On the measurement of dose in-air for small radiation fields: choice of mini-phantom material

The purpose of this work was to determine the effect of choice of mini-phantom material on the measurement and calculation of in-air output factors (Sc) in small fields. Monte Carlo simulations in conjunction with a theoretical determination of Sc were used to validate previously reported measurements. Options for alternative mini-phantom materials were compared. A 6 MV beam from a Varian Novalis linear accelerator operating in stereotactic (SRS) mode was modelled. Phase-space data were used to determine the theoretical value of Sc. To validate previously reported Sc measurements the data were used to model the fibre-optic detector and brass mini-phantom. The impact of mini-phantom material was investigated by comparing the energy spectra of electrons entering the detector volume as a function of field size, and comparing the simulated Sc-measurement to the theoretical calculation. In order to determine factors leading to changes in Sc with field size, the origins of particles in the beam as incident on the mini-phantom were determined. Sc values derived from simulated measurements using a brass mini-phantom on a fibre-optic detector agreed with the measured Sc to within 0.7%. For simulation of measurement for all other mini-phantom materials, Sc values agreed with the theoretically calculated values to within 0.6%. The dominant processes responsible for a decrease in Sc with field size is occlusion of the focal and primary collimator contributions, while the secondary scatter, from the flattening filter and cone collimators, has minimal effect. The secondary electron spectrum is affected by the choice of mini-phantom material, but is almost independent of field size. For cone-collimated small fields in the Novalis beam (<30 mm), the decrease in Sc with field size is primarily due to collimation of the focal radiation beam and scatter from the primary collimator. A fibre optic detector with either a brass, gold or lead mini-phantom with at least d(max) equivalent height is suited to measure Sc for small SRS fields. The use of materials with higher electron/physical density can be used to reduce the size of the mini-phantom and reduce spatial averaging.