Dosimetric modeling of the microselectron high-dose rate 192 Ir source by the multigroup discrete ordinates method
The DANTSYS multigroup discrete ordinates computer code is applied to quantitatively estimate the absorbed dose rate distributions in the vicinity of a microSelectron 192 Ir high-dose-rate (HDR) source in two-dimensional cylindrical R-Z geometry. The source is modeled in a cylindrical water phantom...
Gespeichert in:
Veröffentlicht in: | Medical physics (Lancaster) 2000-10, Vol.27 (10), p.2307-2319 |
---|---|
Hauptverfasser: | , , , |
Format: | Artikel |
Sprache: | eng |
Online-Zugang: | Volltext |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Zusammenfassung: | The DANTSYS multigroup discrete ordinates computer code is applied to quantitatively estimate the absorbed dose rate distributions in the vicinity of a microSelectron
192
Ir
high-dose-rate (HDR) source in two-dimensional cylindrical
R-Z
geometry. The source is modeled in a cylindrical water phantom of diameter 20 cm and height 20 cm. The results are also used for evaluation of the Task Group 43 (TG-43) dosimetric quantities. The DANTSYS accuracy is estimated by direct comparisons with corresponding Monte Carlo results. Our 210-group photon cross section library developed previously, together with angular quadratures consisting of 36
(S
16
)
to 210
(S
40
)
directions and associated weights per octant, are used in the DANTSYS simulations. Strong ray effects are observed but are significantly mitigated through the use of DANTSYS’s stochastic ray-tracing first collision source algorithm. The DANTSYS simulations closely approximate Monte Carlo estimates of both direct dose calculations and TG-43 dosimetric quantities. The discrepancies with
S
20
angular quadrature (55 directions and weights per octant) or higher are shown to be less than ±5% (about 2.5 standard deviations of Monte Carlo calculations) everywhere except for limited regions along the Z axis of rotational symmetry, where technical limitations in the DANTSYS first collision source implementation makes adequate suppression of ray effects difficult to achieve. The efficiency of DANTSYS simulations is compared with that of the EGS4 Monte Carlo code. It is demonstrated that even with the 210-group cross section library, DANTSYS achieves two-fold efficiency gains using the the
S
20
quadrature set. The potential of discrete ordinates method for further efficiency improvements is also discussed. |
---|---|
ISSN: | 0094-2405 2473-4209 |
DOI: | 10.1118/1.1308279 |