Skip to content

Dosimetric characterization of an encapsulated waterproof silicon carbide detector with UHDR electron and proton beams for FLASH radiotherapy

MetadataDetails
Publication Date2025-09-17
JournalPhysics in Medicine and Biology
AuthorsG. Milluzzo, Damiano Zitelli, Andrea Cavalieri, Mariagrazia Celentano, Fabio Di Martino
InstitutionsUniversity of Pisa, Istituto Nazionale di Fisica Nucleare
AnalysisFull AI Review Included
  • Application Domain: FLASH radiotherapy (FLASH-RT) dosimetry, targeting ultra-high dose rate (UHDR) electron beams (> 40 Gy s-1) and proton beams up to several MGy s-1.
  • Detector Innovation: A 10 µm thick, 4.5 mm2 active area silicon carbide (SiC) PIN junction detector encapsulated in a waterproof 15 mm diameter PLA cylindrical housing (eSiC), enabling measurements under reference dosimetric conditions.
  • UHDR Electron Performance: Linearity verified from 1.8 Gy/pulse to 12.4 Gy/pulse (R² = 0.99) corresponding to an instantaneous dose rate (IDR) of 3.1 MGy s-1 using a 9 MeV ElectronFLASH LINAC.
  • UHDR Proton Performance: Dose-rate independence demonstrated from 50 to 530 Gy s-1 at 228 MeV; total dose independence from 1 to 30 Gy.
  • Clinical Validation: Percentage depth dose (PDD) measurements in liquid water with a 180 MeV clinical proton beam showed excellent agreement with the IBA PPC05 reference ionization chamber.
  • Key Advantage Over Ion Chambers: Solid-state SiC bypasses charge recombination issues that limit standard TRS-398 ionization chambers in UHDR pulsed electron beams.
  • Strategic Goal: Advance clinical translation of FLASH-RT by providing a reliable, real-time alternative dosimeter that operates in liquid/solid water phantoms without offline readout.
ParameterValueUnitContext
SiC active thickness10µmPIN junction active layer
Detector active area4.5mm2Circular geometry (2.5 mm diameter)
p+ layer doping (NA)1 × 1019cm-3Top emitter layer, 0.3 µm thick
n- layer doping (ND)8 × 1013cm-3Drift layer
n+ substrate thickness370µmHighly doped substrate
n+ substrate doping (ND)5 × 1018cm-3Conductive handle wafer
Housing diameter15mmWaterproof PLA cylindrical capsule
Housing wall thickness1mmPLA + epoxy resin build-up
Wire bond thickness25µmFront-side signal connection
Dose-per-pulse range tested1.8 to 12.4Gy/pulse9 MeV UHDR electrons
Maximum IDR tested (electrons)3.1MGy s-1ElectronFLASH LINAC
Linearity R² (electrons)0.99—Fit across UHDR range
Residual deviation (electrons)≤ 1.7%Maximum from linear fit
Total dose range tested (protons)1 to 30Gy228 MeV proton beam
Average dose rate range (protons)50 to 530Gy s-1228 MeV cyclotron beam
Proton beam energies tested228 / 180MeVUHDR experimental + clinical lines
Electron beam energy9MeVElectronFLASH LINAC
Beam uniformity correction~3%Expected for 2.5 mm exposed area
Calibration uncertainty (PPC05)1.0%IBA PPC05 reference chamber
Calibration depth (electrons)12mmSolid water, Dmax position
Calibration depth (protons)305mmIsocenter of clinical beamline
  1. Detector Fabrication: 10 µm PIN SiC sensor (4.5 mm2 circular active area) mounted with conductive paste onto a customized 3D-printed plastic PCB.
  2. Encapsulation: Sensor embedded inside a 15 mm diameter, 1 mm thick PLA cylindrical cup; internal cavities filled with bi-component epoxy resin for electrical isolation and waterproofing.
  3. Electrical Routing: 25 µm wire bond connects sensor front side to charge-collection pad; signal and bias routed via two LEMO cables (signal and HV return).
  4. Calibration Geometry: eSiC calibrated against an Advanced Markus chamber (electrons) and IBA PPC05 chamber (protons) in solid/liquid water phantoms at Dmax.
  5. UHDR Electron Irradiation: 9 MeV ElectronFLASH LINAC at CPFR (Pisa); DPP linearity tested from 1.8 to 12.4 Gy/pulse.
  6. UHDR Proton Irradiation: 228 MeV cyclotron beam at Trento Proton Therapy facility; beam currents from 50 to 500 nA scanned for dose-rate independence tests.
  7. Clinical Proton Validation: 180 MeV pencil beam scanned to a 10 × 10 cm2 field; PDD measured in IBA motorized water phantom against PPC05 reference chamber.
  8. Charge Readout: Keithley 6517B electrometer (eSiC); IBA DOSE2 electrometer (PPC05).
  9. Data Analysis: Linear regression with R² goodness-of-fit; residual analysis for dose-rate dependence.

Applicable Materials

  • Optical/Electronic Grade Single Crystal Diamond (SCD) — for groups replicating this dosimetry work using diamond, our SCD films (0.1 µm to 500 µm thickness, polished to Ra < 1 nm) provide the radiation hardness and dose-rate linearity required for FLASH-RT reference dosimetry.
  • Heavy Boron-Doped Polycrystalline Diamond (BDD) — an alternative solid-state detector option for high-dose-rate environments, leveraging diamond’s wide bandgap and carrier mobility advantages analogous to those demonstrated by the SiC device here.
  • Intrinsic/Undoped PCD Plates — suitable substrates for hybrid detector assemblies or PCB-mounted sensor prototypes requiring mechanical stability up to 125 mm wafer formats.

Customization Potential

  • The encapsulated sensor concept used here (circular 2.5 mm active area, integrated housing) is directly transferable to diamond. 6CCVD provides laser cutting and custom shaping to produce SCD or BDD elements in non-standard geometries (small discs, strips, or arrays).
  • Custom metalization schemes (Ti/Pt/Au, Cr/Au, or full-stack variations) are available in-house to support wire bonding and PCB integration, replicating the 25 µm wire-bond interface used in the paper.
  • For multilayer detector stacks, our substrate thicknesses up to 10 mm allow robust mechanical carriers for encapsulated waterproof detector housings.

Engineering Support

  • 6CCVD’s in-house PhD engineering team can advise on material selection for FLASH-RT dosimetry, radiation-hard sensor development, and detector-grade diamond characterization — covering charge collection efficiency, dose-rate linearity, and leakage-current benchmarking against the SiC results reported here.
  • We support projects across the workflow: from bare wafer supply and custom polishing, through metalization and dicing, to integration-ready detector chips.

Call to Action

For custom specifications, detector-grade diamond consultation, or to discuss a FLASH-RT dosimetry collaboration, visit 6ccvd.com or contact our engineering team directly.

View Original Abstract

Abstract Silicon carbide (SiC) detectors have been widely demonstrated to be suitable alternative detectors for dosimetry in FLASH radiotherapy, showing radiation hardness and dose-rate independence at the FLASH radiotherapy instantaneous dose rates (IDRs). However, the practical use of such devices in the preclinical/clinical environment still requires the development of special handy housing enabling the quality assurance (QA) measurements under the reference dosimetric conditions. A 10 μ m thick, 4.5 mm 2 area SiC detector produced by the STLab company was recently embedded at the INFN-Catania Division inside a plastic waterproof 15 mm diameter cylindrical housing. This encapsulated version of SiC (eSiC) allows the measurement of the dose in reference conditions and of the dose profiles in liquid/solid water phantoms for assuring high accuracy dosimetry QA procedures. Dosimetric characterizations were performed with both electron and proton beams at conventional and ultra-high dose rates (UHDR). A first experiment was carried out at the Centro Pisano for Flash Radiotherapy using UHDR 9 MeV electron beams to confirm the linearity of the charge response as a function of the dose per pulse after the encapsulation procedure. A linearity from 1.8 Gy/pulse up to about 12 Gy/pulse, corresponding to an IDR of 3 MGy s −1 , was found. The percentage depth dose (PDD) distribution in water of 9 MeV electron beams was also measured and compared with the PDD measured with a Freiburg Physikalisch-Technische Werkstätten Dr. Pychlau GmbH (PTW) flash diamond detector, used as reference dosimeter. The eSiC detector was also tested with proton beams accelerated by the IBA Proteus 235 cyclotron at the Trento Proton Therapy facility. A response independence on the total delivered dose (1-30 Gy) and average dose rate (50-530 Gy s −1 ) was found using the UHDR 228 MeV proton beam available along the experimental beamline. The depth dose distribution measured with the eSiC within a liquid water phantom was successfully compared with the one simultaneously measured by the IBA PPC05 reference chamber, using 180 MeV clinical proton beams. The excellent results demonstrated that this first realized eSiC prototype can be used to accurately perform reference and relative dosimetry with UHDR electron and proton beams, contributing to support the clinical translation of FLASH radiotherapy.

  1. 2023 - A prototype low-cost secondary standard calorimeter for reference dosimetry with ultra-high pulse dose rates [Crossref]
  2. 2023 - Evaluation of single-fraction high dose FLASH radiotherapy in a cohort of canine oral cancer patients [Crossref]
  3. 2022 - The probe-format graphite calorimeter, aerrow, for absolute dosimetry in ultrahigh pulse dose rate electron beams [Crossref]
  4. 2023 - Charge collection efficiency of commercially available parallel-plate ionisation chambers in ultra-high dose-per-pulse electron beams [Crossref]
  5. 2021 - Spread-out Bragg peak proton FLASH irradiation using a clinical synchrocyclotron: proof of concept and ion chamber characterization [Crossref]
  6. 2020 - FLASH radiotherapy with electrons: issues related to the production, monitoring and dosimetric characterization of the beam [Crossref]
  7. 2022a - A new calculation method for the free electron fraction of an ionization chamber in the ultra-high-doseper-pulse regimen [Crossref]
  8. 2022b - A new solution for UHDP and UHDR (Flash) measurements: theory and conceptual design of ALLS chamber [Crossref]
  9. 2023 - Architecture, flexibility and performance of a special electron linac dedicated to Flash radiotherapy research: electronFlash with a triode gun of the centro pisano flash radiotherapy (CPFR) [Crossref]
  10. 2020 - Physics and biology of ultrahigh dose-rate (FLASH) radiotherapy: a topical review [Crossref]