Evaluation of different methods for determining the magnitude of initial recombination in ionization chambers

Transcription

1 Evaluation of different methods for determining the magnitude of initial recombination in ionization chambers Klaus Derikum Physikalisch-Technische undesanstalt undesallee 100, raunschweig, Germany

2 Mechanisms of Ion Recombination initial recombination volume recombination positive ion negative ion recombined independent on dose rate depends on dose rate to be corrected for

3 Motivation misleading statement in IAEA TRS 398 (section ) continuous beams: k S = ((V 11 /V /V 22 ) 22-1) 1) / ((V 11 /V /V 22 ) 22 -(M 11 /M /M 22 )) )) This relation is isbased on on a linear dependence of of 1/M on on 1/V 2.. The presence of of initial recombination disturbs this linearity but this is isnormally a small effect which may be beneglected. NE 2571 (200V) (200V) at at 1 Gy/min volume recombination ΔQ/Q = initial recombination ΔQ/Q =

4 M(100V) / M(U) M s / M = 1 + b/u + β j d 4 / U 2 0,1 Gy/min 0,4 Gy/min 1,1 Gy/min 1,6 Gy/min 3,0 Gy/min NE 2571/ kv U -1 kv -1-1

5 does not correct N D 60, w, Co for recombination losses N 60 Dw,, Co : absorbed dose to water calibration factor for 60 Co radiation N D,w accounts for ion loss independent on dose rate (i. e. ion diffusion and initial recombination)

6 Ion Loss in Pulsed eams Derikum, Roos, PM (1993) Qs/Q = 1, ,15 d 2 q / U Seuntjents et al. MP (2000) measurements NE 2571 : 1,0021 1,0022 NE 2611A : 1,0022 1,0023 PR06-C : 1,0018 1,0015 Exradin A12 : 1,0020 1,0019 Formula describes result of of extrapolation within 0.1 %.. Constant term attributed to to initial recombination..

7 Measuring the amount of initial recombination Method 1: fitting M/ M s = 1 - b/u - β j d 4 / U 2 ( 60 Co) simultaneously for various dose rates Method 2: fitting M/ M s = 1 - b/u - β j d 4 / U 2 ( 60 Co) at dose rate < 0.1 Gy /min Method 3: 3.1: fitting M/ M s = 1 - a/u (Linac) for various doses per pulse 3.2: fitting M 1 / M s = 1 - b/v 1 - c q ( parameter - variable )

8 Results from fitting simultaneously M/ M s = 1 - b/u - β j j d 4 // U 2 chamber b β NE NE 2561/244 0,27 V 7, V 22 /(A m) m) NE NE 2561/293 0,27 V 7, V 22 /(A m) m) NE NE 2561/297 0,27 V 6, V 22 /(A m) m) 0,28 0,28 V urns, Rosser PM 35 (1990) NE NE 2571/977 0,24 V 9, V 22 /(A m) m) NE NE 2571/2906 0,22 V 7, V 22 /(A m) m) FC FC 65-G/771 0,17 V 7, V 22 /(A m) m) oag (1987) :: 6,7 6, V 22 /(A /(A m) m)

9 ΔM/M 0 1 % V -200V NE E 0 = 20 MeV FS = 15x15 cm 2 z = 22 mm -100V -50V -40V t / min Run Nr 14277

10 M(200V) / M NE 2561/244 D i = 0,25 mgy U -1 / kv -1 ksne

11 charge deficit at 100 V 5 % NE % plane parallel d = 2 mm C/m 3 3 pulse charge density pulse charge density ksq 10-5 C/m 3 x = electron beam, = photon beam

12 Linac vs. 60 Co f i = 1 - b/u b / V pulsed beam NE (0.28) NE PR06C W23331/ Roos-type FC65-G Co

13 Conclusions Consistent results obtained by different methods. (uncertainties < 0.05 %, k = 2, prelim. guess) Initial recombination in high energy beams does not depend on beam quality. ( 60 Co, pulsed photon, pulsed electron) Charge loss is negligible if corrected appropriately. (needs coordinated practice)

14 thank you