Intercomparison of JAERI Torso Phantom lung sets

Transcription

1 Intercomparison of JAERI Torso Phantom lung sets Gary H. Kramer and Barry M. Hauck Human Monitoring Laboratory, Radiation Protection Bureau, 775 Brookfield Road, Ottawa, Ontario K1A 1C1 Canada INTRODCTION The International Atomic Energy Agency (IAEA) sponsored a coordinated research proposal (CRP) in The CRP was to perform an international Intercomparison of In Vivo counting systems using a Reference Asian Phantom (1). The CRP featured a commercially available phantom designed at the Japan Atomic Energy Research Institute (JAERI) to represent the torso of an average Asian male. The JAERI phantom is realistic to better simulate the interaction of low energy photons (< 200 kev) with bone, cartilage, muscle and adipose tissues. The torso plate is constructed of an adipose-muscle substitute mixture and contains synthetic bone, and cartilage. The phantom core contains a full rib cage, spine and scapula in the rear of the torso. The overlay plates are constructed of different adipose-muscle substitute mixtures. The torso cavities contain lungs, heart, liver and other organs. All internal organs but the lungs are constructed of muscle substitute material; the lungs are constructed of lung substitute material. The simulated lungs provided with the phantom were replaced by lungs made specifically for the CRP using the polyurethane-based lung tissue substitute developed at the Lawrence Livermore National Laboratory for use in the JAERI Phantom. Six sets of radioactively labelled lungs were provided: natural thorium (120 Bq), natural uranium (1,160 Bq), uranium with a 3% enrichment (1,550 Bq), Am (488 Bq), and two Pu sets with significantly different radioactivity contents (4,630 and 40,800 Bq). An additional blank set was provided for background measurement. The CRP had nine participants: Australia, Bangladesh, Canada, China, India, Japan, South Korea, Malaysia, and the SA. The IAEA obtained its JAERI phantom from Kyoto Kagaku Company in The Human Monitoring Laboratory (HML), which operates the Canadian National Calibration Reference Centre for In Vivo Monitoring (2), also has a JAERI phantom, which was obtained in 1995, and several lung sets. This paper describes the measurements made on the IAEA JAERI phantom with the IAEA lung sets and the lung sets made for and by the HML to assess their similarity, or differences. METHODS AND MATERIALS The JAERI Phantom and lung sets: The JAERI phantom, as supplied by the manufacturer, consists of a torso core, a chest plate cover, sliced lungs, other internal organs, and six overlay plates. The adipose content of the JAERI phantom is either 10%, 20% or 30% depending on which overlay series is used. The phantoms details are shown in Table 1. The phantom was accompanied by five lung sets that were manufactured for the IAEA by the niversity of Cincinnati. The HML also has five lung sets that were manufactured by Pacific Northwest National Laboratories (PNNL) and three lung sets ( Am, 152 Eu, and -nat) made in-house as described elsewhere (3). The lung sets are summarised in Table 2. Muscle Equivalent Chest Wall Thickness (MEQ-CWT): The MEQ-CWT is the thickness of muscleequivalent-absorber that reduces the photon flux from the lungs by the same amount as the actual combination of muscle and adipose tissue in the chest plate and overlay plates. This method has been successfully used elsewhere (4). The function is shown below: MEQ CWT CWT µ adp AMF µ msc 100 µ msc (100 AMF) 100 (1) Where MEQ-CWT is the muscle equivalent chest wall thickness (mm), CWT is the chest wall thickness (mm), AMF is the adipose mass fraction, µ adp is the linear attenuation coefficient at a given energy for adipose tissue (cm -1 ), and µ msc is the linear attenuation coefficient at a given energy for muscle tissue (cm -1 ). Both µ adp and µ msc are obtained from the literature (5) by multiplying the mass attenuation coefficients by the appropriate density (adipose or muscle). The values in Table 1 were supplied by the IAEA. Germanium lung counter: The lung counting system at the HML consists of four large area germanium 1

2 detectors obtained from EG&G Ortec. Each detector, which is cooled by a 17 liter Dewar, is 70 mm in diameter and 30 mm thick. The beryllium entrance window is 0.5 mm thick. The detectors are housed in a counting chamber constructed of 20 cm thick low background steel. The shield s interior is covered with a lead liner approximately 0.6 cm thick. Spectra acquired from the individual Ge detectors were stored and analyzed using EG&G s GDR software custom modified for the HML. Counting Protocol: The JAERI phantom was measured with each lung set. Detector #1 was placed above the lower portion of the left lung and above the heart (lower left); detector #2 was placed above the upper portion of the left lung (upper left); detector #3 was placed above the lower portion of the right lung (lower right), and detector #4 was placed above the upper portion of the right lung (upper right). The X ray and gamma photon energies of interest were 17.1, 17.5, 20.5, 26.3, 38.7, 43.5, 45.3, 59.5, 63.3, 92.5, 121.8, 129.3, 143.8, 163.3, 185.4, 205.3, 209.4,.6, 244.7, and kev. Each phantom was counted with the chest plate alone and with each overlay plate. Count times were long (3600 to 50,000 seconds) to obtain the best possible counting statistics within a reasonable time. There was not sufficient counting time for the Pu(low) lung set. The range of counting statistics was :-nat 0.7% to 2.2%; - enr 0.4% to 1.4%; 232 Th 1.9% to 3%; Pu(low) 3% to 5%; Pu(high) 0.6% to 1%. RESLTS AND DISCSSION The counting efficiencies for the four-detector germanium array is shown in Table 3 for lung sets measured (PNNL, HML, IAEA) as a function of energy for each overlay plate configuration. Fig. 1 show the counting efficiency as a function of energy for the JAERI torso phantom with no overlay plate. Plots for the other counting configurations follow the same trend and are not shown here. Except for the Am (59.5 kev) and occasionally one of the 232 Th (209 kev) photopeaks, the PNNL and IAEA lung sets show good agreement. The IAEA Am lung set gives a counting efficiency that appears about 25% too high for all overlay plate configurations. This was observed by other participants and no explanation has yet been found for this. It does, however, exemplify that the manufacture of tissue substitute lung sets is still something of a black art. Despite all precautions, this lung set is either inhomogeneous or has had the wrong activity added. Heterogeneity can lead to an error in the activity estimate of a factor of three (6) if the activity was severely localised due to improper mixing. A factor of 1.25, which appears to be the discrepancy, could easily be explained in this way. It will not be known for some time, however, what the true reason is as the participants are still waiting for the destructive analysis of this lung set to determine the true activity. The poor agreement of 232 Th (209 kev) may simply be due to counting statistics despite the fact that this lung set received the most count time (23,000 to 50,000 sec). The counting efficiency appears to be too low when using no overlay plate, CZ10879, 20853, 30826, and Counting efficiency values obtained with overlay plates CZ11577 and CZ31541 are in agreement with the others. An examination of the count data showed no reason why these deviations should have occurred. CZ11577 received the most counts and CZ31541 received the least counts. The kev photopeak from 232 Th gave much better agreement with the other data. The sliced lungs ( Am, 152 Eu, and -nat) manufactured by the HML are also in excellent agreement with the PNNL and IAEA lung sets. This finding is in agreement with work published elsewhere (3). The advantages of sliced lung sets and planar sources are manifold. Activity can be distributed in a known and reproducible manner to mimic either a homogeneous or heterogeneous distribution in the lung. Short lived radionuclides can be used. Cost is much less than purchasing or manufacturing lung sets that have the activity homogeneously distributed throughout the tissue substitute material. Many different sources can be used with a single lung set. Table 4 shows the averages and coefficients of variation for selected (i.e., those that had three or more values) counting efficiency data at different photon energies. The IAEA Am lung set has been eliminated as an outlier from this analysis. The coefficient of variation shows that agreement between lung sets depends on the photon energy. In general, the lower the photon energy the worse the agreement. The 17.7 and 26.4 kev photopeaks show the highest coefficients of variation. This could be due to several factors: the distribution of the radionuclide in the lung insert, the positioning of the detector, the analysis of the photo peak. In the latter case, the HML finds that the analysis software often gives poor fits to the photopeaks requiring an operator assisted analysis. Once the photon energy rises above 60 kev the coefficient of variation is always less than 6 %. This means that at the 2 level one can expect counting efficiencies derived from lung sets that have the activity homogeneously distributed through the tissue substitute material to be within 12% of each other. This range is well within the guidelines outlined in North American performance criteria (7,8). However, at lower photon energies the agreement is only to within 30%, and this may present a problem for lung intercomparison programs that use low energy emitting photons. 2

3 REFERENCES 1 Shirotani T. Realistic torso phantom for calibration on in vivo transuranic-nuclide counting facilities. J. Nucl. Sci. and Tech. 25(11), (1988). 2 Kramer, G.H.; Limson Zamora, M. The Canadian National Calibration Reference Centre for Bioassay and In-Vivo Monitoring: A Program Summary. Health Phys. 67(2), (1994). 3 Kramer, G.H; Hauck, B.M; Lee, T-Y; Chang, S-Y. Comparison of sliced and whole lung sets for the LLNL and JAERI Torso Phantoms using Ge Detectors. Health Phys. 76(5), (1999). 4 Newton D., Wells A.C, Mizushita S. Toohey R.E., Sha J.Y., Jones R., Jefferies S.J, Palmer H.E, Rieksts G., Anderson A.L, Campbell G.W. The Livermore phantom as a calibration standard in the assessment of plutonium in lungs. In: Assessment of radioactive Contamination in Man, IAEA Vienna ; (1985). 5 International Commission on Radiation nits and Measurements. Tissue substitutes in radiation dosimetry and measurement. Bethseda: ICRP; ICRP Report 44 (1989). 6 Kramer, G.H., Burns, L.C. and Yiu, S. Lung Counting: evaluation of uncertainties in lung burden estimation arising from a heterogeneous deposition using Monte Carlo code simulations. Rad. Prot. Dosim. 74(3), (1997). 7 Atomic Energy Control Board. Regulatory Standard: Technical and quality assurance standards for dosimetry services in Canada. Ottawa: Atomic Energy Control Board; Consultative Document S-106 (1998). 8 Health Physics Society. Performance criteria for radiobioassay. McLean: Health Physics Society; HPS N (1996). 3

4 Table 1: The IAEA JAERI phantom s characteristics. Torso + Overlay Adipose content MEQ-CWT (cm) None CZ CZ CZ CZ CZ CZ Table 2: Lung sets used in the intercomparison( PNNL = Pacific Northwest National Laboratory, oc = niversity of Cincinnati, HML = Human Monitoring Laboratory). Numbers in parentheses represent the percentage uncertainty on the activity. Manufacturer Nuclide Activity (Bq) Ref. Date Weight (g) oc 1158 (0.7) 53 (0.7) 18 Jul (0.7) 310 (0.7) 18 Jul Th 120 (0.8) 22 Oct Am 488 (1.0) 11 Jul Pu (0.7) 17 Sept PNNL Am 152 Eu (2.8) (1.1) 20 Nov Pu (2.1) 3 Mar Am (3.5) 4 Mar (0.2) 401 (0.2) 4 Mar (0.3) 4 Mar HML 152 Eu (1.5) 18 May Am (1.0) 27 June (0.2) 467 (0.2) 12 July

5 Table 3: Counting efficiencies (cnt/photon) of the JAERI with lung sets manufactured by PNNL, HML (sliced), and IAEA as a function of energy for no overlay, CZ10879, and CZ Energy No Overlay CZ10879 CZ20853 CZ30826 CZ11577 CZ21559 CZ31541 (kev) x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x

6 Table 4: Average values of the counting efficiency (cnt/photon) and their coefficients of variation (1 ) for selected photopeaks. Energy (kev) no overlay CZ10879 CZ20853 CZ x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x CZ11577 CZ21559 CZ x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x

7 Fig. 1 Counting efficiency of the JAERI phantom with PNNL, HML, and IAEA lung sets as a function of energy using no overlay plate. 7