Radiation Protection Dosimetry (2006), Vol. 121, No. 2, pp. 179–185 doi:10.1093/rpd/ncl011 Advance Access publication 10 March 2006 RADIOLOGICAL CHARACTERISATION OF DISPOSED PHOSPHOGYPSUM IN BRAZIL: EVALUATION OF THE OCCUPATIONAL EXPOSURE AND ENVIRONMENTAL IMPACT A. J. G. Santos1, P. S. C. Silva1, B. P. Mazzilli1,Ã and D. I. T. Fa´varo2 1Instituto de Pesquisas Energe´ticas e Nucleares, Laborato´rio de Radiometria Ambiental, IPEN, Brazil 2Laborato´rio de Ana´lise por Ativaca¸ ˜o, Av. Prof. Lineu Prestes, 2242, Cidade Universita´ria, CEP 05508-000, Sa˜o Paulo, Brazil Phosphogypsum, a waste by-product derived from the wet process production of phosphoric acid, represents a serious problem facing the phosphate industry in Brazil. This by-product (mainly calcium sulphate dihydrate) precipitates during the reaction of sulphuric acid with phosphate rock and is stored at a rate of about 4 Â 106 kg per day on several piles in Cubata˜o, Brazil. Contents of natural radionuclides from thorium and uranium series were measured in Brazilian phosphogypsum samples from disposal piles, using high-resolution gamma ray spectrometry and instrumental neutron activation analysis (NAA). These phosphogypsum piles present a potential threat to the surrounding environment and to the individual occupationally exposed. The results obtained in this study show that radionuclides, although present in relatively high concentrations in phos- phogypsum, do not imply in significant doses for individuals occupationally exposed. The results obtained for the water activity in the monitor wells showed that the run-off of the piles is influenced by the activity present in the piles, giving indication of a possible groundwater contamination. Sediments from rivers in the area of influence of the pile presented higher concentrations of 238U and 232Th when compared with reference values. INTRODUCTION Brazilian regulatory agency (Comissa˜o Nacional de Energia Nuclear—CNEN) has recently publi- Phosphogypsum, a waste by-product derived from shed a regulatory guide concerned on mining and the wet process production of phosphoric acid, milling of natural occurrence radioactive material, represents a serious problem facing the Brazilian which may generate enhanced concentrations of phosphate industry. By taking into account that the radionuclides, under the radiological protection ratio phosphogypsum/P O produced is 5:1, this 2 5 point view (Requisitos de Seguranca¸ e Proteca¸ ˜o industry has produced, since 1991, 390 Â 106 Radiolo´gica para Instalaco¸ ˜es Mı´nero-industriais metric tonnes of phosphogypsum in Brazil(1). This CNEN-NN-4.01)(2). Such activities may include, by-product (mainly calcium sulphate dihydrate) for instance, the mining and processing of ores precipitates during the reaction of sulphuric acid as well as storage of raw material, products, with phosphate rock and is stored at a rate of by-products, residues and wastes containing about 4 Â 106 kg per day on several piles in Cubata˜o, radionuclides of the 238U and 232Th series, simult- Brazil. Two main producers, responsible for these aneously or separately, which may incur undue gypsum piles, are named in this paper as A and C. exposures of members of the public and occupation- Both industries use as raw material an igneous ally exposed. phosphate rock from Catala˜o I carbonatite complex. During the reaction of phosphatic rock with The fresh rock is a phoscorite cut by abundant sulphuric acid the radioactive equilibrium between carbonatitic veins. This complex is located in the 238U, 232Th and their decay products is disrupted State of Goia´s, Central Brazil and belongs to a and the radionuclides migrate according to their group of Mesozoic alkaline rocks bordering the solubility: uranium isotopes form highly soluble Parana´ Basin. The main problem associated with compounds with the ion (PO )3- while Ra isotopes, this material concerns the relatively high levels of 4 210Pb and 210Po, concentrate into phospho- natural uranium and thorium decay series and gypsum(3). Thorium behaviour is dependent upon other impurities, which could have an impact on the chemical reaction and is uniformly distributed the environment and prevent its commercial use. in the two reaction products(4). No Brazilian regulation in terms of radiation Phosphogypsum produced during the process is protection principles has been applied to these filtered off and pumped as slurry to nearby ponds, industries so far. Therefore, no data are available where it stays for a period sufficient to allow com- so far concerning the real occupational and plete deposition. The water is not released to the environmental impact of such industries The environment, since the plants operate in closed circuit. The phosphogypsum waste is then moved ÃCorresponding author: [email protected] to nearby storage areas, the so-called gypsum piles. Ó The Author 2006. Published by Oxford University Press. All rights reserved. For Permissions, please email: [email protected] A. J. G. SANTOS ET AL. Figure 1. Location of industries A and C (comprising pond and piles) and sampling points. Areas 1 and 2 represent phosphogypsum piles from industry A; 4 represents phosphogypsum piles from industry C; 3 represents ponds from industry C; MW ¼ monitor wells; BR ¼ Bugre River and MR ¼ Mogi River. 180 RADIOLOGICAL CHARACTERISATION OF DISPOSED PHOSPHOGYPSUM IN BRAZIL In Figure 1 the location of the two industries energy gamma rays is highly dependent upon sample (comprising pond and piles) and the surrounding composition. The approach used was that sugges- aquatic system are depicted. It can be seen that ted by Cutshall et al.(5) and the absorption correc- there is just one small river, Bugre River (BR), tion was carried out for each sample. Typical lower which is close to gypsum piles from industry C. limits of detection for gamma spectrometry were This river is tributary of Mogi River (MR). 4.5 Bq kgÀ1 for 226Ra, 3.1 Bq kgÀ1 for 228Ra and Besides the obvious waste of potentially valuable 25 Bq kgÀ1 for 210Pb, for a counting time of 50 000 s. by-product, the main problem associated with phos- The 238U and 232Th determination by NAA was phogypsum storage is the potential threat to the carried out by irradiation of 150 mg of each sam- surrounding environment and to the individual ple, for 8 h at a neutron flux of 1012 ncmÀ2sÀ1,at occupationally exposed. This paper is mainly con- Instituto de Pesquisas Energe´ticas e Nucleares cerned with the radionuclide characterisation of (IPEN) research reactor IEA-R1. The induced radio- the phosphogypsum piles from industries A and C, activity was measured with a Ge-hyperpure detector, in Cubata˜o region, and with the evaluation of the Intertechnique, with 2.1 keV resolution for the 1332 consequent environmental and occupational impact. keV 60Co photopeak. The concentration of the The aquatic environment near the disposal area was analysed elements was determined by comparing assessed by measuring natural radionuclide activities activities obtained in the samples with certified ref- in monitor wells, river water and sediment samples in erence materials, Buffalo River Sediment (NIST the vicinity of gypsum pile from industry C. As for SRM 2704) and Soil-7 (IAEA). the individual occupationally exposed, the pathways The aquatic environment near the disposal area considered were internal exposure due to inhalation was assessed by measuring natural radionuclides of radon emanated from phosphogypsum piles and activities in monitor wells, river water and sediment external gamma and beta exposures due to immer- samples in the vicinity of the gypsum pile from sion in the radioactive plume and due to direct irra- industry C. One sediment sample was collected in diation from phosphogypsum. Bugre River (BR1) and nine in Mogi River (MR1– MR9). Water samples were collected in four monitor wells around the gypsum piles from industry C MATERIALS AND METHODS (MW1–MW4), in BR and in MR. The location of Contents of natural radionuclides from thorium and the sampling points is depicted in Figure 1. uranium series (226Ra, 210Pb and 228Ra) were meas- The activity concentration in river sediment ured in phosphogypsum samples from piles, samples, in water samples from monitor wells in using high-resolution gamma spectrometry. the vicinity of the gypsum piles and in soil in the 238U and 232Th characterisation were performed by reference area were determined by gamma spectro- instrumental neutron activation analysis (NAA). metry and NAA, in the same way as already Samples of phosphogypsum were randomly col- described for the analysis of phosphogypsum. lected by coring from piles of industries A and C. Activity concentration of 228Ra, 226Ra and 210Pb Activity concentrations of 226Ra, 228Ra and 210Pb in water samples, collected in the rivers and in the were measured in 33 samples of phosphogypsum reference area, was determined by a radiochemical and in soil from the reference area, by gamma spec- procedure, since the radionuclide content was too trometry with a hyper-pure germanium detector, low to be measured by gamma spectrometry. The EGNC 150-190 R, from Eurisys Measures, with res- procedure consisted of an initial precipitation of Ra 60 olution of 1.8 keV for the 1332 keV Co photopeak, and Pb with 3M H2SO4, dissolution of the precipit- for 4096 channels. The detector was calibrated using ate with nitrilo-tri-acetic acid at basic pH, precipita- natural soil, rock and water spiked with radionuc- tion of Ba(Ra)SO4 with ammonium sulphate and 210 lides certified by Amersham. Samples were packed in precipitation of 30% PbCrO4 with sodium chro- 100 cm3 cans and sealed for about 4 weeks prior to mate. The 226Ra and 228Ra concentrations were the measurement in order to ensure that equilibrium determined by gross alpha and beta counting of the 226 210 has been reached between Ra and its decay pro- Ba(Ra)SO4 precipitate and the Pb concentration ducts of short half-life. The 226Ra activities were through its decay product, 210Bi, by measuring the 210 determined by taking the mean activity of three gross beta activity of the PbCrO4 precipitate.