Diamond photodiodes for x-ray application
At a Glance
Section titled “At a Glance”| Metadata | Details |
|---|---|
| Publication Date | 2024-02-05 |
| Journal | OSTI OAI (U.S. Department of Energy Office of Scientific and Technical Information) |
| Authors | J. R. Distel, J. Smedley, Jeffrey W. Keister, Erik Müller, Jean Jordan - Sweet |
| Institutions | Case Western Reserve University, A.S. Watson (Netherlands) |
| Analysis | Full AI Review Included |
Executive Summary
Section titled “Executive Summary”- Application: Development of single-crystal diamond (SCD) photodiodes for X-ray detection, calibrated at synchrotron beamlines (NSLS).
- Material Tested: “Detector grade” SCD (Element Six), 4×4 mm plates, 0.25-0.5 mm thick, [001] surface, < 4 nm roughness.
- Performance Achieved: Linear response across 10-11 orders of magnitude of incident flux (100 pW to 1 W) and a broad energy range (0.2-28 keV).
- Key Electrical Parameter: Mean pair creation energy (w) of 13.3 ± 0.5 eV, yielding predictable responsivity modeling.
- Contact Engineering: Three contact regimes evaluated—oxygen-terminated (blocking), carbide-forming (Ti, Mo, Pt annealed), and non-carbide forming.
- Thermal Finding: Oxygen termination is lost above 600 °C, converting blocking contacts to injecting (photoconductive gain) contacts—triggered by temperature, not radiation.
- Process Innovation: In-situ XRD monitoring of contact formation, defining transition temperatures (e.g., Mo₂C forms at 835 °C; Pt grain growth > 400 °C).
Technical Specifications
Section titled “Technical Specifications”| Parameter | Value | Unit | Context |
|---|---|---|---|
| Diamond Material | Detector-grade SCD | — | [001] surface, [110] edge normal |
| Plate Dimensions | 4 × 4 | mm | Standard sample footprint |
| Plate Thickness | 0.25 - 0.5 | mm | Two thickness regimes evaluated (260 µm and 300 µm) |
| Surface Roughness | < 4 | nm | As-received from manufacturer |
| Mean Pair Creation Energy (w) | 13.3 ± 0.5 | eV | Derived from responsivity model |
| Photon Energy Range | 0.2 - 28 | keV | Across NSLS beamlines U3C, X6B, X8A, X15A, X28C |
| Flux Range | 10⁻¹⁰ - 1 | W | Linear dynamic range |
| Linear Dynamic Range | 10 - 11 | orders of magnitude | Validated for blocking and annealed Pt contacts |
| Bias Field (Hole Collection) | 0.1 | MV/m | Full hole collection threshold (≈ 25 V across 0.25 mm) |
| Bias Voltage (Typical) | 200 | V | Used for spatial maps at 19 keV |
| Pulsed Bias Duty Cycle | 95 - 99 | % | At 1 kHz, for soft X-rays < 1 keV |
| Oxygen Loss Temperature | 600 | °C | Threshold where blocking contacts become injecting |
| Mo₂C Transition Temp | 835 | °C | Ramp rate 3 °C/s in He ambient |
| Pt Grain Growth Onset | 400 | °C | [111] fiber texture, grain size 50 Å → 200 Å |
| Pt Contact Thickness | 30 - 40 | nm | Sputtered, blocking vs. annealed comparison |
| Ti/Pt Contact Thickness | ~30 (Ti) + capping | nm | Carbide-forming contact stack |
| Expected Response (19 keV) | 2.2 | µA/W | Reference value for spatial map normalization |
| Beam Size (Spatial Map) | 0.2 × 0.2 | mm² | Used at X15A |
| Carbon K-edge Energy | 284 | eV | Causes responsivity dip due to surface recombination |
| Pt M-edge Energy | 2120 | eV | Causes absorption feature in responsivity curve |
| Charge Collection Incomplete Depth | < 1 | µm | Near metallized layer for sub-keV photons |
Key Methodologies
Section titled “Key Methodologies”- Surface Preparation
- Chromic acid etch to remove non-diamond carbon.
- Alternative: Ozone lamp exposure (several hours) for oxygen termination.
- Verified via NEXAFS oxygen-edge analysis.
- Metallization
- Sputter deposition (~30 nm) through a 3 mm-diameter shadow mask on both sides.
- Metals evaluated: Cu, Pt, Nb, Mo, Ti (with Pt capping layer).
- Thermal Annealing Protocol
- He-filled oven with in-situ XRD at beamline X20C.
- Ramp rate: 3 °C/s.
- Monitored transitions: Mo → Mo₂C (835 °C); Pt grain coarsening (> 400 °C).
- Calibration Setup
- Five NSLS beamlines used for cross-validation (U3C, X6B, X8A, X15A, X28C).
- Reference detectors: Si photodiode (low flux/soft X-ray), ion chamber (mid flux), Cu calorimeter (high flux).
- Vacuum environment for soft X-ray lines; dry N₂ for high-flux to suppress ozone.
- Electrical Readout
- Pulsed and DC bias modes (square wave: duty cycle, frequency, amplitude, polarity).
- Current measurement via Keithley 617 and 6517 electrometers.
- Radiation Hardness Check
- Focused white beam at X28C (100 W/cm², 5-15 keV) for 1 min.
- Post-irradiation NEXAFS showed no oxygen termination loss—confirming thermal (not radiation) origin of contact change.
- Responsivity Modeling
- Equation: S = (1/w) × [CE / (1 − e^(−Adia·tdia) · e^(−Ameta·tmeta))] · derived from pair creation energy, thicknesses, and absorption lengths.
- Empirical fit to Monte Carlo results used for carbon K-edge region.
6CCVD Solutions & Capabilities
Section titled “6CCVD Solutions & Capabilities”This paper demonstrates that detector-grade SCD with carefully engineered metallization can deliver calibrated, linear X-ray detection across an exceptional dynamic range. 6CCVD offers the precision manufacturing and engineering depth to support such demanding detector programs.
Applicable Materials
- Optical-Grade SCD (single crystal diamond): Directly applicable. The paper’s [001]-oriented, 4×4 mm, 0.25-0.5 mm detector-grade SCD is well within our SCD thickness range (0.1 µm - 500 µm) and aligns with our capability for custom dimensions. Our optical-grade SCD achieves Ra < 1 nm surface roughness, exceeding the < 4 nm requirement reported here, which is advantageous for minimizing surface recombination losses near the carbon K-edge.
Customization Potential
- Metallization: 6CCVD provides in-house metallization with Au, Pt, Pd, Ti, W, and Cu—all metals tested in this study are within our standard stack options. Ti/Pt bilayer (carbide-forming + capping) and 30-40 nm Pt contacts are directly reproducible with our sputtering processes.
- Contact Geometry: The 3 mm-diameter masked contacts and alternative geometries (interdigitated, guard-ring, or full-area patterns) can be fabricated to spec.
- Laser Cutting: For non-standard footprints or detector array geometries, our laser cutting service enables custom shaping of SCD plates without compromising the crystal structure.
Engineering Support
- Surface Termination Control: Achieving and verifying oxygen termination is critical for blocking-contact behavior. 6CCVD’s in-house PhD engineers can advise on acid-etching vs. ozone-UV termination protocols and downstream thermal budget constraints (the 600 °C oxygen-loss threshold identified here).
- High-Purity Material Selection: For synchrotron and radiometric applications where low defect density is essential to avoid localized photoconductive gain (as seen with annealed Pt on strained regions), 6CCVD can recommend material grades optimized for low inclusion density and strain.
- Thermal Management: For high-flux environments (analogous to X28C’s 100 W/cm² conditions), our engineering team can assist with substrate integration, metallization stack selection, and contact stability assessments for elevated-temperature operation.
Call to Action
For custom specifications, metallization stack consultation, or material selection for your X-ray detector program, visit 6ccvd.com or contact our engineering team directly.
View Original Abstract
Single crystal high purity CVD diamonds have been metallized and calibrated as photodiodes at the National Synchrotron Light Source (NSLS). Current mode responsivity measurements have been made over a wide range (0.2-28 keV) of photon energies across several beamlines. Linear response has been achieved over ten orders of magnitude of incident flux, along with uniform spatial response. A simple model of responsivity has been used to describe the results, yielding a value of 13.3 {+-} 0.5 eV for the mean pair creation energy. The responsivity vs. photon energy data show a dip for photon energies near the carbon edge (284 eV), indicating incomplete charge collection for carriers created less than one micron from the metallized layer.
Tech Support
Section titled “Tech Support”Original Source
Section titled “Original Source”- DOI: None