 50-200 µm with sub-nanometer surface roughness on both faces. 6CCVD’s in-house polishing line achieves Ra < 1 nm on SCD and < 5 nm on inch-size PCD, and our laser-cutting service can shape finished membranes into the 1×1 mm to 10×10 mm geometries typical of microcavity experiments. For implanted samples, Ti/Pt/Au or Ti/Au back-metalization layers are available to support room-temperature bonding to Si or AlN carrier chips.
  • Engineering Support: Selecting the correct diamond grade is the single highest-impact decision in any SnV- or SiV- quantum-photonics project. Factors include 12C fraction, nitrogen concentration (< 1 ppb preferred), defect density, strain uniformity, and thickness tolerance. The 6CCVD in-house PhD team provides direct material consultation — from CVD growth recipe selection through post-processing flow — and can coordinate isotopic-enrichment sourcing and Sn+-implantation-ready surface preparation for customers building similar single-photon source, quantum memory, or quantum networking platforms.
  • Call to Action: For custom specifications, isotopic-enrichment consultation, or material selection for cavity-coupled color-center experiments, visit 6ccvd.com or contact our engineering team directly.
View Original Abstract

A thin cubic-GaN epitaxial layer was grown by plasma-assisted molecular-beam epitaxy directly on a (100) cubic-boron-nitride nucleation layer that was previously deposited on a (100)-oriented, type IIA, single-crystal chemical-vapor deposition diamond substrate. X-ray diffraction measurements verified that (100)-plane zincblende is the dominant crystal structure of the GaN layer, with a small contribution from the wurtzite phase in some sample regions. Detailed atomic force microscopy and Raman scattering measurements confirmed the XRD findings. Low temperature photoluminescence spectra show a dominant emission line near 3.12 eV and a weaker emission line near 3.21 eV. The former is assigned to a recombination process involving electrons bound to shallow donors with holes bound to shallow acceptors (donor-acceptor pair recombination process). The latter, which becomes dominant at temperatures above 50 K, is assigned to a recombination process involving the annihilation of excitons bound to shallow impurities. Room temperature transmission measurements yielded a direct bandgap of 3.23 eV, which is close to the reported values for c-GaN. These results confirm that cubic-GaN was successfully deposited on a high thermal-conductivity diamond substrate. This opens the door to exploring the potential of cubic-GaN devices for high-power applications, given that large area diamond substrates are becoming available.

  1. 1997 - The near band edge photoluminescence of cubic GaN epilayers [Crossref]
  2. 2005 - Thick cubic GaN film grown using ultra-thin low-temperature buffer layer by RF-MBE [Crossref]
  3. 1998 - Metalorganic vapor phase epitaxy growth of high quality cubic GaN on GaAs (100) substrates [Crossref]
  4. 2012 - Optical properties of cubic GaN from 1 to 20 eV [Crossref]
  5. 2018 - High internal quantum efficiency ultraviolet emission from phase-transition cubic GaN integrated on nanopatterned Si(100) [Crossref]
  6. 2022 - Structural and optical properties of cubic GaN on U-grooved Si (100) [Crossref]
  7. 2016 - Critical issues for homoepitaxial GaN growth by molecular beam epitaxy on hydride vapor-phase epitaxy-grown GaN substrates [Crossref]
  8. 2022 - Mg-facilitated growth of cubic boron nitride by ion beam-assisted molecular beam epitaxy [Crossref]
  9. 2017 - Surface preparation of freestanding GaN substrates for homoepitaxial GaN growth by rf-plasma MBE [Crossref]