An accurate computation of the collection efficiency due to recombination has a capital importance in order to perform high-precision dose measurements with ionization chambers. The two-voltage method, developed to compute charge collection efficiency in gas chambers, cannot be directly applied to liquid-filled ionization chambers (LICs) due to the ionized charge strong dependence on the collection electric field, which is caused by initial recombination. It is shown that in order to apply the two-voltage method to parallel-plate LICs it is necessary to introduce explicitly the slope of the ionization charge yield with respect to the polarization voltage applied in the chamber. A three-voltage method is also proposed to avoid the introduction of the slope parameter. That method allows decoupling initial and general recombination. It leads to a nonlinear system of equations which can be solved using numerical methods.

A liquid isooctane (C(8)H(18)) filled ionization linear array for radiotherapy quality assurance has been designed, built and tested. The detector consists of 128 pixels, each of them with an area of 1.7 mm x 1.7 mm and a gap of 0.5 mm. The small pixel size makes the detector ideal for high gradient beam profiles such as those present in intensity modulated radiation therapy (IMRT) and radiosurgery. As the read-out electronics we use the X-ray Data Acquisition System with the Xchip developed by the CCLRC. Studies concerning the collection efficiency dependence on the polarization voltage and on the dose rate have been made in order to optimize the device operation. In the first tests, we have studied dose rate and energy dependences. Dose rate dependence was found to be lower than 2.1% up to 5 Gy min(-1), and energy dependence lower than 2.5% up to 20 cm depth in solid water. Output factors and penumbras for several rectangular fields have been measured with the linear array and were compared with the results obtained with a 0.125 cm(3) air ionization chamber and radiographic film, respectively. Finally, we have acquired profiles for an IMRT field and for a virtual wedge. These profiles have also been compared with radiographic film measurements. All the comparisons show a good correspondence. The device has proved its capability to verify on-line therapy beams with good spatial resolution and signal-to-noise ratio.

This work is devoted to studying the influence of chamber response functions on the standard IMRT verification for the different detector technologies available on commercial devices. We have tested three of the most used 2D detector arrays for radiotherapy dosimetry verification, based on air-ionization chambers and diode detectors. The response function has been carefully simulated using the Monte Carlo method and measured through slit and pinhole collimators. Although the response function of air-ionization detectors is considerably different with respect to that of standard diodes, the impact on a verification based in the gamma function with tolerances 3 mm and 3% is quite limited. The results show that the standard air-ionization detector arrays perform in a similar way whenever the tolerances for the gamma function are not lowered below 1.5 mm and 1.5%. Additionally, the sensitivity of these devices to fluence perturbations was measured by intentionally modifying some leaf positions in the multileaf collimator. The wider response function of air-ionization chamber arrays made them slightly more sensitive to random fluence perturbations, although silicon diode arrays are more accurate to describe the dose distribution in a point by point basis.