Year

2013

Degree Name

Master of Science - Research

Department

Centre for Medical Radiation Physics

Abstract

The aim of this study is the characterisation of an innovative silicon 2D array Detector named Magic Plate 512 for dosimetry in four-dimensional small field external beam radiotherapy. The need for such a device arises from the inability of current commercial dosimetric devices to account for thoracic movement during 4- dimensional radiation therapy (4DRT). 4DRT represents the next step from intensity modulated radiation therapy (IMRT), where organ movements during treatment are no longer neglected as the treatment beam will actively follow a target volume. Dynamic multi-leaf collimators (DMLCs) are one solution to this issue of patient respiration as a DMLC will continuously reshape the treatment aperture to compensate for target motion [1].

The Magic Plate 512 (M512) is a 2D dosimeter array utilizing 512 silicon detectors on a bulk p-type substrate, which has been designed as a high resolution quality assurance tool to determine penumbral variations as a result of simulated patient respiration by a phantom. This variation will be compared with the 4DRT software to determine the accuracy and effectiveness of the DMLC treatment.

Test structure diodes have been fabricated on two sets of silicon substrates: one from an American silicon wafer provider (named CMRP) with resistivity of 10 cm and 1 cm, the second set from an European silicon facility (named KDB) with resistivity of 10 cm, diodes were fabricated from both substrates with, and without an n+ guard ring. Electrical characterisation as well as radiation damage studies from both photons and photoneutrons were undertaken on these test structures.

The purpose of these tests was to determine the optimal configuration regarding ion implantation, substrate type and resistivity as well as determining the effect of a guard ring on the charge collection efficiency and dose response. Increased concentrations of ion implantations were found to decrease both leakage current and device capacitance as well as showing a smaller response decrease as a result of both photon and neutron damage. The CMRP substrate demonstrated a smaller response decrease than the KDB substrate, whilst it was found the guard ring had little to no effect on either leakage current or capacitance.

Two Magic Plate 512 arrays were constructed as a result of these test structure characterisations and subjected to electrical characterisations and radiation damage studies. The two magic plate arrays utilized a 10 cm(CMRP) substrate and 100 μC ion implantation (array #10) and another with the 10 cm(KDB) substrate and 30 μC ion implantation (array #24). The electrical characterisation of these two devices found the leakage current was considerably higher for array #24, whilst a comparison of diode position to leakage current found that the position of the diode in the array has no effect on the magnitude of leakage current. The radiation damage tests found results consistent with those of the test structures, although the absolute response of the diodes fabricated on the CMRP substrate was so low that the calculated uncertainties are too large to make any conclusions from the data. As such it was determined that array #24 with the diodes fabricated on the KDB substrate provides a more stable response to radiation damage and presents as the more viable option for small field radiation dosimetry.

Materials and findings in this research project were included partially to a paper submitted to IEEE TNS Characterisation of an innovative p-type epitaxial diode for dosimetry in modern external beam radiotherapy A.H. Aldosari, A. Espinoza, D. Robinson, I. Fuduli, C. Porumb, S. Alshaikh, M. Carolan, M.L.F. Lerch, V. Perevertaylo, A.B. Rosenfeld, Member, IEEE, M. Petasecca, Member, IEEE. Data relative to MP512 will be used for a publication in Medical Physics which is currently in preparation.

Download

Share

COinS
 

Unless otherwise indicated, the views expressed in this thesis are those of the author and do not necessarily represent the views of the University of Wollongong.