Key comparison BIPM.RI(I)-K4 of the absorbed dose to water standards of the METAS, Switzerland and the BIPM in 60 Co gamma radiation C. Kessler 1, D.T. Burns 1, S. Vörös 2 and B. Hofstetter-Boillat 2 1 Bureau International des Poids et Mesures, F-92312 Sèvres Cedex 2 Federal Institute of Metrology (METAS), Bern, Switzerland Abstract An indirect comparison has been made of the standards for absorbed dose to water in 60 Co radiation of the Federal Institute of Metrology (METAS), Switzerland and of the Bureau International des Poids et Mesures (BIPM). The measurements at the BIPM were carried out in June 2013. The comparison result, based on the calibration coefficients for two transfer standards and evaluated as a ratio of the METAS and the BIPM standards for absorbed dose to water, is 1.0001 with a combined standard uncertainty of 5.2 10 3. The results are analysed and presented in terms of degrees of equivalence for entry in the BIPM key comparison database. 1. Introduction An indirect comparison has been made of the standards for absorbed dose to water in 60 Co radiation of the Federal Institute of Metrology (METAS), Switzerland and of the Bureau International des Poids et Mesures (BIPM) to update the previous comparison result of 2000 (Allisy Roberts et al 2003) published in the BIPM.RI(I)-K4 key comparison database (KCDB 2014). The comparison was undertaken using two ionization chambers of the METAS as transfer standards. The measurements at the BIPM were carried out in June 2013 and final results were supplied by the METAS in July 2014. 2. Details of the standards The METAS determination of absorbed dose to water is based on a water calorimeter (Domen 1994). The BIPM primary standard is a parallel-plate graphite cavity ionization chamber positioned at the reference depth in a water phantom (Boutillon et al 1993). 3. Determination of the absorbed dose to water At the BIPM the absorbed dose to water rate is determined using a cavity ionization chamber, with measuring volume V, by the relation I W D a w, BIPM ( µ en ) w,c sc, aπk i, (1) V e where a I W a a is the density of air under reference conditions, is the ionization current measured by the standard, is the mean energy expended in dry air per ion pair formed, 1/10
e is the electronic charge, (µ en / ) w,c is the ratio of the mean mass energy-absorption coefficients for water and graphite s c,a k i is the ratio of the mean mass stopping powers for graphite and air, and is the product of the correction factors to be applied to the standard. The values for the physical constants, the correction factors entering in equation (1), the volume of the BIPM primary standard and the associated uncertainties for the standard are given in Table 1 (Allisy-Roberts et al 2011). Table 1. Physical constants, correction factors and relative standard uncertainties for the BIPM ionometric standard for absorbed dose to water Symbol Parameter / unit Value Physical constants Relative standard uncertainty (1) 100 s i 100 u i a dry air density (0 C, 101.325 kpa) / kg m 3 1.2930 0.01 ( en / ) w,c ratio of mass energy-absorption coefficients 1.1125 (2) 0.01 (2) 0.14 (2) s c,a ratio of mass stopping powers 1.0030 0.11 (3) W a /e mean energy per charge / J C 1 33.97 Correction factors k p fluence perturbation (includes ( en / ) w,c ) 1.1107 (2) 0.05 (2) 0.17 (2) k ps polythene envelope of the chamber 0.9994 0.01 0.01 k pf front face of the phantom 0.9996 0.01 k rn radial non-uniformity 1.0056 0.01 0.03 k s saturation 1.0017 0.01 0.01 k h humidity 0.9970 0.03 Measurement of I / effective volume of CH4-1/ cm 3 6.8810 0.19 0.03 I ionization current (T, P, air compressibility) 0.02 short-term reproducibility (including positioning and current measurement) 0.02 Combined uncertainty of the BIPM determination of absorbed dose to water rate quadratic summation 0.20 0.21 combined relative standard uncertainty 0.29 (1) expressed as one standard deviation. s i represents the relative uncertainty estimated by statistical methods, type A u i represents the relative uncertainty estimated by other methods, type B. (2) the ratio (µ en / ) w,c is included in k p and likewise its uncertainty is included in the uncertainty for k p. (3) uncertainty value for the product s c,a W/e. At the METAS, absorbed dose to water is based on a water calorimeter, described by Medin et al (1999). The detector is a glass vessel filled with ultra-pure water, in which 2 NTC probes measure the temperature increase due to the absorbed radiation energy, positioned in a thermally isolated cubic water phantom of 30 cm side length. The absorbed dose to water is determined using the equation D w, METAS C p T k i (2) 2/10
where C p ΔT k i is the specific heat capacity at constant pressure, is the measured temperature increase, and is the product of the correction factors to be applied to the standard. The relative standard uncertainty of the METAS absorbed dose to water determination is given in Table 2. Table 2. Relative standard uncertainties for the METAS standard for absorbed dose to water Symbol Parameter / unit Value Relative standard uncertainty (1) 100 s i 100 u i Water calorimetry Thermistor calibration - 0.20 Thermistor self-heating / K W 1 2 10 3 0.05 Calibration of bridge sensitivity - <0.01 <0.01 Absolute temperature from PT100 - <0.01 0.01 Irradiation duration ( 60 Co) - <0.01 <0.01 k p Field perturbation ~1.0030 0.05 k rho Density 1.0000 0.07 k c Conductive heat flow 0.9980 0.15 k v Convection 1 k t Transient thermistor response 1.0000 <0.01 Depth in water - 0.03 Lateral position - 0.01 Field size - 0.01 k HD Heat defect 1.0000 0.30 C w Specific heat capacity of water / J kg 1 K 1 4204.84 0.05 Reproducibility 0.02 Combined uncertainty of the METAS determination of absorbed dose to water rate quadratic summation 0.02 0.41 combined relative standard uncertainty 0.41 (1) expressed as one standard deviation. s i represents the relative uncertainty estimated by statistical methods, type A u i represents the relative uncertainty estimated by other methods, type B. The correction factors, k i, applied in (2) are the following: k p k c is the effect of the perturbation of the radiation field due to the presence of the glass vessel and NTC probes; it was measured for each different vessel using a diode located at the reference depth, corrects for conductive heat flow due to the presence of glass, which has a different heat capacity from that of water, and to the presence of temperature gradients due to non-uniformities in the absorbed dose distribution; it was calculated using finite elements method software, 3/10
k HD is the heat defect due to endothermic or exothermic chemical reactions in the water, which has been calculated using various models to be zero in pure water saturated with H 2, N 2 or Ar (Klassen and Ross 1991; Klassen and Ross 1997). The measurement conditions are chosen in such a way that the remaining factors are taken to be strictly unity, or 1.0000 with the uncertainty stated in Table 2: k rho is the effect of the different water densities in calorimetric measurements at 4 C and in secondary standard measurements at 20 C (the depth of measurement being determined in the units g cm 2 at both temperatures and hence no additional correction is needed), k v corrects for convective heat flow due to the temperature gradients in the water (convection does not take place when measuring at 4 C, the temperature of maximum water density), k t is the effect of transient behaviour of the thermistor just after the end of irradiation, which might affect the post-irradiation drift curve (data during the first 20 s after irradiation are discarded from the fit to avoid this effect). Reference conditions Absorbed dose is determined at the BIPM under reference conditions defined by the Consultative Committee for Ionizing Radiation (CCEMRI 1985): the distance from the source to the reference plane (centre of the detector) is 1 m; the beam size in air at the reference plane is 10 cm 10 cm, the photon fluence rate at the centre of each side of the square being 50 % of the photon fluence rate at the centre of the square; and the reference depth in the water phantom is 5 g cm 2. The reference conditions at the METAS are the same as those at the BIPM. The transfer standards used in this comparison were measured in the same water phantom as that used for water calorimetry, a cubic tank of 30 cm side length. The source to reference distance is 100 cm with the depth to the reference point in the tank set to 5 g cm 2, including the Polystyrene window of 2.6 mm thickness as a water-equivalent thickness in g cm 2. The field size at this reference point is 10 cm x 10 cm. There is 10 cm of Styrofoam insulation in the beam path outside the tank that is not included in the 5 g cm 2, but is the same for reference and transfer measurements. Reference values The BIPM reference absorbed dose to water rate D w, BIPM is taken as the mean of the four measurements made around the period of the comparison, corrected to the reference date of 2013-01-01, 0 h UTC, as is the ionization current of the transfer chambers. The half-life of 60 Co is taken as 1925.21 days (u = 0.29 days) (Bé et al 2006). The value of D w, METAS used for the comparison was obtained as the mean of calorimetric measurements made between December 2012 and August 2013 with the water calorimeter, all calorimetric and ionometric measurements corrected to the reference date 2014-01-01. The half-life is also taken as 1925.21 days (u = 0.29 days). 4. Experimental method The comparison of the METAS and BIPM standards was made indirectly using the calibration coefficients N for two transfer chambers given by Dw 4/10
, (3) ND w, lab Dw,lab Ilab where D w, lab is the water absorbed dose rate at the METAS or the BIPM and I lab is the corresponding ionization current for a transfer chamber measured under reference conditions at that laboratory. The experimental method for measurements at the BIPM is described by Allisy-Roberts et al (2011); the essential details for the determination of the calibration coefficients N Dw for the transfer chambers are reproduced here. The ionization chambers NE 2571 and NE 2611, serial numbers 2807 and 147, respectively, both belonging to the METAS, were used as transfer chambers for this comparison. Their main characteristics are listed in Table 3. These chambers were calibrated twice (in May 2013 and January 2014) at the METAS. Each chamber was inserted in its own PMMA waterproof sleeve 1 mm thick, provided by the METAS. Table 3. Characteristics of the METAS transfer chambers Characteristic/Nominal values NE 2571 NE 2611 Dimensions Inner diameter / mm 6.4 7.4 Wall mass thickness / g cm 2 0.065 0.090 Cavity length / mm 24.1 9.2 Tip to reference point / mm 13.0 5.0 Electrode Length / mm 21.0 6.4 Diameter / mm 1.0 1.7 Air cavity Volume / cm 3 0.6 0.3 Wall Material graphite graphite Voltage applied / V Density / g cm 3 1.7 1.7 Negative polarity to outer electrode at the BIPM Positive polarity to collector at the METAS 250 250 Positioning At each laboratory the chambers were positioned with the stem perpendicular to the beam direction and with the same orientation (line and text on the stem of the chambers and line on the waterproof sleeves facing the source). Applied voltage and polarity A collecting voltage as indicated in Table 3 was applied to each chamber at least 30 min before any measurements were made. No polarity correction was made as both laboratories apply the polarity in an equivalent way. Recombination loss Volume recombination is negligible at a dose rate of less than 20 mgy s 1 for these chamber types at this polarizing voltage, and the initial recombination loss will be the same in the two laboratories. Consequently, no correction for recombination was applied. Radial non-uniformity correction At the METAS, the correction for the radial non-uniformity of the beam over the section of the transfer chambers, relative to the section of the water calorimeter, is taken as 0.9980 and 0.9983 for the NE 2571 and NE 2611, respectively, with an uncertainty of 5 10 4. At the 5/10
BIPM, the corresponding corrections applied to the ionization current for the transfer chambers NE 2571 and NE 2611 are 1.0008 and 1.0002, respectively, with an uncertainty of 2 10 4. Phantom window and waterproof sleeve Both laboratories use a horizontal radiation beam and the thickness of the PMMA (BIPM) and Polystyrene (METAS) front window of the water phantom is included as a water-equivalent thickness in g cm 2 when positioning the chamber. In addition, the BIPM applies a correction factor k pf (0.9996) that accounts for the non-equivalence to water of the PMMA in terms of interaction coefficients. Two waterproof sleeves of PMMA were supplied by the METAS for use with the chambers. The same sleeves were used at both laboratories and, consequently, no correction for the influence of the sleeve was necessary at either laboratory. Charge and leakage measurements The charge collected for each transfer chamber was measured using a Keithley electrometer, model 642 at the BIPM and model 6517 at the METAS. The chambers were pre-irradiated for at least 30 min ( 10 Gy) at the BIPM before any measurements were made. At the METAS, the pre-irradiation lasted only 10 min to achieve the same 10 Gy. The ionization current measured for each transfer chamber was corrected for the leakage current; at both laboratories, this correction was less than 4 10 4 in relative value. Ambient conditions During a series of measurements, the water temperature is measured for each current measurement and was stable to better than 0.01 C at the BIPM and to better than 0.05 C at the METAS. The ionization current is corrected to the reference conditions of 293.15 K and 101.325 kpa at both laboratories. Relative humidity is controlled at (50 5) % both at the BIPM and at the METAS. Consequently, no correction for humidity is applied to the transfer chamber ionization current measured at either laboratory. 5. Results of the comparison The NE transfer chambers were set-up and measured in the BIPM 60 Co beam on three separate occasions. The reproducibility of the results was around 1 10 4. At the METAS, the transfer chambers were measured in May 2013, just before sending them to the BIPM and in August 2013. On the first occasion, 5 series of measurements were performed, and 3 series on the second occasion, each series consisting of 10 leakage current measurements followed by 25 measurements during irradiation and finally 10 more leakage current measurements, each measurement lasting for 60 s. The reproducibility of the results over the 8 months separating the METAS measurements was around 1 10 3. The result of the comparison, R Dw Dw, is expressed in the form R N (4) D METAS, BIPM, w, N Dw in which the average value of measurements made at the METAS before and after those made at the BIPM is compared with the mean of the measurements made at the BIPM. The results for each chamber are presented in Table 4. Contributions to the relative standard uncertainty of combined standard uncertainty u c for the comparison result N D w, lab are listed in Table 5 and the R D is presented in Table 6. w 6/10
Table 4. Final result of the METAS/BIPM comparison of standards for 60 Co absorbed dose to water Transfer chamber N D w, METAS / Gy µc 1 N D w, BIPM R / Gy µc 1 Dw u c NE 2571-2807 45.306 45.272 0.9999 0.0052 NE 2611-147 103.99 103.96 1.0003 0.0052 Mean values 1.0001 0.0052 Table 5. Uncertainties associated with the calibration of the transfer chambers BIPM METAS Relative standard uncertainty 100 s i 100 u i 100 s i 100 u i Absorbed dose to water rate 0.20 0.21 0.02 0.41 Ionization current for the transfer chambers 0.01 0.02 (1) <0.01 0.06 Temperature <0.01 0.03 Pressure <0.01 0.04 Humidity 0.03 Distance 0.02 Lateral position 0.03 Depth in water 0.02 0.06 0.09 Field size 0.03 Short-term stability 0.01 <0.01 Recombination loss k s,tr 0.09 Radial non-uniformity 0.02 0.05 Relative standard uncertainties of N D w, lab Quadratic summation 0.20 0.22 0.02 0.44 Combined uncertainty 0.30 0.44 (1) includes the uncertainty of the BIPM temperature and pressure normalization. Table 6. Uncertainties associated with the comparison result Relative standard uncertainty 100 s i 100 u i N N 0.20 0.48 (1) Dw, METAS Dw, BIPM Relative standard uncertainties of (1) R D, w 0.20 0.48 Takes account of correlation in type B uncertainties (k s and k h ) u c = 0.52 7/10
The comparison result is taken as the unweighted mean of the results for the two transfer chambers, R D, w = 1.0001, with a combined standard uncertainty for the comparison of 0.0052 demonstrating the agreement between the two standards for absorbed dose to water. 6. Degrees of equivalence Following a decision of the CCRI, the BIPM determination of the dosimetric quantity, here D w,bipm, is taken as the key comparison reference value (KCRV) (Allisy-Roberts et al 2009). It follows that for each NMI i having a BIPM comparison result x i with combined standard uncertainty u i, the degree of equivalence with respect to the reference value is the relative difference D i = (D wi D w,bipmi )/ D w,bipmi = x i 1 and its expanded uncertainty U i = 2 u i. The results for D i and U i are usually expressed in mgy/gy. Table 7 gives the values for D i and U i for each NMI, i, taken from the KCDB of the CIPM MRA (1999) and this report. These data are presented graphically in Figure 1. Table 7. Degrees of equivalence For each laboratory i, the degree of equivalence with respect to the key comparison reference value is the difference D i and its expanded uncertainty U i. Tables formatted as they appear in the BIPM key comparison database BIPM.RI(I)-K4 - SIM.RI(I)-K4 (2002) - EUROMET.RI(I)-K4 (2005 to 2008) D i U i D i U i Lab i / (mgy/gy) Lab i / (mgy/gy) MKEH -1.7 9.6 CIEMAT -4.9 7.3 PTB -3.9 7.4 CMI -4.0 23.6 VSL -7.4 9.8 RMTC -5.3 12.0 ENEA -0.1 8.8 SSM -1.4 10.0 NPL -2.0 12.8 STUK -3.9 8.5 BEV -0.4 8.8 NRPA 3.2 8.8 VNIIFTRI -2.4 8.6 SMU -4.7 24.7 NRC -2.0 10.4 IAEA -0.4 10.0 NMIJ -4.0 9.2 HIRCL 3.0 12.4 ARPANSA -2.7 10.6 ITN -7.1 13.0 LNE-LNHB -2.9 7.8 NIST -0.6 11.1 METAS 0.1 10.4 LNMRI 1.0 15.0 CNEA 12.0 17.9 ININ 3.9 23.0 When required, the degree of equivalence between two laboratories i and j can be evaluated as the difference D ij = D i D j = x i x j and its expanded uncertainty U ij = 2 u ij, both expressed in mgy/gy. In evaluating u ij, account should be taken of correlation between u i and u j Following the advice of the CCRI(I) in 2011, results for D ij and U ij are no longer published in the KCDB. 8/10
MKEH PTB VSL ENEA NPL BEV VNIIFTRI NRC NMIJ ARPANSA LNE-LNHB METAS ININ CIEMAT CMI RMTC SSM STUK NRPA SMU IAEA HIRCL ITN NIST LNMRI CNEA Metrologia 52 (2015) Tech. Suppl. 06002 Figure 1. 35 Graph of the degrees of equivalence with the KCRV BIPM.RI(I)-K4, 2002 SIM.RI(I)-K4 and 2005 to 2008 EUROMET.RI(I)-K4 Degrees of equivalence for absorbed dose to water 25 15 Di / (mgy / Gy) 5-5 -15-25 -35 N.B. Black squares indicate results that are more than 10 years old. 7. Conclusions A key comparison has been carried out between the METAS and the BIPM standards for absorbed dose to water in 60 Co gamma rays, using two ionization chambers as transfer instruments. The comparison result is evaluated as the ratio of the calibration coefficients measured by the METAS and the BIPM. The comparison result is 1.0001 (52) and so the METAS standard is in agreement with the KCRV within the standard uncertainty for the comparison. The result of the comparison made in 2000 was 0.9999 (54), in good agreement with the present result. When compared with the results for the other laboratories that have carried out comparisons in terms of absorbed dose to water at the BIPM, the METAS standard for absorbed dose to water is in good agreement. Note that the data presented in the tables, while correct at the time of publication of the present report, become out of date as laboratories make new comparisons with the BIPM. The formal results under the CIPM MRA are those available in the BIPM key comparison database. 9/10
References Allisy-Roberts P J, Burns D T and Stucki G 2003 Comparison of the standards for absorbed dose to water of the METAS and the BIPM for 60 Co gamma radiation Rapport BIPM- 03/02 Allisy-Roberts P J, Burns D T and Kessler C 2011 Measuring conditions and uncertainties for the comparison and calibration of national dosimetric standards at the BIPM Rapport BIPM-11/04 Allisy P J, Burns D T and Andreo P 2009 International framework of traceability for radiation dosimetry quantities Metrologia, 2009, 46(2), S1-S8. Bé M M, Chisté V, Dulieu C, Browne E, Baglin C, Chechev V, Kuzmenco N, Helmer R, Kondev F, MacMahon D and Lee K B Table of Radionuclides (Vol. 3 A = 3 to 244) Monographie BIPM-5 Boutillon M and Perroche A-M 1993 Ionometric determination of absorbed dose to water for cobalt-60 gamma rays Phys. Med. Biol. 38 439-454 CCEMRI 1985 Constantes physiques pour les étalons de mesure de rayonnement BIPM Com. Cons. Etalons Mes. Ray. Ionisants (Section I) 11 R45 (Paris : Offilib) CIPM MRA: Mutual recognition of national measurement standards and of calibration and measurement certificates issued by national metrology institutes, International Committee for Weights and Measures, 1999, 45 pp http://www.bipm.org/en/cipm-mra/cipm-mradocuments/cipm-mra-online.html. Domen S R 1994 A sealed water calorimeter for measuring absorbed dose J. Res. Nat. Bur. Stand. 99 121-141 KCDB 2014 BIPM key comparison database 60 Co absorbed dose to water comparisons BIPM.RI(I)-K4 Klassen N V and Ross C K 1991 Absorbed dose calorimetry using various aqueous solutions Radiat. Phys. Chem. 38 95-104 Klassen N V and Ross C K 1997 Water calorimetry: The heat defect J. Res. Natl. Inst. Stand. Technol. 102 63-74 Medin J, Seuntjens J, Klassen N, Ross C K and Stucki G 1999 The OFMET sealed water calorimeter Proceedings of NPL Workshop on Recent advances in Calorimetric Absorbed Dose Standards NPL Report CIRM 42 10/10