The influence of defect levels on the dose rate dependence of synthetic diamond detectors of various types on exposures to high-energy radiotherapy beams

• Dose rate linearity index Δ was evaluated for various diamonds using two beam types.
• Only Δ values of a HPHT diamond were independent of both beam energy and type.
• Presence of C–H and N3 defect-centres linked to variation of Δ with beam energy.
• Δ Values of diamonds with low defect levels show ≤2.2% change with bias voltage.

The linear response of a radiation dosimeter with absorbed dose rate is a principal requirement in radiotherapy. Fowler's model for electrical conductivity, σ of a solid-state detector and absorbed dose rate, Dr is of the form σ∝DrΔ where Δ is the linearity index that can take on a range of values around unity. Utilising synthetic diamond detectors of various types as sensors, this study investigates the influence of defect levels on the Δ values of the sensors and the dependence of Δ on bias voltage, beam energy and type in the dosimetry of high-energy photon and electron therapy beams. One main objective of the study was to establish whether for a given diamond detector, Δ could be determined only once for any given beam energy and then used for other beam energies of clinical interest. In order to attain the ICRU overall ±5% uncertainty of absorbed dose delivery in radiotherapy, ±2% accuracy was considered. The study was conducted on one HPHT and eight CVD synthesised diamonds of optical grade (OG) and detector grade (DG) qualities using 6 and 15 MV photon, and 7 and 12 MeV electron energies. Values of Δ ranging from 0.79–1.03 to 0.85–0.96 were obtained for the electron and photon beams, respectively for all the diamond sensors at 1 kV/cm. The Δ values were found to change with various defect levels present within the crystals as characterised by Raman spectroscopy, ESR, FTIR spectroscopy and TL emission, and it was observed that the Δ values of crystals with high defect levels varied strongly with bias voltage. Whereas the Δ values of the HPHT diamond were found not vary with the electron and photon energies, only those of three CVD samples of a given class showed a variation within 2% between the two energies of each beam type. However, for all the crystals tested Δ showed a maximum variation of 3.4% between the photon energies unlike the electron energies where a very strong variation (>5%) was observed for three OG CVD crystals. The results of this study have suggested that differences in crystal quality due to the presence of defects could cause Δ to vary with bias voltage, beam energy and type.