Home About us Editorial board Search Ahead of print Current issue Archives Submit article Instructions Subscribe Contacts Login 
Home Print this page Email this page Small font size Default font size Increase font size Users Online: 40

 Table of Contents 
Year : 2016  |  Volume : 39  |  Issue : 2  |  Page : 96-100  

Calibration of Phoswich detector for the measurement of natural uranium in lungs

Radiological Safety Division, Electronics Instrumentation and Radiological Safety Group, Indira Gandhi Center for Atomic Research, Kalpakkam, Tamil Nadu, India

Date of Web Publication13-Sep-2016

Correspondence Address:
M Manohari
Radiological Safety Division, Electronics Instrumentation and Radiological Safety Group, Indira Gandhi Center for Atomic Research, Kalpakkam, Tamil Nadu
Login to access the Email id

Source of Support: None, Conflict of Interest: None

DOI: 10.4103/0972-0464.190391

Rights and Permissions

The occupational workers of fuel fabrication and reprocessing facilities at Indira Gandhi Center for Atomic Research (IGCAR) have a potential for internal exposure to natural uranium which is hazardous both in chemical and radiological aspects. Hence, in vivo monitoring of the radiation workers has to be carried out to ensure safe working conditions. In IGCAR, the in vivo monitoring of natural uranium is being carried out using Phoswich-based lung monitor. The measurement and quantification of internal exposure due to natural uranium is done using 63 and 93 keV photons emitted by 234 Th, immediate daughter of 238 U. Realistic anthropomorphic Lawrence Livermore National Laboratory (LLNL) phantom is used for the calibration of the system. As natural uranium-loaded lung set is not available currently in the laboratory, lung set loaded with indigenously prepared natural uranium source capsules were used. Efficiency curve for 238 U was established for varying muscle equivalent-chest wall thickness (MEQ-CWT) and the efficiency values were in the range of 5.604E-03 to 8.601E-03 CPS/Bq. Simulation results of LLNL voxel phantom having uniform lung distribution of natural uranium agreed within 9% with the measured one which is comparable with the error associated with measurement. This confirms that the distribution pattern of 12 capsules in each lung in the given geometry closely resembles the uniform distribution. The efficiency and minimum detectable activity (MDA) values for Indian population were found to be ranging from 1.011E-03 to 7.931E-03 CPS/Bq and 12–18 Bq, respectively. The efficiency value for 238 U established from the measurement using Japan Atomic Energy Research Institute (JAERI) phantom having uniform source distribution agreed with that of LLNL phantom measurement having capsule source distribution for the same MEQ-CWT thickness (33.8 mm) within 3%. This further reaffirms that the adopted capsule distribution is close to uniform distribution.

Keywords: Efficiency values, hole matrix lung set, JAERI, LLNL phantom, natural uranium, Phoswich detector, source capsules

How to cite this article:
Manohari M, Deepu R, Mathiyarasu R, Rajagopal V, Jose M T, Venkatraman B. Calibration of Phoswich detector for the measurement of natural uranium in lungs. Radiat Prot Environ 2016;39:96-100

How to cite this URL:
Manohari M, Deepu R, Mathiyarasu R, Rajagopal V, Jose M T, Venkatraman B. Calibration of Phoswich detector for the measurement of natural uranium in lungs. Radiat Prot Environ [serial online] 2016 [cited 2019 Jun 25];39:96-100. Available from: http://www.rpe.org.in/text.asp?2016/39/2/96/190391

  Introduction Top

Radiation workers in fuel fabrication and reprocessing plants at Indira Gandhi Center for Atomic Research (IGCAR) handle a large quantity of natural uranium in either powder or liquid form. Such workers have a potential for internal exposure to natural uranium, mainly through inhalation. Such internal exposure has to be measured and quantified to ensure that they are within the annual dose limits as per regulatory requirements.

Natural uranium contains 99.28%238 U, 0.72%235 U, and 0.005%234 U, which are alpha emitters. Their alpha activities will be in ratio of 50.57%, 2.3%, and 47.11%, respectively. In freshly separated Uranium, the daughter products which are typically used to measure the uranium content in the environmental samples such as 226 Ra,214 Pb, and 214 Bi will not be present. Hence, quantification of uranium by the measurement of these daughter products is not possible. Furthermore, the high-energy gamma radiation emitted by the immediate daughters such as 234 Pa have low yield (<1%)[1] and lesser efficiency, which results in higher MDA values. Hence, the natural uranium is generally measured using the 63.3 (4.8%), 92.4 (2.8%), and 92.8 (2.8%) keV gammas emitted by 234 Th, the immediate daughter of 238 U.[2] NaI (Tl) detector-based Phoswich lung counting system is utilized for the monitoring of Uranium in the lungs of the workers. For the calibration of Phoswich system for low-energy photon measurements, Lawrence Livermore National Laboratory (LLNL) phantom with activity tagged lung sets is used. At present, there is no natural uranium-loaded lung set available at our laboratory. However, a blank hole matrix lung set with empty source capsules is provided along with the LLNL phantom. The natural uranium source capsules prepared in Radiological Safety Division (RSD) were inserted into the hole matrix lung set of the phantom and was utilized to perform the experimental calibration. The activity of natural uranium in each source capsule is verified using a coaxial High-purity germanium (HPGe) detector-based gamma ray spectrometer. Efficiency curve for 238 U was established for various muscle-to-fat compositions using the different chest wall plates provided with the LLNL phantom. The net counts in the 40 to 120 keV region covering both 63 and 93 keV gammas were used for calculation of the efficiency. Comparison was also done with Monte Carlo N-Particle simulated efficiency values obtained using LLNL voxel phantom having uniform activity in the lungs and measurement made with Japan Atomic Energy Research Institute (JAERI) phantom loaded with uniformly distributed uranium lung set.

  Materials and Methods Top


The Phoswich detector having a 3 mm thick NaI (Tl) crystal sandwiched with a 51 mm thick CsI (Tl) crystal of 203 mm diameter and a 0.5 mm beryllium entrance window along with pulse shape discrimination electronics is used [3] in this study. The detector is housed inside a 200 mm thick, low background steel room (pre-war steel) with graded lining (3mm Pb + 2 mm Cd + 1 mm Cu). The system provides an analysis energy range from 10 to 200 keV, thus having the capability for direct measurements of actinides deposited in the body.

Lawerence Livermore National Laboratory phantom

LLNL phantom is an anthropomorphic phantom representing the torso of an adult North American male with a height of 177 cm and a body weight of 76 kg.[4] The calibration factor of a lung monitoring system depends on the thickness and the composition of the chest wall of the subject/phantom. To simulate variation in the chest wall thickness (CWT) and composition, the phantom is provided with three sets of overlays of CWT ranging from 16 to 36 mm, with varying muscle-to-fat ratio of 13:87 (Set A), 50:50 (Set B), and 100:0 (Set C). The LLNL phantom available at RSD, IGCAR is used for calibration.

The CWT of the phantom is not a constant and it varies over the chest region. Manufacturer has provided CWT in their data sheet only for the region commonly used for HPGe-based lung monitoring system. Since, the Phoswich detector covers larger area than the HPGe-based systems, the physical CWT values at various points (300 locations) in each overlay plate were physically measured using digital vernier caliper. From the measured thicknesses, the effective CWT values, which take the exponential attenuation of gamma photons into account,[5] were calculated using Equation 1.

where Xi is the measured thickness(cm), µL is the linear attenuation coefficient for the materials (cm −1), and N is the number of sampling points on the torso or overlay plate in the area being measured. The “µL” values [6] used for calculation are provided in [Table 1].
Table 1: Linear attenuation coefficient values

Click here to view

The effective CWT values were calculated separately for 63 and 93 keV photons. The Muscle equivalent-CWT (MEQ-CWT) which takes into account both the variation in composition and thickness of the chest wall was calculated [3] for all the overlays and is also provided in [Table 2].
Table 2: Averaged muscle equivalent-chest wall thickness (mm) for 63 and 90 keV photons

Click here to view

Preparation of natural uranium lung set

The LLNL phantom was provided with a blank hole matrix lung set (left and right lungs). The left lung has 13 holes and the right lung has 14 holes. Empty capsules, made of tissue equivalent material, to be inserted into the holes are also provided. [Figure 1]a shows the hole matrix lung set along with source capsules and [Figure 1]b shows the source capsule.
Figure 1: (a) Hole-matrix lung set. (b) Source capsule

Click here to view

Since natural uranium tagged lung set was not available with present LLNL phantom, it was decided to prepare natural uranium source capsules and place them in the hole matrix lung set for calibration. Considering the difficulties such as leakage and contamination associated in handling of liquid sources, the capsules are filled with uniformly dispersed uranium in a solid matrix of polyacrylamide. Uranium standard solution was prepared by dissolving 1.206 g of uranium nitrate in 2 ml of water. A monomer solution of acrylamide was prepared by dissolving 3 g of acrylamide in 2 ml of water. Exactly 100 μl of Uranium solution containing 714 Bq was taken in the capsule and 50 μl of acrylamide monomer solution was added and mixed thoroughly using ultrasonic agitator. The capsule was sealed and then subjected to gamma irradiation (200 mGy) for in situ polymerization [7] of acrylamide at 10°C. Contamination over the capsule was checked using swipe sample. In a similar way, 12 capsules of uranium source with same activity were prepared. The uranium activity in each source capsule was further confirmed by HPGe-based gamma ray spectrometer measurement. All capsules with a total 238 U activity of 4344 ± 130 Bq were loaded in one lung of the phantom.

Calibration with uranium tagged lung set

The 12 source capsules were first loaded in the left lung keeping the right lung inert. Measurements were carried out for all the chest plates of various muscle-to-fat ratios using the Phoswich detector. The measurement geometry used is shown in [Figure 2]. The detector was kept straight, parallel to the chest at its center and outer edge of detector tangential to the suprasternal notch. The distance between the detector and the phantom is kept constant at 1 cm throughout the measurements. The counting time was fixed as 3600 s, so that the error due to counting statistics is <1%. The measurements were repeated with the source capsules loaded into the right lung. The counts obtained for both the lungs in the energy region of 40–120 keV [Table 3] were used to get the efficiency values.
Figure 2: Counting geometry of Lawrence Livermore National Laboratory phantom

Click here to view
Table 3: Measurement details

Click here to view


The validated Phoswich detector was modeled over the LLNL voxel phantom [8] with MEQ-CWT of 16.24 mm and having uniformly distributed uranium lung set in the same geometry as in the measurement [Figure 3]. Spectrum was simulated for the energies 63 and 93 keV. Since the activity level is of the order of 10 kBq, the uranium K-X-rays above 93 keV emitted in the 234 Th decay chain are also included in the simulation. The efficiency was estimated from the simulated spectrum.
Figure 3: Monte Carlo N-Particle simulation – Lawrence Livermore National Laboratory phantom with Phoswich detector

Click here to view

Japan Atomic Energy Research Institute phantom measurement

Earlier, RSD, IGCAR, had access to JAERI phantom, having a uniformly distributed natural uranium lung set, as part of IAEA intercomparison exercise and made measurements for uranium estimation in lungs using the same Phoswich system. The JAERI Torso phantom represents an Asian reference man. The phantom is manufactured using polyurethane and epoxy.[2] It had a CWT of 3.47 cm and the muscle-to-fat ratio is 80:20. The MEQ-CWT for the JAERI phantom is estimated to be 3.38 cm. The efficiency value corresponding to this MEQ-CWT of JAERI phantom is obtained from efficiency calibration curve of LLNL measurement. Both the efficiency values agreed within 3%. This further supports the fact that the present source distribution in the hole matrix lung set closely represents the uniform distribution in the lung.

  Results and Discussion Top

Using the measurements of LLNL phantom with hole matrix lung set, efficiency curve for 238 U as a function of MEQ-CWT is established for the 40–120 keV along with the error bar [Figure 4]. The error was calculated as error propagation with 3% error in the activity and 5% error in the counting statistics. The efficiency (ε) values varied from 5.604E-03 to 8.601E-03 CPS/Bq. The fitted equation is given in Equation 2:
Figure 4: Efficiency curve for Phoswich system for uranium measurements

Click here to view

For the Indian population, the CWT varies from 10 to 25 mm and the muscle-fat composition varies from 9% to 40%.[9] Considering this variation in the Indian population, the efficiency values could vary from 7.931E-03 to 1.011E-2 CPS/Bq for the calculated MEQ-CWT ranging from 5.7 to 20.19 mm while the MDA variation is estimated to be in the range of 12–18 Bq.


The simulated and the measured spectra are compared in [Figure 5]. The small difference seen is due to the PSD of the high-energy photons which was not incorporated in the simulation. The simulated efficiency value of 9.1 CPS/kBq (torso plate alone) corresponding to the MEQ-CWT of 16.24 mm closely matching with experimental value of 8.6 CPS/kBq is observed with hole matrix lung set. This proves that the present activity distribution in the hole matrix lung set gives efficiency closer to that of uniform distribution and can be used for regular calibration of Phoswich system.
Figure 5: Comparison of simulated and measured spectrum

Click here to view

Comparison with Japan Atomic Energy Research Institute phantom

The efficiency value established with JAERI phantom was 6.67E-03 CPS/Bq. The efficiency corresponding to 3.38 cm MEQ-CWT was inferred from the efficiency curve of LLNL phantom to be 6.53E-03 CPS/Bq, which is 3% lesser than JAERI. This close matching of efficiency values further confirms that the present 12 source plug distribution in each lung is equivalent to the uniform distribution for the Phoswich lung counting geometry.

  Conclusions Top

Through this work, natural uranium source capsules suitable for hole matrix lung set of the LLNL phantom have been successfully prepared for calibration purpose at Whole body counting facility, RSD, IGCAR. Phoswich-based lung monitoring system has been calibrated using this lung set and the efficiency curve for various MEQ-CWT established. The close agreement between the simulated and the measured efficiency values confirmed that the presently adopted source distribution in the hole matrix lung set is equivalent to that of uniform distribution. This fact was further verified by comparing with the JAERI phantom measurement.


The authors sincerely thank Dr. S. A. V. Satya Murty, Director, IGCAR, for his support, Shri V. Santhanakrishnan, BARCF, for his technical support. Mr. Kevin Capello, HML, Canada, is also acknowledged for providing LLNL voxel phantom. Authors also acknowledge the technical support of Dr. D. Ponraju, Head, PCS, SED, IGCAR, in preparing uranium sources.

Financial support and sponsorship


Conflicts of interest

There are no conflicts of interest.

  References Top

Clark D. U235: A Gamma Ray Analysis Code for Uranium Isotopic Determination. UCRL-ID-125727, LLNL, University of California; 1996.  Back to cited text no. 1
International Atomic Energy Agency. Inter Calibration of in-vivo Counting Systems Using an Asian Phantom. Results of Co-Ordinate Research Project 1996-1998, IAEA TECDOC-1332; 2003.  Back to cited text no. 2
Manohari M, Mathiyarasu R, Rajagopal V, Meenakshisundaram V, Indira R. Calibration of Phoswich based lung counting system using realistic chest phantom. Radiat Prot Dosimetry 2011;144:427-32.  Back to cited text no. 3
Griffith RV, Anderson AL, Anderson SW. Fabrication of a Set of Torso Phantom for Calibration of Transuranic Nuclide Lung Counting Facilities. 6th International Congress of IRPA, Berlin, West Germany; 7-12 May, 1984.  Back to cited text no. 4
Kramer GH, Hauck BM. Chest wall thickness measurements of the LLNL and JAERI torso phantoms for germanium detector counting. Lawrence Livermore National Laboratory and Japanese Atomic Energy Research Institute. Health Phys 1997;73:831-7.  Back to cited text no. 5
Hubble JH, Seltzer SM. Tables of X-ray Mass Attenuation Coefficients and Mass Energy Absorption Coefficients from 1 keV to 20 MeV for Elements Z = 1 to 92 and 48 Additional Substances of Dosimetric Interest, Report No: NISTIR 5632; 1996.  Back to cited text no. 6
Collinson E, Dainton FS, McNaughton GS. The Polymerization of Acrylamide in Aqueous Solution. Transactions of the Faraday Society 53; 1957. p. 476-88.  Back to cited text no. 7
Manohari M. Simulation of in-vivo Monitors and VOXEL Phantoms for Establishing Calibration Factors, PhD Thesis, HBNI; 2014.  Back to cited text no. 8
Nadar MY, Sing IS, Chaubey A, Kantharia S, Bhati S. Ultrasonic Measurements of Chest Wall Thickness for the Assessment of Internal Contamination Due to Actinides in Indian Radiation Workers. Recent Trends in Radiation Physics Research. Proceedings of NSRP-18; 2009. p. 284-5.  Back to cited text no. 9


  [Figure 1], [Figure 2], [Figure 3], [Figure 4], [Figure 5]

  [Table 1], [Table 2], [Table 3]

This article has been cited by
1 Lung tissue substitute: synthesis, characterization and attenuation studies for low energy photons
Shailesh Joshi,J. S. Brahmaji Rao,K. Sivasubramanian,R. Kumar,V. Jayaraman,B. Venkatraman
Journal of Polymer Research. 2017; 24(5)
[Pubmed] | [DOI]


Similar in PUBMED
   Search Pubmed for
   Search in Google Scholar for
 Related articles
Access Statistics
Email Alert *
Add to My List *
* Registration required (free)

  In this article
Materials and Me...
Results and Disc...
Article Figures
Article Tables

 Article Access Statistics
    PDF Downloaded85    
    Comments [Add]    
    Cited by others 1    

Recommend this journal