Uncertainty evaluation in radon concentration measurement using charcoal canister G. Pantelić a,n , M. Eremić Savković b , M. Živanović a , J. Nikolić a , M. Rajačić a , D. Todorović a a University of Belgrade, Vinča Institute of Nuclear Sciences, Mike Petrovića Alasa 12-14,11001 Belgrade, Serbia b Serbian Institute of Occupational Health "Dr Dragomir Karajović", Deligradska 29, 11001 Belgrade, Serbia HIGHLIGHTS  Measurement uncertainty budget for radon activity concentration established.  Three different methods for ROI selection are used and compared.  Recommend to use one continuous ROI, less sensitive to gamma spectrometry system instabilities. article info Available online 18 November 2013 Keywords: Radon Charcoal Gamma-ray spectrometry Uncertainty abstract Active charcoal detectors are used for testing the concentration of radon in dwellings. The method of measurement is based on radon adsorption on coal and measurement of gamma radiation of radon daughters. The contributions to the final measurement uncertainty are identified, based on the equation for radon activity concentration calculation. Different methods for setting the region of interest for gamma spectrometry of canisters were discussed and evaluated. The obtained radon activity concentra- tion and uncertainties do not depend on peak area determination method. & 2013 Elsevier Ltd. All rights reserved. 1. Introduction Radon levels in houses and in workplaces have been investigated and monitored worldwide very extensively in the last four decades. The charcoal canister method of radon concentration estimation is the most widely used method of screening (Eremić-Savković et al., 2002; Manic et al., 2006; Tondeur et al., 2011; Yu et al., 1992). This method is sufficiently precise and shows significant correlation with results from etch track detectors (Cortina et al., 2008). Almost all published data for radon concentration obtained by charcoal canister measurement are without measurement uncer- tainty. The influence of counting statistics as a major part of the uncertainty was indicated in Zhukovsky et al. (2010) and the total uncertainty of radon concentration measurements conducted by a single charcoal canister was estimated to be between 25% and 35% (k ¼ 2). Many factors can influence the combined uncertainty of radon measurements. One example is the influence of exposure conditions (Ronca-Battista and Gray, 1988). In this paper we identify all the contributions to the total measure- ment uncertainty. An integral part of the charcoal canister method is the gamma spectrometry of canisters. We show that spectral region of interest (ROI) in gamma spectrometry does not influence the measurement uncertainty. 2. Material and methods Active charcoal detectors are used for radon concentration screening in dwellings in accordance with the US EPA protocol 520/5-87-005 (Grey and Windham, 1987). The measurement method is based on radon adsorption on charcoal and measurement of gamma radiation of radon daughters ( 214 Pb and 214 Bi). It is based on the high affinity of activated charcoal for several gases and vapors, one of which is 222 Rn. Activated charcoal is used to construct passive radon charcoal canisters in the following way: a metal can is filled with low-activity activated charcoal. A metal grid, with 30–50% of the total area in holes, is placed over the charcoal. The can is closed with a padded metal lid, and adhesive vinyl tape is used to seal the canister. The canister is exposed by removing the metal lid. The exposure should last no less than 48 h. During the exposure, radon is adsorbed to and desorbed from the activated charcoal, and it also undergoes radioactive decay. After the exposure, the canister is re-sealed by replacing the metal lid and vinyl tape. Measurements are then made Contents lists available at ScienceDirect journal homepage: www.elsevier.com/locate/apradiso Applied Radiation and Isotopes 0969-8043/$ - see front matter & 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.apradiso.2013.11.012 n Corresponding author. Tel.: þ381 113408240; fax: þ381 116308437. E-mail address: pantelic@vinca.rs (G. Pantelić). Applied Radiation and Isotopes 87 (2014) 452–455