X-ray-based analysis of quantum-grade homoepitaxial single-crystal diamond fabricated by MPCVD
Paper details
Section titled “Paper details”| Field | Value |
|---|---|
| Journal | Functional Diamond |
| Published | 2026-07-30 |
| Authors | Luke Chia Wei Min, Siu Hon Tsang, Apoorva Chaturvedi, Edwin Hang Tong Teo |
| DOI | 10.1080/26941112.2026.2705818 |
The lowest HRXRD (004) rocking-curve full width at half maximum (FWHM) compiled for homoepitaxial CVD single-crystal diamond is 0.0027° (9.7 arcsec). The review proposes a screening threshold below 20 arcsec (0.0056°), while warning that X-ray data certify the host lattice rather than NV spin coherence. A defensible acceptance process therefore combines structural measurements with EPR, FTIR, photoluminescence, ODMR, and device tests.
Four properties define the structural screen
Section titled “Four properties define the structural screen”The review organizes qualification around four properties: crystal orientation, bulk crystallinity, epilayer-substrate misalignment, and surface roughness. X-ray methods fit these requirements because they are non-contact, quantitative, and need little sample preparation. No single method resolves all four. The proposed workflow qualifies a seed before growth, monitors the MPCVD process, and rechecks the finished layer.
HRXRD provides the primary crystallinity metric
Section titled “HRXRD provides the primary crystallinity metric”During an HRXRD rocking-curve measurement, the sample rotates through the Bragg condition while the detector records diffracted intensity. The FWHM of the resulting peak is the principal crystallinity metric. A four-crystal monochromator reaches about 10 arcsec instrumental width, so slit settings and sample bending must accompany every FWHM. Using 5° rather than 2.5° Soller slits raises intensity by about 40% but broadens the peak.
Bauer et al. reported 0.0043° (15.5 arcsec) for freestanding CVD films, including a CO2-added sample. Their later off-axis study reached 0.0027° (9.7 arcsec), still two to three times the theoretical Darwin width. Commercial CVD substrates commonly give 30 to 100 arcsec, and 91 arcsec has been achieved on 2-inch CVD wafers. The review uses 20 arcsec as a proposed target, not a validated community standard.
Peak separation between epilayer and substrate estimates tilt. The highest-quality regions have recorded tilts below 1 arcsec, while typical CVD epilayer tilts are 1 to 10 arcsec.
Wavelength controls the sampled depth. Cu Kα is absorbed about 10.2 times more strongly than Mo Kα because absorption scales with wavelength cubed. A symmetric reflection probes through a 100 nm to 1 µm film, so its peak mixes film and substrate. Grazing-incidence or asymmetric geometry is needed to isolate such a layer.
Topography resolves dislocations that FWHM averages away
Section titled “Topography resolves dislocations that FWHM averages away”Synchrotron X-ray topography maps strain contrast from dislocations, stacking faults, and sector boundaries. Diamond’s Cu Kα absorption coefficient of about 0.17 cm⁻¹ permits transmission imaging through millimetre-thick crystals. Rocking-curve imaging adds local measurements, with about 1 µm spatial resolution and 0.001° angular resolution at synchrotron sources. Laboratory versions reach about 0.1 to 1 mm spatial resolution.
Imaging under multiple diffraction vectors identifies Burgers vectors through extinction. HPHT diamond has reached densities below 47 cm⁻². The review proposes below 10⁵ cm⁻² for quantum material and below 10³ cm⁻² for ideal material. State-of-the-art CVD diamond typically shows 10⁴ to 10⁵ cm⁻². Below about 10⁴ cm⁻², rocking-curve width is unreliable because bending, polishing damage, and instrument resolution dominate.
Reciprocal space mapping separates tilt from lattice-parameter strain. Earlier diamond measurements resolved about 10 ppm lattice-spacing variations and about 1 arcmin tilts within growth sectors. Unlike a one-dimensional rocking curve, the two-dimensional map identifies whether a shift comes from orientation or strain.
Orientation and roughness control NV suitability
Section titled “Orientation and roughness control NV suitability”Pole figures provide the full orientation distribution and quantify twinning. They are confirmatory for (001) single crystal and essential for (111) and (113) material. Reported preferential NV alignment reaches 94 to 99% on {111} and about 73% on {113}, compared with none on {100}. The review links this alignment to an ensemble magnetometer sensitivity gain of roughly four.
X-ray reflectometry measures roughness over a 0 to 2 nm range with about 0.1 nm precision. Shallow NV centres lie within 10 nm of the surface, so the review cites an upper limit of 0.2 to 0.5 nm rms and proposes 0.3 nm rms. AFM supplies local maps, while XRR also detects buried-interface roughness. For polishing damage, iterative HRXRD or GIXRD found that removing 6 µm was needed to suppress diffuse scattering wings.
MPCVD growth parameters leave measurable signatures
Section titled “MPCVD growth parameters leave measurable signatures”Substrate misorientation is the strongest reported growth lever. Studies of (001) material associate 3 to 5° off-axis angles with step-flow growth and lower FWHM. Step flow requires methane below 0.15%, unless misorientation exceeds 3 to 4°. For NV density control, the review describes 2 to 4° misorientation with methane below 0.5% as standard practice.
Nitrogen addition has two effects. It can double the growth rate and suppress hillocks, but it also increases vacancy-related defects and internal stress. The practical throughput window cited is 0.3 to 0.6% N2. Above 0.8 to 1.2% N2, growth changes from step flow to bidimensional nucleation. For non-NV layers, nitrogen below 1 ppb is targeted. Plasma-purified growth produced 0.007° FWHM with no detectable NV emission in photoluminescence.
Seed quality also propagates into the epilayer. CVD seeds produced about half the dislocation density of HPHT seeds after 1 mm of growth. Pre-growth etching of at least 5 to 6 µm removes the polishing-damaged layer that can nucleate interface dislocations.
A proposed qualification protocol
Section titled “A proposed qualification protocol”The review proposes the following gates. These are review recommendations, not yet community-validated specifications.
| Property | Primary X-ray method | Proposed threshold |
|---|---|---|
| Crystal orientation | HRXRD, four- or eight-azimuth protocol | Miscut within ±0.05° of target |
| Crystallinity | HRXRD (004) rocking curve | FWHM below 20 arcsec (0.0056°) |
| Dislocation density and type | Synchrotron XRT with Burgers-vector analysis | Below 10⁵ cm⁻² for quantum material; below 10³ cm⁻² for ideal material |
| Epilayer-substrate tilt | RSM or HRXRD peak-position map | Below 0.01° |
| Surface roughness | XRR | Below 0.3 nm rms |
| Polishing damage | Iterative HRXRD or GIXRD | No diffuse wing enhancement above background |
First-pass X-ray results certify crystallinity, defect topology, epilayer alignment, near-surface morphology, and macroscopic strain. Second-pass tests must establish ODMR linewidth, T2*, T2, T1, charge-state stability, photoluminescence contrast, charge collection, and device sensitivity. The authors recommend describing a sample as “X-ray-qualified for quantum-device fabrication” until the functional pass is complete.
Limits and the need for device data
Section titled “Limits and the need for device data”X-ray metrics also show why device data remain necessary. Balasubramanian et al. reported T2 = 1.8 ms at room temperature in 12C-enriched diamond with a paramagnetic defect concentration below 10¹³ cm⁻³. Herbschleb et al. later reported T2 = 2.4 ms in 99.998% 12C-enriched, phosphorus-doped n-type CVD diamond grown on (111) Ib substrates. A separate CVD series spanning 168 to 1 ppb NV⁻ had T2 values from 45.5 to 549 µs. None of these coherence values can be inferred from an FWHM alone.
Four gaps block a stronger claim. HRXRD averages square-millimetre regions, while individual NV centres respond to nanometre-scale strain. Laboratory instruments cannot reliably verify the lattice quality of delta-doped layers below 5 nm. Published (111) and (113) datasets remain sparse, and no complete four-property protocol has been reported for heteroepitaxial material.
The review also notes that the depth-to-FWHM relation for polishing damage remains qualitative. A controlled series linking dislocation density to T2* has not been published. Dark-field X-ray microscopy reaches about 200 nm resolution, while X-ray free-electron laser methods can image angstrom-scale surface topography. These methods address resolution gaps, but they do not replace calibrated ODMR, EPR, FTIR, photoluminescence, and device tests.
Materials and support
Section titled “Materials and support”NV quantum sensors use homoepitaxial MPCVD single-crystal diamond as the host lattice for nitrogen-vacancy centres and the shallow implantation surface. The paper requires orientation, crystallinity, tilt, defect, and roughness checks on both seed and finished layer before device fabrication. The 6CCVD diamond materials catalog lists MPCVD-grown single-crystal plates and wafers up to 125 mm, with thicknesses from 0.1 to 500 micrometers, while engineering support can define a custom qualification plan.
View Original Abstract
Homoepitaxial single-crystal diamond grown by MPCVD is increasingly recognised as the optimal substrate for NV centre-based quantum sensors. Establishing eligibility for quantum applications requires evaluating four properties that exceed standard semiconductor criteria: crystal orientation, bulk crystallinity, epilayer-substrate misalignment, and surface roughness. X-ray techniques are uniquely suited to this task, being non-destructive, quantitative, requiring minimal preparation, and spanning laboratory and synchrotron facilities bridging wafer-scale screening and device-region certification. This review critically surveys X-ray methods addressing all four properties: high-resolution X-ray diffraction (HRXRD) rocking curve analysis, synchrotron and laboratory topography, reciprocal space mapping (RSM), X-ray reflectometry (XRR), grazing incidence diffraction (GIXRD), pole figure analysis, and Bragg diffraction imaging. Reported HRXRD (004) rocking curve FWHM values reach 0.0027° (9.7 arcsec) for the best CVD films, approaching the theoretical Darwin width. Synchrotron topography enables Burgers vector identification and dislocation density mapping below 10⁵ cm-²; RSM separates lattice tilt from strain; XRR resolves the sub-nanometre roughness required for shallow NV implantation. We examine how MPCVD growth parameters like substrate misorientation, nitrogen addition, temperature, and methane concentration imprint on X-ray signatures, and establish quality thresholds linking X-ray metrics to NV coherence (T₂ up to 2.4 ms in ¹²C-enriched, phosphorus-doped n-type CVD diamond), distinguishing T₂ from T₂*. Critically, X-ray observables are necessary-but-not-sufficient structural proxies: they certify the host lattice, while spin-bath-limited T₂, zero-phonon-line homogeneity, charge-state stability, and charge-collection efficiency require EPR, SIMS/FTIR, photoluminescence, ODMR, and device-level metrology. We present a “ready reckoner” cross-mapping each X-ray observable to the metrics it can and cannot predict and recommend a standardised HRXRD/XRT-RCI/RSM/XRR protocol with defined pass/fail thresholds for community-wide quantum-grade qualification.
High Concentration Quantum Grade Single Crystal Diamond
View Specs →
Free Standing Boron Doped Diamond Pcd Film
View Specs →
Boron Doped Single Crystal Diamond Plate
View Specs →References we cover
Section titled “References we cover”- Effect of Nitrogen on Growth and Optical Properties of Single-Crystal Diamond Synthesized by Chemical Vapor Deposition
- Studies of Dislocations in Type Ib, Type IIa HPHT and CVD Single Crystal Diamonds
- Nitrogen related paramagnetic defects - Decoherence source of ensemble of NV− center
- Surface Morphology and Spectroscopic Features of Homoepitaxial Diamond Films Prepared by MWPACVD at High CH4 Concentrations
- Creation of nitrogen-vacancy centers in chemical vapor depositionn diamond for sensing applications