Bureau International des Poids et Mesures

Bureau International des Poids et Mesures Comparison of the standards for absorbed dose to water of the ARPANSA and the BIPM for 60 Co gamma radiation P J Allisy-Roberts and D T Burns BIPM J F Boas, R B Huntley and K N Wise ARPANSA Otober 2000 Pavillon de Breteuil, F-92312 SEVRES edex

Comparison of the standards for absorbed dose to water of the ARPANSA and the BIPM for 60 Co gamma radiation by P J Allisy-Roberts and D T Burns Bureau International des Poids et Mesures, Sèvres J F Boas, R B Huntley and K N Wise Australian Radiation Protetion and Nulear Safety Ageny, Lower Plenty Road, Yallambie, Vitoria 3085, Australia Abstrat A omparison of the standards for absorbed dose to water of the Australian Radiation Protetion and Nulear Safety Ageny and of the Bureau International des Poids et Mesures (BIPM) has been arried out in 60 Co gamma radiation. The Australian standard is based on a graphite alorimeter and the subsequent onversion from absorbed dose to graphite to absorbed dose to water using the photon fluene saling theorem. The BIPM standard is ionometri using a graphite-walled avity ionization hamber. The omparison result is 1.0024 (standard unertainty 0.0029). 1. Introdution An indiret omparison of the standards of absorbed dose to water of the Australian Radiation Protetion and Nulear Safety Ageny (ARPANSA), Vitoria, Australia and of the Bureau International des Poids et Mesures (BIPM) was arried out in 60 Co gamma radiation. The Australian primary standard for absorbed dose is a graphite alorimeter as desribed in [1] and [2], the absorbed dose to water for 60 Co radiation being derived from the absorbed dose to graphite using the photon fluene saling theorem. The BIPM primary standard is a graphite avity ionization hamber of panake geometry as desribed in [3]. This absorbed dose to water omparison is the first suh omparison made between the two laboratories. The omparison was undertaken using two ionization hambers belonging to the ARPANSA as transfer standards. The hambers were alibrated at the ARPANSA before and after the measurements made at the BIPM in April 1997. The result of the omparison is derived from the ratios of the alibration fators of the transfer hambers determined at the two laboratories. 1

2. Determination of absorbed dose to water 2.1 The BIPM standard At the BIPM, the absorbed dose rate to water is determined from where I/m W e s,a = ( I m)( W / e) s Ψ ( µ ρ ) ( 1+ ε ) Πk, (1) Dw, BIPM /,a w, en w, w, i is the mass ionization urrent measured by the standard, is the mean energy expended in dry air per ion pair formed, is the eletroni harge is the ratio of the mean mass stopping powers of graphite and air, Ψ w, is the ratio of the photon energy fluene in water and graphite at the referene point in the water phantom, ( µ en ρ) w, is the ratio of the mean mass energy-absorption oeffiient for water to that in graphite averaged over the unperturbed and perturbed photon spetrum at the referene point, ( 1+ ε ) is the ratio of the absorbed dose to the ollision omponent of kerma, at the w, referene point in water to the same ratio at the referene point in graphite, and Π k i is the produt of the other orretion fators to be applied to the standard. The ionometri method is desribed fully in [3], whih also gives the values of the physial onstants and the orretion fators k i for the BIPM standard together with their unertainties. The ombined relative standard unertainty in the absorbed dose is 2.9 10 3, the detailed unertainty budget being given in Table 1 of this report. Absorbed dose is determined at the BIPM under the referene onditions defined by the Consultative Committee for Ionizing Radiation (CCRI, previously the CCEMRI) [4]: the distane from the soure to the referene plane (the entre of the detetor) is 1 m; the field size in air at the referene plane is 10 m 10 m, the photon fluene rate at the entre of eah side of the square being 50 % of the photon fluene rate at the entre of the square; the referene depth in water is 5 g m 2. 2

Table 1. Physial onstants, orretion fators and relative standard unertainties for the BIPM ionometri standard for absorbed dose to water Quantity BIPM value BIPM relative standard unertainty (1) 100 u i ; (v i ) 100 u j Dry air density (2) / (kg m 3 ) 1.293 0 0.01 W/e /(J C 1 ) 33.97 s,a 1.003 0 k av (air avity) 0.990 0 0.03 0.04 ( en ρ) w, µ 1.112 5 0.01 0.14 Ψ w, (photon fluene ratio) 1.006 5 0.04 0.06 (1+ε) w, (dose to kerma ratio) 1.001 5 0.06 k ps (PMMA envelope) 0.999 9 <0.01 0.01 k pf (phantom window) 0.999 6 0.01 k rn (radial non-uniformity) 1.005 1 <0.01 0.03 k s (reombination losses) 1.001 6 <0.01 0.01 k h (humidity) 0.997 0 0.03 Volume (4) /m 3 6.881 0 0.19 0.03 I (ionization urrent) 0.01 ; (7) 0.02 Quadrati summation 0.20 0.21 Combined relative standard unertainty of D w,bipm 0.29 0.11 (3) (1) In Tables 1 to 3 and 8, u i represents the Type A relative standard unertainty u A(x i ) / x i estimated by statistial means; v i represents the number of degrees of freedom; u j represents the Type B relative standard unertainty u B(x i ) / x i estimated by other means. (2) At 0 C and 101.325 kpa. (3) Combined unertainty for the produt of ( W / e) s, a (4) Using standard hamber serial number CH4-1. 3

2.2 The ARPANSA standard 2.2.1 Method of measurement At the ARPANSA the absorbed dose to water is derived from the alorimetri determination of the absorbed dose to graphite by the relation where D w Ψ D ( µ ρ) w w w w, ARPANSA en i = D Ψ β Πk, (2) is the absorbed dose to graphite at the referene point in graphite, is the ratio of the photon energy fluene at the referene points in water and graphite, 1 ( µ ) w en ρ is the ratio of the mean mass energy-absorption oeffiients for water and graphite for the photon energy spetra at the orresponding referene points, 1 w β = β w β where β is the ratio of the absorbed dose to the ollision omponent of kerma, at the referene point 1 in water (w) or in graphite (), and Πk i is the produt of the orretion fators to be applied to the standard. The fators ( ) w w w µ en ρ and β are derived by alulation [5], and the photon fluene ratio Ψ is determined using the "dose ratio" method [6, 7] as desribed in setion 2.2.3. Absorbed dose to water is determined at the ARPANSA under the following referene onditions whih are slightly different from those at the BIPM: the distane from the soure to the surfae of the water phantom is 1 m; the field size at the referene plane (nominally 1.05 m from the soure) is 10.5 m 10.5 m; the referene depth in water is 5 g m 2. 2.2.2 Determination of absorbed dose to graphite at the ARPANSA The ARPANSA standard of absorbed dose to graphite is a Domen-type alorimeter [8] onstruted by the Österreihishes Forshungszentrum Seibersdorf (ÖFZS) as desribed by Witzani et al. [9]. The absorbed dose rate to graphite, D, at the referene point in graphite is given by ( P m) kgapkzkrn kan t D = / k. (3) 1 The different notation ompared with that of equation (1) reflets the different experimental arrangements at the BIPM and at the ARPANSA. 4

The physial quantities and orretion fators in equation (3) are as desribed below and are listed with their relative standard unertainties in Table 2. Table 2. Physial quantities, orretion fators and relative standard unertainties for the determination of absorbed dose to graphite at the ARPANSA Quantity ARPANSA ARPANSA relative standard unertainty value 100 s i ; (ν i ) 100 u i P (power alulation) 0.08 ; (152) 0.04 Repeatability 0.05 ;(13) m (ore mass) /g 1.5622 0.01 k gap (alorimeter gaps) 1.0074 <0.01 0.04 k z (graphite depth) 0.9934 0.01 ; (4) 0.03 k rn (radial non-uniformity) 1.0026 0.02 ; (80) 0.04 k an (axial non-uniformity) 1.0000 <0.01 ; (5) 0.05 k t (soure deay) 0.01 Quadrati summation 0.10 0.09 Combined relative standard unertainty of D 0.13 The radiation power absorbed in the graphite ore, P This is alulated [1, 2] from voltage and resistane measurements. The alorimeter is normally operated in the quasi-isothermal mode [9] in whih the eletrial power input to the alorimeter ore in the absene of radiation is mathed as losely as possible to the antiipated radiation power. The eletri heating is swithed off at the same time as the radiation soure is swithed on, so that the rate of heating of the ore remains approximately onstant. The radiation power input an thus be determined readily against the ARPANSA working standards of resistane and voltage. The mass of the alorimeter ore, m This was measured at the ÖFZS and orreted for impurities and buoyany. Corretion fator for the alorimeter gaps, k gap The differene between the absorbed dose rate at the entre of the alorimeter ore and that at the same position in a solid graphite phantom is alulated using Monte-Carlo odes [5] and the orretion fator, k gap, is derived from these results. The ARPANSA gap orretion is alulated for a 35 mm radius field (ross-setion approximately equivalent to a 65 mm 65 mm field) inident on the alorimeter with the entre of the ore at a depth in graphite of 30 mm (5.37 g m 2 ) and at a distane of 650 mm from the soure. 5

Corretion fator for the referene depth in graphite, k z The desired referene depth is not ahieved exatly with the available ombination of graphite build-up plates. Thus a orretion is applied whih is obtained by interpolation from attenuation measurements. Corretion fator for radial non-uniformity of the 60 Co beam over the alorimeter ore, k rn This was obtained experimentally by measuring the radial profile of the beam using an NE 2561 thimble ionization hamber. Corretion for the axial non-uniformity of the 60 Co beam over the alorimeter ore, k an This was obtained from the departure from linearity of the measured depth-dose distribution over the alorimeter ore. Normalization fator for the referene date and time, k t The alorimeter measurements were orreted to the referene date and time of 1997-03-15 at 12:00 Australian Eastern Daylight Time 2. The half life of 60 Co was taken as 1 925.5 d, σ = 0.5 d [10]. 2.2.3 Conversion of absorbed dose to graphite into absorbed dose to water by the "dose ratio" method at the ARPANSA w In the "dose ratio" method, the photon fluene saling theorem [11] is used to determine Ψ for a point soure of radiation by saling the phantom dimensions, measurement depths and distanes from the soure in the inverse ratio (0.619 58) of the eletron densities of water and the graphite (using the tabulated value of water density at 20 C, 0.998 22 g m 3, the measured graphite bulk density, 1.790 g m 3, and the Z/A ratios). Under these onditions and assuming that Compton sattering is the only interation mehanism, the ratio of primary to sattered radiation energy fluenes will be the same at orresponding points, i.e. the energy spetra will have the same shape. Furthermore, the ratio of the primary photon energy fluenes at these mapped referene points will be in the inverse ratio of the square of their distanes from the soure: w 2 ( r rw ) kpgkn Ψ =, (4) where r and r w are the distanes from the soure to the referene points in graphite and water respetively, and k pg k n are orretions needed for failure to ompletely satisfy the requirements of the photon fluene saling theorem as explained in the following paragraphs. 2 AEDT = UCT+ 11 h, where AEDT is Australian Eastern Daylight Time and UCT is Universal Coordinated Time 6

Distanes from the soure to the referene points in graphite and water, r and r w The position of the soure entre was approximated by appliation of the inverse square law to ionization hamber urrent measurements in air along the beam axis, orreted for air attenuation, as desribed by Wise [5]. The distane from the soure to the referene point in water is fixed by the referene onditions at 1.050 m, and from the saling theorem, the distane to the referene point in graphite is 650.56 mm. The unertainty in the graphite density of 0.005 g m 3 leads to an unertainty in the saled graphite referene distane of 1.82 mm. The saled referene depth in graphite is (31.03 ± 0.09) mm. The omponents of the unertainty in (r /r w ) 2 are onsidered in [1] and inlude ontributions from the water phantom whih also enter into the unertainty in the water-depth orretion for the transfer hamber. Corretion fators for the limitations of the photon fluene saling theorem as applied to the graphite and water phantoms, k pg and k n The fator k pg orrets for the failure to sale the graphite phantom geometry to that of the water phantom. The fator k n orrets for the small number of non-compton interations whih are not proportional to the eletron density. These fators have been evaluated by Wise [5] using Monte-Carlo simulations with orrelated sampling. Thus D w an be alulated from D provided that r, r w, k pg, k n and the physial quantities in (2) and (3) are known. In pratie, two additional orretions are applied and the absorbed dose to water at the referene point is given by D w w ( µ en ρ) β kwin air w w, ARPANSA DΨ k =. (5) The physial quantities and the orretion fators k win and k air are desribed below and are listed in Table 3 together with their relative standard unertainties. Ratios of mass energy-absorption oeffiients ( µ ) w en ρ and of the quotients of absorbed w dose to the ollision omponent of kerma, β These were obtained using published data for the oeffiients [12] and Monte-Carlo simulations of the energy spetra at the referene points in the water and graphite phantoms as desribed by Wise [5]. Corretion fator for the differene in attenuation of the front window of the water phantom and that of the same thikness of water, k win For the ARPANSA water phantom used for measurements with 60 Co radiation, the beam passes through a PMMA 3 window of thikness 2 mm and the orretion fator was evaluated theoretially [5]. Corretion fator for air attenuation over the distane between the graphite alorimeter and the water phantom, k air This was derived from the simulated energy spetrum of the radiation beam and the attenuation oeffiients of air [12] at the appropriate energies. 3 Polymethylmetharylate 7

Table 3. Physial quantities, orretion fators and relative standard unertainties for onversion of absorbed dose to graphite to absorbed dose to water by the "dose ratio" method Quantity ARPANSA value ARPANSA relative standard unertainty 100 u i ; (ν i ) 100 u i D (see Table 2) 0.10 0.09 (r /r w ) 2 0.383 88 0.05 k n (non-compton interations) 0.9996 <0.01 0.02 k pg (phantom geometry) 0.9998 <0.01 0.02 ( µ ) w en ρ 1.1131 0.01 0.14 β w, (dose to kerma ratio) 1.0003 <0.01 <0.01 k win (water equivalene) 0.9988 0.03 k air (attenuation) 0.9972 0.01 Quadrati summation 0.10 0.18 Combined relative standard unertainty of D w,arpansa 0.20 3. Proedure for the omparison 3.1 The use of transfer hambers The omparison of the ARPANSA and BIPM standards was made indiretly by means of the alibration fators N for the ARPANSA transfer hambers given by Dw = D, (6) N D w, lab w, lab I lab where D w,lab is the absorbed dose rate to water and I lab is the ionization urrent of a transfer hamber normalized to the same referene date, eah measured at the ARPANSA or the BIPM. Current measurements are orreted for the effets and influenes desribed in this setion. The D w, BIPM value is the mean of measurements performed at the BIPM under the referene onditions over a period of three months before and after this omparison. By onvention it is given at the referene date of 1997-01-01, 0:00 UCT, as is the value of I BIPM (The value for the half life of 60 Co reommended by the IAEA [10] is used at both laboratories). The relative standard unertainty of the mean of these eight measurements is 10 4. The D w, ARPANSA value is derived from the mean of 14 alorimeter measurements made sine the installation of a new 60 Co soure at the ARPANSA in Marh 1995 and orreted to a 8

referene date of 1997-03-15 at 12:00 AEDT. The relative standard unertainty of the distribution of these 14 measurements is 4.5 10 4. The ionization hamber urrents are the mean of measurements made before and after the measurements at the BIPM, normalized to the same referene date and time. The two laboratories determine absorbed dose by methods that are quite different and the only signifiant orrelation is that due to the use of ( µ en ρ) w, or ( µ ) w en ρ. Note that these ratios are slightly different but are assumed to be fully orrelated. The unertainty of the result of the omparison is obtained by summing in quadrature the remaining unorrelated unertainties of D w,bipm, D w, ARPANSA and the ontributions arising from the use of transfer standards. The ARPANSA transfer standards used for this omparison are graphite avity ionization hambers manufatured by Nulear Enterprises (Type NE 2561, serial numbers 070 and 328). Their main harateristis are listed in Table 4. Table 4. Charateristis of the transfer hambers type NE 2561 Charateristi Serial numbers 070 and 328 Dimensions Inner diameter / mm Wall thikness / mm Cavity length / mm Cavity entre from tip / mm 7.35 0.5 9.22 5.00 Eletrode Diameter / mm 1.00 Volume Air avity / m 3 0.325 Wall Build-up ap Material Density / (g m 3 ) Material Thikness / mm graphite < 0.01% impurity 1.80 Delrin 3.87 Applied voltage Negative polarity / V 210 The alibration proedures are desribed briefly below and are disussed in more detail in [1, 2] and [13] for the ARPANSA and the BIPM, respetively. 3.2 Measurement onditions and orretions The ARPANSA water phantom is a ube of side length 300 mm, with a wall thikness of 12 mm exept for the beam entrane window whih is a 60 mm diameter irle of thikness 2 mm. The BIPM phantom is a ube of similar size with a square window of side length 150 mm and 4 mm thik. Positioning of transfer hambers in the water phantoms At the ARPANSA, the depth of the entre of the transfer hamber was adjusted to 50 mm (standard unertainty 0.05 mm) using a depth gauge and a omputerized motion ontrol system; this distane inludes the front window of the phantom. The ionization urrent is orreted to a depth of 5 g m 2 using tabulated depth dose data [14] whih give a relative 9

gradient of 4 10 3 mm 1. At the BIPM, the entre of the transfer hamber is plaed at 5 g m 2 aounting for the window (its thikness and density, its distortion due to the water pressure and its non-equivalene to water in terms of interation oeffiients) and for the water density at its measured temperature, and so no orretion is required for depth. In eah laboratory, any small differene (not more than 1 mm) between the referene distane (1 m and 1.05 m for the BIPM and ARPANSA respetively) and the distane at whih the measurement is made, is orreted by the inverse square law. Waterproof sleeve Eah hamber was supplied with a thin-walled (0.5 mm) PMMA waterproof sleeve manufatured at the ARPANSA. At the ARPANSA, a orretion fator of 1.0003 is applied to the urrent measured using the transfer hamber in the water phantom to aount for the inreased attenuation of the PMMA sleeve over that of the same thikness of water. For onsisteny the same orretion was applied to the measurements of urrent at the BIPM. Humidity, temperature and pressure During alibration, the relative humidity at the BIPM was between 48 % and 52 %; the air temperature was around 21 C and was stable to better than 0.01 C during a series of measurements. At the ARPANSA the relative humidity an vary between 30 % and 90 %; the temperature was around 23 C and was stable to within 0.02 C during a series of measurements. At both laboratories, the temperature of an air avity in the water phantom is measured with a alibrated thermistor. The measured ionization urrent is normalized to a temperature of 293.15 K and a pressure of 101.325 kpa, as part of eah laboratory's measurement system. At the ARPANSA, orretions of between 0.02 % and 0.09 % (standard unertainty 0.01 %) were made to orret the results to 50 % relative humidity. These orretions were alulated from an empirial fit, as desribed in [1, 2]. No suh orretion is required at the BIPM as humidity is losely ontrolled. Colleting voltage A olleting voltage of 210 V (negative polarity) was supplied at eah laboratory. The olleting voltage was applied at the BIPM for at least 30 minutes before measurements were made. At the ARPANSA, measurement results showing an initial drift were exluded. Measurement of harge The harge Q olleted by the hambers was measured using the loal measurement system at the BIPM and at the ARPANSA. The hambers were irradiated for at least 30 minutes before measurements ommened at the BIPM. The measured urrent was orreted at the BIPM for the leakage urrent of about 0.02 %. No orretion was made for this at the ARPANSA, as the ombined effet of bakground, leakage and bias urrent was typially less than 0.01 %. The unertainty due to this is inluded in the unertainty for ionization hamber measurements. To detet any gross hanges in the transfer hambers during transport, a hek soure of 90 Sr belonging to the ARPANSA (NE2562, serial number 024) was used to irradiate the hambers in a onstant geometry. The onsequent ionization urrents measured at both the ARPANSA and the BIPM were normalized for temperature and pressure, to 50 % relative humidity as desribed above, and for radioative deay to a ommon referene date of 1997-03-15. The half life of 90 Sr was taken as 28.8 years. The mean result for the two hambers, expressed as a urrent ratio R I = I ARPANSA / I BIPM, was 1.0010 [15]. As disussed in [1, 2], variations within a range of about 0.3 % have been noted for 90 Sr measurements with the NE2561 hambers at 10

the ARPANSA. At this level of unertainty, no hanges due to transport were observed and no differene between the measurement systems of the two laboratories ould be identified. Reombination and polarity The value of the reombination orretion fator for the NE 2561 hambers measured and applied at the ARPANSA is 1.0017 [1, 2]. For onsisteny, as initial reombination predominates and is independent of the dose rate, the same value was applied at the BIPM, the measurement onditions being similar. The transfer hambers were used with negative polarity at both laboratories and no orretion fator was applied for the polarity effet. Radial non-uniformity No orretion to the measurements was made at either the ARPANSA or the BIPM for the radial non-uniformity of the beam over the ross-setion of the sensitive volume of the transfer hambers. For NE 2561 hambers this effet is less than 0.02 % at eah laboratory. The transfer hamber orretion fators and their assoiated unertainties are summarized in Table 5. Table 5. Corretions and ombined relative standard unertainties for transfer hamber measurements at the ARPANSA and at the BIPM in their respetive water phantoms Measurement Value applied at the ARPANSA and the BIPM Relative standard unertainty 100 u /x Combined unertainty (1) of the ratio ARPANSA BIPM I BIPM /I ARPANSA Referene distane - 0.01 0.02 0.02 Depth position and orretion 0.9996 (2) 0.03 0.02 0.04 Waterproof sleeve 1.0003 0.01 0.01 Temperature, pressure and humidity orretions 0.03 0.02 0.04 Charge measurement 0.03 0.02 0.04 Reombination 1.0017 0.03 0.03 Radial non-uniformity 1 0.01 0.01 0.01 Deay orretion 0.01 0.01 Combined relative standard unertainty 0.06 0.05 0.07 (1) Unertainties of the orrelated values have been removed. (2) At the ARPANSA, the orretion shown is to the referene depth of 5 g m 2, the ombined unertainty arises from the atual depth set using the gauge (0.02 %), the front window thikness (0.02 %) and the ambient water temperature, 17 C to 23 C (0.01 %). At the BIPM the hamber is plaed at the referene depth so no orretion is needed; the ombined unertainty arises from the phantom window, the water density orretion and the depth position measurement unertainty. 11

4. Results of the omparison Reproduibility of measurements The short-term relative standard deviation of the mean ionization urrent, measured by eah transfer hamber, was 3 10 4 at the ARPANSA (three series eah of 50 measurements of 400 pc for eah hamber) and 2 10 4 at the BIPM (four series eah of 30 measurements of 1000 pc for eah hamber). The differenes in the urrents using the 90 Sr hek soure measured by the ARPANSA before and after alibration at the BIPM are ompatible with these short-term variations as shown in Table 6. Table 6. Stability of the ARPANSA transfer hambers and hek soure measurements I Sr, ARPANSA, 1997-03-15 /pa Ratio Date Chamber 070 Chamber 328 070/328 Marh 1997 23.681 22.645 1.0457 May/June 1997 23.703 22.660 1.0460 Relative differene +0.09 % +0.07 % +0.03 % Table 7 gives the relevant values for the alulation of omparison,, expressed in the form R Dw Dw Dw,ARPANSA N D, w using (6), and the results of the R = N N. (7) Dw,BIPM In the stated unertainty u of R Dw, the orrelated unertainties arising from the use of transfer hambers and from ( µ en ρ) w, have been removed. Table 8 summarizes the unertainty omponents. The omparison result, 1.0024, is taken from the mean value for both transfer hambers, with a ombined standard unertainty of 0.0029. The differene (3 10 4 ) between the results RDw for the two hambers is ompatible with the statistial unertainty (4 10 4 ) of the harge measurements. Table 7. Calibration fators and the results of the omparison Transfer hamber D w,arpansa / (mgy s 1 ) I ARPANSA / pa N Dw,ARPANSA/ (Gy µc 1 ) D w,bipm / (mgy s 1 ) I BIPM / pa N Dw,BIPM/ (Gy µc 1 ) RDw u NE2561 070 4.679 45.751 102.27 4.617 45.260 102.01 1.0025 0.0029 NE2561 328 4.679 45.397 103.07 4.617 44.896 102.84 1.0022 0.0029 mean values 1.0024 0.0029 12

Table 8. Estimated relative standard unertainties of the alibration fator, of the transfer hambers and of Quantity R D w Relative standard unertainty ARPANSA BIPM N Dw, Combined unertainty of R D w 100 u i 100 u j 100 u i 100 u j 100 u i 100 u j Absorbed dose rate to water (Tables 1 and 3) 0.10 0.18 0.20 0.21 0.22 0.18 Transfer hamber measurements (Table 5) 0.06 0.05 0.07 Relative standard unertainties of Relative standard unertainty of N 0.21 0.29 Dw, lab R 0.29 Dw The two transfer hambers were also used in a onurrent omparison of air kerma standards [15], from whih the result was R K = 1.0028. Consequently it is possible to obtain the ratio N D w N K, as shown in Table 9, to hek the onsisteny of the hambers. The values obtained at eah laboratory, orreted for the fators desribed in this report, reflet the onsisteny of the hambers (3 10 4 ), and the differene between the ARPANSA and the BIPM values (3 10 4 ) reflets the differene in the omparison results for absorbed dose to water (1.0024) and air kerma (1.0028). Table 9. Comparison of N D w N K ratios at the ARPANSA and the BIPM Laboratory Transfer hamber N Dw / (Gy µc 1 ) N K / (Gy µc 1 ) N D w N K ARPANSA 070 328 102.27 103.07 93.86 94.62 1.0896 1.0893 BIPM 070 328 102.01 102.84 93.59 94.38 1.0899 1.0896 13

5. Conlusion The ratio of the ARPANSA and BIPM primary standard determinations of absorbed dose to water is 1.0024, with a standard unertainty of 0.0029. The result will be used as the basis for an entry to the BIPM key omparison database and the determination of degrees of equivalene between the ten national metrology institutes (NMIs) whih have finalized suh omparisons. The standard unertainty of the distribution of the results of the BIPM omparisons for these ten NMIs, shown in Figure 1 is 2.5 10 3. Figure 1. International omparison of absorbed dose to water standards 1.020 in 60 Co γ radiation 1.015 Dw, NMI / Dw, BIPM 1.010 1.005 1.000 0.995 0.990 0.985 0.980 ARP BEV BNM ENEA LSDG NIST NMi NPL NRC PTB National Metrology Institute Aknowledgements We would like to thank L H Kotler (ARPANSA) for modelling the ARPANSA 60 Co soure and for the initial Monte-Carlo simulation of the spetrum, and M. Boutillon and J. Miles for their helpful improvements to the text. Referenes [1] Huntley R B, Boas J F and Van der Gaast, H (1998) The 1997 determination of the Australian standards of exposure and absorbed dose at 60 Co, ARL/TR 126, ISSN 0157-1400ARL, Vitoria, Australia, 60 pages. [2] Huntley R B, Wise K N and Boas J F (2000) The Australian Standard of Absorbed Dose, Proeedings of the NPL Calorimetry Workshop, 8-10 Deember 1999, in press. [3] Boutillon M and Perrohe A-M (1993) Ionometri determination of absorbed dose to water for obalt-60 gamma rays, Phys. Med. Biol. 38 439-454. 14

[4] BIPM (1979) Comparaisons d'étalons de dose absorbée, BIPM Comité Consultatif pour les Etalons de Mesure des Rayonnements Ionisants, Setion (I) vol 4 (Paris/ Offilib) p RI(6). [5] Wise K N (2000) Monte Carlo methods used to develop the Australian absorbed dose standard, ARPANSA Tehnial Report ARPANSA/TR (in preparation). [6] Burns J E and Dale J W G (1990) Conversion of absorbed dose alibration from graphite into water, NPL Report RSA (EXT) 7 April 1990 Teddington U.K., 188 pages. [7] Burns J E (1994) Absorbed dose alibrations in high energy photon beams at the National Physial Laboratory: onversion proedure, Phys. Med. Biol. 39 1555-1575. [8] Domen S R and Lamperti P J (1974) A heat-loss ompensated alorimeter: theory, design and performane, J. Res. NBS 78A 595-610. [9] Witzani J, Duftshmid K E, Strahotinsky Ch and Leitner, A (1984) A graphite absorbed dose alorimeter in the quasi-isothermal mode of operation, Metrologia 20 73-79. [10] IAEA (1991), X- and gamma-ray standards for detetor alibration, IAEA-TEC-DOC-619, 157 pages. [11] Pruitt J S and Loevinger R (1982) The photon-fluene saling theorem for Compton sattered radiation, Med. Phys. 9 176-179. [12] Higgins P D, Attix F H, Hubbell J H, Seltzer S M, Berger M J and Sibata C H (1992) Mass energy-transfer and mass energy-absorption oeffiients, inluding in-flight positron annihilation for photon energies 1 kev to 100 MeV, NIST Report NISTIR-4812, Gaithersburg MD, 69 pages. [13] Boutillon M (1996) Measurement onditions used for the alibration of ionization hambers at the BIPM, Rapport BIPM-96/1, 19 pages. [14] BJR/IPEMB (1996) Central axis depth dose data for use in radiotherapy: Brit. J. Radiol. Suppl. 25, 188 pages. [15] Allisy-Roberts P J, Boutillon M, Boas J F and Huntley R B (1998) Comparison of the air kerma standards of the ARPANSA and the BIPM for 60 Co γ rays, Rapport BIPM-98/4, 10 pages. (Otober 2000) 15