Physical dosimetry and spatial dose distribution

In conventional hadron therapy, the total dose is applied to the tumor volume in fractions of 10⁷-10⁹ particles per voxel within a ms typically. Using laser driven accelerated (LDA) particle pulses the same number of particles will be delivered in one pulse of a duration in the order of ns. Hence for any clinical application of LDA beams control monitors and dosimetric devices must be able to handle these ultrashort pulses of highest intensity. 

State of the art of detecting LDA particle pulses are non-electronic detectors like radiochromic films (RCF), Image Plates (IP) or nuclear track detectors (CR39). Although these kind of detectors are widely used in LDA-experiments, they are not suited for any clinical control monitor as all the relevant information about delivered dose is not available in real time.
Ionization chambers, which are generally used in conventional therapy facilities, are most probably not suited to both, monitoring and dosimetry of LDA beams, because of saturation effects. Nevertheless these conventional systems will be tested under LDA beam conditions.  
For the preclinical studies of project D.3.3 at the Tandem accelerator dose verification have to be done. As no dosimetry protocol is available for proton beams of 25 MeV, where the water equivalent range is limited to only few mm, Gafchromic films will be used to verify the dose after the irradiation. These films will be placed directly behind the beam exit window, where effects due to quenching, as they appear when using these films in the Bragg peak region, are negligible.

Main topic in this project is the attempt to use pixel detectors for detection of LDA beams in real time. As each pixel represents a small detector unit in itself, only a small fraction of the whole LDA beam will be detected by each pixel and so problems due to detector saturation might be overcome by this new approach. Due to excellent spatial resolution of pixel detectors additionally information about the spatial dose distribution is maintained.
First tests of commercially available systems show good linearity between integrated detector signal and particle fluence as well as negligible blooming between adjacent pixels.
In future experiments commercially available pixel detector systems will be tested at the MLL Tandem accelerator in Garching as well as at the MPQ to characterize the performance of these systems under LDA beam conditions.