Radiation detection with polycrystalline CVD diamond films - effects of film thickness and irradiation on detector performance
Paper details
Section titled “Paper details”| Field | Value |
|---|---|
| Journal | Journal of Instrumentation |
| Published | 2026-02-01 |
| Authors | Bohumír Zaťko, M. Varga, Andrea Šagátová, Vladimír Kršjak, Ondrej Szabó |
| DOI | 10.1088/1748-0221/21/02/c02034 |
Polycrystalline CVD diamond films detected 5.5 MeV alpha particles at zero bias and were characterized after 4 MeV proton fluences up to 10¹⁶ p/cm². A 30 μm film with a 3 mm contact pad also detected 15.5 MeV fast neutrons, with an estimated detection efficiency of about 0.077%.
Film growth and detector construction
Section titled “Film growth and detector construction”Polycrystalline CVD diamond films were grown by microwave plasma-enhanced chemical vapor deposition on polished boron-doped p-type silicon. The substrates measured 1 × 1 cm² and 730 μm thick and had a (100) orientation. Before growth, each substrate received 40 min of ultrasonic agitation in a suspension of 5 nm nanodiamond powder. Growth used hydrogen at 500 sccm and methane at 25 sccm.
The chamber pressure was 120 mbar, microwave power was 5 kW, and average deposition temperature was approximately 950 ± 30 °C. The films ranged from 0.55 μm to 55 μm thick.
Each film received 5 nm Ti and 50 nm Au by evaporation. Circular top contacts measured 1, 1.5, 2, or 3 mm across, while the silicon rear side received a blanket Ti/Au layer. These films and contacts formed the pCVD detector structures on silicon used for all particle tests.
Alpha particle response
Section titled “Alpha particle response”Testing used 5.5 MeV alpha particles from an Am-241 source in vacuum. The detectors operated at zero bias because earlier testing found that applied bias increased current and noise without improving detection. For spectrometric measurements, the authors used a CAEN A1422 charge-sensitive preamplifier and a CAEN HEXAGON digital amplifier.
The S/N ratio was defined by the detection peak position relative to the noise level. As film thickness increased from 0.55 μm to 4.7 μm, the S/N ratio decreased. The S/N ratio also fell as contact diameter increased from 1 mm to 3 mm. Smaller pads were associated with higher S/N and lower capacitance.
The alpha-particle range in diamond was about 15 μm in earlier work by the authors, exceeding the films used in the thickness comparison. The authors did not therefore expect the lower peak position observed for thicker films. Roughness increased with thickness and was their leading explanation, but they stated that further analysis was required. They proposed lapping thicker films to improve detection properties.
Proton irradiation changes the material
Section titled “Proton irradiation changes the material”The degradation study used films 0.55, 2.7, and 55 μm thick. Samples were irradiated with 4 MeV protons at fluences from 10¹³ to 10¹⁶ p/cm². The 75 μm projected proton range exceeded every tested film thickness and was expected to damage the film volume uniformly.
SEM images of 2.7 μm films showed no noticeable morphology change after exposure to 10¹³ or 10¹⁶ p/cm². Raman spectra detected more non-diamond sp² carbon near 1600 cm⁻¹. The sp² signal increased in direct proportion to fluence, which the authors interpreted as mild graphitization of diamond grains.
PALS on 55 μm films separated annihilation events into defect-free lattice, vacancy-type, and larger void components. About 80% of positrons annihilated in diamond, while 20% penetrated the silicon substrate, which had only a slight impact on the results.
Mean positron lifetime fell from 246 ps before irradiation to 238 ps at 10¹⁶ p/cm², rather than rising as expected. One point at 10¹⁵ p/cm² departed from the monotonic trend and remains unexplained. The short defective component rose from about 208 ps to approximately 238 ps while its intensity fell from 70% to 64%.
The authors interpreted this as growth of existing vacancy clusters rather than formation of new clusters. They also suggested that fewer larger cavities or suppressed positronium formation could explain the third component’s decline at high fluence.
Irradiation changes detector response
Section titled “Irradiation changes detector response”Alpha spectra were compared before and after proton irradiation. For a 0.55 μm film with a 1 mm pad, 10¹³ and 10¹⁴ p/cm² shifted the peak strongly toward lower channels. The same shift occurred with 1.5 mm pads, and at 10¹⁵ p/cm² the peak fell below the noise limit.
With a 2 mm pad, only the 10¹³ p/cm² peak remained identifiable. A 2.7 μm film with a 1 mm pad became unusable after 10¹³ p/cm², because its initial S/N was lower. The initial S/N set the highest proton fluence for which a peak remained measurable in each test.
Prior work cited by the authors exposed pCVD detectors to 70 MeV, 800 MeV, and 24 GeV protons up to 2 × 10¹⁵ p/cm². That work reported about 50% signal relative to unirradiated samples. The authors warned that comparisons require the damage factor for each proton energy.
Fast neutron proof of concept
Section titled “Fast neutron proof of concept”A preliminary test used a 30 μm pCVD film with a 3 mm contact pad. Neutrons came from a deuterium-tritium generator with an energy of 15.5 MeV. The measured spectrum separated neutron pulses from the generator-off noise trace. This confirmed detection of 15.5 MeV fast neutrons by the pCVD film.
For a 30 μm film at 15.5 MeV, the estimated detection efficiency was about 0.077%, based on the neutron cross-section of ¹²C. An exact comparison with the estimate was not possible because several source parameters were uncertain. The target-to-film distance was 5 to 6 cm, and the generated flux was 10⁴ to 10⁵ n/cm²/s.
The authors present this as proof of concept for fast-neutron monitoring. The paper did not report a calibrated efficiency or tests with lower-energy neutrons.
Engineering selection and unresolved limits
Section titled “Engineering selection and unresolved limits”The authors chose pCVD to address the high cost, small size, and limited availability of high-quality single-crystal substrates. Earlier work had reported pCVD dosimetric properties similar to single-crystal diamond, including good linearity and dose-rate dependence.
Film thickness and contact geometry should be selected together because both affected alpha-particle S/N. For 5.5 MeV alpha particles, thin films and small pads gave the strongest response in this study. Surface condition matters because thicker films became rougher, and the authors recommend lapping to improve detection.
Radiation tolerance depends on the starting S/N, so a film that resolves particles before irradiation may still lose its peak at a lower fluence. Raman and PALS data show changes not obvious in SEM, including increased sp² carbon and altered positron lifetimes. Material selection should account for charge-particle energy, proton fluence, noise, and required detector size.
Zero-bias operation avoided the added current and noise seen under bias, but the study did not show improved detection with bias. The carbon reaction has an approximately 6.2 MeV threshold. Thermal neutrons require a ¹⁰B or ⁶Li conversion layer. Within these limits, pCVD films can serve as passive alpha monitors and high-energy neutron detectors.
Materials and support
Section titled “Materials and support”This study’s fast-neutron monitor and alpha sensor use a polycrystalline CVD diamond film with patterned Ti/Au contacts. 6CCVD’s diamond materials catalog includes MPCVD-grown polycrystalline diamond plus Au and Ti metallization, covering the study’s contact materials. The engineering support page provides custom fabrication support for the required film geometry and contact pattern.
View Original Abstract
Abstract In this work, radiation detectors based on a polycrystalline diamond film were studied. Polycrystalline chemical vapor-deposited diamond films with varied thickness up to 55 μm were prepared on silicon substrates. Circular Ti/Au electrical contact pads with diameters between 1 and 3 mm were processed on the diamond films. The detector structures were tested with 5.5 MeV alpha particles generated by 241 Am radioisotope. They were irradiated with 4 MeV protons to a maximum fluence of 10 16 p/cm 2 . The irradiated samples were analysed using scanning electron microscopy, Raman spectroscopy, and positron annihilation lifetime spectroscopy. Additionally, the polycrystalline diamond detectors were used to detect fast neutrons with an energy of 15.5 MeV, which demonstrated their potential for use in neutron monitoring.
Free Standing Boron Doped Diamond Pcd Film
View Specs →
Boron Doped Single Crystal Diamond Plate
View Specs →Boron Doped Diamond Electrode On Silicon
View Specs →References we cover
Section titled “References we cover”- Properties of Diamond-Based Neutron Detectors Operated in Harsh Environments
- Diamond Detectors for Radiotherapy X-Ray Small Beam Dosimetry
- A study of the radiation tolerance of poly-crystalline and single-crystalline CVD diamond to 800 MeV and 24 GeV protons
Original source
Section titled “Original source”References
Section titled “References”- 2024 - A study of particle detectors based on single crystal diamond substrate [Crossref]