• TLD

Thermoluminescence dosimeters (TLDs) are solid state detectors widely
used in conventional
radiation detection and dose verification.
The development of ion beam cancer therapy, and the research in
radioprotection in space, stimulated the use of the TLDs for heavy ions
dosimetry.
The main advantages of this kind of detector, compared, e.g., to ionization
chambers, are the small dimensions, ease of handling, no interference on the
radiation field and the usability in solid state phantoms.
However the response of TLDs with dose is non-linear. It can be supralinear
and is affected by saturation effects as well.
The response of these detectors when irradiated with particle beams depends
also strongly on the quality of the radiation.
For this reason, in order to use TLDs with particle beams, and specifically
to get a prediction of their response in a treatment plan, a model that can
reproduce the behavior of these detectors in different conditions is needed.

In literature, several models describing the TLDs behavior already exist
and in this work we start briefly introducing some of them. In particular, we
focus on an extension of `local effect model' (LEM).
This model stems from an amorphous track structure model and assumes
the knowledge of both the radial dose distribution around heavy ion
trajectories and the detector response to reference radiation.
Even though the LEM was originally developed for predicting the
response of biological systems following ion irradiation, it can also be
extended for efficiency calculations of solid state detectors, such as TLD.

In this context a new, simple and completely analytical algorithm for the
calculation of the efficiency dependence on ion charge Z and energy E has been
developed.
The response of the whole detector has been evaluated starting from the
response to a single ion of the beam.
The dose contributions coming from neighbouring tracks are assumed to add
up linearly.
This approach is realistic in a low dose approximation, but we nevertheless
analized its limits of validity.
Its main advantage is that, being fully analytical, it is
computationally fast and can be efficiently integrated in treatment
planning verification tools.
Furthermore, it is robust against modifications of the radial dose
distribution of a single ion in the detector, as well as to different detector
response models as we critically evaluated in our work.

The calculated values of the efficiency have been compared with experimental
data, as well as to other calculated values provided by different approaches.
Moreover, after implementing our model data in the treatment
planning code TRiP98, we performed signal calculations on macroscopic target
irradiated with an extended Carbon ion field. Also these results were compared
with recent experimental measurements as well as with alternative calculations.
The results, both for pure efficiency calculations and for their propagation
in macroscopic dose response prediction, achieve a level of accuracy which is
comparable to previous calculations, while needing a much lighter computational
effort.