PIXE analysis of U and Pu from hot particles: K-lines vs L-lines

Accepted Manuscript PIXE analysis of U and Pu from hot particles: K-lines vs L-lines M.C. Jiménez-Ramos, J. García López, M. Eriksson, J. Jernstrom, R. GarcíaTenorio PII: S0168-583X(11)00698-7 DOI: 10.1016/j.nimb.2011.07.054 Reference: NIMB 58258 To appear in: Nucl. Instr. and Meth. in Phys. Res. B Please cite this article as: M.C. Jiménez-Ramos, J. García López, M. Eriksson, J. Jernstrom, R. García-Tenorio, PIXE analysis of U and Pu from hot particles: K-lines vs L-lines, Nucl. Instr. and Meth. in Phys. Res. B (2011), doi: 10.1016/j.nimb.2011.07.054 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. PROGRAM ID: O24 1 PIXE analysis of U and Pu from hot particles: K-lines vs L-lines M.C. Jiménez-Ramos (a,b), J. García López (a,b), M. Eriksson (c), J. Jernstrom (d) and R. GarcíaTenorio (a,b) (a)Centro Nacional de Aceleradores, Universidad de Sevilla. Thomas Alva Edison 7, E-41092 Sevilla, Spain. [email protected]; [email protected]; [email protected] (b) Applied Nuclear Physics Research Group, University of Seville, Av. Reina Mercedes, 41012 Sevilla, Spain. (c) IAEA-MEL, 4 Quai Antoine 1er, MC 98000, Monaco. M.Erikss[email protected] (d) Radiation Research Division, Risoe National Laboratory for Sustainable Energy, Technical University of Denmark (DTU), DK-4000 Roskilde, Denmark. [email protected] Abstract Transuranic nuclides have been released into the environment since the beginning of the nuclear age. In many of the areas so contaminated, a significant fraction of the plutonium appears under the form of radioactive particles (also called ‘hot’ particles). In previous works [1, 2], radioactive particles from aircraft accidents in Palomares and Thule have been characterized. Using micro-Proton Induced X-ray Emission (µ-PIXE) and confocal X-ray fluorescence microprobe (µ-XRF), the L-lines of the main components in these particles, U and Pu, were analyzed. In this work, the key goal is the possibility to determine elemental ratios and concentration of transuranic elements in hot particles from the same areas but using their K-lines, because of their advantages in comparison with the study of L-lines. Indeed, the U and Pu K-lines present not overlapping between the main peaks in the spectra and the X-ray absorption in the sample is much lower than for L-lines. PROGRAM ID: O24 2 Keywords: Hot particles, K-lines, PIXE, RBS Introduction Plutonium was synthesized for the first time in 1940. Since then, contamination from plutonium has resulted from a number of above-ground nuclear tests and nuclear incidents, including military accidents where nuclear weapons have burned. In many of the areas so contaminated, a significant fraction of the plutonium appears under the form of radioactive particles (also called ‘hot’ particles). These hot particles can be defined as localized aggregates of material containing radioactive atoms, having diameters larger than 0.45 µm, that give rise to an inhomogeneous distribution of radionuclides significantly different from that of the bulk matrix. Characterization of the elemental composition and knowledge about the Pu distribution in these particles can provide useful information to asses their radioecology impacts. In previous works [1, 2], radioactive particles from Palomares and Thule have been characterized by applying microscopic nuclear techniques. In both places, following an aircraft accident some thermonuclear weapons were released, with the transuranic contamination of these areas in the form of hot particles and aerosols as main consequence. The experiments were carried out with the 3 MV Tandem accelerator of the National Accelerator Centre (CNA) in Seville and the ANKA synchrotron facility in Karlsruhe. MicroProton Induced X-ray Emission (µ-PIXE) and confocal X-ray fluorescence microprobe (µXRF) were used to analyze the L-lines of the main components in these particles, U and Pu. In this work, hot particles from Palomares and Thule were analyzed by PIXE, focusing on the possibility to determine elemental ratios and concentration of transuranic elements using their K-lines, because of their advantages in comparison with the study of L-lines. Indeed, the U and Pu K-lines present not overlapping between the main peaks in the spectra and the X-ray PROGRAM ID: O24 3 absorption in the sample is much lower. High Energy PIXE (HE-PIXE) using 18 MeV protons from our Cyclotron and up to 6 MeV protons from our Tandem was employed to get a high X-ray production cross section for U and Pu K-lines, together with a LE(Ge) detector to have good efficiency in the 100 keV region. Due to the increase of nuclear reactions with proton energy, a study of the signal to background ratio was performed to find the optimal experimental conditions. For small particles (about 100 µm), the RBS spectra were recorded simultaneously to obtain further information about the particle size and composition. In that case, a microprobe was used to acquire transuraniuc elemental maps using the PIXE and RBS signals. Experimental Tandem accelerator µ-PIXE measurements have been carried out using the 3 MV tandem accelerator of the National Accelerator Centre (CNA) in Seville, which is described in detail elsewhere [3]. The microprobe focussing system and the data acquisition set-up are based on an Oxford Microbeams end station OM2000 and on the Oxford Microbeams DAQ system [4], respectively. The samples, tilted 45º, were irradiated with a 4, 5 and 6 MeV proton beam with 4x4 µm2 of size and beam currents up to 1 nA. A LEGe detector, with 50 mm2 active area, 25 µm thick Be window and placed at 90º was used, together with a Canberra 2020 amplifier, for µ-PIXE analysis. In order to reduce the background signal due to the pile-up from the Pu and U L-lines and from the light elements that form the soil which recover the particle, and also to eliminate the noise created by the backscattered protons that could reach the detector, the measurements were perform using a 3 mm thick Al filter. A light tight Si detector (Bu-CAM300) placed at 143º in Cornell geometry was employed for the simultaneous RBS measurements. A 3 mm diameter collimator was placed in front of the particle detector to avoid large kinematics spread. PROGRAM ID: O24 4 The elemental maps for U and Pu were determined from the intensities of Kα-lines of these two elements and also from the U+Pu RBS signal. Pu and U peak areas from PIXE spectra have been obtained with the GUPIX software [5] and they were corrected by the corresponding cross-section and detector efficiency (ε). The X-ray production cross sections have been calculated from the values of the ionization cross sections given by the VIBA-LAB software [6] (extrapolating for the Pu element) and the corresponding fluorescence yields (wK (U)= 0.972, wK (Pu)= 0.973, [7] ). The factor correction εU/εPu = 1.05, due to the different detector efficiency for Kα1 lines of both elements, has been obtained using the GUCSA program [5] from the detector parameters. Cyclotron HE-PIXE measurements using 18 MeV protons from our Cyclone 18/9 accelerator have been carried out to study macroscopic particles (> 1 mm). Our beam transport line is mounted in a vault adjacent to the cyclotron hall whose concrete walls are 2 m thick. Along the external beamline several elements are installed, as quadrupoles, steerers, graphite variable slits and Faraday cups, to control the final beam size (~1 mm) and current (~ pA). The PIXE spectra were recorded using the same LEGe detector and filter described above, placed at 135º. It is important to remark that the PIXE measurements were performed in air and this point has turned out to be essential to reduce the background signal. Indeed, the limits of detection in vacuum for U and Pu are one order of magnitude higher in comparison with the measurements performed in air, probably because of the increase of Compton gamma rays that reach the detector after hitting against the vacuum chamber walls. Results In this study of hot particles, the main parameters to take into account are the proton range, the values for the X-ray production cross section and the photon autoabsorption. PROGRAM ID: O24 5 Table 1 shows the transmission for Pu and U L and K-lines from different thicknesses taking composition and density from reference [1].Although in previous works [1,2,8,9], the relative concentration of U and Pu was found through the use of their L-lines, it is obvious from Table 1 that, when the particle size is bigger than 200 µm, problems with the autoabsorption of the L X-ray appear, therefore, the analysis of K-lines is essential in order to obtain information from the whole particle. Fig. 1 shows the variation of U L-lines and K-lines ionization cross section with proton energy. As seen, the K-lines cross sections are three orders of magnitude lower than for Llines, and to get significant U and Pu K-lines production, high energy protons are necessary. In addition, the use of high energies is critical to increase the depth of analysis and therefore, to provide information representative from the whole particle. The ranges for protons at 18, 6 and 4 MeV in this kind of samples are around 3.2, 0.5 and 0.3 mm respectively. However, it is important to notice that not always the higher energy available is the best choice to characterize hot particles, as the limit of detection depends on different parameters, as the background level. In fact, as observed in Fig. 2, which shows PIXE spectra for sample HP1_1 recorded at 6 and 18 MeV, the signal to background ratio achieved with protons at 6 MeV is 3 times better in the photon range between 90 and 120 keV. The quantitative results obtained from three hot particles originated from the two bombs released in Palomares are shown in Table 2. At 6 MeV the Pu/U relative concentration obtained for the particles from the same bomb (HP1_1 and HP2_1) and from the other bomb (HP3_2) is quite different. To asses the homogeneity of this big particles, analyses at different energies are required. For these samples, with sizes about 1.5 mm, protons at 6 MeV give information up to the first 500 µm, therefore higher energy is necessary to provide an average Pu/U ratio in the whole particle. For that, HP1_1 and HP2_1 were also studied with 18 MeV protons, and the results clearly indicate the non homogeneous relative distribution of Pu and U into these samples. PROGRAM ID: O24 6 In view of the relative simplicity of the U and Pu K-lines spectra obtained for these big particles, we have checked if the same methodology can be of interest for the study of microparticles from Palomares and Thule. The measurements in this case were carried out in our 3 MV tandem, where the nuclear microprobe is installed. As seen in Fig. 3, although the cross section for U K-lines increases slightly between 4 and 6 MeV, the background signal (given in counts per seconds) rises faster with beam energy, due to the higher probability for nuclear reactions at higher energies. From this, the lower limit of detection using K-lines (below 1% at.) was found for 4 MeV protons. Further information can be obtained if we record simultaneously the RBS spectra (Fig. 4). Although the mass resolution is not good enough to separate the U and Pu contributions, at 4 MeV we can clearly distinguish a signal due to the light elements (C and O) and a signal due to the medium and heavy elements from the particle. By increasing the beam energy the mass resolution improves and, at 6 MeV, it is possible to observe the signals from the C and O, a peak from the Si and other elements from the soil and the signal due only to Pu+U. From this peak the absolute amount of Pu+U can be obtained, while the PIXE measurements provide the Pu/U elemental ratio. Then, from both results, we can know the concentrations of Pu and U separately. Moreover, if we compare the thickness in at/cm2 found by RBS with the physical thickness obtained by other methods (by SEM, for instance), we can estimate the density of the particle, which is an important magnitude very difficult to determine. Conclusions The best experimental conditions to study this kind of samples depend mainly on the particle size. For microparticle (<200 µm) L-lines and K-lines can be used to provide information of Pu and U concentration in the sample. In this case, it is worth noting that the same Pu/U ratios have been obtained with both lines. However, L X ray autoabsorption presents an unavoidable drawback for the analysis of bigger particles, where the use of K-lines is mandatory. For PROGRAM ID: O24 7 particles around 1mm the use of several proton energies is required to study the depth profile distribution of Pu and U which is an essential information to assess the radioecological impact of this samples into the environment. Acknowledgement We acknowledge the Junta of Andalucía for the financial support through the Project 2009/RNM-138. References [1] J. García López, M.C. Jiménez-Ramos, M. García-León, R. García-Tenorio. Nucl. Instr. and Meth. B 260 (2007) 343. [2] M.C. Jiménez-Ramos, M. Eriksson, J. García-López, Y. Ranebo, R. García-Tenorio, M. Betti, E. Holm. Spectrochim. Acta Part B 65 (2010) 823. [3] J. García López, F.J. Ager, M. Barbadillo Rank, F.J. Madrigal, M.A. Ontalba, M.A. Respaldiza, M.D. Ynsa. Nucl. Instr. and Meth. B 161–163 (2000) 1137. [4] G.W. Grime, M. Dawson, M. Marsh, I.C. McArthur, F. Watt. Nucl. Instr. and Meth. B 54 (1991) 52. [5] http://pixe.physics.uoguelph.ca/gupix [6] I. Orlic, S. Zhou, J.L. Sanchez, F. Watt and S.M. Tang, Nucl. Instr. and Meth. B 150 (1999) 83. [7] M.O. Krause. J. Phys. Chem. Vol. 8 Nº 2 (1979) 307. [8] . Eriksson, J. Osan, J. Jernstrom, D. Wegrzynck, R. Simon, E. Chinea-Cano, S. Markowicz, A. Bamford, G. Tamborini, S. Torok, G. Falkenberg, A. Alsecz, H. Dahlgaard, P. Wobrauscheck, C. Streli, N. Zoeger and M. Betti. Spectrochim. Acta Part B 60 (2005) 455. [9] P.A. Burns, M.B. Cooper, K.H. Lokan, M.J. Wilks, G.A. Williams. Appl. Radiat. Isot. 46 (11) (1995) 1099. PROGRAM ID: O24 8 Figure captions Fig. 1. Ionization cross section for U K and L-lines as a function of proton energy. Fig. 2. PIXE spectra showing the K-lines from a hot particle obtained with 6 and 18 MeV protons. Fig. 3. Comparison between the background signal and the ionization cross section for U Klines for protons of 4, 5 and 6 MeV. Fig. 4. RBS spectra from a microparticle obtained with 4 and 6 MeV protons.