The direct determination of dose-to-water using a water calorimeter

Experimental determination of beam quality factors,kQ, for two types of Farmer chamber in a 10 MV photon and a 175 MeV proton beam

Absorbed doses determined with a sealed water calorimeter operated at 4 degrees C are compared with the results obtained using ionization chambers and the IAEA TRS-398 code of practice in a 10 MV photon beam (TPR(20,10) = 0.734) and a 175 MeV proton beam (at a depth corresponding to the residual range, R(res) = 14.7 cm). Three NE 2571 and two FC65-G ionization chambers were calibrated in terms of absorbed-dose-to-water in (60)Co at the Swedish secondary standard dosimetry laboratory, directly traceable to the BIPM. In the photon beam quality, calorimetry was found to agree with ionometry within 0.3%, confirming the k(Q) values tabulated in TRS-398. In contrast, a 1.8% deviation was found in the proton beam at 6 g cm(-2) depth, suggesting that the TRS-398 tabulated k(Q) values for these two ionization chamber types are too high. Assuming no perturbation effect in the proton beam for the ionization chambers, a value for (w(air)/e)(Q) of 33.6 J C(-1) +/- 1.7% (k = 1) can be derived from these measurements. An analytical evaluation of the effect from non-elastic nuclear interactions in the ionization chamber wall indicates a perturbation effect of 0.6%. Including this estimated result in the proton beam would increase the determined (w(air)/e)(Q) value by the same amount.

Absolute dosimetry in a d(14 MeV)+Be fast neutron beam

Since 1978, the Universitätsklinikum in Essen operates a d(14 MeV) + Be fast neutron beam for patient treatment. Dosimetric studies were performed in a rectangular 40 x 40 mm2 neutron/photon field using a transportable water calorimeter, which had been developed at the Physikalisch-Technische Bundesanstalt. The water calorimeter allowed small dosimeters to be directly calibrated in units of absorbed dose-to-water in a cylindrical phantom of 50 mm in diameter. Also, the twin detector method was applied in order to determine the photon and the neutron dose separately. By making use of a calibrated ionization chamber, the absorbed dose-to-water calibration in the cylindrical water phantom was transferred to a water phantom, a cube 300 mm on a side. Experiments and Monte Carlo calculations covering the neutron producing target, the collimator and the influence of the water calorimeter on the spectral neutron fluence at the measurement position allow the relative uncertainty of the absorbed dose-to-water determination to be reduced to 2.6% (1 SD). This direct absorbed dose-to-water determination by calorimetry has shown that the treatment planning system underestimates the physical dose to tissue by 9%. For clinical purposes, the statement of the prescribed dose had to be increased by 9% in order that the absolute absorbed dose remains constant and that the same biological endpoints are reached.

Water calorimetry for radiation dosimetry

Calorimetry has a long history as a technique for establishing the absorbed dose, and graphite calorimetry has often been used to establish absorbed dose standards for use in radiation therapy. However, a conversion process is necessary to convert from dose to graphite to dose to water, which is the quantity of clinical interest. In order to more directly measure the dose to water, considerable effort has been devoted in the last fifteen years to the development of water calorimetry. This article reviews these developments and summarizes the present status of water calorimetry. Absorbed dose standards based on water calorimetry and with a relative standard uncertainty of 0.5-1% now seem achievable.

Deduction of the air w value in a therapeutic proton beam

Utilization of air-filled ionization chambers with 60Co-based reference calibrations in proton dosimetry requires application of water to air stopping power ratios and the mean energy required to produce an ion pair (W or w). Accepted uncertainties in current w values for protons leads to a dosimetric uncertainty of 4 per cent when ionization chambers are employed to measure absorbed dose. For this reason, proton dosimetry protocols recommend the use of calorimetry as the absorbed dose standard. We used calorimetry in conjunction with an ionization chamber with 60Co reference calibrations to deduce the proton w value in the entrance region of a 250 MeV proton beam: 34.2 +/- 0.5 eV. Application of this w value, with its 1.5 per cent uncertainty, allows determination of dose in therapeutic proton beams, with uncertainties comparable to photon and electron values.

Water calorimeter dosimetry for 160 MeV protons

The absorbed dose to water from a 160 MeV proton beam as determined by a flexible, temperature regulated, sealed glass core, water calorimeter was compared to that determined from ionization chambers used in accordance with AAPM Report 16. The ratios of these doses as obtained from two experiments done over four months apart, are 0.992 +/- 0.004 and 0.990 +/- 0.004. As there are no radiation dependent parameters required for the water calorimeter, these data add to the growing body of evidence which supports the use of the calorimeter as a reliable absorbed dose standard. They also support the use of 60Co-calibrated ionization chambers used in accordance with AAPM Report 16 for the dosimetry of proton beams.

The thermal defect of A-150 plastic and graphite for low-energy protons

The thermal defects of A-150 plastic and graphite referenced to aluminium were determined for 800 keV protons scattered by a 2 micron nickel foil (mean transmitted energy = 550 keV). Composite cores of Al/A-150, graphite/A-150 and Al/graphite which could be irradiated from one side or the other were employed. The temperature increase of a core caused by 30-100 s of irradiation (3-6 nA of proton beam current) was detected by two thermistors mounted in opposite legs of a Wheatstone bridge. The thermal defect of A-150 plastic was determined to be 0.0421 +/- 0.0036 (SE) referenced to aluminium and 0.0402 +/- 0.0034 referenced to graphite. The thermal defect of graphite referenced to aluminium is 0.0043 +/- 0.0034. No change in the thermal defect of A-150 plastic was detected for accumulated doses up to 8 x 10(5) Gy.

Construction of a calorimeter prototype with a high sensitivity pulsed signal detection circuit

A calorimeter based on a Wheatstone bridge detector is considered to be an ideal absolute absorbed dose measuring device. One drawback of the calorimeter is that its output signals are extremely small. The signal size can be increased by increasing the bridge excitation voltage, which, however, may lead to excess self-heating of a temperature-sensing thermistor in the calorimeter detector and may require corrections. The use of pulsed excitation was investigated in this study in place of a conventional DC excitation to induce higher bridge output voltage while keeping the average self-heating of a thermistor to a reasonably low value. Performance evaluations of our prototype pulsed calorimeter are presented.

A direct comparison of water calorimetry and Fricke dosimetry

Considerable effort has been devoted to measuring the absorbed dose to water using water calorimetry. Most of these efforts have been hampered by a lack of adequate knowledge of the heat defect of water. We argue that there is now sufficient information to establish with considerable confidence the heat defect of high-purity water containing various dissolved gases. For the present work we used water saturated with a 50/50 mixture of H2 and O2 gases, for which the heat defect is calculated to be -2.1%. As a test of this assignment, we have compared the absorbed dose to water as measured using water calorimetry with that obtained from Fricke dosimetry. The water calorimeter consisted of a small sealed vessel containing 100 ml of stirred water saturated with a 50/50 mixture of H2 and O2 gases. It was irradiated with 20 MV x-rays at a dose rate of about 0.4 Gy s-1. The same vessel was then filled with Fricke dosemeter solution, and irradiated under identical conditions. Our Fricke dosimetry is based on the Svensson and Brahme value of epsilon G (3.515 x 10(-3) 1 cm-1 J-1) and agrees to within 0.2% with the dose to water for 60Co gamma-rays obtained via graphite calorimetry. We find that for 20 MV x-rays, the dose to water determined by water calorimetry is 1.006 +/- 0.004 times the dose determined by Fricke dosimetry. Within 0.6(+/- 0.4)%, this result supports the calculated heat defect of -2.1% for water saturated with a 50/50 mixture of H2 and O2 gases.