The design of a new insert for calibration of LDR and HDR 192 Ir sources in a well-typeionization chamberA. E. C. Pineda, E. P. Q. Alcón, and C. E. deAlmeida Citation: Medical Physics 30, 1566 (2003); doi: 10.1118/1.1578773 View online: http://dx.doi.org/10.1118/1.1578773 View Table of Contents: http://scitation.aip.org/content/aapm/journal/medphys/30/7?ver=pdfcov Published by the American Association of Physicists in Medicine Articles you may be interested in Measurement of absorbed dose-to-water for an HDR 192Ir source with ionization chambers in a sandwich setup Med. Phys. 40, 092101 (2013); 10.1118/1.4816673 Air kerma standard for calibration of well-type chambers in Brazil using I 192 r HDR sources and its traceability Med. Phys. 36, 953 (2009); 10.1118/1.3056462 Assaying 192 Ir line sources using a standard length well chamber Med. Phys. 29, 2692 (2002); 10.1118/1.1517046 A parallel plate chamber for calibration of 192 Ir LDR and HDR sources Med. Phys. 26, 2438 (1999); 10.1118/1.598762 Evaluation of a new sealed reentrant well chamber for HDR and LDR brachytherapy calibrations Med. Phys. 25, 719 (1998); 10.1118/1.598242

The design of a new insert for calibration of LDR and HDR 192Ir sourcesin a well-type ionization chamber

A. E. C. Pineda, E. P. Q. Alcon, and C. E. deAlmeidaLaboratorio de Ciencias Radiolo´gicas, Rio de Janeiro State University, Rio de Janeiro, Brazil

~Received 9 November 2002; accepted for publication 9 April 2003; published 16 June 2003!

Some well-type ionization chambers, present a very small sweet spot that are sufficient for smallHDR sources. However, if a longer HDR source or LDR wires are calibrated, the positionaluncertainty increases and an approximated correction factor must be applied, resulting in an in-creased uncertainty. One of the ways to avoid this problem would be to flatten the well chamberresponse by increasing its sweet spot region. This work uses the Monte Carlo code PENELOPE tosimulate the response of a well-type chamber HDR-1000, with its original insert, by using an HDR192Ir source and proposes a new insert design that increases its flatness region from 1.0 cm toapproximately 4.0 cm (62.0 cm about the peak response!. © 2003 American Association of Physi-cists in Medicine. @DOI: 10.1118/1.1578773#

Key words: well-type ionization chamber calibration, brachytherapy, Monte Carlo simulation

I. INTRODUCTION

High dose rate~HDR! afterloading sources using192Irsources have received considerable worldwide attention as amore economical and safer option for brachytherapy treat-ments than low dose rate brachytherapy systems. The na-tional laboratories do not currently offer a direct calibrationof ionization chambers for high dose rate192Ir gamma rays.The American Association of Physicists in Medicine~AAPM!1 has adopted the method proposed by Goetschet al.2 as an interim method that uses an exposure weightedaverage192Ir energy of 397 keV, which is approximately halfway between the137Cs gamma rays and 250 kV x rays(HVL53.2 mm of Cu, effective energy of 146 keV x rays!.The International Atomic Energy Agency,~IAEA !3,4 in orderto enhance the metrological traceability has recommendedthe calibration laboratories to perform free in-air measure-ments and the hospital users to use a well-type ionizationchamber. Although the well-type ionization chamber requiresa calibration against a192Ir source previously calibrated inair, the slight increases in the uncertainty due to this proce-dure may be compensated by other factors such as the repro-ducible source positioning geometry and its long-term stabil-ity ~typically better than 0.5% a year!.5,6 One of the designlimitations of some well chambers is their rather small sweetspot. This becomes more critical when192Ir wires are cali-brated in this chamber. A new insert design is proposed inthis work that reduces the positional uncertainty of low andhigh dose rate192Ir sources hence improving the overall ac-curacy of the measurement.

II. METHODS AND MATERIALS

To simulate the response of a well-type chamber HDR-1000 plus, the energy imparted to the sensitive volume wascalculated by using the Monte Carlo~MC! code PENE-LOPE. The characteristics of this code has been described indetail by Baro´ et al.7 hence, just a brief description is pro-

vided here. The code is implemented in Fortran 77 and itsstructure is based on a set of subroutines that are invokedfrom a main program written by the user. It is applicable toenergies ranging from 1 keV to 1 GeV for photons and from0.1 keV to 1 GeV for electrons. Electron and positron histo-ries are generated on the basis of a mixed algorithm thatcombines detailed simulation of hard events with condensedsimulation of soft events. The well chamber modelling wasperformed using the geometry definition package PEN-GEOM2 that is based on the combination of surfaces~repre-sented by quadrics functions! to form more complex struc-tures such as bodies and body sets~modules!. Thecompositions by weight, for the diverse materials that com-pose the chamber and the192Ir source, were taken from thePENELOPE database. A variance reduction technique wasperformed in order to reduce the simulation time and thestatistical uncertainties, forcing the photons to interact in thechamber sensitive volume by redefining its inverse mean freepath. The MC calculated values, around the sweet spot of thechamber, were obtained by simulating up to 108 histories foreach point.

The bare192Ir radionuclide spectra used as input spectrumfor the MC calculations was taken from Ballesteret al.8 The192Ir microSelectron-HDR source was modelled basically byright cylindrical geometries following the approach adoptedby Borg and Rogers9 that disregard its small elliptical shapedend welds. This consideration has proven to be a good ap-proximation of the real geometry, since due to the energy ofthe photons the source attenuation at the small ends is notsignificant. The source was assumed to consist of a 0.36 cmlong cylinder with a diameter of 0.065 cm of pure Ir metalwith radioactive192Ir uniformly distributed in it. Around thiscore there is a capsule with outer diameter of 0.09 cm madeof AISI 316L steel, and connected to a 0.2 cm long steelcable with diameter 0.07 cm. In this work, all steel materialswere approximated to the composition of AISI 304 steel,10

including the capsule.

1566 1566Med. Phys. 30 „7…, July 2003 0094-2405 Õ2003Õ30„7…Õ1566Õ3Õ$20.00 © 2003 Am. Assoc. Phys. Med.

The well chamber was assumed to consist of a sensitivevolume of 245 cm3 (12.2 cm30.46 cm30.175 cm) of aircontained between two cylindrical walls; the outer with 2.0cm thickness of Al and the inner with 0.09 cm of thickness ofbutyrate and 0.01 of thinned Al. For the upper end of thesensitive volume an Al disc of 0.32 cm thickness was as-sumed and for the lower end, the chamber base, several disksof different materials and dimensions were considered. Thisbase and others parts not described in this report were mod-elled according to the data supplied by the manufacturer, andonly a brief description is provided here, since the completedesign details and characteristics is property of Standard Im-aging Inc. Inside the sensitive volume there is an electrode ofaluminum of 0.09 cm thickness. The source insert fits snuglywithin the central thin-walled aluminum tube. That insert is aremovable assembly leaving a well void of a 3.6 cm diameterand a 12.1 cm deep. The standard insert has 12.1 cm ofheight and 3.5 cm of diameter.

Most of the components of the well chamber were mod-eled as Al 6061, except the electrode material, the insert walland the chamber component near the base, which were con-sidered as Al 3003. The numbers beside Al means the type ofaluminum alloy. The insert base and the chamber bottommade of plastic were considered as composed of PMMA andpolycarbonate, respectively. For a realistic simulation theStyrofoam insert, used to reduce the temperature inside thechamber, was also considered. Thus, the well chamber re-sponse was obtained by considering the complete arrange-ment such as the MicroSelectron-HDR source with its cable,the well chamber detailed geometry and its insert with theStyrofoam included.

To flatten the sweet spot region of the well-type ionizationchamber the new insert proposed was based on thickeningthe original insert provided by the manufacturer. This insert,that guides the source inside the chamber, is made of an Altube with a wall thickness of 0.11 cm and an external radiusof 0.16 cm. The new insert was designed by taking accountthe gamma ray attenuation processes and the192Ir bare spec-tra modification. The modification is mainly due to attenua-tion of g rays in the source core and its encapsulation, thethickened insert wall with Styrofoam and the materials lo-cated around the sensitive volume. Figure 1 shows a com-parison of the original bare192Ir radionuclide spectra used asinput spectrum for the MC calculations and the modifiedspectrum obtained at the entrance of the sensitive volume. Inthe low energy region a significative presence of scatteredphotons is observed as a result of the scatterer media, whichalso contributes to the response of the well chamber.

The new insert was modelled as a set of four cones of Alof different slopes and thicknesses on each side of one cen-tral ring of 0.5 cm high and 0.54 cm radius. This central ringwas located at the position of the chamber’s maximal re-sponse. The radius of the cones was of 0.20, 0.27, 0.35, and0.45 cm, respectively, with a hight of 0.5 cm each. Such radiiwere selected to avoid a step function in the response curve.Figure 2~a! shows the Penelope’s 2D display of the new in-sert design and Fig. 2~b! shows an expanded view of a regionthat includes the four cones and a half of the central ring,

note that the four cones actually make up the one coneshown. In our simulations two setups were considered acomplete and a simplified geometry. For the complete case,the well chamber was modelled including all its details pro-vided by the manufacturer, that is, next to the reality. For thesimplified case, shown in Fig. 2~a!, the geometry complexitywas reduced in order to speed up our simulations, consider-ing only materials situated between the source and the sen-

FIG. 1. The192Ir bare spectra~a! and the MicroSelectron-HDR192Ir photonfluence obtained at the entrance of the sensitive volume.

FIG. 2. New insert design approach illustration using cones of differentslopes and thicknesses.~a! Penelope’s 2D view of the simplified geometry.~b! A detailed zoom of the new insert.

1567 Pineda, Alco n, and deAlmeida: Chamber 1567

Medical Physics, Vol. 30, No. 7, July 2003

sitive volume and eliminating some components as the upperparts, the lower base of the well chamber and the outer wallof the sensitive volume.

III. RESULTS

Figure 3 shows the MC calculated values for the relativeresponse of the well chamber HDR-1000, as a function of thesource position within the well chamber, for both geometries.They are compared to experimental data of Hoet al.11 TheMC calculated values for the complete and simplified geom-etry cases, both using the old insert, were simulated for atotal of seven points from 3 to 9 cm with intervals of 1 cm.The MC simulated relative response curve for the completegeometry case compared to the experimental data shows anexcellent agreement. Also, there are no differences in thesimulation results between both geometry cases over a largepart of the curve except at the ends. Thus, the use of a sim-plified geometry set up will not significantly alter our results.

Figure 4 shows the MC calculated values for the newinsert using the simplified geometry. This curve was obtainedfor a total of 11 points from 2.0 to 8.0 cm with an intervalstep of 0.5 cm and 1 cm at both ends. This curve is comparedto the MC calculated values obtained for nine points from2.0 to 6.5 cm with intervals of 0.5 cm and interval of 1 cm atthe left end, for the old insert case. Both curves are alsocompared to the experimental data of Hoet al.11

IV. CONCLUSIONS

As result of the proposed modification, the sweet spot ofthe chamber was increased considerably from a flat responseof 1.0 cm with the original insert to approximately 4.0 cm(62.0 cm about the peak response! with the new insert. Thisnew insert will certainly improve the accuracy of the calibra-

tion of HDR sources though its main impact is in the cali-bration of low dose rate~LDR! wires of different lengths,avoiding approximated corrections to be used.

ACKNOWLEDGMENTS

We would like to acknowledge Standard Imaging Inc. forproviding the details of the chamber HDR 1000 plus. Thiswork was supported by grants from the IAEA and CNPq.

1AAPM Task Group 40, ‘‘Comprehensive QA for radiation oncology: Re-port of AAPM Radiation Therapy Committee task Group 40,’’ Med.Phys.21, 581–618~1994!.

2S. J. Goetsch, F. H. Attix, L. A. De Werd, and B. R. Thomadsen, ‘‘A newre-entrant ionization chamber for the calibration of irridium-192 highdose rate sources’’ Int. J. Radiat. Oncol., Biol., Phys.24, 167–170~1992!.

3International Atomic Energy Agency~IAEA !, IAEA-TECDOC-1079,Calibration of Brachytherapy Sources~IAEA, Vienna, 1999!.

4International Atomic Energy Agency~IAEA !, IAEA-TECDOC-1074,Safety Measures to Address the Year 2000 Issue at Medical Facilitieswhich use Radiation Generators and Radioactive Materials~IAEA, Vi-enna, 1999!.

5S. J. Goetsch, F. H Attix, D. Pearson, and B. Thomadsen, ‘‘Calibration of192Ir high-dose-rate afterloading systems,’’ Med. Phys.18, 462–467~1991!.

6L. A. De Werd and B. R. Thomadsen, ‘‘Brachytherapy physics,’’ AAPMSummer School, San Diego, California, 18–22 July, 1994, pp. 541–546.

7J. Baro, J. Sempau, J. M. Fernandez-Varea, and F. Salvat, ‘‘PENELOPE:an algorithm for MC simulation of the penetration and energy loss ofelectrons and positrons in matter,’’ Nucl. Instrum. Methods Phys. Res. B100, 31–46~1995!.

8F. Ballester, C. Herna´ndez, J. Pe´rez-Calatayud, and F. Lliso, ‘‘MonteCarlo calculation of dose rate distributions around192Ir wires,’’ Med.Phys.4, 1221–1228~1997!.

9J. Borg and D. W. O. Rogers, ‘‘Spectra and air-Kerma strength for en-capsulated192Ir sources,’’ Med. Phys.26, 2441–2444~1999!.

10J. Williamson and Z. Li, ‘‘Monte Carlo aided dosimetry of the MicroSe-lectron pulsed and high dose-rate192Ir sources,’’ Med. Phys.22, 809–819~1995!.

11A. K. Ho, C. H. Sibata, C. N. deSouza, R. Castiglione, and K. H. Shin,‘‘Evaluation of a well-type ionization chamber for calibration of HDLand LDR brachytherapy sources,’’ Med. Dosim20, 31–34~1995!.

FIG. 4. The MC calculated well chamber response of the new insert designfor a 192Ir HDR source as a function of the distance from the bottom of thechamber compared with the old insert and the experimental data of Hoet al.

FIG. 3. The MC calculated relative axial response of the HDR-1000 cham-ber for a192Ir HDR source as a function of the distance from the bottom ofthe chamber.

1568 Pineda, Alco n, and deAlmeida: Chamber 1568

Medical Physics, Vol. 30, No. 7, July 2003