Simulation evaluation of NIST air‐kerma rate calibration standard for electronic brachytherapy

Purpose: Dosimetry for the model S700 50 kV electronic brachytherapy (eBT) source (Xoft, Inc., a subsidiary of iCAD, San Jose, CA) was simulated using Monte Carlo (MC) methods by Rivard et al. [“Calculated and measured brachytherapy dosimetry parameters in water for the Xoft Axxent x‐ray source: An electronic brachytherapy source,” Med. Phys. 33 , 4020–4032 (2006)] and recently by Hiatt et al. [“A revised dosimetric characterization of the model S700 electronic brachytherapy source containing an anode‐centering plastic insert and other components not included in the 2006 model,” Med. Phys. 42 , 2764–2776 (2015)] with improved geometric characterization. While these studies examined the dose distribution in water, there have not previously been reports of the eBT source calibration methods beyond that recently reported by Seltzer et al. [“New national air‐kerma standard for low‐energy electronic brachytherapy sources,” J. Res. Natl. Inst. Stand. Technol. 119 , 554–574 (2014)]. Therefore, the motivation for the current study was to provide an independent determination of air‐kerma rate at 50 cm in airK ̇air( d = 50 cm ) using MC methods for the model S700 eBT source. Methods: Using CAD information provided by the vendor and disassembled sources, an MC model was created for the S700 eBT source. Simulations were run using the mcnp6 radiation transport code for the NIST Lamperti air ionization chamber according to specifications by Boutillon et al. [“Comparison of exposure standards in the 10‐50 kV x‐ray region,” Metrologia 5 , 1–11 (1969)], in air without the Lamperti chamber, and in vacuum without the Lamperti chamber.K ̇air( d = 50 cm ) was determined using the *F4 tally with NIST values for the mass energy‐absorption coefficients for air. Photon spectra were evaluated over 2 π azimuthal sampling for polar angles of 0° ≤ θ ≤ 180° every 1°. Volume averaging was averted through tight radial binning. Photon energy spectra were determined over all polar angles in both air and vacuum using the F4 tally with 0.1 keV resolution. A total of 10 11 simulated histories were run for the Lamperti chamber geometry (statistical uncertainty of 0.14%), with 10 10 histories for the in‐air and in‐vacuum simulations (statistical uncertainty of 0.04%). The total standard uncertainty in the calculated air‐kerma rate determination amounted to 6.8%. Results: MC simulations determined the air‐kerma rate at 50 cm from the source with the modeled Lamperti chamber to be (1.850 ± 0.126) × 10 −4 Gy/s, which was within the range ofK ̇air( d = 50 cm ) values (1.67–2.11) × 10 −4 Gy/s measured by NIST. The ratio of the photon spectra in air and in vacuum were in good agreement above 13 keV, and for θ < 150° where the influence of the Kovar sleeve and the Ag epoxy components caused increased scatter in air. Below 13 keV, the ratio of the photon spectra in air to vacuum exhibited a decrease that was attributed to increased attenuation of the photons in air. Across most of the energy range on the source transverse plane, there was good agreement between the authors' simulated spectra and that measured by NIST. Discrepancies were observed above 40 keV where the NIST spectrum had a steeper fall‐off towards 50 keV. Conclusions: Through MC simulations of radiation transport, this study provided an independent validation of the measured air‐kerma rate at 50 cm in air at NIST for the model S700 eBT source, with mean results in agreement within 3.3%. This difference was smaller than the range (i.e., 23%) of the measured values.