On the presence of enriched amounts of 235U in hot particles from the terrestrial area affected by the Palomares accident (Spain)

Depósito de investigación de la Universidad de Sevilla https://idus.us.es/ Esta es la versión aceptada del artículo publicado en: This is an accepted manuscript of a paper published in: Environmental Pollution (2007): 24/01/2025 DOI: https://doi.org/10.1016/j.envpol.2006.08.002 Copyright: © 2006 Elsevier Ltd. All rights reserved. El acceso a la versión publicada del artículo puede requerir la suscripción de la revista. Access to the published version may require subscription. “This is an Accepted Manuscript of an article published by Elsevier in [Environmental Pollution] on [2007], available at https://doi.org/10.1016/j.envpol.2006.08.002” ON THE PRESENCE OF ENRICHED AMOUNTS OF 235U IN HOT PARTICLES FROM THE TERRESTRIAL AREA AFFECTED BY THE PALOMARES ACCIDENT (SPAIN) M.C.Jiménez-Ramos, H.Barros, R.García-Tenorio*, M.García-León, I.Vioque y G.Manjón Applied Nuclear Physics Group University of Seville Spain. ABSTRACT The characterisation by ICP-MS of an isolated Pu-U hot particle originating from the nuclear weapons accident in Palomares (Spain) shows, for the first time, that its uranium content is highly enriched in 235U. The enrichment has been confirmed by independent analyses of two surface soil samples collected in a heavily contaminated area close to the impact point of one of the bombs. This finding clarifies better the composition of the weapons involved in the accident and is of importance when the inventory of U and Pu in the contaminated area are to be calculated. Capsule: The nuclear material released in the Palomares accident contains enriched amounts of 235U Keywords: Palomares accident, hot particle, plutonium, enriched uranium - - - - * To whom correspondence should be addressed Address: Departamento de Física Aplicada II, E.T.S.A., Avenida Reina Mercedes-2, 41012-Sevilla, Spain. Phone: +34-954556625 Fax: +34-954557892, e-mail: [email protected] INTRODUCTION On 17th January 1966, four nuclear weapons were released over the village of Palomares (SE Spain), following an accident involving two US military aircraft. While two of the weapons were recovered intact, the chemical explosives in the other two were triggered on impact on land, causing the dispersion of a large amount of fissile material over an area of approximately 230 hectares. Although different clean-up operations were carried out immediately after the accident in order to reduce/dilute actinide contamination levels (Iranzo et al., 1986; Gutierrez et al., 1993), some of the fissile material dispersed during the accident was not recovered and can still be found in soils from the affected area. A considerable fraction of this residual contamination is present in particulate form, i.e., as hot particles (Jiménez-Ramos et al., 2006). In a recent study carried out by our laboratory, a number of these hot particles were isolated from surface soil samples collected around the impact point of one of the bombs. The isolated particles were identified and characterized using different radiometric (gamma-ray spectrometry) and nonradiometric techniques (scanning electron microscopy, x-ray microanalysis) (Jiménez-Ramos et al., 2006). From these analyses, it was concluded that: (a) Palomares hot particles are variable in size (up to several tens of micrometres) and are generally partially coated by soil material; and (b) particle surfaces are mainly composed of plutonium and uranium. The presence of significant amounts of uranium in particle surfaces originating from the accident prompted us to carry out a more detailed investigation on the isotopic composition of this element in the particles. In this study we focussed on the ICP-MS determinations of 238U/235U atom ratios in an isolated hot particle and contaminated soil samples from Palomares. The ratios obytained unequivocally show the presence of 235U in enriched amounts in the weapons involved. This result, which gives additional information about the composition of the exploded bombs, should not be considered as surprising. Uranium–plutonium hot particles released in Thule (Greenland) in an accident similar to that in Palomares also contained 235U in enriched amounts (Eriksson, 2002). Although thermal ionisation mass spectrometry has generally been the technique of choice for 238U/235U atom ratio measurements due to its high precision, during the last years inductively coupled mass spectrometry (ICP-MS) has become a powerful tool for these determinations thanks to its sensitivity and short measurement time. Nowadays, the determination of 238U and 235U in environmental samples by using conventional quadrupole ICP-MS systems (instruments widely used in environmental studies for trace elements and long-lived radionuclides), is a well-established procedure that allows the determination of the 238U/235U atom ratios in environmental samples (Becker and Dietze, 1999; Uchida et al., 2000 ; Yoshida et al., 2000; Howe et al., 2002). However precise uranium analysis at trace levels (at the nanogram level), as expected in the measurement of isolated particles, is a difficult task since the 238U/235U ratio could easily be biased by naturally occurring uranium from the reagents, the walls of the beakers and/or from the laboratory equipment used (Eriksson, 2002). The 238U/235U atom ratio determination at these low levels therefore demands the use on ultra-pure acids as well as unused clean beakers and filtration systems on one hand and the decontamination of the ICP-MS system before each measurement on the other, as an attempt to prevent, or at least minimize, cross-contamination. MATERIAL AND METHODS Samples analysed A total of 3 samples have been analysed. One sample corresponded to one Palomares hot particle that was isolated from a superficial soil sample, while the remaining two were 0.5 gram aliquot soil samples, collected from the Palomares area clearly affected by the accident. The soils were sampled by laying a frame of 0.25 m2 on the ground and collecting all the soil to a depth of 5 cm. After drying and removal of grass, gravel and stones, physical homogeneisation was performed on the samples. Aliquots of 0.5 grams were sampled and screened on an HPGe detector for determination of high activity 241Am samples since elevated concentrations of 241Am reflect the presence of elevated concentrations of Pu (Jimenez-Ramos et al., 2006). Sample preparation The isolated hot particle and the two soil aliquots were each totally digested with a mixture of HNO3 (65%) and HF (50%) in a microwave oven. After this treatment, the two soil samples were filtered in order to assure the removal of undesired remaining silicates. The sample containing the hot particle was not filtered, but total dissolution was checked by ensuring uniform distribution of 241Am (by gamma measurements) in a number of aliquots of the digested sample. All three samples were reduced in volume by evaporation to near dryness and conditioning into an appropriate medium (2% HNO3) for further analysis. The step of reduction and then conditioning into 2% HNO3 was repeated several times to remove the fluorides that could cause problems in later radiochemistry steps, since Pu and U easily form complexes with F. The solutions corresponding to the three samples were divided into several aliquots: an aliquot from each sample, without any radiochemical separation, was devoted to ICP-MS measurements, while another aliquots, in this case only from the two soil samples, were devoted to the determination of the 239+240Pu concentrations and associated 238Pu/239+240Pu activity ratios by using alpha spectrometry following radiochemical separation. The process of plutonium isolation and determination by alpha spectrometry is described in detail elsewhere (Vioque et al., 2002). The digestion of the samples and the preparation of the solutions for the ICP-MS determinations were performed in a clean laboratory in order to minimize the risk of U cross-contamination. Unused beakers and containers, which were previously acid-washed before use, and acids with ultra pure-quality were used. Blank samples were also measured to evaluate the possible magnitude of the contamination. ICP-MS determinations A quadrupole inductively coupled plasma mass spectrometry system (ICP-MS, Thermoelemental X7 Series) was used for U-isotopes and Pu-isotopes analysis of the selected samples. The system was tuned and optimised for the measurement of high mass isotopes by checking that the sensitivity for masses higher than 235 u was constant for the detector employed. This sensitivity was determined precisely in each set of measurements by measuring a 238U calibration sample. It was also confirmed that the detector dead time did not affect the determination of the 238U/235U atom ratio by measuring a large set of natural (uncontaminated) soil samples. For these natural samples, an average value of 138 ± 1 was obtained for the 238U/235U atom ratio, which compares very well with the expected value of 137.5 ± 0.5 for natural uranium (Meinrath et al., 2003). In addition, the possible isobaric interference in the mass 239, due to the counting of 238UH, was evaluated by measuring a set of 238U calibration samples which covered a wide range of concentrations. 238UH/238U counting ratios of 10-4 were obtained in the established measurement conditions. Taking into consideration that the intention was to measure U in nanogram amounts in some samples, a defined protocol was followed for the ICP-MS measurements, in order to minimize and/or evaluate the possible equipment and lab cross-contamination. Before each Palomares sample measurement, the system was cleaned by injecting 2%HNO3 solution into it, and afterwards a blank sample was analysed. RESULTS AND DISCUSSION The ICP-MS measurements performed on the two superficial soil samples permitted the determination of their 238U, 235U and 239Pu levels (see Table 1). No signals from other U-isotopes (234U, 236U), while looking for additional evidences supporting the presence of enriched uranium, nor from 240Pu were detected, since the concentrations of these radionuclides in the measured aliquots were under the limit of detection of the conventional ICP-MS system used. The 239+240Pu concentrations and associated 238Pu/239+240Pu activity ratios are also shown in Table 1, and were determined in aliquots of the two soil samples using alpha spectrometry following radiochemical separation. The 238U concentrations in blank samples were negligible in comparison with those found in the Palomares soil aliquots, thereby reinforcing our conclusions obtained from the analysis of the results which are detailed below. The two soil samples analysed are clearly contaminated with material from the nuclear weapons which were dispersed during the Palomares accident, as evidenced by the observed 239+240Pu levels which are several orders of magnitude higher than those expected for global weapons test fallout in Europe (Bunzl and Kracke, 1988; Holgye and Figas, 1995) and, particularly in the south of Spain (Chamizo et al., 2006). The 238Pu/239+240Pu activity ratios measured by alpha spectrometry as well as the 240Pu/239Pu atom ratios deduced from the combination of the Pu determinations performed by alpha-spectrometry and by ICP-MS, are also indicative of weapons-grade plutonium (Cooper et al., 2000) and therefore support the argument that the origin of this Pu is the aforementioned mentioned accident, although the 240Pu/239Pu atom ratio value obtained for sample P6 has a relatively high uncertainty due to the difficulties of accurately determining its 239Pu levels by ICP-MS. The 240Pu/239Pu atom ratios deduced in the soil samples are also in agreement with those determined by accelerator mass spectrometry in other Palomares soils heavily affected by the accident (Chamizo et al., 2006). In contrast, measured 238U activities are typical for soils with high percentage of silica and carbonates and low organic matter content as the existing ones in Palomares (UNSCEAR, 1977), even though uranium could represents a considerable proportion (in weight) of the hot particles. This fact is not surprising, given the very low specific activity of this radionuclide. It should be noted that a Pu/U mass ratio in the order of one (as it can be found in the Palomares hot-particles, shown later on in this paper) corresponds to a 239+240Pu/238U activity ratio in the order of 105, i.e., hot particles with 239+240Pu levels of 10 Bq would be expected to contribute to the 238U present in the soils with activity levels in the order of 10–1 mBq. However, the 238U/235U atom ratios determined in the two soil aliquots show a clear signal of 235U enrichment, most probably originating from the accident. The obtained U isotopic ratios are in both cases clearly lower than hose from natural Uranium. In addition, the degree of the 235U contamination in the analysed soil samples is clearly correlated with levels of Pu found in these samples since the lowest 238U/235U atom ratio (72) was found in the soil sample with the highest level of 239+240Pu (12 Bq). This reflects a possible common source of the 239+240Pu and 235U contamination in the soils. In the soil sample P6, about half of the 235U has an anthropogenic origin. The ICP-MS results obtained from the hot particle (M10) are shown in Table 2 (no Pu-isotopes determination by alpha-spectrometry was performed). In this case, and due to the concentrations existing in the analysed solution and the good sensitivity reached in the experiment, in addition to 235U, 238U and 239U, it was also possible to determine also 240Pu with sufficient accuracy. The corresponding ICP-MS mass-spectrum is shown in Figure 1. These results confirm the conclusions obtained from the 0.5 g soil analysis, e.g., the presence of 235U in enriched amounts (238U/235U atom ratio of 5.4 ± 0.2) and of high Pu weapon grade Pu levels ( 23.6 Bq of 239+240Pu, 239Pu/240Pu atom ratio of 0.063 ± 0.003). This is a strong indication that the nuclear weapons involved in the Palomares accident also contained Uranium enriched in 235U combined with weapons grade Pu. The ICP-MS determined 240Pu/239Pu atom ratio in the hot particle is in excellent agreement with the value deduced in the sample P6 and is also in very good agreement with the values obtained by accelerator mass spectrometry in other Palomares soils heavily affected by the accident (Chamizo et al., 2006). However, the determined 238U/235U atom ratio cannot be considered with confidence as totally representative of the nuclear fuel dispersed in the accident because it could be slightly biased by the presence of some soil material coating the hot-particle, and also by the contribution of some 238U cross-contamination which is very difficult to prevent at the levels of U we are measuring. From determinations performed in blank samples, we estimate this contribution to be 5%. The 239Pu/235U atom ratio determined in the hot particle (2.7) can be considered more representative since both nuclides are clearly less sensitive to possible contaminations. CONCLUSIONS ICP-MS determinations performed on one hot particle collected in Palomares (Spain) give definitive evidence of the presence of enriched amounts of 235U in the material forming the nuclear weapons which suffered its conventional chemical explosion in the notorious 1966 aircraft accident. This conclusion was confirmed through the additional determination of the 238U/235U atom ratio in two Palomares soil aliquots which had been heavily contaminated by the accident. ACKNOWLEDGEMENTS This work has been financed through the EU 5th Framework programme, project FIGE-CT20000108 (ADVANCE). The assistance of the staff from Electron Microscopy Central Service, University of Seville, is gratefully acknowledged. REFERENCES Becker J.S., and Dietze H.-J., 1999. Precise isotope ratio measurements for uranium, thorium and plutonium by quadrupole-based inductively coupled mass spectrometry. Frisenius Journal of Analytical Chemistry 364, 482-488. Bunzl K. and Kracke W., 1988. Cumulative deposition of Cs-137, Pu-238, Pu-239+240 and Am241 from global fallout in soils from forest, grassland and arable land in Bavaria (FRG). Journal of Environmental Radioactivity 8, 1-14. Chamizo E., García-León M., Synal H.-A., Suter M. and Wacker L., 2006. Determination of the 240Pu/239Pu atomic ratio in soils from Palomares (Spain) by low-energy accelerator mass spectrometry. Nuclear Instruments and Methods in Physics Research B 249, 768-771. Cooper, L.W., Kelly J.M., Bond L.A., Orlandini K.A. and Grebmeier J.M., 2000. Sources of the transuranic elements plutonium and neptunium in artic marine sediments. Marine Chemistry, 69, 253-276. Eriksson M., 2002. On weapon plutonium in the Artic environment (Thule, Greenland). PhD thesis, Riso, Denmark, Riso R-1321, 150 pp Gutiérrez J., Iranzo C.E., Espinosa A. and Iranzo E. 1993. Spanish experience in intervention at an accidentally contaminated site. Proceedings of the conference: Remediation and restoration of radioactive contaminated sites in Europe. DG-11/DG-12, CE Brussels/Louxemburg. Howe S.E., Davidson Ch.E. and McCartney M., 2002. Determination of uranium concentration and isotopic composition by means of ICP-MS in sequential extracts of sediment from the vicinity of a uranium enrichment plant. Journal of Analytical Atomic Spectrometry 17, 497-501. Holgye Z. and Figas R., 1995. Inventory of Pu-238 and Pu-239+240 in the soil of Czechoslovaquia in 1990. Journal of Environmental Radioactivity 27, 181-199.